Independent Evaluators
James Long, Associate Professor of Political Science, University of Washington, US.
(jdlongiv@gmail.com) 
Pranab Panday, Professor, Public Administration, University of Rajshahi, Bangladesh 
Abstract
This document presents the results from the evaluation of 
World Vision’s (WVI) Citizen Voice and Action (CVA) project 
components as a part of their larger “Nobo Jatra” (“New 
Beginning”) program in Bangladesh.
EVALUATION RESULTS OF “CITIZEN
VOICE & ACTION” NOBO JATRA
PROGRAM, WORLD VISION
BANGLADESH
1
Contents
1. Document Purpose 5
1.1 Organizational Outline 5
2. Executive Summary 6
3. Background to the Evaluation 11
3.1 “Nobo Jatra” (New Beginning) (NJP) overview 11
3.2 Citizen Voice and Action in Nobo Jatra overview 11
3.3 Overview of the Evaluation Approach 12
3.3.1 Summary of evaluation framework 13
4. Evaluation Scope, Questions & Objectives 15
4.1 Citizen Voice and Action goals 15
4.2 Phases and Activities of Citizen Voice and Action in NJP 16
4.2.1 Activities 17
4.3 Intended Outcomes of Citizen Voice and Action 18
4.3.1 Evaluation objectives 19
4.3.2 Evaluation questions 19
4.4 Political Economy Mapping and Theory of Change 21
4.5 Summary of Results from Formative Research and Inception Visit 24
5. Evaluation Methodology and Research Design 26
5.1 Evaluation sample 26
5.2 Intermediate and final outcomes 26
5.3 Measures 27
5.4 Data collection and fieldwork 27
5.4.1 FGDs with CVA working groups and Community Health committees 27
5.4.2 FDGs with VDC working groups, Scorecard Sessions & Women’s Groups 27
5.4.3 FGDs with WATSAN committees 28
5.4.4 KIIs with Community Clinic Staff 28
5.4.5 KIIs with Union Council Chairperson & Upazila Council Chairperson 28
5.4.6 KIIs with Union and Upazila Agricultural Extension Officer 28
5.4.7 KIIs with the Union and Upazila Health Officer 29
5.4.8 KIIs with NGO Facilitators & NJP Field Staff 29
5.4.9 KIIs with National level stakeholders of NJP, WVB, and Line Departments 29
6. Results from Fieldwork and Data Collection 30
6.1 Summary of Results 30
2
6.1.1 Outline and Organization of results 32
6.2 Quantitative Results on Indicators from CVA Database 33
6.3 Pre‐CVA 38
6.4 CVA Formation and Activities 46
6.5 CVA Process 50
6.6 Government Response 57
6.7 Participants 60
6.8 Service Outcomes 62
6.9 Interagency Process 68
6.10 Most Important Issue 69
6.11 Sustainability 72
7. Sustainability Through Systems Strengthening 75
7.1 Overview 75
7.2 Policy Dialogue & Actions of CVA in NJP 76
7.2.1 Overview 76
7.2.2 The Political Economy of Governance in Bangladesh 77
7.2.3 Evaluation Results on Policy Dialogue 78
7.3 What Does Sustainability Mean for Social Accountability? 95
7.3.1 Evaluation Results of Systems Strengthening and Sustainability 97
8. Discussion 104
8.1 Did the evaluation meet its objectives? 104
9. Conclusion and Recommendations 108
9.1 Summary 108
9.2 Considerations for Plans of Action for Future Sustainability 108
9.3 Recommendations 111
9.3.1 To NJP/WVB 111
9.3.2 To CVA Working Groups 114
9.3.3 To VDCs, NGOs, other committees, and community people 115
9.3.4 To WVB and the Government of Bangladesh 116
9.3.5 To Donor 117
9.3.6 To WVI 118
Annex A 119
3
List of Abbreviations
AHI Assistant Health Inspector
AIRP Arsenic Iron Removal Plant
ANC Antenatal Care
BADC Bangladesh Agriculture Development
CC Community Clinic
CDP Community Development Plan
CG Community Group
CHCP Community Health Care Provider
CSO Civil Society Organization
CVA Citizen Voice and Action
DAE Department of Agriculture Extension
DPHE Department of Public Health Engineering
FGD Focus Group Discussion
GMP Growth Monitoring program
HH House Hold
HI Health Inspector
KII Key Informant Interview
LGSP Local Government Support Project
LSP Local Service Provider
MP Member of Parliament
NGO Non‐Governmental Organization
NJP Nobo Jatra Project
PNC Postnatal care
RTI Right to Information
SAAO Sub Assistant Agriculture Officer
UAO Upazila Agriculture Officer
4
UHFPO Upazila Health and family Planning Officer
UHFPO Upazila Health and Family Planning Officer
UP Union Parishad
USAID United States Agency for International Development
UZP Upazila Parishad
VDC Village Development Committee
VGD Vulnerable Group Development
VGF Vulnerable Group Feeding
WASH Water and Sanitation
WATSAN Water and Sanitation
WVB World Vision Bangladesh
WVI World Vision International
5
1. Document Purpose
This document provides a summary of results of the evaluation of World Vision’s (WVI) Citizen Voice and
Action (CVA) project component to their “Nobo Jatra” (“New Beginning”) Program (NJP) in Bangladesh.
The present document follows from the completion of the Evaluation Framework and Methodology
document by the evaluation team, the evaluation team’s inception field visit to Bangladesh in November
2019, and their follow‐up data collection visit in January 2020. The present document summarizes the
results of the evaluation, and discusses how the results relate to sustainability and systems strengthening
in Bangladesh through social accountability. It concludes with lessons learned and policy
recommendations for stakeholder knowledge‐sharing regarding the extension of NJP in Bangladesh and
WVI’s CVA programming globally.
1.1 Organizational Outline
This report is organized as follows. Section 1 describes the document purpose and Section 2 provides an
Executive Summary of the evaluation. Section 3 reviews relevant background to the evaluation necessary
to understand the findings and conclusions in the present document (for more information on
background, readers may wish to consult the Evaluation Framework and Methodology document).  
Section 4 providesthe Evaluation Scope, list of objectives and indicators, and the theory of change. Section
5 summarizes the methodology and data collection tools (full discussion regarding the tools and copies of
the tools can be found in Evaluation Framework and Methodology). Section 6 presents the findings from
the data collection and field work central to the evaluation based on quantitative tracking indicators from
the CVA database, focus group discussions, and key informant interviews with relevant CVA working
groups, community members, government officials, service providers, and departmental employees
working in community clinics, water and sanitation, and agriculture extension.  
Following the summary of these quantitative and qualitative findings in Section 6, Section 7 leverages the
resultsfrom the data collection, further document review, policy dialogue, and interviews conducted with
NJP staff and national‐level officials to provide a more in‐depth evaluation of systems strengthening and
policy analysis for social accountability in Bangladesh. This aspect of the evaluation incorporates a more
refined and comparative theory of change that operates at meso and national levels while speaking to
scaling in Bangladesh and other contexts. Section 8 summarizes the quantitative and qualitative findings
against the evaluation objectives presented in Section 4.  
Section 9 concludes and provides recommendations that arise from the evaluation for key stakeholders
involved in CVA and NJP in Bangladesh, World Vision and its partner social accountability projects in other
countries, and to donors.  
6
2. Executive Summary
The design of the Citizen Voice and Action (CVA) module of the Nobo Jatra Program (NJP) in Bangladesh
aimed to harness, mobilize, and improve aspects of the behaviors and lives of citizens while also enhancing
their linkages with important institutional actors to promote government service provision through social
accountability. In evaluating CVA, we can therefore summarize impact and effectiveness by looking at
attributable changes with respect to citizens, institutions, and accountability.
Broadly speaking, the conclusions we can draw from the evaluation involve whether CVA was working to
meet its objectives and, whether this was or was not the case, the reasons why.  
We summarize our first finding to say that CVA did succeed in the NJP project areas with respect to the
processes and outcomes it intended to improve, as confirmed by quantitative indicators and qualitative
assessments. Second, we summarize our findings to say that the linkage we identified as being most
relevant in explaining why CVA was working involves overcoming inertia on the part of citizens, duty
bearers and service providers, as well as other relevant institutional actors. Insofar as the evaluation
revealed “systems strengthening,” we believe it did so by overcoming inertia for i) citizens, who because
of CVA, are now using their voice and providing new actions via robust social accountability while gaining
the correct information, tools, and platforms to do so; ii) institutions, which are performing better from
improving the work among institutional (agency) actors as well as from communities via social pressure
exerted from the outside; iii) accountability, which resulted from strengthening both old and new formal
and informal mechanisms of accountability that make inertia harder while rewarding positive
performance and action. We also documented numerous instances of policy dialogue and actions that
may have downstream and/or lasting effects beyond the actions of CVA at a local level, including
important linkages with divisional and national‐level stakeholders.
More specifically, using quantitative indicators compiled from the CVA database regarding CVA activities,
the evaluation found that all CVA working groups in the project area were formed and conducting
activities from 2018 to the present. These groups had completed activities, including scorecard sessions,
interface meetings, action plans, and follow‐up, in all three development sectors (community clinics,
water and sanitation, and agriculture extension). Baseline indicators regarding the monitoring standards
at community clinics, Union WATSAN committees, and Union agricultural services units found many gaps
and shortcomings in meeting the requisite standards for service provision.
In the first monitoring standard session before CVA activities, compliance of standards for community
clinicsshowed that 58 clinics(49%) met between 76‐100% ofstandardsin 2017, and by the second session
in 2020 after CVA activities, 105 clinics (88%) met between 76‐100% of standards. Compliance of
monitoring standards in water and sanitation in the first session in 2018 found that 3 (7.5%) of the Union
WATSAN committees met between 76‐100% ofstandards, and by the second session, 4 committees(10%)
had met between 76‐100% of standards (but a further 32, or 80%, met between 51‐75% of standards
compared to 20, or 50%, that met between 51‐75% in the first session). Compliance of monitoring
standards for the Union Agriculture Service Unit in the first session in 2018 found that 12 (30%) unions
met between 76‐100% of standards, and by the second session in 2020, 22 (55%) of units met between
76‐100% of standards.
Moreover, citizen assessments reported in surveys revealed generally “fair” or “poor” ratings for service
delivery in the three sectors. By 2020, CVA working groups across the sampled unions could report
important progress on developing, implementing, and following‐up on action plans.  
7
In conjunction with the quantitative indicators, from the qualitative data collection and assessment, we
can further summarize the general evaluation findings as they pertain to citizens, institutions, and
accountability (some of which overlap) as follows.1
  
● We summarize for citizens:
Finding #1: CVA improved service provision and citizens’ attainment of relevant information for social
accountability
Baseline: CVA itself did not involve any real direct spending, programming, or service provision to help
meet the community’s needs; rather it is a social accountability tool that promotes citizens’ voice and
action to inspire service providers and duty bearers to perform in their roles and improve their
performance.  
Problem Identified: As such, the problem of service delivery identified in Bangladesh was not simply one
of resources, but rather inertia in providing them on the part of relevant actors as well as information
about what should be available and what was available on the part of citizens.
Impact: As a result of CVA, service provision is improving along many dimensions and indicators, and
community people have gained relevant information that is necessary for social accountability.
Finding #2: CVA improved citizens’ voice and action to better articulate their demand for services
Baseline: CVA itself does not invent “voice” for citizens  ‐‐ they already have a voice; but CVA helps to
activate civil society, social networks, and community groups to better articulate their demands for
services from relevant stakeholders.
Problem Identified: The problem identified was not that citizens lacked a voice in Bangladesh, but rather
that people needed inspiration to take action by gaining confidence in organizing and making their
individual voices and the community’s voice resonate through dialogue and interface, while also
witnessing positive results after articulating their demands.  
Impact: As a result of CVA, voice is improving in form and content to allow citizens as service seekers to
demand what they want and present whether things are going well by utilizing techniques such as the
scorecards, joint action plans, and follow‐up.  
Finding #3: CVA has helped to activate civil society and coordinate citizen mobilization for collection
action  
Baseline: CVA itself does not “create” or generate civil society; people already had numerous avenues to
mobilize for coordinated actions had they so desired.  
Problem Identified: The problem identified was that even so, no actions were being taken to actually
activate civil society in Bangladesh and coordinate the community’s collective action; additionally, there
was inertia among the community people who needed a coordinating mechanism or platform to inspire
social accountability.
1 It is important to note that while we summarize these findings here, our extended discussion of systems
strengthening and sustainability in Section 7 and our recommendationsin Section 9 demonstrate thatsome of these
positive results are fragile and may or may not sustain depending on a variety of factors.
8
Impact: CVA provided such a mechanism while helping to activate civil society and coordinate collective
action, with both especially important in contexts where NGOs are not direct service providers.
● We summarize for institutions:
Finding #1: CVA has inspired institutional actors to overcome inertia and find motivation to improve
their job performance
Baseline: CVA itself does not directly create new legislation, mandates, or institutional mechanisms to
provide services; but through social accountability, it aims to inspire actions on the part of institutional
actors responsible for service provision to address and pursue ways to improve their job performance.
Problem Identified: The problem the evaluation identified was not a lack of resources on the part of
institutions to effectively provide services in Bangladesh, but rather both bureaucratic inertia on the part
of institutional actors to perform their duties and provide services as mandated by the government and a
lack of will to perform well (or sincerely).
Impact: As a result of CVA, service provision improved by “reminding” duty bearers and service providers
to do their jobs while providing them with the motivation necessary to overcome their inertia.
Finding #2: CVA has helped institutional actors to better organize to achieve coordinated actions to
improve service delivery
Baseline: The units, agencies, and institutionsresponsible forservice provision already existed before CVA,
but CVA aimed to help those actors better coordinate their actions.
Problem Identified: The problem identified in Bangladesh wasthe degree ofreal devolution of government
service provision as mapped on to local governmentstructures, a lack of clarity in roles and responsibilities
horizontally (with Union parishad officials) and vertically (with Upazila parishad officials and government
line departments), along with a lack of proper human resources to staff and oversee institutional actors.  
Impact: As a result of CVA, these actors were inspired to organize and meet more regularly with each
other, clarify their roles in mutually agreed upon actions to improve job performance, and approach
human resource constraints in creative ways. This pertained not only to local service providers and
government officials, but has “trickled” up in many ways to Upazila, divisional, and national‐level actors
as well.
Finding #3: CVA improves monitoring and works cohesively with other aspects of NJP programming
Baseline: In the absence of CVA, institutional actors already had mechanisms, finances, and tools to their
jobs and provide services; but inertia and a lack of oversight and monitoring did not inspire institutional
actors to perform adequately within their roles.
Problem Identified: The evaluation found that the problems in Bangladesh related to a lack of demand
articulation by citizens in terms of priorities and the allocation of expenditures, inertia among officials at
Union parishad and Upazila levels to their job (inertia among principals to monitor agents at all levels),
and a lack of resources and will to work properly within extant monitoring systems.
Impact: As a result of CVA, these mechanisms and tools improved and were assisted by other aspects of
NJP, which CVA helped to contribute to and reinforce; CVA also provided a unique tool for monitoring.
9
Positive results of policy actions and dialogue at meso and national levels have leveraged knowledge‐
sharing and linkagesfrom CVA to contribute to other government programming and partnershipsthat will
contribute to systems strengthening.
● We summarize for accountability:
Finding #1: CVA helps promote formal institutional accountability at various levels of government and
within service delivery
Baseline: The government of Bangladesh already had its own monitoring system to manage institutional
actors and ensure service delivery, as well as government circulars and policies already designed, issued,
and/or tabled by the national government to address many service delivery gaps.
Problem Identified: The problem identified was that the government’s various ways to monitor their
agents were not activated or used, and were not coordinated with local government units and across the
line departments. New legislation also took time to vest, with returnsto new policies or regulations taking
a long time to measure or assess. Overall, there was a lot of inertia at all levels of agency principals and
agents, an overlapping or a lack of clarity on roles and responsibilities among these principals and agents,
and additional human resource constraints that reduced the ability of principals to properly monitor and
hold agents accountable.
Impact: CVA helped to promote accountability within formal institutions at various levels of government,
aided by matchmaking, coordination, and feedback across departments at the Union parishad level; it also
provided inspiration for Union parishad officials and Upazila and line department to work more
cohesively, new forms of citizen‐activated direct accountability for institutional actors (e.g, posting of
phone numbers and activation of committees; results “trickling up” from the local to meso and national
levels), and some documented actions to improve policies, budgets, and coordination at the Upazila,
district, and national levels.
Finding #2: CVA helps promote social accountability, or accountability that arises from “outside” of
institutions
Baseline: Citizens played no formal role in institutional accountability vis‐a‐vis public officials responsible
to themselves (whereas ideally decision‐making at the top should “trickle down” to the bottom &
information from the bottom should “trickle up” to top).
Problem Identified: The problem identified in Bangladesh was that “measures on paper” for institutional
actors existed, but that there was inertia among bureaucracy and bureaucratic actors to oversee and
monitor behavior; further human resource constraints inhibited monitoring, with everyone busy with
activity of the moment without building stronger or more sustainable systems.
Impact: CVA was a pathway to improving institutional performance and accountability even from outside
of institutions via social accountability; it also allowed pathways to work with/within institutions that
provide support to institutional actors (e.g., WATSAN committee) at local, meso, and national levels.
10
Finding #3: CVA helps promote formal political accountability
Baseline: Citizens play a formal role in electoral accountability and in the selection of public officials that
occupy elected offices; while this makes them directly responsible to voters, it also suggests that voters
can indirectly influence the accountability of unelected officials who work alongside elected officials.
Problem Identified: The problem identified in Bangladesh was not the absence of formal accountability or
democratic elections, which exist on paper and thus did not prevent the exercise of formal political
accountability, but rather it was the limitations on the sanctioning mechanism of elections given the
nature of party politics as well as limits to robust participation and candidates’ ability to skirt legal and fair
ways to gain office; formal accountability mechanisms were therefore weak, creating an important space
for social accountability to vest and thrive.  
Impact: Political pressure can still matter as activated through social accountability by convincing   officials
when and how they benefit from CVA. Moreover, the results ofsocial accountability that “trickle up” from
the local level to meso and national actors can contribute importantly to strengthening formal lines of
institutional accountability for actors that are “nested” in the service delivery process, some of whom are
subject to elections and some of whom are not, when citizens exert social accountability at one level and
it influences actions at another.
11
3. Background to the Evaluation
3.1 “Nobo Jatra” (New Beginning) (NJP) overview
In 2015, World Vision embarked on “Nobo Jatra” (“New Beginning”) Program (NJP), a five year USAID
Food for Peace Title II Development Security Activity based in Bangladesh and conducted in partnership
with the World Food Program, Winrock International, and various non‐governmental partner
organizations. In 2019, USAID agreed to a two‐year cost extension of the NJP programming to run through
2022. Further, in 2020, due to the double impacts of COVID and Cyclone Amphan, NJP requested a 12
month bridge phase to help communitiesin the project area recover. If thisis approved, the cost extension
and sustainability phase will be pushed to FY22. This will allow more breathing space in terms of
strengthening CVA groups and actualizing recommendations, discussed in Section 9.
The program seeks to improve food security (including gender equity in food security), nutrition, and
resilience in the Southwest Region of Bangladesh. To achieve its objectives, the program includes
interventions and activities across developmentsectors,specifically integrated programming on maternal
and child health, water and sanitation (WASH), agriculture and alternative livelihoods, disaster risk
reduction, good governance and social accountability, and gender.  
Inherent within the NJP’s “theory of change” is the recognition that addressing pervasive challenges
related to food security, nutrition, and resilience require broader coordination and long‐term solutions
that can only be achieved by strengthening linkages with relevant state and non‐state actors at the
national and regional (“meso”) levels in Bangladesh. Sub‐national governance, social accountability
activities, and national policy work aimed at improving the accountability of government ministries
comprise a central focal point of the NJP, effectively serving as a binding theme across the program.
3.2 Citizen Voice and Action in Nobo Jatra overview
Under NJP’s “Purpose Four”‐‐whose aim is “improved social accountability and national policy
engagement for service provision for vulnerable men and women”‐‐NJP is undertaking advocacy at the
national level, along with civic engagement and good governance initiatives at the local level.  
NJP has prioritized the implementation of cross‐cutting interventions and activities to address Purpose
Four, including the deployment of Citizen Voice and Action (CVA) programming (specifically in the health,
water & sanitation [WASH], and agriculture sectors) in concert with World Vision.
The CVA component of NJP focuses on good governance and social accountability. Building on WVI’s
rigorously tested global social accountability approach, CVA’s aim is to improve the efficacy and quality of
public services. While Purpose Four of NJP does highlight the importance of improved social accountability
and national policy engagement, it isimportant to note that while the original inception and programming
for NJP proposed CVA, there was a lag prior to implementation. Most CVA working groups in the project
area were not constituted until 2018, and many initial CVA activities were not launched until 2018 or 2019.
They continued during the period of the present evaluation.
In a general model, CVA seeks to improve government output; bureaucratic performance; and citizen
engagement, voice, and demand articulation for improved services, with an emphasis on targeted
measures thatspan across local, meso/regional, and national levels, institutional levels, and development
sectors. As a social accountability approach, CVA plays a pivotal role in creating opportunities for dialogue
between government and citizens while extending networks and collaboration between government
organizations, NGOs, and community stakeholders. Social accountability programs are meant to
12
complement‐‐not replace or substitute‐‐other programs that strengthen formal lines of institutional
accountability (including electoral and political accountability, and bureaucratic accountability).   
The CVA process, as instituted by WVI in other settings and as adopted by NJP in Bangladesh, focuses on
raising awareness on basic civil rights, government policies, entitlements, and legislation among citizens
prior to empowering communities to compare existing government services against those pledged to
them by local government authorities. It includes a provision to score and monitor government services
based on a community scorecard and set of monitoring standards. With the support and involvement
from numerous stakeholders, CVA strives to then influence government institutions to increase and
improve service delivery to eliminate gaps between commitments and performance.
3.3 Overview of the Evaluation Approach
The CVA evaluation for NJP was launched with WVI, WVB, and its evaluation team in September 2019. An
inception visit took place in November 2019. A second field visit for data collection occurred in January
2020.
The general purpose of the evaluation is to assess progress and impact under Purpose Four of NJP for CVA
and social accountability. The evaluation documents changes produced by the program for major
stakeholder groups. It specifically tracks shifts in health service standards achieved in community health
clinics, water and sanitation, and agriculture extension services, including resulting benefits (if any) for
service users and providers. It does so using quantitative tracking data from WVI’s CVA database, and
original qualitative data collected by the evaluation team. The evaluation assesses the nature, extent, and
process of changes produced by CVA for different groups of service users and providers, program staff,
and government decision makers. The evaluation also documents key learnings and areas for
improvement in order to help strengthen similar programs in the future.
Broadly, assessing and understanding the impact of CVA required anchoring the program within the
contextual and institutional setting of the country, region, and localities of the project area‐‐in this
instance, those selected by NJP. This informed the particulars of the evaluation methodology, but also
provided guidance as to the comparative relevance of knowledge gained to other settings.  
This approach required first highlighting the political economy factors, the historical and institutional
legacies, and the contemporaneous institutional matrix that are likely to shape the politics of service
provision in Bangladesh. The evaluation team specified much of this work in the Evaluation Framework
and Methodology
Second, the approach required delving systematically into the actors, processes, and outcomes with
respect to NJP and CVP program inputs and outputs across localities in Bangladesh. This involved both
retrospective document review and quantitative data tracking ofspecific indicators, along with qualitative
data collected from formative, inception, and field‐based data collection during the evaluation period.  
Third, the evaluation was anchored within NJP’s larger programming, which was launched with CVA in
mind but whose implementation was delayed. As a result, the evaluation included development activities
and interventions that occasionally overlapped with and therefore potentially influenced CVA‐specific
activities. Though often separate, these interactions may nonetheless have impacted CVA effectiveness.  
Fourth, the evaluation team was tasked with not only examining the direct impact of CVA activities on
certain indicators and outcomes, but also examining whether and how CVA contributed to systems
strengthening and sustainability, and if so, in what ways. This required the evaluation team to further
utilize a sustainability theory of change that was not previously included as part of the original NJP theory
of change.
13
3.3.1 Summary of evaluation framework
Given the nature of the scope of the evaluation, CVA activities within NJP in Bangladesh, and a desire to
produce comparative findings to inform knowledge across settings and country contexts (including
regarding systemsstrengthening and sustainability), we pursued an evaluation approach whose proposed
empirical activities were guided by synthesizing the theoretical concerns of both the “vertical” and
“horizontal” approaches to accountability.  
The inception report and evaluation framework established the importance of investigating vertical “top‐
down” hierarchy, which in the context of the evaluation, focused on performance‐oriented actions and
outcomes shaped by the overall political process and the nesting of bureaucratic interests within public
sector management (given local government structures, line departments, and other directives issued by
the government of Bangladesh). This outlook prioritized identifying and leveraging principal‐agent
dynamics within government ministries across various sectors, including an emphasis on community
clinics, water and sanitation, and agriculture extension. Measuring and evaluating bureaucratic provision
of services requires agency principals – from ministers and various permanent secretaries down to
regional and district managers – to exert oversight and monitoring, as well as positive performance pay
or sanctioning, to the frontline government actors responsible for services at the local level. Vertical
hierarchies allow things to “trickle down” and “trickle up.”
NJP programming generally, and CVA specifically, aimed to improve vertical lines of accountability.
Although the government’s 2009 Local Government Act provided for devolved government structures
related to fiscal functioning and service provision to the most local administrative government units
(Union Parishad) nested within larger units (Upazila Parishad). The evaluation team found through their
inception research that the government policy can be understood as only “partial” devolution,
“deconcentrated decentralization,” or a hybrid model where local units have some fiscal and service
provision responsibilities. These are often, although not always, overlapping with some of those of line
departments in form and content. NJP has therefore had to wrestle with mapping its activities onto local
government structures in design and practice with this hybrid, but not full, devolution, facing the reality
that that top‐down bureaucratic functioning still plays an important role in the performance of CVA at the
union and Upazila levels. As a result, the evaluation approach understood that the type and degree of
decentralization must be considered as a key variable in understanding both the potential for and
performance of social accountability models like CVA (and what may “trickle down” as well as “trickle
up”).  
The inception report and evaluation framework also discovered the importance of investigating
horizontal dimensions of accountability focused on the direct linkages between citizens (clients) with
public service providers. This view leverages greater local community mobilization to increase individuals’
and groups’ direct participation in the processes and outcomes ofservice providersthrough CVA, including
activities such as participatory budgeting, community monitoring, and other satisfaction surveys (e.g.,
community scorecards). Horizontal linkages allow things to “trickle across.”
This perspective is also resonant in the Bangladesh context and informed NJP programming, given that
the Local Government Act of 2009 empanels local government units with fiscal functioning and service
provision devolution and empowers citizens and communities to better demand government
performance. As a result, accountability (if not in form than at least in function) should work horizontally
between citizens, duty bearers, and service providers at the union and Upazila levels. Hence, these
structures proved critical to assessing the impact of CVA, even though the hybrid model of devolution has
often prevented local communitiesfrom operating as designed. And the hybrid model of devolution is not
the only barrier to community engagement in the governance of local bodies. Other contributing factors
14
included a lack of awareness of community members, the reluctance of local government representatives
to perform their duties as mandated by the Local Government Act of 2009 and other legislation and
mandates. This often included non or partial implementation of the act, along with either absent or weak
monitoring of the implementation of the act’s different provisions, which prevented the communities
from operating as designed. As a result, elite capture also had to be considered as a relevant impediment
to citizen engagement, which hindered the participation of the community in governance (and what
may “trickle across”).
Finally, the evaluation also focused on the “vast space in the middle”2 between vertical and horizontal
lines of accountability. Thisspace includesthe levels, institutions, and actorsthat are often either assumed
to lack agency or the possibility of reform in the first approach, or those that are ignored regarding scaling
and sustainability in the second approach. This allows things to “trickle around/back and forth.”
The reality of the political economy of service provision governance in many developing countries,
including Bangladesh,suggeststhat while “top‐down” hierarchies often fail to deliverservicesin the levels
or quality required to sustain many citizens’ livelihood, these countries vary in success across sectors or
locations. Yet even after the roll‐out of large systems‐oriented programs like NJP the persistence of
significant challenges may limit possibilities for reform. Moreover, while a “horizontal” linkage may help
to provide information about rights and responsibilities to citizens who otherwise lack this knowledge
while helping communities to mobilize to articulate demands, beyond a direct linkage with service
provision, such linkages may be limited at scale, in the event of institutional disruption, or when occurring
alongside with political change.  
Accordingly, there are many layers of institutional actors within the sectors between the top‐levels of
policymaking and the day‐to‐day frontline of service delivery that require attention as “the place where
much of the politics of service provision plays out and thus where many opportunities for achieving gains
in performance might be found.” This pertains both to the rigor of assessing progress on CVA within NJP
in Bangladesh, but also to the understanding of system‐level strengthening and comparative insights for
export to other country contexts. Given hybrid devolution, the potential for elite capture, and a lack of
clarity on roles and responsibilities regarding local governance in Bangladesh, these “meso” levels of
accountability were similarly relevant (and what “trickled around/back and forth”).
  
2 Levy, Brian, and Michael Walton, 2013, "Institutions, incentives and service provision: Bringing politics back in."
Working Paper, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2386655.
15
4. Evaluation Scope, Questions & Objectives
4.1 Citizen Voice and Action goals   As a component to NJP, CVA in Bangladesh includes a package of civic education, community services
scorecards, social audits, and advocacy training and implementation that it provides to communities
through the formation of CVA working groups. Three sectors specific to government service units are
related to project activities to improve the quality of government services: 1) Mother and Child Health
and Nutrition (Community Clinic), 2) Water, Sanitation, and Hygiene Promotion (Union WATSAN
Committee), and 3) Agriculture and Livelihood (Union Agricultural Service Unit). Under Purpose Four of
NJP, CVA specifically targets government services in 119 community clinics, 40 Union agricultural services
Units and 40 Union Water and Sanitation (WASH) committees.
With the realities of the political economy of service provision in Bangladesh and the CVA tool’s adoption
and validation by WVI in othersettings, NJP incorporated a CVA approach into its programming to improve
service delivery in the country. CVA is led by local communities and emphasizes community development
practices such as participation, inclusion, ownership, and sustainability. While the practice
implementation of CVA should be flexible and respond to local situations, it contains a number of linked
core elements that are pursued in all contexts, including: information, voice, dialogue, and
accountability. We review those here and discusstheir applicability for Bangladesh to contextualize better
the evaluation scope.
Information: Information is a critical element of CVA since citizens have a right to know information about
the services that their government delivers in order to assess their individual, household, and community
needs. This information should be clear, easily understood, transparent, and made freely available to the
public by public authorities. CVA facilitates access to this information through a range of mediums,
including public meetings where government officials share information, through broader social and
traditional media, theater, radio ads, brochures, etc. depending on local contexts.  
In Bangladesh, the selected approaches to improve the availability and quality of information included
public meetings, the dissemination of civic education brochures, and the circulation of key civic education
information through local media segments. Moreover, CVA working groups directly sought input from
community people in their wards (including VDC members), and relayed information back to the
community.  
Voice: Voice is defined as the capacity of all citizens (including the poor and most marginalized) to express
their views and interests and demand action from those in power who have a duty to provide public
services. Through these processes of awareness raising and empowerment citizens’ voices are
strengthened and amplified.  
In Bangladesh, the use of community scorecards/scorecard sessions(disaggregated by gender, age, ethnic
group, and disability status) helps promote the feedback of a wide cross‐section of community members
on how services are or are not functioning in the NJP areas, while also presenting their proposals to
improve services. These views were collected in small focus group scorecard sessions facilitated by NJP
and partner NGOs to ensure the participation of women, children, and other marginalized groups. Then,
in later public meetings, the views of the different groups were shared either directly by participants or
anonymously depending on the concerns and needs expressed by the participants. Ensuring inclusion and
participation in services via these structured evidence‐based facilitation processes is at the heart of CVA.
Dialogue: CVA aims to provide an opportunity for different stakeholders, especially service users and
service providers, to share their views about the delivery of basic services. Through dialogue, mutual
16
understanding between stakeholders increases and effective partnerships are established. As a result,
relationships within the community are both repaired and strengthened. Through such dialogue it is
intended that citizens, together with service providers and other stakeholders, can then identify ways to
improve service delivery. And opportunities for dialogue can help positive results from the local level
trickle up to meso and national‐level actors.
In Bangladesh, the main forum through which their views were shared as a part of CVA was via an interface
meeting at a service facility (e.g., community clinics) where service users, providers, and government duty
bearers (responsible for the resourcing of the facilities) were present. The key government performance
standards for the service were shared at the meeting alongside assessments concerning facility
performance related to these standards. Community feedback and priorities for improved services were
also presented. During this meeting an agreed plan of action for the facility was developed by the
governmentservice providers and community attendees. Improving voice and dialogue has been pursued
as an aspect of the formation of the CVA working group (giving voice to the joining members), including;
the inclusion of the community’s perspectives into the CVA working group’s activities; scorecard sessions
that allow community members to rate the performance of service provision; and a variety of interface
and dialogue meetings with CVA members, government officials (including Union Parishad and Upazila‐
level officials), and other duty bearers that allow citizens to express their demands. This helps the people
to raise their voices and articulate demandsfor the servicesthat they are entitled to from the government,
with dialogue platforms used to follow‐up on and monitor the delivery and quality of those services. The
activation of dialogue for CVA at local levels also inspired dialogues at Upazila, divisional, and national
levels that built on social accountability practices and evidence.
Accountability: CVA works to increase the accountability of service providers and those in power, as well
as the accountability of citizens around the delivery of basic services. Accountability establishes a
relationship between power holders and those who can hold them to account for their actions. To ensure
the action plan (described above) is implemented, a series of community‐led advocacy activities are
undertaken to ensure government accountability. Where issues are systemic and need higher level
advocacy, WVI works with citizen representatives to ensure their views are heard and actions advocated
at the national level.
In Bangladesh, improving accountability in CVA was designed to arise from follow‐up on action plans and
institutional performance at the Union Parishad and Upazila levels, and in concert with elected and non‐
elected members at those units. Moreover, it provided opportunities for policy dialogue and actions at
meso and national‐levels.
4.2 Phases and Activities of Citizen Voice and Action in NJP
We can understand information, voice, dialogue, and accountability in CVA to occur in three phases, with
each managed within NJP programming with the facilitation of NJP staff and partner NGOs in Bangladesh.
Phase 1: Enabling Citizen Engagement, which involved: researching and understanding public policy;
preparing local materials and resources; education and mobilizing local citizens; building networks and
coalitions; establishing local relationships and connections between service users and service providers.
In many ways, Bangladesh provided some existing institutional structures at the local level to draw from
in this regard outside of NJP, such as village development councils (VDCs) and ward shava budget
meetings. However, these have typically not been activated or mobilized for robust action.
Phase 2: Engagement via Community Gathering, which involved: forming CVA working groups; convening
initial meetings with local service users and providers; holding monitoring standards sessions with local
service users and providers; facilitating score cards sessions with local service users and providers;
17
convening interface meetings with service users, providers, and local authorities (monitoring standards
and scorecard results) including at Union Parishad and Upazila; developing a local CVA action plan as
agreed by local stakeholders. Informal engagement with community gathering had occurred within NJP
after its launch and before CVA became a more formal and formalized process.  
Phase 3: Improving Services & Improving Policy, which involved: all or part of the CVA action plan
implemented by local stakeholders (not by WVI); implementation of CVA action plans monitored by WVB
staff at least once per quarter (e.g., visiting CVA Working Groups to check progress, which is in addition
to Action Plan monitoring undertaken by local CVA Working Groups and others); connecting CVA Working
Groups or other action plan implementers with partners; forming coalitions to influence government
authorities (at any level, including Union Parishad, Upazila Parishad, divisional, and national levels). It is
important to note that these activities, while supported and funded by CVA in NJP, did not take place in
isolation from other NJP programming aspects, and many were inchoate and/or ongoing at the time of
the present evaluation. That provided both positive aspects of potentialsystemsstrengthening, but urged
caution about the degree to which they may sustain depending on a variety of factors.
4.2.1 Activities
Community mobilization, citizen education, monitoring standards for service units, and participants for
initial meetings: NJP formed the CVA working group from the previously organized Village Development
Committee (VDC) members in each Union Parishad. The group consists of one representative from each
VDC under the union who leads their union’s CVA process with the support of NJP. The good governance
and social accountability component arranged necessary orientation and training of the CVA working
group members, along with special emphasis on facilitation of Monitoring Standards, Scorecard Sessions,
and Interface Meetings (facilitated by NJP staff) to equip them to lead the CVA process in the community.
Until the formation and training of CVA working groups, partner’sstaff, along with community facilitators,
facilitated the CVA process in the community. To support the wide dissemination of information, simple
1‐2 page communication brochures in local languages were developed highlighting the key services or
standards of the community clinic, Union Agriculture Service Unit and Union WATSAN Committee. These
brochures were shared through all sectoral outreach activities by NJP sectoral leads and staff to ensure
the widest possible dissemination of information and to maximize the opportunities under the full range
of NJP activities to the population target. A range of local governance participatory mechanisms are
needed to support the civic education sessions, including the involvement of government officials, local
service providers and Union Parishad Standing Committees.
At the outset of CVA implementation, the Good Governance and Social Accountability team of NJP worked
together to go through the government policies relevant to CVA  as well as state documents to identify
the government’s published monitoring standards for the selected service units meant to be actioned
within CVA. Then the draft standards are shared with respective sectoral staff at the Union
Parishad/Upazila of the project to give their feedback. Having incorporated the feedback of the sectoral
lead, the standards are then translated into Bengal and shared with the respective sectoral government
staff at the Upazila level to incorporate their feedback and ideas. After incorporating the government
official’s feedback, the standards are shared with community people as part of civic education. Before
facilitating the Monitoring Standard Session with the service providers, the standards are shared with
respective Upazila Government officials with the tentative schedule of Monitoring Standard Session. Data
from these meetings was recorded in the CVA database, and discussed further in Section 6.
18
Interface meeting, building networks, and coalitions: During interface meetings, all key stakeholders
were brought together. The expected number of total participantsfor interface meetings was 60‐100. The
main participants of the interface meeting were categorized as: i) service providers: the participants of
monitoring standard session; ii) community participants or service users: community members of all the
CVA activities are service users with actual experience of using the service, especially for the scorecard
sessions (broader members of the community are also involved in other activities as well given they or
family members have an interest as prospective service users); iii) government representatives:
government officials are only involved in the interface meeting and additional advocacy activities (for
example, Union Parishad Member, Union Parishad Chairperson, Upazila Health and Family Planning
Officer, Assistant Engineer‐DPHE, Upazila Agriculture Officer).
NJP took the initiative to involve District Level Government Officialslike Civil Surgeon, Executive Engineer‐
DPHE and Deputy Director‐DAE in atleast one interface meeting in each Upazila fortheirrespective sector.
NJP also created opportunities for the community to have close relationships with key stakeholders
such as the Village Development Committees (VDCs), Existing Government Committee Members,
Union Parishad, Local Club, Upazila government officials and local elites. Though NJP emphasizes
forming the union‐wide CVA working group consisting of VDC leaders who will maintain coordination
among the stakeholders to strengthen citizen voice, the NJP project staff facilitated the process of
networking and coalition‐building until the CVA group was formed and capacitated.
Advocacy and influencing policy: NJP supported or facilitated community advocacy activities to Union,
Upazila, and sub‐district levels. The action points of the action plans developed during the interface
meeting by service providers and service receivers were to be followed up by the people mentioned in
the action plan. The necessary action points that were beyond the capacity of community people were
shared with the Union Parishad before their yearly planning and budgeting during the sharing the
Community Development Plan (CDP) by VDCs. Those action points are to be shared with Upazila and
District administration during the Coordination meeting as well as sharing of the formative research
report.  
However, it is important to note that nothing in the CVA guidelines or NJP programming specifically
mentions or outlines strategies regarding neither meso, national level policy dialogue or action items, nor
does it provide specifics on “systems strengthening” and how that maps onto CVA activities in the
Bangladesh/NJP context. These factors were reinforced during the inception visit. Nonetheless, even
though advocacy and influencing policy beyond the Upazila level was often either theoretical or inchoate
and ongoing at the time of the evaluation, the evaluation also found many positive occurrences of policy
dialogues, linkages, and partnerships that will likely contribute to systems strengthening.  
4.3 Intended Outcomes of Citizen Voice and Action
At a broad level, the intended outcomes/results of CVA include increasing and improving the information
and knowledge that community people have about their rights and entitlements from the government;
and increasing and improving fora that inspire and promote voice and dialogue between community
people, government officials, duty bearers, and service providers. Specifically, CVA means to improve
service delivery, including increased accessto health facilities; improve relations between health staff and
community/women; assist service facilities to meet government standards where they were previously
deficient (including in water and sanitation and agriculture extension); improve infrastructure (e.g,
buildings, maternity wards); and improve communities’ knowledge and willingness to engage with
government.  
19
Specific CVA indicators began to be developed and recorded by NJP in the CVA database in 2018 and
tracked through 2020, with many activities commencing in 2018 or 2019. (There has not been adoption,
development, or tracking of CVA specific indicators in WVI’s larger monitoring and evaluation process for
NJP; rather, the evaluation found that NJP field staff developed specific CVA indicators that it followed
and imputed data in WVI’s global CVA database.) The evaluation objectives and questionsthat follow were
developed after NJP and CVA programming started for the purposes of guiding the present evaluation,
which includesthose indicatorsin the CVA database, but were not limited to them. Instead, the evaluation
team recognized the importance of conducting its own qualitative field‐based data collection to augment
quantitative data for the purposes of meeting the evaluation objectives.
4.3.1 Evaluation objectives
Evaluation objectives included the following:
1. Documenting and analyzing change and impact produced by the project and the project’s
effectivenessin driving change. Change and impact can be understood asthe immediate and longer‐term
effects produced by the program, both directly or indirectly, intended or unintended, and positive or
negative. Effectiveness can be understood as the extent to which program objectives and deliverables
were achieved, taking into account their relative importance. Sections 6 and 7 document and analyze
change and impact.
2. Documenting and analyzing key factors, including promising practices, innovations and approaches
that produced positive and sustainable change for different groups of stakeholder groups, and the
processes that enabled them. Sustainability can be understood as the probability of continued long‐term
benefits for community members and key groups after program completion. To assess the success of
social accountability’s sustainability, it is important to understand if community pressure contributed to,
and more importantly, influenced, government decision‐making. Specifically, did local politicians or
government bureaucrats play a role in addressing gaps in local service standards in response to CVA
activities or any follow up independent collective action by communities? Section 7 analyzes
sustainability.
3. Develop recommendations to improve the quality of similar programs in the future using a
comparative evaluation framework as described in the evaluation methodology. Sections 7 and 9
develop these recommendations.  
4. Develop recommendations for NJP cost extension process to improve systems to sustain key
outcomes. This matters for the next two years for NJP given the 2 year cost extension through 2022, and
the possibility to continue aspects of CVA programming included within that.  Section 9 develops these
recommendations.
4.3.2 Evaluation questions
To meet the evaluation objectives, the following evaluation questions and sub‐questions were developed
as specific lines of investigation to help contribute towards meeting the specific evaluation objectives as
enumerated above. While listed here on their own, Section 8 matches them against the findings from
Sections 6 and 7.
1. To what extent have community health clinics met at least one or more additional minimum service
standards and did CVA contribute?
1.1. What percentage of program health clinics met at least once or more additional minimum policy
standards?
20
1.2. What is the average number of additional service standards met by health clinics, WASH, and
agriculture committees/extension workers at the evaluation period as compared to the baseline/initial
monitoring standards session?
1.3. Are there patterns in improvements and lack of improvements? Is there variation in change based on
geographic location or the domain targeted? (e.g., more success in obtaining additional agricultural
extension workers than in addressing stock outs in health facilities).
1.4. Is there evidence of system strengthening, changing power dynamics or women’s empowerment?
1.5. What role, if any, did government decision‐makers and other duty bearer groups (including
politicians) play in achieving service improvements, additional minimum policy standards or other
governance changes in clinics/WASH/agriculture?
1.6. Is there evidence or to what extent could it be said that responsible government decision makers
responded to community lobbying/pressure through the activities?  
2. In what ways, if at all, did CVA activities affect health, WASH, and agriculture service users and
providers?
2.1. What changes regarding health service quality and standards did users and staff observe? How has
this affected them?
2.2. Did the program produce any unexpected outcomes, positive or negative? Who was affected and
how?
3. To what extent were major program components achieved?
3.1. What role did local leaders (political and bureaucratic) play? Were there any differences in the
responses of local leaders? Were they receptive or hostile to the approach? What were some of the
political and social factors that might have affected their response? To what extent did they make
decisions in response to the project activities?  
3.2. Did program teams, community or government stakeholders implement additional or innovative
strategies or activities to achieve the program goal?
3.3 Given the findings of the formative research, have the functioning of standing committees in
NJP target areas improved? If not, why not? If so, to what extent did NJP activities contribute to
improved functioning? Is there any evidence that standing committees, supported and trained by NJP,
played a role in improving services?  
4. In what ways did the program contribute to the sustainability of project outcomes?
4.1. What evidence, if any, indicates that communities and local government will be able to maintain or
expand project benefits, especially improved service quality and standards in the program health
clinics/WASH/agriculture?
4.2. Given the findings of the formative research on elite capture of participatory processes such
as ward shava, has this persisted or is there evidence of greater inclusion of citizens?  
4.3. What else, if anything could have been done to strengthen the sustainability of improved service
quality and standards in program health clinics/WASH/agriculture?
21
4.4. Given existing weaknesses of current institutionalized social accountability spaces and entities (as per
the NJP formative research, e.g, standing committees, ward participatory budgeting, etc.), what do
national and local leaders think of the CVA approach? Do they see opportunities for elements of CVA to
be institutionalized by the government, and which elements do they deem feasible? Given a lack of clear
definition of sustainability or specified theory of change on systems strengthening, what avenues for
policy dialogue and action did CVA and NJP pursue (even if implicitly) that provides evidence on systems
strengthening?  
4.4 Political Economy Mapping and Theory of Change   
To be able to evaluate whether CVA met these objectives, the evaluation approach was anchored in a
political economy mapping that built on NJP and CVA’s theories of change, and then specified an
institutional matrix reflective of the “ideal type” governance ofservice provision that operates “in theory”
or “on paper” in Bangladesh (as per legislation) to identify parameters and institutional points of contact
for citizens and other stakeholders as well as to denote the institutional actors within the governance
structures and the theories of change responsible for improving service provision. Such an approach
accords with USAID’ssustainability framework as outlined in its 2014 report, “Local Systems: A Framework
for Supporting Sustained Development,” with respect to “leveraging systems thinking” to “engage local
systems.”
With this foundation, the matrix helped develop a set of hypotheses that informed the evaluation’s
methodology for data collection. Specifically, and following upon the inception visit and further explained
in the Evaluation Framework and Methodology, the evaluation team worked towards developing a
political economy mapping of allstakeholders, actors, points of delegation (as per Bangladesh’slegislation
and public sector mapping); formal accountability linkages (e.g, elections and parties) and informal
linkages, and outcomes (per standard organizational “org charts” or political economy mappings of
institutions in public sector management) to guide our methodology in this realm. The formative research
and inception visit allowed a drafting of this mapping, and helped frame the methodology and evaluation
strategy to finalize data collection tools and measures. (Further, it allowed the development of a second
theory of change related to systems strengthening and sustainability, explained further in Section 7.)
At each level, to do this mapping, we specified the actors and institutions that were relevant to CVA as
possible change agents. These actors included citizens, politicians, public officials, external stakeholders.
We considered institutions as sets of rules, government bodies, monitoring and enforcement
arrangements (can be formal and informal), as well as the information that assists monitoring, and the
formal and informal sanctions available to support enforcement. We also considered the strategic
interaction among the various actors within and from outside of these institutions, as shaped by the
nature of the prevailing institutional arrangements and the possibility for managerial and worker
discretion at each level. Whether this scope for discretion is present in practice – and how it is used – is
an empirical question that will vary across contexts and for which the present evaluation seeks to answer
in Bangladesh.  
From this, our approach: i) identified the correct unit of analysis for a specific service, sector, or service
providers; ii) described and measured that provider’s performance over time , and iii) explained/charted
the course of the observed pattern of performance on service provision and outcomes overtime asrelated
to the “ideal type.” This informed possible ways of improving performance and system strengthening
outside of the study context to other realms ofservice provision in Bangladesh, other regions not affected
by the program, and beyond the Bangladeshi context to CVA knowledge‐sharing globally.  
22
This approach also helped to distinguish between at least three “vertical” levels that influenced
organizational performance of local governance and service delivery in Bangladesh: “i) the top level,
where the enabling policies for the sector and organization are set as a function of the substantive
orientation and structural features of the political settlement, including the cognitive maps of political
elites and sectoral leaders; ii) the intermediate or meso governance and managerial levels in the sectoral
chain of service provision, that constitute the core drivers of organizational behavior; iii) the service
provision front line.”  
From this mapping, we can think of citizens having the ability to influence organizational behavior in three
ways. First, they can do so as a formal mechanism through voting. The political context of Bangladesh
makesthis difficult, however, but we did not discount the possibility and also considered more specifically
whether local formal accountability linkages exist, what those might be, and how party linkages might still
operate across some of the programming regions.  
Second, citizens can interact by directly engaging with frontline service providers. This not only operates
for every citizen of Bangladesh seeking a governmentservice in their local area, but also has been adjusted
through CVA as per above in terms of new forms, modalities, and platforms of contact provided by CVA
activities.  
Third, citizens can interact by engaging as individuals and communities with a variety of stakeholders
embedded within the organizational structure not only at the local level, but intermediate and national
levels as well. CVA and NJP provide modalities for how this can take place. In this third realm, it is less
obvious how citizens can directly influence organizational behavior and reform as compared to the first
two as well as other formal and informal mechanisms. This specifically informed the evaluation activities
proposed below, but also touches on systems strengthening lessons from other contexts.   
With the political economy mapping of the institutional matrix of local governance and service provision
in Bangladesh, the evaluation was designed to meet the objective of tying the top‐down approach to  
horizontal and meso‐level approaches by focusing on how CVA impacts the politics of service provision at
the local level while also understanding how it strengthens systems at a meso and national level. This is
captured well in the figure below.
23
Referencing this figure,3 CVA programming provides tools, methods, and platforms to improve local
service delivery and bureaucratic functioning resulting from citizen’s expressing their voice, mobilizing for
action, and overseeing actions and output.  
In its design, this is meant to work through multiple inputs: first, i) structured and transparent processes
organise collective opinion (which is more effective than an individual opinion); ii) use of standards and
community score cards make the criteria for judgements transparent (which facilitates discussion
between people with different views or roles); iii) use of standards relates those judgements to existing
government policy (which increases perceived authority or justification for proposals); iv) increasing
legitimacy of claims based on the articulation of group demands(not individual); v) For maternal and child
health, significant to changing gender politics – women/mothers and children ‘heard’; vi) CVA processes
bring multiple levels of decision‐makers in to the process: different forms of authority are available to
address different issues; vii) multiple types of power brought to the table ‐ the “authorising” effect: access
to higher levels of government; access to media; representing donor organisation. These remain
particularly important in the early stages of CVA, and elsewhere where context is less supportive.  
These changes should have broader institutional spillovers with regards to systems‐strengthening as
related to the figure (Westhorp, et al 2018) below by: i)shifting the conceptualization of the role of citizens
as one of customer or client to an important source of information flows and a “political claimant with
rights” that is integrated into governance structures; ii) leveraging information flows to inform decision‐
making regarding system‐level inputsrelevant to localservice provision (including government monitoring
standards, budgets, and health challenges); iii) supporting more cooperative mechanisms to improve the
relationship between citizens/communities, service providers, and other relevant duty bearers; iv)
maintaining, increasing, and improving necessary resources for adequate service delivery (including
budgeting, provision, and scaling); v) improving advocacy and representation of communities and
community members into the planning and prioritization of their local development programming; vi)
strengthening positive feedback loops and sources of accountability that help to maintain and grow
institutional actors and structures, and further promote the diffusion of effective governance and
3 As developed by Westhorp, G & Ball D, 2018, Citizen Voice and Action for Government Accountability and Improved
Services: Maternal, Newborn, Infant and Child Health Services, Final Evaluation Report, World Bank’s Global
Partnership for Social Accountability, Community Matters Pty Ltd, October.   https://gpsaknowledge.org/wp‐
content/uploads/GPSA‐Indonesia‐CVA‐Project‐Final‐Evaluation‐Report‐2018.pdf.  
24
processes.
Source: Westhorp, G & Ball D 2018 Citizen Voice and Action for Government Accountability and Improved
Services: Maternal, Newborn, Infant and Child Health Services, Final Evaluation Report.
4.5 Summary of Results from Formative Research and Inception Visit
Formative research about NJP programming and CVA was conducted as it was launched in 2018. These
results help to contextualize the progress of CVA at the beginning of the evaluation development phase
(and before the evaluation framework had been written or finalized). To summarize some of those results,
the formative research found:
● Increased knowledge of local community and service provider on services standards
● Systemic issues being raised at higher levels
● Increased ownership and community contribution to improve services
● Local elected representatives are making public commitments to improve services
● 25% of clinics with improved infrastructure (water facilities, sanitary latrine, repairing of the
community clinic building)
Activity Number to date (as of
August 10, 2020 )
Social audits of service standards for community health clinics 119 (x2)
Community services scorecards to rate quality of community health clinics 476
Interface meetings for community clinics 119
Social audits for WATSAN committees 40 (x2)
Scorecards for WASH committees 360
Interface meetings for WASH 40
Social audits for Union agricultural extension services 40 (x2)
25
Scorecards to rate extension workers service 160
Interface meetings for agricultural services 40
Upazila and divisional dialogues for advocacy 22
Study tours for government officials 3
Policy briefs 3
National Dialogue on Community Clinics 01
As a first step towards conducting the evaluation, in November 2019 the evaluation team organized an
inception visit to Bangladesh. The inception visit began with members of the evaluation team visiting with
WVB national staff in Dhaka. Then, a field visit took place in the Dacope and Satkhira upazilas, where the
evaluation team liaised with NJP project staff and conducted visits to CVA working groups, village
development community, unions, and upazila, community clinic, agriculture extension, and WASTAN.
They also met with NJP staff and NGO implementing partners. Results from that were critical toward
informing the political economy mapping, refining the theory of change, and completing the research
design and evaluation methodology as described in subsequent sections. Further results from the
inception visit are reported in the Evaluation Framework and Methodology.
26
5. Evaluation Methodology and Research Design
The evaluation approach draws on methodologies deemed relevant to providing a strong analytical
framework for examining processes and what results were achieved, for whom, and in what
circumstances. The framework for the methodology of the evaluation utilized a mixed‐methods approach
to the application of inputs, processes, and outputs that is gender‐sensitive and longitudinal in nature.
The evaluation assesses the degree to which gender, power, politics, and other institutional relationships
and contextual factors give rise to inequities, discrimination, or a lack of access may have been addressed
by specific CVA components with a view toward the promotion of social change and the production of
knowledge for improved project development. Moreover, special attention was paid to whether and how
CVA results “scale‐up” geographically and institutionally to affect policy dialogue and policy change across
meso/regional and national levels in Bangladesh, leveraging systems thinking to engage the realities of
the political economy of governance in the country context. This involved both within the sampled areas
of the NJP program (the core focus on data collection), as well as the potential to explore the prospect of
scaling‐up outside of the intervention areas (with secondary and national‐level analyses).  
Specific methodological activities by the evaluation team included program and government document
review; institutional mapping and a political economy analysis of the political and governance structure
of service provision; CVA and NJP database analysis; data collection from key informant, in‐depth and
semi‐structured interviews; focus groups discussions; retrospective “most significant change” analysis;
and causal‐process tracing and policy analysis.
5.1 Evaluation sample
We performed the fieldwork data collection from the NJP project areas within the Khulna Division. In the
Khulna Division, two districts – Khulna and Satkhira – host NJP, and within those districts, two upazilas
each. In Khulna District, the one project upazila selected was Dacope Upazila (within which Dacope Union
and Koilashgonj Union selected for fieldwork). In Satkhira District, the project upazila selected was Kaliganj
Upazila (within which South Streepur Union and Compaful Union selected for fieldwork). While the larger
evaluation included all four upazilas, especially for the quantitative data tracking, the qualitative fieldwork
was focused on unions in Dacope and Kaliganj Upazilas.  
5.2 Intermediate and final outcomes
We perform analysis on the monitoring and standards for community clinics, water and sanitation, and
agriculture extension.
CVA’s three main final outcomes that support the objective of the project included:
1. Improved citizen voice and action
2. Increased demand articulation and attending improvements in service delivery across targeted
sectors (not only as a one‐time occurrence but also continuously).
3. Improved community and government support, acceptance and commitment to sustaining
progress on targeted sectors and service delivery, and policy dialogue
The project also consists of numerous intermediate outcomes that included:  
1. Mobilization of the groups/committees at the local level
2. Consistent conduct of the community meetings
3. Attendance by diverse and targeted members into the meetings
4. Filling in scorecards at meetings  
5. Relaying scorecards in interface meetings to stakeholders
6. Increased acquisition of critical services from interface meetings and follow‐up
7. Promote positive social norms that encourage inclusion, diversity, participation
27
5.3 Measures  
In order to define the most relevant and feasible measures and data collection tools, consultation with
NJP and relevant stakeholders was performed during the inception phase of the evaluation and finalized
thereafter. Tools and analysis plans were included in full in the research protocol as appears in Evaluation
Framework and Methodology.  In brief to review here, measures and tools included the following:
● Official enrollment data  
● Number and location of meetings
● Scorecard data  
● Other Survey data
● Qualitative/FGD with communities, stakeholders,  
● Quantitative/output/survey data on services  
● Improvements in feeling thermometers
● Most significant change analysis
● Policy analysis of policy change, adoption, implementation and outcomes
● Process‐tracing of policy change per institutional/political economy mapping
● Community surveys, focus group discussions, and key informant interviews to gain quantitative
and qualitative understanding of community perception towards CVA
5.4 Data collection and fieldwork
Quantitative tracking measures on some of the above were obtained through the CVA platform database
and in turn analyzed and summarized by the evaluation team for the present evaluation. Qualitative data
collection involved the following activities(data collection tools can be found in the Evaluation Framework
and Methodology Annexes).
5.4.1 FGDs with CVA working groups and Community Health committees
We performed FGDs with CVA working groups and community health committees in the selected unions.
The objective of the FGD was to probe the CVA working groups and community health committees on
topics generally related to how the CVA working group members perceived whether and how their CVA
working group and the CVA process has improved information, voice, dialogue, and accountability. This
covered both the past and extant activities of the working group that were the “most important change”
or “most important challenge” within the community clinic, agriculture extension, and WATSAN sectors
(community clinics established specifically for the community health committees). It also covered
activities and perceptions of union and upazila level institutions and actors, identifying mechanisms for
reform and barriers to change, as well as the working group’s thoughts on sustainability, system
strengthening, and ideas for CVA programming in the next few years.   
5.4.2 FDGs with VDC working groups, Scorecard Sessions & Women’s Groups
We performed FGDs with VDCsin only a subset of the selected CVA working groupsin the selected unions.
While VDCs are important institutions in their own right as actors relevant for the CVA process, the
purpose of their inclusion was to collect perceptions and views from the “community people” who
constitute these VDCs so they are able to share how they and the community view CVA and the CVA
working groups. The objectives of the FGD with VDC members was to understand the linkages between
VDC creation, operation, functioning, and how this does or does not overlap with the CVA working groups.
It covered many of the same topics as the FGD with CVA working groups, but focused specifically on
28
community member perspectives, how active members are within VDCs, and how they view their role in
CVA.  
5.4.3 FGDs with WATSAN committees
We performed FGDs with the WATSAN committees in the selected unions. The objectives of the FGD with
the WATSAN committee was to focus on topics generally related to whether and how the WATSAN
committee formed and been constituted and what their past and extant actions had been, including the
“most important change” or “most important challenge” within the water and sanitation sector. It also
covered activities and perceptions of union and upazila level institutions and actors related to WATSAN,
identified mechanisms for reform and barriers to change, and the committee’s thoughts on sustainability,
system strengthening, and ideas for CVA programming in the next few years related specifically to
WATSAN.
5.4.4 KIIs with Community Clinic Staff
We performed FGDs and KIIs with the Community Clinic Staff for each union paired with each CVA working
group in the sample. The objectives of the FGD included probing the community clinic staff about topics
related to CVA working groups and the community clinic, including whether the CVA process had
specifically improved information, voice, dialogue, and accountability for the community clinics and the
performance of health officials at the union and upazila levels. This covered the past and extant activities
of the CVA working group that are the “most important change” or “most important challenge” within
community clinic, including learning about monitoring and standards (through their posting, which makes
those entitlements known to community people and area residents/patients at the clinic), and tangible
changes of the quality ofservices provided by the clinic. It also covered activities and perceptions of union
and upazila level institutions and actors, identifying mechanisms for reform and barriers to change, as
well as the community clinic’s thoughts on sustainability, system strengthening, and ideas for CVA
programming in the next few years.  
5.4.5 KIIs with Union Council Chairperson & Upazila Council Chairperson
We performed FGDs and KIIs with the Union Council and Council Chairperson for each union. The
objectives of the FGD and KII included probing the union officials about topics related to CVA working
groups and the community clinic, agricultural extension officer, and WATSAN committee. We heard from
their perspective on whether the CVA process had improved information, voice, dialogue, and
accountability for the community, as well as how union officials have in turn responded.  
5.4.6 KIIs with Union and Upazila Agricultural Extension Officer
We performed KIIs with the agricultural extension officers at the union and upazila levels. The objectives
of the KII included probing the officers about what they know regarding CVA working groups on topics
related to the work of the agricultural extension officer and the work they are meant to do, and how they
perceive the CVA working group members and their activities regarding the agricultural sector. We also
probed about their past and extant activities at the union or upazila level in agriculture that CVA was
specifically involved in, including the “most important change” or “most important challenge” within
agriculture extension from their perspective. We also covered activities and perceptions of other union
and upazila level institutions and actors, identifying mechanisms for reform and barriers to change in
addition to the officials’ thoughts on sustainability,system strengthening, and ideasfor CVA programming
regarding agriculture in the next few years. And we probed about this issue regarding political concerns
around LSP suppliers, monitoring of LSPs, etc.
29
5.4.7 KIIs with the Union and Upazila Health Officer
We performed KIIs with community clinic and other relevant health officials at the union and upazila
levels. The objectives of the KII included probing the officers about what they know regarding CVA working
groups on topics related to the work of the community clinic officer and the work they are meant to do,
and how they perceive the CVA working group members and their activities regarding the community
clinic sector. We also probed about their past and extant activities at the union or upazila level in
community clinic that CVA specifically was involved in, including the “most important change” or “most
important challenge” within community clinic from their perspective. It also covered activities and
perceptions of other union and upazila level institutions and actors (including the other health services
providers and relevant ministries), identify mechanisms for reform and barriers to change, and the
officials’ thoughts on sustainability, system strengthening, and ideas for CVA programming regarding
community clinics in the next few years.  
5.4.8 KIIs with NGO Facilitators & NJP Field Staff
We performed FGDs with the NGO facilitators who work across the selected unions in both upazilas. The
objectives of the FGD with the NGO facilitators focused on NGO and NJP staff supporting CVA activities
and CVA working groups, including facilitation, organization, oversight, monitoring, and activities (like
scorecard session, interface meetings, dialogue sessions, follow‐up).  
5.4.9 KIIs with National level stakeholders of NJP, WVB, and Line Departments
Members of the evaluation team had the opportunity to conduct key informant interviews with NJP and
WVB staff, and other national policy makers in relevant line departments, in Dhaka. NJP national staff
included the monitoring and evaluation team, the NJP head, chief of party, and policy/advocacy officer.
Line department members were drawn from the relevant health ministries, local government ministry,
and department for health engineering. The objectives of KII with NJP and WVB staff  for monitoring and
evaluation focused on their perspective of CVA and how it scaled to national level dialogues and policy
changes. With KII of line departments, this process assessed how they seek community support and
involvement in line department activities, the opportunities and challenges they face delivery services to
the local level, and how they perform oversight and monitoring of their line department members at the
Upazila and Union level.
30
6. Results from Fieldwork and Data Collection
6.1 Summary of Results
We first summarize the results of the evaluation, before exploring individual topics and results from
specific data collection tools below in various subsections. This involves summarizing the results while
identifying the factors and mechanisms thatseem most relevant to CVA impact and system strengthening
and the CVA components that speak to sustainability. Section 7 further explores system strengthening.
Section 8 tracks these summaries of results against indicators as listed in Section 4.
Quantitative indicator tracking provides systematic data on the CVA activities as per the CVA database.
These standards are collected via CVA and inputted in the CVA database. The program first collects the
officially delineated and published standards by the government for the relevantsectors. CVA then checks
the reality of the situation of the standards at the relevant unit (for example, a community clinic) against
the official standards, which are captured in the CVA database. This generates baseline compliance to the
standards originally before CVA activities that are then tracked against progress on those standards.
The evaluation found that all CVA working groups in the project area were formed and conducting
activities. They had completed activities, including scorecard sessions, interface meetings, action plans,
and follow‐up, in all three development sectors (community clinics, water and sanitation, and agriculture
extension). Baseline indicators regarding the monitoring standards at community clinics, Union WATSAN
committees, and Union agricultural services units (taken before social audits via CVA) found many gaps
and shortcomings in meeting the requisite standards for service provision.  
In the first monitoring standard session before CVA activities, compliance of standards for community
clinicsshowed that 58 clinics(49%) met between 76‐100% ofstandardsin 2018, and by the second session
in 2020 after CVA activities, 105 clinics (88%) met between 76‐100% of standards. Compliance of
monitoring standardsin water and sanitation in the firstsession found that 3 (7.5%) of the Union WATSAN
committees met between 76‐100% of standards, and by the second session, 4 committees (10%) had met
between 76‐100% of standards (but a further 32, or 80%, met between 51‐75% of standards compared to
20, or 50%, that met between 51‐75% in the first session). Compliance of monitoring standards for the
Union Agriculture Service Unit in the first session found that 12 (30%) unions met between 76‐100% of
standards, and by the second session in 2020, 22 (55%) of units met between 76‐100% of standards.
Moreover, citizen assessments reported in surveys revealed generally “fair” or “poor” ratings for service
delivery in the three sectors. By 2020, CVA working groups across the sampled unions could report
important progress on developing, implementing, and following‐up on action plans asreported in the CVA
database.  
To support these quantitative results, we also summarize from qualitative data collection in the field.  
First, we can draw important conclusions regarding how CVA impacted citizens. A first finding is that
service provision improved and citizens gained information. The situation before CVA was such that CVA
did not involve any real direct spending, programming, or service provision to help meet the community’s
needs. However, the problem identified was not resources, but rather inertia in providing them as well as
information regarding what should be available and what was available. As a result of CVA, service
provision is improving along many dimensions and indicators, with community people also having gained
information.
A second finding is that citizens’ voice and action are improving and helping to articulate demand for
services. CVA does not invent “voice” for citizens ‐‐ they already have a voice. But the problem identified
31
before CVA was not lacking a voice, but rather giving people confidence in their voice and an organizing
voice while making voice matterthrough dialogue, interface, and the observation of positive changes after
articulating demands. As a result of CVA, we can say that voice is improving in form and content to allow
citizens/customers to demand what they want and present whether or not things are going well by
utilizing techniques  such as the scorecards, joint action plans, and follow‐up. This arises from a greater
understanding of the standards themselves, as well as government constraints in providing them
previously. From that, CVA operates like an informal targeted civic education platform that then reveals
processes and points of contact for citizens to leverage with institutional actors to improve services.  
A third finding is that civil society and citizen mobilization have been activated and has helped citizens
coordinate collective action. Before CVA, citizens had the ability to mobilize for civilsociety, and CVA does
not generate civil society de novo. But the problem identified was that even so, no actions were being
taken to activate civil society, and there was inertia among the community people who needed a
coordinating mechanism. The impact is that CVA provided such a mechanism while helping  to activate
civil society and coordinate collection action, with both especially important when NGOs are not direct
service providers.
Next, we can draw important conclusions regarding how CVA impacted institutions and institutional
actors. A first finding that we document is that institutional actors have overcome inertia and found
motivation to improve their job performance. CVA itself did not create new legislation, mandates, or
institutional mechanisms to provide services, but the problem the evaluation identified was not a lack of
resources, but rather bureaucratic inertia performing duties and providing services as well as a lack of will
to perform well (or sincerely). As a result of CVA, service provision improved by “reminding” duty bearers
and service providers to do their jobs while providing them with the motivation to do so.
A second finding is that institutional actors have better organized to achieve coordinated actions to
improve service delivery. Before CVA, the units, agencies, and institutions responsible for service
provision already existed. The problemsidentified related to the degree of real devolution, a lack of clarity
in roles and responsibilities horizontally (union) and vertically (upazila, line departments), and a lack of
proper human resources. As a result of CVA, these actors got in a room together and started talking to
each other, helping to clarify rolesfor mutually agreed upon actionssubsequently helped them to do their
jobs while approaching human resource constraints in creative ways.  
A third finding is that CVA improved monitoring while working cohesively with other aspects of NJP
programming.  In the absence of CVA, institutional actors already had mechanisms, finances, and tools to
do their jobs and provide services. The evaluation found that problems related to a lack of demand
articulation by citizens in terms of priorities and the allocation of expenditures, inertia among officials at
Union and Upazila levels to their job (inertia among principals to monitor agents at all levels), and a lack
of resources and will to work properly within extant monitoring systems. As a result of CVA, these
mechanisms and tools were improving due to other aspects of NJP, but CVA helped to reinforce them
while also providing a unique tool for monitoring.
Last, we can draw important conclusions regarding how CVA improved accountability. A first finding is
that CVA helped promote formal institutional accountability. Before CVA, the government already had
its own monitoring system to manage institutional actors and ensure service delivery. The problem was
rather that new legislation took time to vest, with much inertia at all levels of principals and agents,
overlapping or a lack of clarity on roles and responsibilities, and other associated human resource
constraints. The impact from CVA is that accountability was aided by matchmaking, coordination,
feedback along with new forms of citizen‐activated direct accountability (e.g, the posting of phone
numbers and activation of committees; results “trickling up”).  
32
Second, CVA helped promote social accountability, or accountability that arises from “outside” of
institutions. Citizens played no formal role in institutional accountability as public officials are responsible
to themselves (whereas ideally decision‐making at the top should “trickle down” to the bottom &
information from the bottom should “trickle up” to top). But even so, “measures on paper” for
institutional actors were accompanied by inertia among bureaucracy to oversee and monitor along with
human resource constraints. Moreover, officials were busy with activity of the moment without building
stronger or more sustainable systems. The impact of CVA has been to activate social accountability as a
pathway to improving institutional performance and accountability even from outside of institutions,
which also allows pathways to work with/within institutions that provide support to institutional actors
(e.g.,  WATSAN committee).   
A third finding is that CVA helped promote formal political accountability. Citizens play a formal role in
electoral accountability and public officials are responsible to voters. But the problem is not the absence
of formal accountability but rather that it was limited by the nature of party politics in Bangladesh, where
elections have only been one mechanism (and a weak mechanism). The evaluation found that political
pressure can still matter as activated through social accountability by convincing officials when and how
they benefit from CVA.
6.1.1 Outline and Organization of results
In the following subsections of Section 6, we present the findings from the data collection and field work
with respect to the qualitative investigation central to the evaluation. These findings are based primarily
on quantitative indicators as well as focus group discussions and key informant interviews with relevant
CVA working groups, community members, government officials, service providers, and departmental
employees working in community clinics, water and sanitation, and agriculture extension.   
To do so, we break down the data analysis and results into the main topics and themes inquired about
across the range of quantitative indicators, interviews, and focus group discussions in the data collection.
These topics proceed longitudinally and thematically in Section 6, with responses from all manner of data
collection tools reflecting information conveyed by the focus group or interview where relevant.  
In subsection 6.2, we begin with quantitative tracking indicators from the CVA database regarding the
formation and activities of the CVA working groups along with tracking indicators regarding community
clinics, water and sanitation, and agriculture extension services. Then, from sub‐section 6.3, we discuss
“Pre‐CVA,” or the conditions that existed before the advent of CVA programming in the NJP areas. We
then proceed to discuss “CVA formation and activities” (6.4) as reported by stakeholders over what has
taken place over the last 2‐3 years for the CVA working groups. “CVA process” (6.5) describes the
contribution of the social accountability platform and mechanism toward improving voice and action and
government performance. This is followed by “Government response,” (6.6) or the reported reaction by
officials to CVA programming. “Participants” (6.7) examines what groups were targeted or excluded in
CVA, and “Services Outcomes” (6.8) reports whether and how CVA contributed to improved service
delivery with respect to community clinics, water and sanitation, and agriculture extension services.
“Interagency Process” (6.9) reviews the interagency process for institutional actors, while “Most
important issue” (6.10) reflects stakeholders’ own evaluation of what they thought was most effective (or
most in need of improvement) in CVA programming as well as its impact on social accountability. Finally,
“Sustainability” (6.11) conveys stakeholders’ reported ideas about how to sustain the positive impact of
CVA in the future (as well as areas for retooling).  
33
For ease of readability in the following results, we adopt a numbering convention for each focus group
discussion or key informant interview that lists the type of respondent and their location with a number
of the interview listed in text (e.g., “#20”) that corresponds to the full list in Annex A.
Following the summary of these quantitative and qualitative findings in Section 6, Section 7 leverages the
resultsfrom the data collection, further document review, policy dialogue, and interviews conducted with
NJP staff and national‐level officials for a more in‐depth evaluation of systems strengthening and policy
analysis. This is produced in light of a more refined and comparative theory of change that operates at
meso and national levels and that speaks to scaling in Bangladesh and other country contexts. Section 8
summarizes the qualitative findings with quantitative indicators and tracking against the evaluation
objectives.
6.2 Quantitative Results on Indicators from CVA Database
The quantitative aspect of the evaluation involves reporting on data systematically collected by the NJP
field staff, and reported in the CVA database. This involves tracking the establishment and activities of
CVA working groups from 2017‐2020, as well as perceptual data on citizens’ views of service provision at
the start of CVA; data on monitoring standards at the start of CVA for community clinics, water and
sanitation, and agriculture extension services; and tracking changes to these indicators over time. With
these quantitative data, the evaluation is able to assess changes over time and in part whether CVA
activities met indicators and evaluation objectives. While the main qualitative data collection for the
evaluation occurred through fieldwork interviews and focus groups, the evaluation team also took a
narrow scope of relevant CVA database indicators to include for the present evaluation.
Summary of Results on Quantitative Indicators from CVA Database:
● Formation and activities of CVA working groups are occurring in all of the sample, and have
met the full list of activities as per Section 4.
● Community perceptions of services before CVA were generally “fair” or “poor” regarding
community clinics, water and sanitation, and agriculture extension services.4
● Monitoring standards were lacking for community clinics, Union WATSAN committees, and
Union agricultural service unit before CVA.
● After CVA and by 2020, compliance of monitoring standards hasincreased for community clinics,
Union WATSAN committees, and Union agricultural service units.  
● CVA working groups are making important progress on developing, implementing, and
following‐up on action plans.  
Finding #1: Formation and activities of CVA working groups are occurring in all of the sample, and have
met the full list of activities as per Section 4.  
4 A second wave of community perceptions of services was not conducted so it is not possible to track changes over
time.
34
Mirroring the qualitative responses by CVA working groups below, the quantitative data confirm that in
the NJP project areas where CVA was implemented. In 2017, NJP facilitated community people to form
Village Development Committees (VDCs) and in 2018, NJP mobilized VDC members to form 40 CVA
working groups in 40 Unions, drawing from representatives of VDC members. The project undertook
capacity‐building initiatives for the CVA working groups members, and they are now functioning across
the 40 Unions (and 360 wards). Further, CVA had first conducted activities for the 119 community clinics
in the first project year, followed by activities in the 40 Union WATSAN committees and 40 Union
agriculture extension offices shortly thereafter. These groups across all the relevant wards and unions
have not only been constituted, but have all completed their community perception ratings, scorecard
sessions, monitoring standards compliance, action plan status, and listed the action item monitoring by
2020.
Finding #2: Community perceptions of services before CVA were generally “fair” or “poor” regarding
community clinics, water and sanitation, and agriculture extension services.
Community perception ratings were taken at the beginning of CVA to generate baseline indicators of the
views of community people toward the service delivery process at community clinics, in water and
sanitation, and agriculture extension.5
  
The survey was conducted across the communities where CVA activities were to commence. Community
people participated in the scorecard sessions of three service units, with Community Clinics (4,789
sample), Union WATSAN Committee (4,004 sample), and Union Agricultural Service Unit (1,603 sample).6
Respondents included subsamples balanced by gender and age.7
  
The survey probed respondents’ ratings on selected itemsrelated to their perceptions ofservice provision
in their community. Answer options were given on a five point scale corresponding to emoticons of
varying degrees of smiley faces that reflect “very good,” “good,” “just okay,” “poor,” and “very poor”
ratings. While the CVA database includes data on numerous indicators and activities regarding various
NJP activities, we focus here on what was relevant for the three sectors and the present evaluation.8
  
For community clinics, the respondents reported an average rating of 2.87 for the “availability of clean
latrines/toilets for patients,” average rating of 2.33 for the “availability/reliable source of safe water at
the facility,” and an average rating of 3.7 for the “presence and accessibility to drugs/medicine.”  These
results demonstrate that from the perspective of citizens in the community who likely seek services at the
community clinic, those clinics were lacking in providing the requisite infrastructure, tools, and medicine
to support patient comfort and treatment. They barely reached a “just fair” rating, and were more likely
viewed as poor or very poor. This suggested the great potential for CVA to plausibly provide impact in
5 The decision was taken not to repeat this survey in 2020, so it is not possible to track changes in community
perception ratings; we therefore only present the ratings as perceived at the beginning of CVA as data on the
baseline (e.g, pre‐CVA).
6 A few of the scorecard participants may have overlapped among the service units but not by that much (and we
lack information on how much of the participants have overlapped among the service sectors).   7 The number of respondents varies for these performance indicators since different groups prioritized and voted
for different sets of performance measures/standards.
8 Each “category” in the scorecard can be denoted as S(X), equal to the value of category X in the scorecard; each
category in the scorecard is then assigned a weight, and the “weighted value” is denoted as W(x), equal to:
[S(X)*Weight(x)], where weight is defined as: Very Bad=1, Bad=2, Just Okay= 3, Good=4, Very Good=5. Therefore,
the total score we focus on here is essentially the weighted average of the scorecard.
35
addressing these issues to improve service delivery at community clinics. These results were also echoed
in the qualitative assessments on the “Pre‐CVA” conditions of the clinics as expressed by CVA working
groups, community members, and clinic service providers discussed below.
For water and sanitation, at baseline, average rating of 2.8 reported by community people for the
presence of “safe drinking water facilities (STW, DTW, Pond and Water Purifier),” an average rating of
3.06 for “household access to sanitary latrine,” and an average rating of 2.06 for the presence of “waste
disposal system/points.” These results demonstrate that from the perspective of citizens reporting on the
conditions of water and sanitation service provision in their communities, those conditions to access safe
drinking water, and basic sanitation and waste disposal were mostly deplorable. This pointed to the need
for Union WATSAN committees to activate and mobilize to address community‐level water and sanitation
needs as a part of the CVA process. These results were echoed in the qualitative assessments provided by
many government officials and service providers, who noted the lack of access to safe drinking water and
proper sanitation across the communities.  
For agriculture extension services, at baseline, the community people surveyed reported an average rating
of 2.91 regarding the “number of service providers for provision of seeds,” an average rating of 3.11 for
“fertilizer and insecticides with proper price,” and an average rating of 3.03 for “modern agricultural
equipment.” These results demonstrate a lack ofsufficient agriculture extension services provided to local
farmers across the communities surveyed, and the need for more robust action by the Union agricultural
service unit. This pointed to the important role that CVA activities could plausibly serve in addressing
service provision for farmers by unions. And these results were echoed in the qualitative assessments
provided by community people, Union officials, and others during the qualitative data collection.
Finding #3: Monitoring standards were lacking for community clinics, Union WATSAN committees, and
Union agricultural service unit before CVA.
Next, we present quantitative results on the baseline measures of monitoring standards (the official
standards provided by the government in formal documentation) for community clinics, WATSAN
committee, and agriculture extension. These monitoring standard measures were taken at the outset of
CVA activities with NJP guidance in the 119 community clinics, 40 Union WATSAN committees, and 40
Union Agricultural Service Unit as reported and tracked in the CVA database.
For community clinics, the baseline assessment of compliance with monitoring standards showed that in
the 119 clinics surveyed, compliance with the government’s 22 monitoring standards were not
consistently met in 2017. 58 clinics (49%) met between 76‐100% of standards in 2018, while 52 clinics
(44%) met between 51‐75% of standards, 7 clinics (6%) met between 26‐50% of standards, and 2 clinics
(2%) only met between 0‐25% of standards. In the first session, compliance was most likely on issues
related to the existence of the clinic building/structure, day and time of service delivery posted, and the
existence of community support group committees. Compliance was least likely on issues related to
proper medical equipment, safe sanitation, and safe drinking water facilities.
For Union WATSAN committees, the baseline assessment of compliance with monitoring standards
showed that overall in the 40 unions surveyed, compliance with the government’s 11 monitoring
standards were not consistently met. 3 (7.5%) of the Union WATSAN committees met between 76‐100%
of standards in 2018, while 20 (50%) met between 51‐75% of standards, 10 (25%) met between 26‐50%
of standards, and 7 (17.5%) only met between 0‐25% of standards. In the first session, compliance was
36
most likely on issues related to holding the monthly meeting,9 necessary initiative to reach the real
beneficiaries, and coordination among the government and non‐governmental organizations for a list of
WASH projects. Compliance was least likely on issues related to access to sanitary latrines, pond per
village, and access to safe drinking water.
For Union agricultural service units, the baseline assessment of compliance with monitoring standards
showed that overall in the 40 unions surveyed, compliance with the government’s 9 monitoring standards
were not consistently met. 12 (30%) of the units met between 76‐100% of standards in 2018, while 18
(40%) met between 51‐75% of standards, 7 (17.5%) met between 26‐50%, and 3 (7.5%) met only between
0‐25% of standards. Compliance was most likely on issues related to having an action plan and the SAAO
having a diary. Compliance was least likely on issues related to field visit and service schedule/time of the
SAAO and staffing/service providers.  
The lack of compliance and adhere to monitoring standards in public health, water and sanitation, and
agriculture extension all demonstrated at baseline the important attention that CVA activities could pay
to improve service delivery at community clinics, Union WATSAN committees, and Union agricultural
service units.  
Finding #4: After CVA and by 2020, compliance of monitoring standards has increased for community
clinics, Union WATSAN committees, and Union agricultural service units.
We present quantitative results on measures of monitoring standards for community clinics, WATSAN
committee, and agriculture extension taken after CVA and reported by 2020 to track progress from
baseline in the same sampled locations asreported in the CVA database.10 The table compares compliance
between the first assessment before CVA and the second session after CVA.
The table demonstrates clear evidence of positive changes in the three sectors in 2020 (endline) as
compared to 2017‐18 (baseline) in the project areas. In the first monitoring standard session before CVA
activities, compliance of standards for community clinics showed that 58 clinics (49%) met between 76‐
100% of standards in 2018, and by the second session in 2020 after CVA activities, 105 clinics (88%) met
between 76‐100% of standards. Compliance of monitoring standards in water and sanitation in the first
session found that 3 (7.5%) of the Union WATSAN committees met between 76‐100% of standards, and
by the second session, 4 committees (10%) had met between 76‐100% of standards (but a further 32, or
80%, met between 51‐75% of standards compared to 20, or 50%, that met between 51‐75% in the first
9 Importantly, this likely reflects members of WATSAN committee attending meetings at the Union Parishad, rather
than constituting and holding the monthly meeting as required by the WATSAN committee specifically.
10 For community clinics, the initial monitoring standard session was conducted between March 2017 and September
2017; the second monitoring standard session was conducted between May 2019 and September 2019. For union
WATSAN committees, the initial monitoring standard session was conducted between October 2017 and December
2018; the second monitoring standard session was conducted between October 2019 and March 2019. For union
Agricultural Service Unit, the initial monitoring standard session was conducted between April 2018 and December
2018; the second monitoring standard session was conducted between October 2019 and March 2020.  
37
session). Compliance of monitoring standards for the Union Agriculture Service Unit in the first session
found that 12 (30%) unions met between 76‐100% of standards, and by the second session in 2020, 22
(55%) of units met between 76‐100% of standards.
Finding #5: CVA working groups are making important progress on developing, implementing, and
following‐up on action plans.  
Action plans were developed and implemented across the sampled areas for community clinics, Union
WATSAN committees, and Union agriculture services units. For community clinics, a total of 1,325 action
items were developed for the 119 community clinics; as of 2020, 679 (51%) had been completed, 537
(41%) were pending, and 109 (8%) were ongoing. For Union WATSAN committees, a total of 734 action
items were developed for the 40 committees; as of 2020, 376 (51%) had been completed, 313 (43%) were
pending, and 45 (6%) were ongoing. For the Union agriculture service unit, a total of 515 action items
were developed for the 40 units; as of 2020, 206 (40%) had been completed, 279 (54%) were pending,
and 30 (6%) were ongoing. These results demonstrate important progress not only on establishing the
action plans for the three sectors, but charting progress on completing the action items.
The CVA database does an excellent job of keeping track of action item monitoring, delineating the
activity, who is responsible for monitoring it, by what date it should achieve completed status, and then
the current status indicated. By way of example, ss shown in the figure below for an example list of action
items regarding community clinics as shown in the CVA database, these are also color coded to show
progress (green=completed, yellow=pending, red=ongoing).
38
6.3 Pre‐CVA
Since the qualitative aspect of the evaluation was limited to retrospective analyses in order to investigate
possible changes in indicators and outcomes due to programming, to assess change we first attempted to
establish baseline conditions as reported by key stakeholders of the pre‐CVA environment. We inquired
to the CVA working groups, service providers, and government agents as to the conditions of service
provision pre‐CVA, as well as how they described processes for influence, dialogue, and accountability
between communities and officials before CVA. Specifically, we discussed the conditionsin the community
before CVA, including the conditions of services in the relevant sectors (community clinic, water and
sanitation, and agricultural extension service unit); we then inquired about experiences with government
officials before CVA; and asked the WATSAN committee, Community Clinic staff, and agricultural officers
their perspectives on the pre‐CVA environment.  
Summary of Findings on Pre‐CVA:
● The gap between service providers and service recipients during pre‐CVA was noticeable in all
three sectors as a matter of “supply”;service providersreported inertia and a lack of motivation
as reasons for not performing up to task, while there were few accountability mechanisms to
encourage good performance.
● A gap in services did not justresult from problems ofsupply but also as a “demand” side problem
on the part of communities and people lacking knowledge; community members were not
aware of their rights and entitlements with regards to seeking services from the government
departments; they also lacked a platform to articulate demands and dialogue and interface with
duty bearers.  
● A lack of accountability in service provision exacerbated the problems of supply and demand
pre‐CVA; without robust monitoring and evaluation from institutional actors or the opportunity
for community people to demand services or inspire accountability, both institutional and social
forms of accountability were not present.
Finding #1: The gap between service providers and service recipients pre‐CVA was noticeable in all three
sectors as a matter of “supply”; service providers reported inertia and a lack of motivation as reasons
for not performing up to task and there were few accountability mechanisms to encourage good
performance.
The first area of the research was intended to gauge communities’ experiences with service delivery in
their areas before the advent of CVA regarding the three government departments most relevant to CVA
programming‐‐including WASH, Community Clinic, and Department of Agriculture Extension. We broadly
summarize the pre‐CVA experience as demonstrating that service delivery of the three departments was
ineffective and inefficient; service recipients were not able to access the government services to which
they are legally entitled; and accountability for service provision was not ensured. Moreover, citizens
report not having a clear understanding of their rights and entitlements to receive services from these
departments nor what services they could realistically access in water and sanitation, at the community
clinic, and from the agricultural extension office.  
39
Interestingly, both service providers and government officials agreed with the citizens providing these
general community assessments regarding their (service provider and government officials) own
performance. These stakeholders similarly agreed that their offices and associated departments had not
adequately provided services or fully understood the rights and entitlements their departments were
meant to provide before CVA.
One of the reasons indicated for this poor performance in service delivery is that citizens reported feeling
a lack of voice, efficacy, and the means to “speak out” to articulate their individual and community
demands to duty bearers before the advent of CVA. Citizens also reported a lack of willingness on the part
of services providers and officials to listen or otherwise engage them in constructive dialogue. Service
providers also reported a lack of motivation or incentive to their jobs well and in line with departmental
standards and mandates. As a result, in sampled communities for the evaluation, all stakeholders and
groups agreed that services were generally lacking pre‐CVA, with specific reference to the three
departments along with relevant information to citizens or oversight of service providers.  
Having discussed the issue of pre‐CVA experiences of the respondents, the researchers found that there
was an obvious gap between the service providers and service recipients as a matter of the relevant
departments’ failure of “supply” (this general trend is also confirmed by quantitative tracking indicators
reviewed in subsection 6.2 regarding the provision and quality of services at community clinics, WASH,
and agriculture extension.) In addition to services lacking that resulted from a failure of the departments
to adequately supply them, stakeholders reported challenges regarding the attitude and outlook of the
LSPs when community members or individuals would try and engage with those agencies. LSPs were
neither cordial to people seeking services nor serious about fulfilling their duties.
Supply Gap at Community Clinics: A consistent finding relayed across the community meetings and other
stakeholder interviews regarding service provision relates to the failure of the public health department
to adequately supply health services at community clinics across the sampled areas. This included a gap
in service provision of mandated services and governmentstandards, proper conditions and infrastructure
of the clinic itself, and the overall management of community clinics at the community level. Although the
quantitative indicators show that most clinics were technically already established and could be located
as existing physical structures before CVA, at that time, these clinics mostly remained closed or open only
during irregular hours and were not active or activated, while the management and attitude of health
care providers at the clinics was reported to have been seriously lacking.  
Across focus groups and interviews, community members and individuals consistently reported that
before CVA, none of the community clinics in the project areas successfully provided health services up
to the government‐mandated standards. A large majority of Bangladesh understandably must rely on
community clinics to receive medicine to treat common ailments, especially with regards to maternal and
child health. Yet clinics were not successfully supplying these services. As reported by CVA working group
members in a focus group discussion in Kaligonj (#38), medicine that had been given to the community
for a six‐month supply would run out within the first month. Indeed, the evaluation research found
consistent reporting of missing or inadequate medicine provided to, or available at, community clinics.
Current medicines would also frequently run out before the next phase of deliveries arrived. There was
often a shortage of supply of medicines at community clinics, and the amount of medicines was
inadequate. The reporting on the problem of medicinal supply was not solely obtained from community
members; health officials at the Union and Upazila levels confirmed these general trends regarding a lack
of proper medicine delivery before CVA. One aspect of inadequate medicinal supply reported to the
evaluation team was a point made by CVA working group members in Kalingonj (#38), which was that
health officials would often open health supplies and medicine at community clinics without the
community group president present, as per regulation. This allowed medicine to go missing or be
40
otherwise unavailable, thusfailing to instill confidence in the community that the medicine had ever really
arrived or been available for their benefit.
Similarly, community clinics did not fulfil their mandate to properly provide the community members and
areas surrounding the clinics with the requisite health‐education promotion campaigns or information.
Public health officials in Dhaka reported that the government continues to engage, as they have
historically, in various health promotion campaigns, which includes the circulation of new government
policies and circulars along with heavier interventions, such as making vaccines available or partnering
with health or international NGOs on certain health campaigns. Upazila and Union officials reported
knowing of many of these campaigns, but reported having not been given the adequate tools by the
central government to spread the word across the communities. A Food and Sanitation inspector in
Kalingonj mentioned the failure of community clinics to disseminate health education related information
with their area (#26), and many community clinic officials reported their inability to fulfill even the basic
mandates provided in the government circulars along with deficient motivation to seek clarification nor
the tools to do so. Community members report that they rarely knew of these campaigns, what
information they contained, nor what services they were meant to make available.  
Individuals who did try to seek health services at the community clinics before CVA not only reported a
lack of adequate attention to their needs, but given the supply gap, they also reported additional
downstream effects where they and others felt demotivated to return or attempt to continue seeking
services at the community clinic. An Upazila Health Inspector in Kaligonj (#22) reported that the
government’s lack of attention to adequate service delivery generally had demotivated many citizens
from seeking specific services that they would otherwise require, such as in health, because they
understood the community clinics to be a government‐sponsored department. Experience with a lack of
adequate medicinal supplies faced by one person is often reported to others, such that by the advent of
CVA, many citizens had stopped bothering altogether. This same Upazila Health Inspector reported that
community clinics widely lacked available medicine to treat even easy to diagnose and dispense ailments,
such as paracetamol for fevers and oral rehydration salts. The demotivating effect of inadequate care was
echoed by many respondents, which may also explain a lack of motivation to participate in other public
health promotions and campaigns.  
Health seekers also noted the absence of treatment‐friendly environments as well as proper
infrastructure at community clinics. Many community clinics lacked basic infrastructure, like not having a
water tank, safe drinking water, or properly allocated latrines to support the livelihoods of citizens during
visits. This was a concern that was obviously very salient for pregnant mothers and young children as well
as older individuals. Clinics also reported a dearth of electricity and electric fans to support some of the
necessary equipment required for service provision as well as patient comfort. While community clinics
had been built and gazetted in nearly all of the sampled locations prior to NJP, the actual location of some
clinics presented challenges for patients, as for example, in Dacope, where the clinic (#7) lacked an
approach road and was only accessible to the evaluation team following the crossing of a river with a
narrow man‐made bamboo bridge. In other locations, community members reported not knowing about
the location of this clinic given its inaccessibility or lack of outreach to patients in the community. At the
same clinic in Dacope, the staff confirmed that before CVA the facility had a toilet, but lacked watersupply,
and that many people complained about the lack of a proper approach road.
Clinics also did not have possession of the requisite equipment to deliver proper medical care in
accordance with government standards, including machines to provide care to individuals. Clinic health
officials (#10) mentioned that a clinic in Dacope did not have a blood pressure monitoring machine,
diabetic measuring machine, nor a weight‐lifting machine for babies. Community members and clinic staff
41
in all of the sampled clinics reported lacking some if not nearly all of the basics necessary to provide
services up to standards.
Respondents also reported a lack of robust communication between health care service providers at the
community clinic and the community members. Respondents from the community perceived negative
attitudes among community clinic staff in their interactions with community people and other individuals
seeking health services from them before NJP (see case study box).  
Case Study: Lack of Communication between Health Care Providers and Community Members:
One CHCP in Kalinganj (#23) opined that the ‘‘CG was not serious about the problems of the community
clinic. Members of it thought that both the CHCP and community clinics belong to the government. What
can CG do for them?’’ Even the staff responsible for providing health services at community clinics,such
as the CHCP, largely agreed with the communities’ assessments regarding their performance before
CVA. Health officials at clinics in Dacope and Kaligonj agreed that in addition to their failure to provide
adequate services, their attitudes towards health‐seeking patients was not up to standard nor did they
express sincerity for their patients or people in the community. An interview with one CHCP in Kaliganj
(#23) testified to this finding by saying: ‘‘I did not have a feeling that providing service was my
obligation.’’  
Moreover, many clinic staff confirmed that they themselves lacked knowledge as to their roles and
responsibilities, including awareness of the government service standards as established by the line
department along with additional health promotion campaigns. The same CHCP admitted thatshe did not
know the government standards or even what her office schedule should have been regarding the days
and times the clinic was meant to be open. Another CHCP reported that they were not able to maintain
an adequate supply of medicine as the result of their own ignorance about what should be and what was
available.  
Gap with Water and Sanitation: The evaluation similarly noted a consistent gap in service provision
regarding the water and sanitation sector, including deficiencies in proper conditions and infrastructure,
along with gaps in the management of the water and sanitation (WATSAN) committee. In addition this
sector also presented a few unique challenges as compared to community clinics. Whereasthe placement
of community clinics centrally in union wards provides community access to health‐seeking individuals,
many water and sanitation issues reside more locally in individuals’ communities and households. That
being said, the WATSAN committee sits at the Union. Therefore, the performance of duty bearers related
to water and sanitation matter not only for individuals but communities, requiring important linkages
between what goes on in communities with how those interests are represented at the Union level and
vice versa. Moreover, before CVA, the government had previously taken numerous actions to address
water and sanitation issues, including establishing ward WATSAN committees and Union level WATSAN
committees. Yet the evaluation found that many of these structures were not functional or activated.
Moreover, as we explore below, before CVA, to the degree that there was delivery of services in relation
to WATSAN and in water and sanitation, there was some discrimination and bias. For example, the
allocation of tube‐wells and sanitary latrines usually was provided to those who have close relations to
the Union Chairperson and representatives. As a result the lack of proper water and sanitation was felt
acutely by all communities in the evaluation area and to the degree that it functioned, it was an elite‐
dominated process.  
42
Across focus groups and interviews, community members, VDCs, and WATSAN members consistently
reported that before CVA, water and sanitation was not a priority area actioned by relevant line
departments as experienced by people at the local level, despite there having been various government
interventions. Few areas, if any, experienced positive results in their attempts to meet government
standards. Furthermore, Union and Upazila officials seemed otherwise inattentive to local water and
sanitation issues as well as the needs in communities, as reported by CVA working group members. As
summarized by a CVA working group in Dacope (#1), despite the important health, welfare, and livelihood
benefits that water and sanitation provides across communities, it was rarely indicated as any sort of
priority by local government structures, including specifically Unions and Upazilas.  
Problematically, what this means is that community people lacked basic knowledge about safe water and
sanitation practices. VDC members in Dacope/Kailashganj (#5) reported that most of the people in their
community were not aware of important hygienic practices, how to access safe drinking water, nor how
to make latrines hygienic. “Among 100 people, only 10 people have some degree of knowledge about safe
and hygienic sanitation facilities.” Community people in Compaful from the scorecard group members
(#36) similarly reported that area residents lacked proper knowledge of basic hygiene issues regarding
how to properly use the latrine, wash hands, and access safe drinking water. An SAAO officer (#18)
confirmed that access to water and sanitation had been abysmal and that people in their union were not
aware of safe water, therefore having led to an increase in suffering from various diseases. The officer
replied that open defecation was common in their locality, and that a number of floating latrines existed
by bodies of water that area residents used for drinking, cooking, and washing. For example, Bangladeshis
who inhabit coastal areas often use pond water as drinking water, yet pond water is also often subject to
human and farming run‐off that pollutes it. The Union chairperson in Compaful (#32) echoed these same
sentiments about the residents of his union not knowing or having basic hygienic practices before CVA.
(In fact, NJP had numerous non‐CVA interventions regarding improving basic hygiene practices,
particularly around raising awareness and community‐knowledge).
While other aspects of NJP programming had worked to activate VDCs and ward WATSAN committees,
CVA specifically aimed to target the activation and performance of the Union WATSAN committees.
Before CVA, the Union WATSAN committees were inactive or barely functioning acrossthe sampled areas.
In unions in Dacope and Satkhira, the Union chairperson and WATSAN committee members reported that
they had not met before CVA (#14/27). Many of these committees were inactive or not functional and
there were no linkages between the Union WATSAN committees and the ward WATSAN committees, nor
the Union WATSAN committee and other Union‐level officials or departments. A Union chairperson in
Dacope (#15) mentioned that the WATSAN committee had not made any annual plans. Both community
members at WATSAN members confirmed that union WATSAN meetings were not held in most cases, and
not even the committee itself was constituted in some cases.  
Moreover, both community and government officials reported that there was poor communication and
operational coordination between union officials on water and sanitation issues; between union and
upazila officials, and all officials and community people, including the ward WATSAN committees. A
consistent finding was that union WATSAN committees members did not understand their roles or
responsibilities, Union chairpersons could not explain or account for why the committees were not
meeting, and Upazila officials had little knowledge about the state of affairs in their areas regarding water
and sanitation. The Union chairperson in Dacope (#15) relayed that there was no agenda from local ward
WATSAN committees tabled before the Union Watsan committee, and that linkages between union and
ward committees were otherwise non‐existent. (He reports that they now meet twice a month). A Union
chairperson in Satkhira reported that the local WATSAN committee had used resources to distribute
resources to members, who then were tasked with distributing them to people in their communities, all
43
without proper oversight or in accordance with government standards (or proper eligibility). This
chairperson also reported that WATSAN members had lacked any training and had not received any
intervention from the government, including those related to the role of the WATSAN committee, its
members, and government standards on water and sanitation. The CVA working group in South
Sreepur/Kalingonj (#38) reported that a local people’s representative had misappropriated funds
allocated for water and sanitation, and that the community people had been unaware of this nor what
the allocation of funds to water and sanitation should have been.
These reports from the field stand in stark contrast to the government line department’s own reporting
and understanding of the role of WATSAN committees, as reported to the evaluation team in Dhaka. In
an interview with an official at the Department of Public Health Engineering (responsible for overseeing
the government’s water and sanitation policies) in Dhaka, they were unable to say how many WATSAN
committees were in operation, and when pressed, were unable to consult any documentation from the
government as to where this information is collected and reported. Moreover, officials in attendance at
this meeting did not appear to know or understand the role of WATSAN members or the role of the
committee other than to note their role in helping to implement any programming mandated by the line
department. Moreover, they could not say specifically what role WATSAN committees had played in
recent interventions nor what the WATSAN committee’s role was vis‐a‐vis the Union and Upazila
structures. Instead, they said that they relied on the department’s own personnel at the Upazila level to
oversee water and sanitation projects within the upazila as directed by the ministry’s programming,
circulars, and employment policies.
Gap with Agricultural Extension:  Agriculture remains an important sector to support the livelihoods of
all communities in the NJP and CVA areas. Thus, unsurprisingly, many households reported the
importance of robust agricultural extension services. Moreover, the government in Bangladesh plays a
vital role with the provision or oversight of some services in the agriculture sector, such as with feed and
fertilizersubsidies. Agriculture extension officers are meant to sit at the Union and Upazila levels(the sub‐
assistant agriculture officer [SAOO] at Union and Upazila Agriculture Officer [UAO] at Upazila), but engage
with community people in the communities they serve. While agricultural extension offices were already
established at the union and upazila levels before CVA, CVA groups, community people, and officials
themselves reported a lack of robust service provision by extension offices. Officials were frequently
unavailable or not at work and there was poor coordination with other relevant local government units.
As a result, there were significant gaps in the supply of agricultural extension services not only at the
Union level but also among services aimed directly at farmers in their communities.
Respondentsreported that before CVA, agriculture extension services were lacking across all of the locales
sampled for the evaluation. In particular, as reported by a VDC in Kailashganj (#5), many farmers in the
area were unaware of the correct application of seeds and fertilizers, and as a result, often used lower
quality inputs that led to inefficiencies in production levels.  
One reason for this wasthat in all of the Unions sampled for the evaluation, various respondentsreported
that the agriculture officer was often neither present nor informed about their roles and responsibilities
before CVA. Community people, Union and Upazila level officials, and department officials themselves all
admitted that before the advent of CVA, sub‐assistant agriculture extension officers were rarely in the
office during the mandated time period, they often failed to perform their duties and do outreach/visits
to farming households in the community, and were otherwise not engaged in the provision of linkages
from the government service standards to service recipients. For example, the scorecard and women's
group (#36) in Chompaful union reported that the SAAO was not cooperative and that community people
did not know who he was or whatservices were available for them at the office of agriculture in the union.
44
Similarly, a focus group with a CVA working group in South Sreepur/Kalingonj (#38) said that community
people did not know the SAAO, hardly ever visited the union officer, and were in fact afraid to visit. They
also reported that from their perspective, local service providers did not appear to know about
government circulars. In general, the lack of necessary human resources at the Union level meant not only
that community members were demotivated to seek agriculture extension services at the Union level, but
also that they were not provided services in their locales.
Government officials at the Union and Upazila levels confirmed this lack of coordination. First, there was
a lack of really robust and realized coordination among line department officials, Upazila officials, and
their Union‐level officials. Occasionally, officials reported some coordination among the staff of line
departments, but agreed that there was a lack of coordination with the SAOO and Union Council
members, as well as a lack of monitoring from Upazila officials. Moreover, there was a lack of coordination
with the Union‐level officials and other Union officials, such as the elected Chairperson and people’s
representatives. Union representatives, including the Chairperson and council members, reported that
they often did not know the work of the extension officer or whether or not they were in attendance.
They also did not make an effort to frequently coordinate the other activities of the Union with extension
work. Overall, this lack of coordination demotivated both extension workers from properly doing their
jobs and community people from seeking agriculture services.
One of the downstream effects of these challenges in agriculture extension involved not only government
officials and farmers, but the localservices providers who were meant to be monitored by the government
line department, including the Union official. In the areas sampled, it was often reported that LSPs could
take advantage of price gouging, which included failings to post the government mandated price for
subsidized seeds and fertilizers. As a result, citizens felt disempowered to try to report on any problems
with LSPs since they could not find the requisite officials in attendance at the union or upazila level and
were otherwise unable to contact them.
Finding #2: a gap in services did not just result from problems of supply but also as a “demand” side
problem on the part of communities and people lacking knowledge; community members were not
aware of their rights and entitlements with regards to seeking services from the government
departments; they also lacked a platform to articulate demands, and dialogue and interface with duty
bearers.
Having discussed the issue of pre‐CVA experiences of the respondents, the evaluation found that there
was an obvious gap between the service providers and service recipients as the result of “demand” side
problems of communities and individuals, which lacked the proper means to articulate their needs and
desires regarding services. As indicated above, part of this problem no doubt arose from a lack of
knowledge about the proper practices, following from a lack of education on the part of citizens. This
appears to have been especially true in water and sanitation, which includes deficiencies in knowledge
regarding proper latrine practices and water usage, as mentioned. But this also involved a lack of
knowledge on the part of citizens about their rights and entitlements. Community people repeated the
importance of not having processes in place to gain information nor express these problems: not only
problems of performance in service delivery, but also a lack of knowledge as to duties and entitlements
along with what role they could take on as civil society to voice their concerns and activate a government
response. Respondents also reported challenges regarding the attitude and service delivery outlook of
the LSPs as hindrances to demand articulation.
45
Lack of knowledge of citizens of rights and entitlement: Uniformly, citizens and other stakeholders
reported that there was a general lack of knowledge among citizens about the services available to them
by the government departments. Additional concern was added with respect to information regarding
services specifically available to be sought at community clinics, such as those on water and sanitation
and in agriculture extension. This finding held across all of the sampled areas. As mentioned above, these
services were not actually adequately available, but even so, a consistent finding in each location was that
citizens were not informed as to their rights and entitlements with regards to what care health‐seeking
patients should expect to receive at community clinics, what communities should do regarding water and
sanitation, and what farmers could expect with respect to extension services. A member of the CVA
working group (#5) in Kailasgonj summarized this point:
“Among 100 people, only 10 people have some degree of knowledge about safe and
hygienic sanitation facilities. People mostly relied on private service providers and quacks.
They did not know that SAAOs are supposed to sit at the Union Parishad level. Most of the
tenants were not aware of the application of quality seed and fertilizers. Use of lower
quality seeds and fertilizers led to a loss [in productivity]. Apart from these, they did not
have the courage to meet with government officials, including the Union chairman.”  
A lack of knowledge regarding the servicesthatshould have been available may be associated with lacking
knowledge as to rights and entitlements and the services due to them in accordance with government
department standards, which were reported to be basically unknown among all the respondents in the
project area. A CVA working group member in Dacope (#1) stated that:
“Before the commencement of CVA activities, community members were not aware of
their rights and entitlements from the Union Parishad or Upazila Parishad in general and
three departments in particular. That is why services were not considered as rights. Thus,
whatever services were made available to use were welcomed but the quality of service
delivery was very poor.”  
The reasons for lack of knowledge about rights, entitlements, and services available are manifold. First,
citizens report that they personally, along with other community members, had feelings of fear to visit
local service providers. As mentioned above, this relates to the perceived attitudes of service providers ‐‐ 
which was often not welcoming, friendly, or sincere. Many service providers did not dispute these claims
of the perceptions of community members, but instead, confirmed them and expressed understanding as
to why this was the case before CVA. For example, a CHCP in Shaktira, Kaliganj (#21) admitted that she
herself was often not sincere in her treatment of patients at the community clinic, and that she felt a lack
of motivation to improve her behavior. Moreover, community people themselves admitted that their own
feelings of disempowerment resulted in them failing to articulate their demands and learn their rights.  
Second, community people were often unaware of the placement, location, or existence of service
provision locations, such as their offices, or where they reside, as mentioned regarding community clinics
above. The remoteness of the clinic facility in Dacope mentioned above (across a body of water and
without a proper feeder road), meant that patients were often not aware of its existence. As WATSAN
committees did not meet, citizens were not made aware. Additionally, because the agriculture officer was
not available at the Union, citizens remained unaware.  
Finding #3: A lack of accountability in service provision exacerbated the problems ofsupply and demand
pre‐CVA; without robust monitoring and evaluation from institutional actors and the opportunity for
community people to demand services or inspire accountability, both institutional and social forms of
accountability were not present.
46
Another challenge in the pre‐CVA environment relayed by respondents involved a lack of accountability
from the perspective of CVA working group and community members regarding government agents,
service providers, and duty bearers. Before CVA, citizens and communitieslacked a mechanism to express
their frustration where service provision was a problem. As such, there was no real ability to articulate
demands nor a process to do so. That is, they lacked a platform like scorecards that would have provided
a feedback mechanism to rate service providers. Therefore, there was no ability to transpose voice into
action, no coordination across committees and other units, and no ability to dialogue with service
providers.
  
Moreover, research revealed a lack of robust interaction or linkages between service users and suppliers,
nor coordination with officials, to fulfill their own monitoring and accountability obligations. Above, we
document how this played out often with WATSAN committees and other Union‐level officials. Responses
indicate that these problems of access and linkages then prevented the participation of citizens in the
development process and governance of service delivery in these sectors. For example, citizens often
reported that they did not have scope for being part of the beneficiary selection process, and that the
WATSAN funds were spent on beneficiaries aligned with committee members. In a few instances,
respondents did mention that they had good relationships with some Union and Upazila officials, but this
was typically when they knew them personally.
Another consistent finding relates to the lack of accountability pre‐CVA due to limits in the ability and
motivation among public officials to talk to or work cohesively together, which included holding
themselves accountable to each other. We explore this further in the discussion of “Interagency” in the
subsection below, yet it is worth reiterating here that few linkages existed across local government units
or in conjunction with the department offices. Responses also indicated that significant bureaucratic
inertia existed, which prevented duty bearers from performing their duties. As a result, in concert with
the findings above, a lack of accountability for health officials meant that health services were not
adequately provided at community clinics; a lack of accountability in the provision of water and sanitation
meant that the WATSAN committee did not function and hygiene was poor; and a lack of accountability
for union and upazila agriculture extension offices meant that farmers were not able to access extension‐
related services and that LSPs had little oversight.
6.4 CVA Formation and Activities
After reviewing the baseline conditions of service provision, citizens’ voice and action, and accountability
prior to the advent of CVA, as per the discussion in the previous subsection, we next track the
establishment and activities of CVA. The evaluation inquired asto CVA working groupsformation and their
launching activities. (It is important to remember that the data collection for the evaluation only occurred
in sampled areas with established CVA working groups.)
Summary of Findings on CVA Establishment and Activities:
● CVA working groups successfully completed their formation and recruitment of members who  
represented broad groups in society.
● Community members’ information about CVA and CVA activities was mixed.
47
● After the formation of CVA working groups, key CVA activities have included scorecard sessions,
interface & dialogue meetings, the development of action plans, and follow‐up; these activities
were met with robust frequency and action across all of the sampled areas.
Finding #1: CVA working groups successfully completed their formation and recruitment of members
who represented broad groups in society.
NJP facilitated the empanelment of VDCs in 2017 followed by CVA working groups starting in 2018, which
began functioning thereafter. However, not all CVAs were formed at the same time. There were slight
differences in the commencements of CVA in different working areas, though they had all been formed
by 2018. After forming, the CVA working groups began their activities, starting with scorecard sessions
focused on community clinics, before moving to water and sanitation and agriculture extension in 2018
and 2019. That these groups were established, made functional, and continue to work is also confirmed
by the quantitative indicators above.
The evaluation found that CVA working groups both represented broad groups in society and were
inspired to join CVA for a variety of reasons (see also subsection 6.7 below). Members responded that
they joined CVA (the working groups as well as the process) because they were inspired by both an
intrinsic desire to help and improve themselves, as well as extrinsic desires to mobilize to improve the
community. Members of working groups report that they joined for different reasons (and drawn from,
or previously served on, VDCs). This is confirmed by results from the VDC  focus groups  ‐‐ some VDC
members knew about CVA and expressed desire to “graduate” to there. The working groups had good
gender and age balance, and all attendees were willing to talk across the numerous focus groups during
data collection.
Many working group members mentioned social and other‐regarding motives for joining CVA, such as
helping to develop the community. As revealed in the inception visit, In Kalinganj, the head of the CVA
working group, a village elder and former teacher, provided inspiration to the other members to join as
they wanted to live by his example. Members mention more personal motivations as well. One female
member in Dacope viewed membership as a good stepping stone to raise her visibility as a local notable
with the hope that one day she may be able to run for local office (Union parishad council). Male and
female youth members in Kalinganj mention the opportunities CVA provides youths to get training,
capacity building, and improved learning regarding the community as an aspect of their education and
future career plans.  
CVA working groups had a positive gender balance and representation. This was not only observed by the
evaluation team during data collection, but was noted by other actors involved in the CVA process. For
example, a health inspector in Dacope (#8) noted regarding women’sinvolvement that “I found more than
50% of the interface meeting participants were women.” Community members further commented on
women’s participation (#1): “Women did notspeak up much earlier. Now they can speak a lot. The courage
and information that they have now due to CVA have created scope for them to be a member of various
committees.”
The frequency of meetings of the CVA working groups is heterogeneous across the sampled areas, which
tended to be driven by programmatic needs and a desire to meet rather than due to directives from NJP.
For example, among the selected CVAs, two conducted their meetings once a month, one sat together
48
quarterly, and one sat together twice a year. Groups reported that they felt asthough they met as needed
and did not think that they met either too few times or too often.
Finding #2: community members’ information about CVA and CVA activities was mixed.
An important finding is that the majority of the CVA members expressed robust understanding of the
different activities of CVA ‐‐ that is, it did not appear that the CVA “existed on paper” or “in name only,”
but that it had real participation of, and contributions from, its associated members. CVA working group
members understood CVA programming in NJP, their roles as individuals and as a working group
committee, and their purview for action and oversight of duty bearers, government actors, and as a
communication relay between local communities and relevant institutional actors. Moreover, all CVA
working group members expressed a desire to continue to work with CVA and were happy to hear of the
NJP extension, a subject we return to in Sections 7 and 9. They additionally provided important insights
and lessons for sustainability as well as possibilities for scaling (discussed in subsection 6.11). In these
respects, the NJP programming aspects of CVA were well‐implemented and understood by the CVA
working group members. CVA working groups’ knowledge of CVA and NJP was attributed acrossthe board
to robust programming, oversight, and organization by NJP and partner NGOs.
However, there was heterogeneity in the degree to which other members of the communities and
committees had awareness of CVA members and CVA activities. One the one hand, during data collection,
some community members reported that they knew about CVA working groups (who were not CVA
members), particularly in the scorecard and VDC sessions. But this was not uniform, and many
respondents in the VDC sessions reported not knowing about CVA. For example, although CVA working
group members were drawn from VDCs as per NJP programming, when meeting with the VDC members
for the evaluation only some reported having knowledge of the CVA working group activities. Scorecard
group members also appeared to have only limited understanding of the CVA process beyond their direct
participation in the scorecard meetings.  
In the case of WATSAN committee members, who were also members of the CVA working group, or Union
officials, who were otherwise engaged in or linked with NJP programming, there was more awareness of
CVA activities. In Kailashganjj and South Shreepur WATSAN committee (#27), participantsreported having
a well‐informed understanding of CVA activities in their area. But often, knowledge of CVA activities were
mostly limited to those who were directly involved. A partner NGO staff member (#20) summarized this
point well:   
“We did not have a previous understanding of CVA. At first, it takes time to understand
the dynamics of CVA activities. Secondly, we mostly discussed the CVA related issues with
CVA members. Understanding the core values of the CVA process took most of the time of
the project. We had very limited time to transfer our ideas, including the understating of
CVA group members.”      
It seems that where CVA members were more directly involved in other groups or structures they were
better able to disseminate information about CVA activities. Unsurprisingly ‐‐ but nonetheless positively ‐
‐ CVA activities were known to local service providers, particularly for those of whom their work was most
intimately linked with specific CVA activities, including the CHCPs, WATSAN committees, and SAAOs.
Moreover, UP Chairpersons, Upazila, and Upazila‐level officers were aware of CVA activities generally, but
demonstrated less comprehensive awareness regarding the CVA activities across the three sectors. These
officials were most likely to be aware of interface and dialogue meetings, as well as NJP activities more
49
broadly, than they were of CVA. Where there was a more limited level of understanding of CVA activities,
stakeholders were less likely to know about the processes and services provided from CVA.  
Finding #3: after the formation of CVA working groups, key CVA activities included scorecard sessions,
interface & dialogue meetings, the development of action plans, and follow‐up; these activities were
met with robust frequency and action across all the sampled areas.
The evaluation found that CVA activities, as reported by the working groups and other relevant
stakeholders, adhered to the NJP programming guidelines. For all groups in all areas, this included
scorecard sessions on community clinics, water and sanitation, and agriculture extension; interface and
dialogue meetings; and the development of action plans and follow‐up. These reports given by
participants in the qualitative fieldwork component of data collection confirm the quantitative indicators
reported in the CVA database.
Scorecard sessions: An important activity reported by all groups in the evaluation revealed that they had
all managed to hold scorecard sessions with community members per NJP programming related to the
government department’s performance on service provision in the three development sectors. They
started with community clinics before then first moving to agriculture extension prior to WATSAN.
Respondents reported that these sessions not only helped to establish the baseline condition of services
in communities, but also helped to give community people a voice and platform for demand articulation
to duty bearers to improve service delivery. The quantitative tracking indicators above confirm that these
scorecard sessions were held for all three sectors across all of the project area.
Interface and dialogue: A second noteworthy activity of the CVA working groups also reported by all
sampled areas involved the interface and dialogue meetings. Such meetings are critical to the success of
CVA and designed to bring community people and duty bearers into a shared platform to address both
the demand and supply side aspects of services and citizens’ voice and action. The evaluation found that
these interface meetings consistently adhered with the guidelines per the NJP programming, with CG
group members, health offices, CHCP, and VDC members participating when relevant while being
organized by NJP staff and partner NGOs. These interface meetings were held at the behest of, and with
the active support of, CVA working group members. In fact, among the different stages of CVA
implementation, the interface meeting has been repeatedly mentioned as the most influential stage of
the process from the perspective of the working group members. The quantitative tracking indicators
above confirm that interface and dialogue sessions were held for all project areas.
These meetings served numerous multiplicative purposes not only in design, but in effect as reported to
the evaluation team. First, the interface meetings worked as an ice breaker and “meet‐and‐greet”
between members of the community, the CVA working group members, and duty bearers. Given the lack
of knowledge about rights, entitlements, and available services by the government departments, thisstep
was a simple, yet important aspect of CVA programming that helped to inform community people about
who was responsible for attending their needs. While some local service providers were better known,
higher level officials, including at union and upazila levels, were less so. Moreover, many officials even
within the same administrative unit, such as the Union or Upazila, often either did not know about each
other or had weak linkages with officials and service providers at the community level.
Second, the interface and dialogue meetings provided an avenue for communication and dialogue where
both the service recipients and service providers were able to interact with each other on different aspects
of these services. Respondents reported that interface meetings played a key role in opening up avenues
for citizens to express their needs and demands.  
50
Third, the interface meetings provided a forum to bring these various groups together in a situation
through which their actions could be mutually supportive while encouraging them to seek support from
each other. This helped with accountability while also keeping everyone focused on next steps and
possible actions. It also acted as a place of resource mobilization for the community clinic and WASH‐
related services. Most of the relevant officers and politicians either expressed that they had participated
in interface meetings or that they had known about the interface meetings and had representatives
attend to report back. While CVA working groups mostly focus activities at the ward and Union levels, we
discuss how interface and dialogue “trickled‐up” to Upazila, divisional, and national levels in Section 7.
Interestingly, these aspects of interface and dialogue did not just affect the performance of the CVA
working groups, but also allowed the CVA groups to provide support to other individuals, institutions, and
structures. Community people, CHCPs, and SAAOs all reported seeking support from CVA members.
During the discussion with a VDC group, community people reported asking the CVA members to stop a
child marriage while community people failed to convince the guardians of a daughter they sought help
form CVA members.  According to a VDC member (#5),   
“Parents and relatives of a girl of Horintana village tried to get her married at an age
younger than 13. At first, a number of VDC members tried to stop the marriage, but the
parents of the girl were adamant about the marriage. Having found no other alternatives,
they informed the case to one of the CVA members, who is also the Chairman of the VDC.
He met her parents and asked them to stop the marriage. He informed them that if they
do not stop the marriage, he would inform other CVA members and an Upazila‐level
officer. And finally, the parents changed their decision. The girl is now studying.”  
Many CHCPs reported having received assistance from CVA groups and that they had felt
comfortable seeking help from them. One CHCP (#25) noted that,  
“Sometimes I asked CVA groups to disseminate information on health‐related issues to
community people. I found them reliable in this regard because they are very active to
improve the quality of lives of the community people and they want the improvement of
the community clinic.”    
Most of the SSAOs also noted that they seek help from CVA members to identify the proper beneficiaries
for agriculture extension services. An SAAO in Compaful (#33) described his experience: “Recently, I asked
the CVA members to provide a list of 76 farmers who will receive the benefits of [government] subsidies.
I rely on their list because they are not politicians and most of them work for the poor people.” This
suggests that the CVA, through the implementation of different stages, has brought about changes in the
behavior of the service recipients and providers as they were able to sensitize both groups.  
Action Plans and Follow‐up: A third important CVA activity involves the development of action plans and
follow‐up. The CVA working groups report that they are making important progress on developing,
implementing, and following‐up on action plans. They have listed areas to fix, delegated who is
responsible for doing so, and plan to monitor. The duty bearers confirmed this as well, and all sides
appreciate the process. (These are being tracked in the CVA database at the project level, as discussed in
subsection 6.2).  
6.5 CVA Process
One of the most important aspects of CVA involves the “process” component to social accountability,
which goes beyond its outcomes. By process, what we mean is that CVA itself provides a platform to
engage citizens to inform them of rights and entitlements, allow “voice” as well as action, and dialogue
51
and interface between citizens and community members on one hand and government officials and duty
bearers on the other. The CVA platform facilitates, manages, and directssocial accountability “processes.”
From the perspective of this evaluation, we might considerthese processes as analogousto “mechanisms”
that link service seekers to service providers. Yet, understandably, evaluating processes/mechanisms is
often difficult as they are intangible (and therefore difficult to measure). It is thus often hard to directly
attribute to them a specific impact. Even when following guidelines in NJP’s programming, the CVA
process can work differently in different places among different actors, often resulting in either
observationally similar or dissimilar outcomes.
Given these limitations, we attempt to evaluate different aspects of the CVA process as reported by
stakeholders to try to uncover as much information as we can about what mechanisms the CVA platform
provided and in what ways it operated (before turning to outcomes in subsequent subsections below).
We therefore inquired from the CVA working groups their views regarding various aspects of the CVA
process, including perceptions of whether and how it provided new information and informational flows;
opportunities to express voice and dialogue; and the ability to articulate demands for services and
improved accountability. We also asked officials and service providers involved in community clinics,
water and sanitation, and agriculture extension their views regarding the CVA process in addition to
relevant government officials at the Union and Upazila levels.
Summary of Results on the CVA Process:
● Activating civil society and social networks was an important part of the CVA process.
● Information dissemination was an important part of the CVA process.
● The CVA process succeeded at providing a mechanism for citizen “voice” through formal and
informal channels.
● The CVA process succeeded at improving services by providing mechanisms of accountability
through “voice.”  
Finding #1: Activating civil society and social networks was an important part of the CVA process.
The first important aspect of evaluating CVA as a process or mechanism arrived from its ability to activate
civil society and social networks in community areas through the mobilization and activities of the CVA
working group (reviewed above). While this is an explicit goal of CVA “on paper,” the evaluation
documented its realized success across the sampled areas. The evaluation found that in many ways  the
potential for civil society activation was already present in Bangladesh, while the strength of social
networks was already extant in many communities. However, what was lacking was the energy to truly
use that potential to be transformed into activation by sparking voice and action indicative of robust civil
society. That energy had laid dormant and lacked a platform or mechanism to organize itself and express
interest, and, as some key informant interviews revealed, the historical success of civil society in
Bangladesh had not come from the activation of social networks as such, but rather the success of non‐
governmental organizations in the area of direct service provision. Regardless, the evaluation notes
specifically, both in the sample communities and overall, that the formation of the CVA working group
served as an important ignition point to help overcome inertia on the part of civil society.
52
This civil society activation is an important part of the CVA process, which, despite its difficulty in
measuring, provided the mechanism and initial spark for CVA at its outset. This is important not only to
assess the value of CVA, but also to recognize that despite this dormant social energy, without a tool to
ignite it, it will remain dormant. Civil society in Bangladesh has otherwise not seen this activation, and in
fact, the government is growing even more leery of civil society in general. CSOs have been successful in
the past, like BRAC and Grameen Bank, but only where these CSOs have acted as direct service providers.
Instead, respondents reported that rather than providing services directly, CVA allowed the “spark”
necessary for citizens to express their demands to service providers and for service providers to better
understand the needs of citizens.  
Finding #2: Information dissemination was an important part of the CVA process.
Once formed, the CVA process is built around informing the CVA working group such that the working
group can then inform the community about their rights and entitlements (also see above). The sampled
communities for the evaluation all reported that this was an important “process” aspect, working to
provide mechanismsthat transfer information to people, and vice versa. For what it is worth, even though
the government had already provided rules and legislation to do things like create community clinics,
empanel WATSAN, and deploy sub‐assistant agricultural officer (SAOO), the government itself had rarely
done the equivalent of civic education to inform the citizens about what was available to them. In fact, in
meetings with line department ministry representatives in Dhaka, across the three sectors, central
government officials could often state what was supposed to exist on “paper,” but uniformly rarely had
any sense of what happened on the ground in terms of reality and what Bangladeshi citizens knew about
their rights and entitlements, as measured by any of these ministries. Instead, CVA working group
members reported that CVA helped to fill this role, not by creating new rights or entitlements, but by
simply “reminding” or “informing” people of what they can demand from the government in ways that
the government has otherwise not been compelled to do (that is, by helping people to build their agency).  
As evidence regarding how their involvement with CVA activities informed them regarding their rights,
service recipients from a scorecard session and women’s group in Champaphool Union, Kaligonj (#36)
summarized the following:     
“After working with the CVA process, they have been able to come to know about different
services available for them at the community clinic and SAAO office. Moreover, while they
visited these offices as a part of CVA activities, they found that there are different types of
posters which have been hung in those offices. They have also been aware of the duties of
respective offices through those posters. In addition, community people have been able to
realize that the community clinic is for them and taking care of it is their responsibility.”  
CVA working groups also provided information to service recipients and service providers. One CVA
member in Shreepur, Kaligonj (#38) described their information dissemination activity the following way:
“CVA group members disseminate authentic information, which has been written regarding government
policies, to the people.” Numerous individuals reported having received information after their
involvement in CVA activities, with specific reference to the reporting arising from the scorecard sessions
in the interface and dialogue meetings. This information transfer likely occurred because of direct input
from CVA since CVA working group members, NJP partner staff, and NGO facilitators worked to collect
service‐related policy documents, including laws, monitoring standards, rules, circulars, and other
announcements as part of CVA programming. But citizens did not usually have this information before
CVA.
53
Therefore, the information transmission mechanism is a very important component of the CVA process.
CVA does not only help with information transmission from duty bearers to citizens, it also improves the
bidirectional nature of information flows by providing direct reporting from communities to duty bearers.
As a part of this organization, NJP and CVA working groups played a key role disseminating information
within a bidirectional flow between service providers and recipients, particularly during the monitoring
standard meetings. As one of the NGO partner staffs reported from Kaligonj, Satkhira:   
“Community level health service providers do not have the proper information regarding
the service delivery process and standards. They do not collect information relying on
authentic documents such as policies, laws, rules, and guidelines. Sometimes, they relied
on the direction of Upazila and District‐level supervisors and their fellow colleagues.  We,
along with CVA, requested them to conduct their activities as per the policy guidelines.
Sometimes CVA provided or showed them the policy documents service standards,
including the related policy documents.”  
As a result of this bidirectional flow of information dissemination, the CVA process played a key role in
improving the knowledge of common people regarding community clinics, water and sanitation, and
agriculture extension. Now people not only know about the existence and location of community clinics,
but also know about the standards of a community clinic and the treatment that they have a right to
receive there. Expressing awareness of the services available at their community clinic, a CVA working
group member in Dacope (#1) said:  
“Information that has been received through the CVA process played a key role in
changing our level of knowledge and it influences our lives in different ways.  Now we are
aware of the functions of a community clinic, the types of medicine available for us in a
community clinic, its schedule, and so forth.”  
Similar findings pertain to services available in agriculture extension. One Union chairperson (#30) noted,  
“Now he (SAAO) comes to the office four days a week. Previously, community people were
less informed about the activities of the SAAO. Now they are able to know whatever they
need to know from the SAAO, either visiting him at the union parishad or making a phone
call to him. His mobile numberis available at the union parishad and all agriculture‐related
grocery shops.”    
Another important aspect of improved information flows involved those between government officers at
local, union, and upazila levels, including the SAAO and CHCP, who used the network activation of CVA to
disseminate their information on programming. In many cases, the statement of the Union Parishad
Chairperson expressed that the cell phone number of the SAAOs has been made available to them, and
that the SAAO and farmers have begun to communicate with each other. However, many unions noted
that having the SAAO more consistently sitting in his/her office at the Union Parishad also broadened the
opportunity for community people to contact them. This arrangement played a significant role in
improving the service delivery related knowledge of farmers, including regarding the prices and subsidies
for seeds and fertilizers as well as government‐sponsored promotional campaigns.  
Agriculture extension officers also attended CVA meetings directly to disseminate information to CVA
members and the community people in attendance, where they also encouraged CVA members to spread
the word to the community afterwards through their social networks. In fact, the development and
activation of networks among community people due to their involvement of CVA is one of the
evaluation’s most important findings, as echoed in many examples we heard from SAAOs. One SAAO (#18)
expressed his information dissemination process in the following way:
54
“Now, I am working with many people, and VDC/CVA played a key role in establishing a
network between me and the service recipients.  Because of this sort of connection, they
communicated with me over the phone and I advised them promptly. I transmit different
information through their CVA/VDC meetings.”
CVA also helped by serving to link  information to government officials, service providers, and other duty
bearers. For instance, the CVA working group in Kailashganj collected Union Parishad related information
to prepare a booklet entitled “UP At A Glance,” a document that played a key role in identifying the
problems in that area. It was then passed along to the Union Chairperson (#19).  
CVA played an additional role in helping to improve information channels for local communities. For
example, in disseminating information related to water and sanitation, the CVA working group, along with
the VDC, played a key role in Kailashgonj (#4/5), where both groups monitored a hanging latrine that had
a connection to rivers and other bodies of water. Both in Kolashgonj/Dacope and Shreepur, VDC members
under the leadership of CVA members occasionally distributed soap to poor people and students in their
areas. For instance, a WATSAN Committee in Kailashgonj (#17) distributed 150 soaps to primary school
children to improve their hygiene practices.  
Finding #3: The CVA process succeeded at providing an important mechanism for providing citizen
“voice” through formal and informal channels.  
A third important aspect of the CVA processinvolves providing community people with a platform in which
“voice” is not only allowed, encouraged, and activated, but also given real meaning of dialogue and
evidence that duty bearers are listening and changing behavior. That is, voice that gets expressed and
heard. Dialogue and interface meetings were the main venues for community interests to be expressed.
Views of one of the CVA members in Kailashgonj (#4) reflect the importance of these practices of voice‐
raising and interaction during interface meetings:  
“We organized an interface meeting where people expressed their demands. Interface
meetings are a platform where people interact with service providers. This kind of
interaction played a key role in improving the quality of services of the community clinic,
water and sanitation, and the agriculture department.”    
Most of the government officials and politicians interviewed report having participated in interface
meetings directly. They also report that such meetings allowed community people to raise their voice,
which was often appreciated by officials. One UHFWO in Dacope (#6) noted,
“CVA is doing good jobs in sensitizing the community people to get services from a
community clinic. We rarely discussed with the CG group and common people [before
CVA], interface meetings open up avenues for this kind of discussion.”   
It was apparent from all manner of respondents that interface meetings and dialogue sessions between
service providers and service recipients brought them togetherin a single platform – this was a very critical
way that CVA provided new processes and mechanismsto facilitate exchange of communications. Though
initially some service providers demonstrated some negative attitudes having to attend interface
meetings, after a few sessions they report having become sensitized and interested in expressing their
views while also remaining respectful to the voice of the CVA and community members. As reported by
an Upazila Chairperson in Kaligonj, Satkhira (#28):  
“CVA process has reduced the gap between citizens and their representatives. Elected
representatives listen to citizens now very carefully. Working with the CVA process has
made people courageous and informative; that in turn has led them to have face‐to‐face
55
interactions with their representatives about things that the Upazila Chairperson controls,
and have a significant influence on their lives.”
In addition to leveraging the formal aspects of the CVA processto find mechanismsfor expressing “voice,”
we note numerous examples of individuals and CVA members reporting other ways that they either
individually or as a group discussed a variety of issues with service providers to express their needs above
and beyond formal CVA mechanisms. For example, they submitted letters or applications to Upazila or
district level officers. One success of CVA and interface meetings is that government officials and elected
representatives did not consider CVA and its activities as a potential threat to their authority, but rather
grew to see them as partners in the development process. The following Box shows an example from a
Community Clinic in Satkhira:  
Case Study: Submission of Application to Improve Infrastructure at Community Clinic
The community group of South Sreepur community clinic (#21) submitted an application to the Union
and Chair of the Upazila Parishad for constructing a new building and wall for their community clinic.  
But they were initially skeptical because they thought that since this type of work is done by the
government, nothing will happen from their initiative. However, after getting involved with CVA, they
have come to believe that they can place such a demand either verbally or in written format to submit
to relevant officials at the upazila level.  
Many respondents discussed the improvement of “voice”, both generally and in other aspects of their
lives. As mentioned above, officials and working group members in Dacope noted the important ways in
which CVA process had improved women’s representation and amplified their voices. Further, during the
formative research visit, women in CVA working groups mentioned that they felt a stronger voice to
engage with non‐CVA related activities in their homes, schools, workplaces, communities, and with
government officials. In these ways, we note positive spillovers from CVA into non‐CVA processes. CVA
process promoted women’s participation at the community level and within local government level
activities. We discuss women’s empowerment further below in subsection 6.7.
Additionally, CVA membersreported thatsuch spillovers also occurred by bleeding into the CVA members’
work in other areas, groups, and committees, such as within the ward shava and ward budget meetings,
which are formally constituted by the government to encourage people’s participation in local
governance. In this regard, one of the Panel Chairman of Koilashgonj Union Parishad (#19) said that VDC
and CVA membersinfluence the ward shava and budget meetings. He added that if the people participate
in these meetings they will be more aware and they will be more capable of ensuring their rights are
realized. Parallel structures like CG, WATSAN, Ward Shava and Budget meetings were found to be more
functional when compared to past meetings. CG members now organize meetings regularly with the
support of NJP staff (#10). A similar situation has been found in the case of WATSAN committees, as
evidenced the following example:    
Case Study: Function of WATSAN Committee (FGD, Union Watsan Committee, #14)
“Before the CVA phase, meetings of the WATSAN committee was not held at all. Union WATSAN
committee did not make an annual plan. No agenda from Ward WATSAN was placed before the meeting
of the Union WATSAN. In fact, there was no link between Ward WATSAN and Union WATSAN. Now
56
Union WATSAN meets twice a month and makes discussion on agenda prepared. Until now, 9 scorecard
sessions have been conducted. They participated in interface meeting, made an action plan, and
reviewed follow‐up of their plan. Ward WATSAN now identifies problems of community and placesthose
in meeting of Union WATSAN committee.”
All told, these results suggest that a positive finding is not just regarding CVA directly making things
happen through programming, but also directly and indirectly activating and motivating other actors,
institutions, and structures (like pre‐existing committees and groups) and then bolstering them and their
efforts. How this “trickles up” to meso and national levels is a subject we return to in Section 7.
Finding #4: the CVA processsucceeded at improving services by providing mechanisms of accountability
through “voice.”
A fourth finding regarding the process of CVA is that it played a role in activating the committees and
platforms that then provided mechanisms of accountability and linkages to improve services in the three
sectors of community clinics, water and sanitation, and agriculture extension. That is, “voice” was not just
important for providing voice, but also by directly linking voice to improve the service delivery process.
More specifically, this occurred as voice was employed to make service providers more accountable.   
A first way to see this is reflected in the fact that CVA activities established a network among the service
providers and service recipients, which some service providers are using to targetservice provision at real
beneficiaries. A statement of an SAAO (#33) notes that,  
“I have good communication with CVA members and use this network at different times.
For instance, I, along with the CVA members, request local service providers to
demonstrate the price list/red cloth (lalshalu). It has an impact on ensuring fair prices.”  
A second way to view this impact manifests from CVA’s mobilization of resources. By mobilizing resources
through interface meetings, CVA improves the situation of community clinic and water and sanitation‐
related issues. During the interface meeting in Kalingonj (#24), people provided a glucometer, but
questioned who would provide the necessary strips. The CG group took the decision that everyone who
set to take the service of diabetics test would pay 20 takas for the strips. They agreed to give 20 taka
because at the nearest private hospital, service providers would take 60 taka (in addition, people would
also have to pay transportation costs). A Union  member also donated a blood pressure machine to the
community clinics in Dhanopoti, Dacope as articulated in the demands of community people during their
interface meeting.  
A third way this outcome is realized is through the monitoring of the activities of local service providers.
A Food and Sanitation Inspector in Kaligonj, Satkhira (#26) described the monitoring activities of CVA:
“there was no action plan and community level monitoring earlier. However, NJP and CVA have started
helping to formulate the action plan now and they have been monitoring the activities of the CHCP.
Because of their monitoring activities, the CHCP has become alert, gives more time, and has started to
open the clinic on time.” One SAAO (#33) expressed how CVA members help to monitor the activities of
LSPs, “I, along with the CVA members, request the LSPs to demonstrate the price list/red cloth (lalshalu).
It has an impact on ensuring fair price.” According to one respondent, “CVA members along with Union
parishad members, have destroyed the hanging latrine adjacent to river and water bodies.” A WATSAN
committee in South Sreepur/Sathirka (#27) identified genuine beneficiaries for logistic support provided
by NGOs working in the locality. Such support includesrings and slabs used for the construction of hygienic
latrine.   
57
Relatedly, the evaluation documented that the attitudes of service providers have improved, as has their
behavior in helping to streamline and improve human resources and management. All CHCPs now wear
their official dress and behave politely with the service recipients. When they fail to provide services as
per the expectation of service recipients, they are much more likely to be reported.   
Different types of challenges have been highlighted by service providers and service recipients. One of the
key challenges for the supply of supply‐side actors is balancing the lack of human and material resources
against the needs of service recipients. An Upazila health inspector in Kaligonj (#22) mentioned that the
CVA process had helped him to work through some human resource problems in his upazila. He
mentioned that they should have 12 union assistant health inspectorsto help him monitor the community
clinics in his area, but they only had 8 positions staffed by the ministry. Because of CVA, they had the idea
to turn 4 health assistants into assistant health inspectors to fulfill those human resource needs. Similarly,
an SAAO in Dacope echoed these sentiments (#16):
“A lack of SAAO against the required positions is one of the important challenges. In
Dacope Union, although there are three posts for SAAO, now only two are available. Out
of two, one is compulsorily required to act as an AAO at the Upazila. That is why she is
hardly able to spend time with farmers at the field level.”  
Field level officers very often deal with community level activities, for which they are supposed to follow
the instruction of their bosses. If they do not stay in the field, sometimes service recipients express their
frustration. A UAO describes his experience in the following way,  
“My field level officers who are working at the field and supposed to be present at the
Union Parishad level. But very often, we have to send statistics/service recipient lists very
urgently because of the instruction from higher authorities. In this case, SAAOs come to
the Upazila office. On the contrary, community people then do not get SAAOs at the field
level and they show their grievance. This situation put pressure on us.”  
In Section 9, we return to the topic of human resources faced by institutional actors and the  
need for greater delegation, monitoring, and accountability mechanisms for the government.    
6.6 Government Response
Local government representatives, including at Unions and Upazilas, are critical actors within the CVA
theory of change, serving important roles in specific CVA activities (e.g., interface and dialogue meetings).
They are responsible for and have purview over a number of service provision aspects in community
clinics, water and sanitation, and agriculture extension. These officials also have access to budgets and a
mandate to address local development issues in these sectors; they work with parallel committees like
the WATSAN committee and community clinics that are a part of CVA; and CVA is also supposed to bring
community people closer to the government, and vice versa. Overall, in the evaluation, government
officials, including people’s representatives, have gradually grown more positive in listening and
responding to citizens’ demands. In this part of the evaluation, we wanted to understand from the CVA
and other community groups how they perceived the government response to community participation
through CVA, as well as hear from government officials themselves regarding how they report having
responded.
Summary of Results on Government Response:
58
● Government officials, including people’s representatives, have grown more and more willing to
listen to and respond to citizens’ demands as expressed through CVA.
● CVA has resulted in a positive change in attitude among office bearers about their roles and
responsibilities, and obligations to citizens.
● CVA helps to reduce the gap and power distance between service clients and office bearers,
working to increase citizens’ confidence.
Finding #1: Government officials, including people’srepresentatives, have grown more and more willing
to listen to and respond to citizens’ demands as expressed through CVA.
  
After becoming engaged in various steps of the CVA process, community people have grown more bold
in placing their demands before government office bearers, including the Union Parishad Chairperson and
council, especially as the result of the interface meetings. The evaluation discovered many examples of
CVA members and other community people expressing incidences of engaging directly with Union chair
and ward representatives, not only attending council sessions but also visiting office hours to demand
services individually or for the community. In response, the evaluation found that government office
bearers have grown more and more willing to listen to and respond to citizens’ demands as expressed
through CVA. The following findings shed additional light on the changes that have taken place among
government office bearers. In Section 7, we discuss CVA working group members’ and community
people’s participation in Upazila, divisional, and national dialogues replicating the method of these local
points of contact at meso and national levels.
Finding #2: CVA has resulted in a positive change in attitude among office bearers about their roles and
responsibilities, as well as their obligations to citizens.
Initially, many office bearers were not enthusiastic about face‐to‐face interactions with citizens or CVA
working groups asthey did not want to look bad nor have these meetings become contentious. Moreover,
they did not want to be held accountable for not having done well in performing their duties (which they
themselves were not shy in admitting). There was also the pervasive view that neither citizens nor NJP
staff, nor any NGOs, should be in a position to hold them accountable. While some chairpersons and
council members reported feeling the pressures of electoral accountability, they did not report feelings
of social accountability before CVA.  
However, because of CVA, we note that their attitude toward both community people and local service
providing organizations has changed. An opinion expressed by a CHCP can be cited as an example (#23):
‘‘I feel very happy that UP Chair comes and makes query about the problems I face.’’ In an interview with
a SAAO in South Sreepur Union (#29), that person stated that ‘‘Union Parishad representatives have come
to know my role by now. They inform the farmers of my role and send them to me whenever needed.
Although my workload has increased, I am enjoying the changes that have occurred.’’ Moreover, the
government’s seriousness about local service providing entities has made some office bearers more
responsive to their duties, demonstrated not only by many findings here in Section 6, but addressed again
with respect to meso and national levels in Section 7.
The extent to which office bearers have been responsive to citizens’ demand can be better understood
from a statement by a participant of a scorecard session and women’s group (#2): ‘‘Since the government
59
is very much serious about community clinics, any queries made by them about community clinic is taken
into consideration by their Union Parishad Chairperson.” This participant added that within the territory
of their union a community clinic exists, people from long distances are often unable to come, so they
demanded a vehicle to their Union Chairperson, and the chairperson has made the promise of providing
a vehicle to a community clinic from next budget allocation. Upazila level officials have also been
sensitized to the performance of local service providers. Such finding can demonstrated through a
statement made by an SAAO (#29), who said that ‘‘The UAO’s supervision has increased from the past. I
am instructed not only to inspect the field but also to provide solution of the farmers if anything is brought
into my notice.’’ In this way, the “bottom‐up” nature of CVA istrickling up to the Upazila‐level in important
ways, discussed further in Section 7.
This documented change in attitude may arise in part from the view that CVA helps government officials
to get the job done more easily and overcome their inertia. Due to the intervention of CVA, getting the
job done as a local service provider is not easy, but it is also free from the vagaries of partisan politics.
One participant of FGD, while speaking with CVA working group members, said that they assist SAAO in
selecting real and relevant beneficiaries for the support given by the government. This, in turn, has made
his job easier. Otherwise, he had to seek advice from someone else who may be a chair or members of
UP that would certainly lead him in the wrong direction, given that he does not know the real state of all
local farmers. One SAAO (#16) stated:  
‘‘There are times when I am provided agriculture‐related items from the government to distribute
free of cost among eligible farmers. In such cases, I ask members of the VDC or CVA working group
to provide me with some name of the farmers who seem to be genuine beneficiaries of government
support rendered.’’
Finding #3: CVA helps to reduce the gap and power distance between service clients and office bearers,
working to increase citizens’ confidence.
Although community people had some contact with office bearers within their Union before CVA, after
and because of CVA this relationship has been strengthened in a number of ways. First, CVA has enabled
community people to place their demands before the Union parishad, while union officials are now more
responsive to those demands. An example that testifies to the influence of citizen engagement and Union
responsiveness comes from the standing committee at the Union Parishad on WASH, which decided that
‘‘people having no hygiene toilet at residence will not be allowed to have been nominated for VGD or VGF
from UP” (#37). Another statement by a participant of the scorecard session and women’s group (#36)
demonstrated that the Union Parishad had become much more responsive compared to the past:  
“There was no electricity in their community clinic. Being informed of this fact, the Union Parishad
has some allocation of Solar Power System, we communicated with CVA working group, which
then made dialogue with the Union Parishad Chair. Finally, the Chair agreed with the CVA working
group and provided a Solar Power to our community clinic.’’  
Second, Union representatives themselves, including the Chairpersons and council members, reported
that they appreciated the work of community people and CVA working groups because they say that it
helped them to stay focused on their jobs and improved their performance. A Union Parishad Chairperson
in South Sreepur (#30) expressed his opinion in this way: “it is a better way to serve the community.” Many
elected membersto the Union‐level, including Chairpersons and ward representatives, reported that they
now understood social accountability to work to improve their performance alongside that of political
accountability and electoral accountability. Such attitudes come across in statements made during
interface and dialogue meetings.
60
Third, CVA has provided community people the opportunity to meet with government officer bearers,
previously a rare practice (particularly regarding the Upazila level), to vocalize their demands through
interface and dialogue. Before CVA, community members did not know government officials nor did they
have a platform with which to interact with them. Now, they know the relevant actors involved in the
service delivery process at the Union and Upazila levels. Moreover, Upazila‐based officials are now serious
about monitoring the performance of local service providers, with many officials now visiting LSPs more
regularly to ensure that they serve the community people. An SAAO from South Sreepur Union (#29),
Kaligong said: ‘‘I wasin the field once. Suddenly the UAO called me and wanted to know my location. Being
informed, UAO told me to stay there and that he is coming. Such things were rare in past.’’  
In addition, officials are now more willing to take queries and complaints into account while also
expressing their willingness to take action. An interview with the CHCP in Dacope (#9) mentioned:
‘‘UH&FPO has declared in Upazila‐based dialogue that if complaints are made with evidence that is
justified, proper legal action will be taken against the accused health worker.’’ These realities have
boosted the confidence of community people that government office bearers are looking out for them
and working on their behalf.
6.7 Participants
Even with the positive aspects of improved service delivery arising from CVA and NJP, the evaluation
sought to examine whether or not any service improvements were disproportionately excluded from any
areas, participants, or groups within the project locations. We therefore queried community people as
well as service providers and government officials as to whether there were any individuals or groups in
the community being excluded from the benefits, and ifso, who and why. In particular, we were interested
to understand whether vulnerable groups,such as women, the disabled, the elderly, those living in remote
areas, or the ultra‐poor, had not achieved inclusion regarding aspects of CVA programming. Thus, the
inclusion of these groups in the CVA working groups and other aspects of the CVA process, like the
scorecard groups, helped to obtain their perceptions and participation. Moreover, for those who were
participating and also receiving services, we investigated whether they thought certain types or groups of
service recipients were more likely to be given favors or better services compared to others, as well as
whether everyone was being treated equally. Last, we assessed the extent to which CVA has accelerated
the process of women's empowerment in local communities.
Summary of Results on Participants:
● CVA documents equal access to service provision in its project areas, including for vulnerable
groups such as women, the disabled, the elderly, and the ultra‐poor.
● Geographical location/disparities in especially hard to reach areas and a lack of transportation
systems hindered efforts to ensure access to services.
● CVA participantsreported very few incidents of malpractice or nepotism in localservice delivery.
● CVA has accelerated the process of women's empowerment.
Finding #1: CVA documents equal accessto service provision in its project areas, including for vulnerable
groups such as women, the disabled, the elderly, and the ultra‐poor.  
61
One of the most notable findings of the study is that service delivery was equitable, with no respondents
reporting that there were any groups or individuals in the community who were missing out on these
services. The CVA process itself specifically targeted for inclusion members of vulnerable groups into
working groups and other activities. Despite some minor issues, one encouraging fact which was
acknowledged by most of the respondents was that recipients were not deprived from services on the
basis of ability, caste, creed, or other arbitrary considerations. In particular, CVA working group members
and service providers reported that vulnerable groups, including women, the disabled, the elderly, and
the ultra‐poor, were all receiving services.
Finding #2: Geographical location/disparities in especially hard to reach areas and a lack of
transportation systems hindered efforts to ensure access to services.  
While commenting on this issue participants of a community group in Dacope (#7) mentioned that “this
is the only community clinic in our (Dacope) union, people from long distance face difficulties while coming
here.”  
Finding #3: CVA participants reported very few incidents of malpractice or nepotism in local service
delivery.
A very important positive finding is that CVA working group members, VDCs, and other citizen
respondents uniformly reported across all of the sampled areas that they had not perceived extensive
incidents of malpractice or nepotism in local service delivery since the advent of CVA. As reported above
in the subsection on Pre‐CVA, malpractice and nepotism were common in service delivery before CVA.
Yet now it is extremely rare, as reported by community people.
That being said, a few occurrences of malpractice were reported. Community participants in South
Shreepur (#35) described the persistence of some patterns of nepotism in the following way:  
“Even though the situation has changed [from CVA], the patrons of political leaders and Union
Parishad representatives, sometimes receive special attention. They and their patrons do these
silently.”
The evaluation found that while pre‐existing patterns of nepotism in the NJP areas have been reduced as
a result of CVA, some small forms of nepotism, though slight, may have persisted, remaining partially
acceptable to some government officials. The following pair of statements of two Upazila level officers
can be linked to the existing situation. “All people do not get services equally. We are supposed to give
special attention to the needs and aspirations of UP Chairmen. Sometimes, UP chair sends the list to us to
give services as per their suggestion,” reported a Union Agriculture Officer in Dacope (#13). “A group of
people always try to get undue privilege. The situation is now far better than the past” (Union Health and
Family Welfare Officer, #6). Part of this may arise from the domination of power and politics on the service
delivery process previous to CVA. Additionally, the evaluation found that the Union Parishad and Upazila
Chairpersons from the ruling party were more successful at exerting importance on the service delivery
process than chairpersons belonging to the opposition.  
Finding 4: CVA has accelerated the process of women's empowerment.
CVA has worked to accelerate the process of women's empowerment. This is first a result of the fact that
CVA working groups are designed to have proper gender balance and representation. Second, CVA is
enabling women’s empowerment by improving their voice in waysthat are unique and new. They are able
62
to express themselves in novel ways. While commenting on the service seekers, the CVA working group
participants in Dacope (#1) stated that:  
“everyone isreceiving services equally. Women did notspeak up much earlier. Now they can speak
a lot. The courage and information that they have now due to CVA have created scope for them
to be a member of various committees.”  
As noted above,  a health inspector in Dacope (#8) noted regarding women’s involvement as “I found
more than 50% of the interface meeting participants were women.” Community members in Dacope
further commented on women’s participation (#1): “Women did not speak up much earlier. Now they can
speak a lot. The courage and information that they have now due to CVA have created scope for them to
be a member of various committees.”  
Third, CVA is helping women’s empowerment by forcing numerous duty bearers, service providers, and
officials to stop pretending to ignore citizens’ concerns by downplaying complaints expressed by or
demands articulated by women. Instead, as these actors must respond to citizen mobilization more and
more, they are now forced to confront women as a part of this program, eliminating their ability to ignore
their concerns.  
Last, the empowerment of women in families, as well as sensitization for their male counterparts, was
another aspect of women having participated in CVA activities. Women reported that men are now more
willing to share in household chores while allowing wives and mothers to be more active in household
decision‐making, such as with childrens’ education decisions.  
6.8 Service Outcomes
A critical aspect to CVA programming derives not only from citizens and communities voicing their
demands and mobilizing for action to improve service delivery, but also from the linkage of a social
accountability platform directly to downstream improvements in services. As designed, after exposure to
the CVA process, government line departments are meant to bring positive improvements to various
aspects of their service delivery process  (including by changing the mindset of government officials). As
a result, significant improvements should be detectable if CVA does in fact operate to positively impact
service delivery. For CVA and NJP, this means the improvement of services related to community clinics,
water and sanitation, and agriculture extension, some of which we touched on above looking at the
quantitative indicators in the CVA database. Here, we rely on reports from participants themselves on
their perception of services outcomes, and we also investigate the factors that have influenced the
process of service delivery in their views.
Summary of Results on Service Outcomes:
● CVA produced positive changes at the knowledge level of the service suppliers and recipients
that matter for services outcomes.
● CVA produced positive changes in the infrastructure and equipment that matter for services
outcomes.
● CVA produced positive changes in the mindset and behavior of the service suppliers and
recipients that matter for services outcomes.
63
● CVA increased accessibility and responsiveness from platform creation for the accountability
that matters for services outcomes.
● CVA improved services outcomes at community clinics, in water and sanitation, and in
agriculture extension.  
Finding #1: CVA produced positive changes at the knowledge level of the service suppliers and recipients
that matter for services outcomes.  
We can document important changes at the knowledge level of the service suppliers and recipients that
matter for service outcomes. Such changes are twofold. First, CVA has made LSPs more knowledgeable
about what they are required to do and the level at which they must perform to properly provide the
services that citizens are entitled to. Accordingly, they now more strictly maintain service standards
because of CVA. Across the program locations, stakeholders (including community people) reported that
the CHCP and SAAO now sit in their respective working places at the community clinic or union level as
scheduled. CHCPs now open clinics at 9:00 and remain open until 15:00 for 6 days each week, while SAAOs
remain present at their offices at the Union Parishad between 3 and 4 days for a few hours each week.
This establishes important conditions to allow for a more complete and extensive provision of services.
Concomitantly, service providers also gained knowledge that helped to improve service outcomes with
respect to their actual duties. Some service providers previously lacked knowledge regarding how to
properly use commonly required equipment in offering services to citizens. CVA has helped to close this
gap. An Assistant Health Inspector in Dacope (#8) noted that a local “CHCP was unaware of how to operate
Growth Monitoring Programme (GMP) as there was no GMP in the community clinic earlier. The CHCP is
now doing such things successfully with the assistance of weight lifting machine.’’  
Second, changes at the knowledge level have come from the community people. The population seeking
services now possess a better understanding of the services that should be made available to them at the
community clinic, agriculture extension office, and WATSAN committee. They can now also differentiate
the various institutional actors at the Union and Upazila levels, ensuring they can identify who is
responsible for specific services, oversight, and accountability. While commenting on the contribution of
CVA, a working group member in Dacope (#4) stated that:
“CVA played a key role to set up the monitoring standard of the community clinic. They
made the people aware about their rights. Thus, the CHCP opens the community clinic
timely and provides services. 26‐27 types of medicines are available at the clinic. If she
fails to offer any service due to incapacity or lack of availability of medicine, she politely
explains that to the service recipients.”
This demonstrates that one important avenue to improved service delivery provided by CVA involves a
feedback loop of knowledge and awareness.
Finding #2: CVA produced positive changes in the infrastructure and equipment that matter for services
outcomes.
The evaluation documents important changes in the infrastructure and equipment that matter for service
outcomes as reported by CVA participants. This includes improvements made to the conditions at the
community clinic, with regards to local water and sanitation issues as well as agriculture extension. These
changes resulted from CVA in that they activated citizens, service providers, and government officials to
work more cohesively and copacetically, including through interface and dialogue.
64
Community Clinic: We document improvements to infrastructure and medical equipment at community
clinics, with service delivery outcomes improving after the introduction of CVA. Community clinics have
improved through infrastructural changes. People who previously visited community clinics had to suffer
from lack of safe drinking water, unhygienic toilets, poorly maintained access roads, and so forth.
Improvements were made in these areas as a result of CVA activities. For example, while interacting in
the interface meetings, community people, who were now sensitized to theirsocial responsibility, worked
to provide various contributions that aided the infrastructural development of community clinics. This
included a water purifying machine, water posts, electric fans, blood pressure monitoring machine, and
diabetic monitoring machines. In addition to these improvements, an arrangement provided a modern
toilet and approach road was made with an allocation of tk. 50000 from LGSP by the Union Parishad Chair,
in addition to securing a water tank from RUPANTOR by the Union Parishad Chair.   
Moreover, government officials at the Union and Upazila levels were made aware of needed
improvements to infrastructure and equipment as a result of discussion at CVA working groups, dialogue
with community clinics, and interface meetings as enumerated above. Workers at the community clinics,
along with other Union and Upazila‐level health officials, began working in concert with each other to
improve existing infrastructure and equipment as needed. In Section 7, we document more specifically
how Upazila dialogues incorporated findings from Union Parishads, and then helped to “trickle‐up” to
divisional and national dialogues on community clinics.  
Water and Sanitation: Community people gained increased access to safe drinking water and sanitation
facilities as a result of CVA activities and improvements in infrastructure and equipment. One participant
with a Union WATSAN Committee mentioned that although 95 percent of households have hygienic
latrines, only 50 percent have safe drinking water. This gap is due to people in the community being
uninformed and lacking necessary economic means. Since buying a water tank is costly, poor people are
heavily dependent on others, including NGOs, the DPHE, and Union Parishads. WATSAN committees in
particular were important for liaising with other Union and Upazila‐level officials, while line ministry
members within water and sanitation, as well as VDC and ward WATSAN committees, began working in
concert with each other to improve the infrastructure and equipment. Improvements occurred because
there were now linkages between Ward WATSAN and Union WATSAN committees. Need identification
was being channeled from Ward WATSAN to Union WATSAN. In cases where beneficiaries are either hard
to select based on need, or where they default in their ability to pay, voice was raised and the process for
selecting beneficiaries became more equitable. While commenting on the activism of CVA, one working
group member from South Sreepur (#38)  stated that:      
“All people including our adolescent group, along with a daughter of an incumbent Union
Parishad member, visited the houses of our community to ensure 100 percent sanitation
of our community. With the support of the Union Parishad, CVA and community people
re‐paired a water line which was non‐functional for a long time. Now it is working as a
source of water for a large number of people, including the poultry farm of this area.”
As this quote indicates, success in improving water and sanitation infrastructure can also be attributed to
government officials at the Union and Upazila levels, who were made aware of needed improvements for
infrastructure and equipment as a result of  CVA working groups, dialogue with WATSAN committees, and
interface meetings. This evidence to action also made its way to the Khulna Water Convention (discussed
in Section 7).
Agricultural Extension: The evaluation documents improvements to infrastructure and equipment in the
provision of agriculture extension services after the introduction of CVA. People who previously did not
bother to visit the extension office at the Union Parishad have begun to do so. Furthermore,  when they
do, they are now finding the SAAO in attendance, as per the guidelines. Moreover, those offices now have
65
important equipment, such as a machine that provides important information about the weather, which
helps the officer to provide the correct advice to area farmers. While commenting on the contribution of
CVA on the agriculture services, one CVA working group participant (#4) stated that:
“Agriculture extension officer, known as SAAO, provides services by sitting at the Union
Parishad allocated office room for him. CVA talked with UP to allocate the room for SAAO.
Due to their demand, the room has been allocated.”
Additionally, the extension officers now monitor price lists of the seeds and fertilizers that are scheduled
to be hung in retailer shops to ensure that the customers can buy each product at the right price.
Important tools were introduced to help relay information, including the posting of the extension officer’s
cell phone number at the Union Parishad and the inclusion of a running register and tally of local farmers
that enables them to call or send messages. These tools were mentioned by many Union officials in the
evaluation sample, and incorporated into the divisional dialogue on agriculture extension (discussed in
Section 7).
Finding #3: CVA produced positive changes in the mindset and behavior of the service suppliers and
recipients that matter for services outcomes.   
Another positive change regarding service provision relates to changes in the mindset of service seekers
as well asthe behavioral patterns of localservice providers, with both becoming more friendly and citizen‐
focused. Service receivers are also now more positive about seeking services having been involved with
the CVA process.  
First, community people, particularly members of community groups, gained a sense of ownership over
their community clinic asthough it were their own organization. As a result, the number ofservice seekers
increased significantly.  
Second, localservice providers became more aware and sensitive to the fact that it wastheir responsibility
to provide servicesto citizens. An SAAO in South Sreepur Union (#29) mentioned that ‘‘having participated
in the interface meeting, I have been able to understand that it is my responsibility to help the farmers by
providing information about matters in relation to agriculture.’’  
This change in mindset helped to drive behavior change. Although many community people had reported
that they previously felt that many service suppliers were rude or insincere, they grew to respect service
seekers more and while increasingly their compassion for them as well as their sincerity in their supply of
services. They were able to realize that people became organized and more aware of their entitlements.
Thus, providers modified the way they behaved with service seekers and improved their behavior,
communication, and actions. While commenting on the change of behavior of service providers, a
participant in the community clinic focus group (#7) stated that:  
‘‘Health workers are now very cordial to people who are visiting them for treatment in the
community clinic. They not only speak with us very politely but also listen to our problems carefully.
In the case of critical diseases, patients are advised to go to the upazila health complex.’’  
The opinion of another service provider (a CHCP from South Sreepur, #23) demonstrates reported change
in behavior because of the CVA process:  
‘‘If any expected mother comes to me, I advise them to take nutritious foods and rest, and not to
do heavy work. I ask them to bring their mother‐in‐law with them the next time if anyone [at their
home] is found to be skeptical about their taking rest. In such cases, I personally motivate them if
they come.”
66
Finding #4: CVA increased accessibility and responsivenessfrom platform creation for the accountability
that matters for services outcomes.   
CVA has led to an increase in accessibility and responsiveness for the accountability that improves
services. As compared to the past, service providers became more responsive to service clients. This
included not only service clients who came physically to service delivery points, but also the entire
community of people that they are meant to serve. As noted, their cell phone numbers were made
available at service delivery points, including on the wall of the community clinic and at the office of the
Union Parishad. Moreover, their number had also been made public through a method of person to
person transfer. An SAAO (#29) said that he was inspired by NJP, which caused him to make his phone
number public so that anyone could call him at any time. He now  receives more than 20‐25 calls from the
farmers regarding different issues each day, but he remains happy about this change. And the posting of
phone numbers was not only reported by such officials, but was confirmed by the evaluation team,
community members, and other Union members, such as chairpersons and council members.
Since community people became aware of their entitlements, they have begun to make queries when
anything goes wrong or is insufficient, opportunities for interface and dialogue support this. CVA has
provided accessibility and points of contact to do so. One of the participants of community clinic focus
group in Sreepur (#21) opined that “the CHCP knows that she will have to face a number of questions
unless she maintains the office schedule.’’ This was corroborated with another statement given by
community ‘‘community people are notified/informed of their absence (CHCPs) if anything that may lead
them to be absent in office tomorrow happens.’’ There is evidence that when they find any non‐
compliance with the government process they are quick to take action to correct these issues. In this
regard, one of the SAAOs interviewed opined that: ‘‘proper action is taken if the complaint is made in
written form with the signature of the applicant.’’ This indicates that CVA has created a platform for
ensuring social accountability (see also case box below).
Case Box on Improved Social Accountability (from NGO Facilitator, #3)
“In an upazila‐based meeting where the deputy director of health was present,
participants from the community made complaints about the medicines that
were supposed to be available at the community clinic. The deputy director, in
his discussion, stated that if this complaint were to be lodged with proper
documentation, legal action would be taken from the government side. This
was only possible due to the CVA process. In fact, face to face interaction scope
has been created. It is not a joking matter that community people can raise their
voice in the presence of a high official of the health department. It is a reality.
It is a CVA process that has made people courageous to be vocal about their
sufferings.”
Finding #5: CVA improved services outcomes at community clinics, in water and sanitation, and in
agriculture extension.  
Services have improved in the three sectors as reported by CVA working group members, other
community people, and confirmed by other officials.
67
First, services outcomes have improved at community clinics. Care is now provided in ways it was not
before. From CVA activities, CHCPs now report to work during the mandated hours and clinics are open
per schedule. Previously, as noted above, CHCPs often came to the clinic only three or four days a week
for a couple hours, but now respondents report that the CHCPs are attending at the clinic 6 days a week
from 9am to 3pm. At clinics, CHCPs are performing in their jobs better and providing more robustservices.
For instance, most of the CHCP did not have a clear understanding of the operation of the Growth
Monitoring Programme (GMP) earlier, but now they have been performing this task successfully using the
weight machine.
The evaluation revealed many instances of significant improvements in the state of infrastructure and
medical supplies provided at community clinics, echoing many of the activities, commitments, and
changes made not only from CVA working groups, but also the Upazila, divisional, and national‐level
dialogues on community clinics as discussed in Section 7. Safe drinking water, proper latrine,
reconstructed approach roads have been improved across the sample in the response to the problems
brought up by CVA and community people during interface meetings. Moreover, CVA working groups
have liaised during interface meetings with health officials and Union members to better provide for
medicine and other equipment for patient comfort and to assist the quality of service at the clinics,
including water purifying machines, water posts and spigots, electric fans, blood pressure monitoring
machines, and diabetic monitoring machines.
Accordingly, changes in the CHCPs behavior from CVA has made them more equipped, sincere, and
trained to provide better services and improve services outcomes at community clinics as reported by
respondents across groups and areas. Services are now more likely being provided in adherence to
government monitoring standards. Medicinal supplies are more likely to be delivered, monitored, and
dispensed per government guidelines. Because of CVA, CHCPs and service providers now open the
medicine box in front of the community people and maintain a proper registrar book. Overall, the
members of the CGs have become more active monitoring the service quality and the environment of
service delivery maintaining proper monitoring standards, and CVA working groups and other community
members all reported improvements in services outcomes. To maintain accountability and the quality of
improved services, community people report that they now are more likely to access Upazila‐level officials
of the health department or visit the office of the Upazila Health and Family Planning Officer (UHFPO) to
lodge any complaints against CHCPs or for advice on a solution to problems of services at the clinics.  
Second, services outcomes have improved in water and sanitation. Access to safe drinking water and
sanitation facilities has increased across the sampled communities. Participation in Union WATSAN
committees has activated their work and made them more responsive to provide local water and
sanitation services, including in coordination with ward WATSAN committees. These committees are now
operating in line with governmentstandards, and successfully finding new budget allocations and projects
to implement under the direction of Upazila‐level officials and the Department of Health Engineering to
address local needs and improve outcomes, such as with identified beneficiaries receiving government
allocationsforthe installation of tube wells and sanitary latrines. Respondentsin the evaluation also noted
that there have been changes in the attitudes and behavior of community people to practice safer
sanitation and hygiene due to the activism of the Union WATSAN committees, which will likely improve
health outcomes. These improvements in outcomes also track with the findings from the Upazila and
Divisional Dialogues on water and sanitation, addressed in Section 7.
Third, services outcomes have improved for agriculture extension. Citizens now have improved
knowledge about the provision of services from the Union agricultural services unit, and respondents
across the sampled areas report improved services outcome from those officials. Unit officials now sit at
the mandated time in the Union offices for farmers to receive advice and services, and those officials are
68
working more closely to reach out to provide information,services, advice, and support directly to farming
households at the community level. Improved monitoring of localservice providers with respect to proper
seed and fertilizer supply and pricing has improved outcomes for farmers, and officials are now more
willing to address any attempts at selling these inputs at inflated prices. CVA has also improved services
outcome by helping to identify the farming households that are the most in need as beneficiaries. Citizens
report improved access to Upazila‐level officials of the agriculture department to make any complaints or
seek advice if there are shortcomings in service provision at the Union level. These results also track with
the findings regarding the commitments made and discussion at the divisional dialogue on agriculture
extension reviewed in Section 7.
6.9 Interagency Process
An important aspect of CVA involved how it aimed to improve the performance of institutional actors, or
what we term the interagency process. CVA programming itself revealed numerous problems in the
interagency process of service provision and local governing structures (Union Parishads and Upazilas). As
noted, bureaucratic inertia on the part of service providers, government officials, and other relevant
institutional actors not only created problems for service delivery, but revealed numerous managerial
roadblocks and pathologies of poor governance in the public sector. We therefore inquired to these
stakeholders whether or not they perceived CVA as helping them perform in their roles as service
providers and duty bearers by strengthening the interagency process within bureaucratic management.
Summary of Results on Interagency Process:
● CVA positively improved the interagency process by clarifying roles and responsibilities for
institutional actors.
● CVA positively improved the interagency process by clarifying delegation and accountability
chains within institutions.
Finding #1: CVA positively improved the interagency process by clarifying roles and responsibilities for
institutional actors.
CVA helped to improve the interagency process by clarifying roles and responsibilities for institutional
actors. A consistent finding that came out in the evaluation was there was pervasive “inertia” on the part
of institutional actors ‐‐ from localservice providersto Union Parishad members ‐‐ to perform well in their
mandated jobs. And this was not only reported by CVA working group members and community people,
it was also reported by institutional actors themselves. While institutional roles and responsibilities are
already delineated by the government, line departments, and per legislation regarding local government,
the evaluation found that before CVA few institutional actors understood their precise roles and
responsibilities individually nor when compared to other institutional actors (especially at meso or
national levels). While CVA itself did not change any of these mandates, many of the institutional actors
interviewed for the evaluation replied that the CVA process encouraged them to seek information and
gain clarity as to their roles and duties in the service delivery process precisely because citizens, through
their voice and action (expressed in interface and dialogue), were requiring them to do so to perform
better in their jobs.  
69
Moreover, institutional actors reported the lack of coordination among themselves in interagency
processes as a reason for failures in service delivery generally. A food and sanitation inspector in Kaligonj
(#26) mentioned that the persistent lack of well‐trained people with requisite technical skills and
adequate resources was a reason for poor services at community clinics. This included a lack of staffing
for the midwives necessary to help pregnant women as well as the persistence of deficient knowledge of
prescription drugs by health officialsto warn patients ofside effects. The inspector noted that these issues
persisted because higher level officials were not properly training community level officials well enough
to ensure they possess adequate skills. While CVA has helped to foster incentives and platforms for
coordination, discussions around the interagency process within institutions revealed many important
avenues and challenges for improvement that the government must take to support service delivery in
Bangladesh above and beyond CVA, a topic we return to in Section 9.
Finding #2: CVA positively improved the interagency process by clarifying delegation and accountability
chains within institutions.
A second positive impact was that after clarifying roles and responsibilities for institutional actors,
government agencies could do a better job delegating vertically and horizontally, while also establishing
better accountability mechanisms within the agencies. Numerous institutional actors noted a tendency
for themselves and their colleagues to frequently shirk responsibility or “pass the buck” when it came to
adequately addressing citizens’ concerns before CVA. The evaluation revealed that this was largely due to
a lack of clear direction and description of roles from line ministries down to Upazila and Union levels, as
well as from Unions to front‐line service providers. Higher level officials, including health and agriculture
officials at the Upazila or Union, admitted that they lacked the proper monitoring tools or resources to
fully oversee officials working within their purview. While some of these problems persist, they also
reported incidents in which CVA successfully helped them to make requests for more resources from the
relevant ministries. CVA helped to improve these interagency processes by first incentivizing these actors
to figure out their roles and responsibilities and purview for action, and second, by encouraging them to
mobilize action to obtain greater resources.
6.10 Most Important Issue
In trying to better locate the specific contribution of CVA to outcomes, the evaluation seeks to delineate
a number of “most important issue” analyses that we probed stakeholders to articulate from their
perspective. These include the most important aspect of CVA process, most important influence, most
important change, and most important challenges. We posed these queries not only to CVA working
groups, but also to government officials, service providers, and other relevant actors.
Summary of Results on Most Important Issue:
● The most important aspect of the CVA process was interface and dialogue (followed by direct
participation of communities in service delivery and information dissemination).
● The most important influence of CVA was improving the commitment and sincerity of service
providers (and communication channels between providers and recipients).
● The most important change resulting from CVA was improving knowledge and awareness on
part of service recipients and sensitization on the part of service providers (including Union
Parishads).
70
● The most important challenge was occasional friction between service providers and recipients.
Finding #1: the most important aspect of the CVA process was the interface and dialogue (followed by
direct participation of communities in service delivery and information dissemination).
Many respondents expressed that they viewed the interface and dialogue meetings asthe most important
mechanism for bringing about changes in terms of the CVA process. The reason for this wasthat it created
scope for citizens to hold face‐to‐face interaction with the service providers, and created the opportunity
for demand and supply side actors to exchange their views and subsequent behavior. The interface
meeting and dialogue opportunities in particular made this happen and created a sense of ownership
between demand and supply side actors. Due to this interaction, citizens were not only able to express
their demands more cogently, but the in‐person social dynamic helped to encourage duty bearers to
improve their actions and responses while bringing about greater accountability.  
Moreover, the interface meetings, action plans, and follow‐up helped to inform Upazila‐Level officials,
who are now more attentive and capable of ensuring the performance of government officials and service
providers across the unions in their respective upazilas. This means that the interface meeting provided
an important mechanism for both bottom‐up and top‐down accountability linkages. As we explore more
in Section 7, interface and dialogue has not only pertained to the level of the ward, clinic, Union, or
Upazila, but has also allowed CVA participants and CVA related evidence to “trickle‐up” to divisional and
national‐level officials.  
The direct participation of community people in service delivery was also mentioned as a most important
aspect of the CVA process. Some community level government officers, such as CHCHs and SAAOs, noted
that they believed that the direct involvement of citizens in the service delivery process was the most
important activity of CVA. As an example, a CHCP in Dacope (#9) said:    
“Involving community people in the service delivery process wasthe most important factor
that has contributed to those changes. It is CVA that helps community people to get them
involved in the service delivery process. In fact, CVA has sensitized community people
towards their responsibilities.”  
Other respondents considered information dissemination as the most important aspect of the CVA
process producing positive change. CVA plays a significant role in making the community people aware
about their entitlements by providing information that they could then use to monitor the service delivery
process, scorecard sessions, and adherence to standards. This arises because of how CVA builds agency
and empowerment of community people in the service delivery process. As one UHFPO officer in Dacope
(#6) noted, “sensitizing of community people by CVA/NJP is a notable activity.”
Finding #2: the most important influence of CVA wasimproving the commitment and sincerity ofservice
providers (and communication channels between providers and recipients).  
Many of the community people who participated in the CVA working groups noted that they viewed the
most important influence of CVA activities as bringing changes in the commitment and sincerity of service
providers. Due to CVA’s influence, respondents believed that these service providers became more
responsive and cordial. The response of most of the participants of focus group discussions can be
reflected in the following way (from a CVA working group in Dacope, #1):  
71
“The most important influences of CVA on service providers includes an increase in their
commitment and sincerity towards their duties and the reduction of their absenteeism. As
a result, people are getting services at their doorsteps, the number of water‐borne
diseases has decreased, and damage in agriculture items hasreduced. These have resulted
in the savings of money, energy, and time.”  
Relatedly, another most important influence mentioned by CVA working group participants and a number
of officials noted improved communication between service recipients and service providers. According
to one SAAO in Kailashganj (#18) “the most important issue is the strong communication between farmers
and CVA. Due to this kind of communication, he, along with the CVA members, monitor whether the service
provider displays the price list/red cloth (lalshalu). These kinds of monitoring activities played a key in
ensuring the fair price of a product.”
Finding #3: the most important change resulting from CVA was improving knowledge and awareness
on part of service recipients and sensitization on the part of service providers (including Union
Parishads).  
All respondents, including service providers and service recipients, expressed the opinion that awareness
of service recipients and sensitization of service providers were the most important change from NJP and
CVA. The following example demonstrates how the state of expressing demands from the government
side now links up with a concomitant service provision (#10) commitment:  
“During the interface meeting, the Union Parishad chairman was committed to allocating
a fund of eighty thousand taka in 2018‐19 financial year for improving the service of
community clinics. A Union Parishad member also donated a BP machine to the
community clinic as per the demand of community people during the meeting.”
Another response regarding the most important change involves the effect that CVA has had on
encouraging Union Parishads to be much more engaged with development processes in the three sectors
in addition to being much more likely to allocate dedicated resources and finances to fund water and
sanitation initiatives and Union WATSAN committees.
Finding #4: the most important challenge was occasional friction between service providers and
recipients.  
One important challenge mentioned during these queries wasthat particularly in the early periods of CVA,
some government officials, including politicians like the Union Parishad representatives, did not approach
CVA cordially nor did they express an explicit will to reform or improve. A CVA working group member in
Kailashganj (#4) commented that, “sometimes they were not taken as positively and they considered us as
a threat.” Since CVA do not have any formal platform or identity, they may face difficulties in the future
after the phasing out of the NJP. For their part, government officers noted that the CVA process
sometimes generated a negative attitude among some service recipients that they found abrasive. Two
officials expressed their opinion in the following way:  
“Due to the intervention of CVA and NJP, negative attitudes regarding our services have
been growing among the service recipients. As per their information dissemination
activity, people know what they are supposed to get from my office, but at this moment
due to having a number of problems, we fail to provide necessary service to farmers. We
do not have enough manpower. As an example, I can note that different agencies are
linked to agriculture related services. For instance, if farmers expect good seed from me, I
cannot be able to offer thisservice because other agencies are responsible forseed related
services, such as BADC and Seed Certification Agency.”  (#13)
72
“Sometimes it presented as negatively to the community people, ignoring our limitations.” (#6)
6.11 Sustainability
In this subsection, we report information regarding sustainability as reported through the perceptions
relayed in the data collection by CVA working groups, community people, officials, service providers, etc.
to gauge their opinion. (We discuss our own evaluation of sustainability and system strengthening in
Section 7). Sustainability is a difficult topic to understand and measure with regards to impact evaluation,
and as a result, a nettlesome question to ask participants in a program like CVA their ideas about.
Nonetheless, we asked each type of respondent how they viewed sustainability (indicators of
sustainability) and probed them regarding their ideas to promote it (since sustainability can only really be
assessed after projects end). To do so, we first framed “sustainability” as an aspect of the particular
activities and modalities of CVA programming. Second, we inquired as to the sustainability of the service
improvements, dialogue, voice, etc. that have already been shown to work and queried how they could
ensure that they continue to work. Third, we considered the sustainability of linkages, committees, and
institutional roles activated by CVA. Last, we asked about the sustainability of accountability.  
It is important to note that we place “sustainability” generally into the context of the present evaluation,
whose timing made the various aspects of sustainability difficult to track (as a part of phase‐out and 2
year cost extension). This created both the opportunity to probe stakeholders regarding their ideas about
sustainability, but also to ask about additional challenges. Up until now, what has been achieved through
the contribution of the CVA process is remarkable. Yet at the same time we should bear in mind that the
achievement is not an end; rather, making these achievementssustainable isthe ultimate goal. Given this,
it was therefore unsurprising that all respondents expressed their concern about sustainability following
the phasing out of the project, which we summarize here from their reports before further exploring
further in Section 7. (It is important to note that because all of the fieldwork took place directly before
the Covid outbreak, no respondents had Covid or the “bridging period” in their minds when providing
responses.)
Summary of Results on Sustainability:
● Participants in CVA desired it to continue and become sustainable.  
● CVA participants provided numerous concrete and specific ideas of their own regarding how to
make CVA sustainable.
● Possible creation of a link between Union Parishad and localservice providers, and other parallel
structures and committees, may help with sustainability.
Finding #1: Participants in CVA desire it to continue and become sustainable.  
Acrossthe board, the interviewees expressed their willingnessto continue working to sustain the changes
resulting from the CVA process, which they are benefiting from. All manner of participants in CVA, from
CVA working groups to service providers and government officials, expressed a desire for CVA to continue
and be made sustainable. They think that people involved with the CVA working group, VDC, and standing
committees have been sensitized and desire to continue their work. They recognize, however, that they
must become habituated with this process to ensure it can be continued without external pressure; until
then, they wish for NJP to continue working as a catalyst to support their actions.
73
The desire for community members to continue in their activism via CVA is expressed by the desire to
sustain and expand activities as per (#35): ‘‘NJP has built an Arsenic Iron Removal Plant (AIRP) by investing
tk. 350000. Now our duty is to maintain it. For doing so, we have formed a ‘maintenance committee’,
which has mobilized fundsfor its maintenance.’’ Further, the desire among duty bearers can be expressed
as wanting CVA to continue since it helps them to perform in their jobs. But many respondents expressed
worry about sustainability given NJP phase‐out.
Finding #2: CVA participants provided numerous concrete and specific ideas of their own regarding how
to make CVA sustainable.
CVA participants provided numerous concrete ideas of their own regarding how to make CVA sustainable
(a topic we return to in Sections 7 and incorporate into our series of recommendations in Section 9). One
idea noted the possibility of registering CVA working groups as legal entities like CSOs or NGOs. CVA
working groups expressed the desire to have their own identity out of a fear that they would otherwise
encounter many problems in the absence of NJP. For instance, they currently receive support from the
NJP staff to fix appointments with the government officials when they need to meet with him to discuss
matters related to different community problems.  
The question now shifts to determining who will fix similar appointments when NJP is phased out. This
perhaps suggests the necessity of developing legal identification for CVA as a group. A legally‐defined
identity and registration as a CVA group in perpetuity would help its members to more easily continue
their activities with fewer hassles. All CVA working groups expressed this commitment, which is echoed
in the following sentiment (#35):  
“Before forming the CVA working group, (I am now a member of the CVA working group), I took a
patient to Union Health Centre and asked the person in‐change to examine and provide medicine.
In response to my request, he asked me many questions including who I am and why he should
listen to me. When I told him that it is his duty, he told me to go home as he is not ready to listen
to me. After the formation of the CVA, the situation has changed as we are 12 in number.’’ He also
added that “if they have a formal organization or identity nobody will ignore them if their demands
are rational and relevant.”  
Similarly a health inspector in Kaliganj (#22) suggested that “those who will not receive antenatal care
(ANC) and postnatal care (PNC) from the community clinic will not be considered for maternal allowance.”  
One alternative to this view was suggested by some respondents, who encouraged NJP to consider
forming an Upazila level CVA comprising members from different CVA who are already working at the
ward level. The Upazila level CVA would work as an umbrella organization for all of the Union Parishads
within its area. Moreover, it would be easier to take registration for Upazila level CVA instead of taking
registration of all union level CVAs.
Another idea that they expressed involved their own income generating activities. The groups provided
and discussed how to incorporate microfinance to sustain and grow the activities of the CVA working
groups. They also discussed seeking donations from others to help support transportation to working
group meetings and supplies as well as seeking contributions from CVA working group members to
provide the necessary logistical and administrative support required for the meetings.
Finding #3: Possible creation of a link between Union Parishads and local service providers, and other
parallel structures and committees, may help with sustainability.
74
For the continuation of changes that have taken place as the result of CVA, respondents mentioned that
strong linkages between the Union Parishad, local service providers, and government line departments
will have to be established and fostered. How this may operate is explored more fully in Section 7,
involving the way that CVA participants and evidence from CVA activities worked to “trickle up” to Upazila,
divisional, and national‐level dialogues and fora. For qualitative data collection, many CVA working group
participants were aware of these meso and national dialogue sessions, and further promoted them as
important avenues for further interface and policy advocacy.
Moreover, the evaluation noted that  because of CVA, parallel structures like the WATSAN committees
and community clinic committees are operational and running at the ward and Union level. Even if CVA
phases out, if these committees have continued support from the Union Parishad and other relevant
stakeholders, their work can continue. We incorporate these insights into our recommendations in
Section 9.
75
7. Sustainability Through Systems Strengthening
7.1 Overview
CVA seeksto improve government output; bureaucratic performance; and citizen engagement, voice, and
demand articulation for improved services. It aims to promote these improvements throughout local,
meso/regional, and national levels and across development sectors and institutional levels. As a social
accountability tool, Section 6 demonstrated with quantitative and qualitative data that CVA has played a
pivotal role in creating opportunities for dialogue between the government and citizens while extending
networks and collaboration between government organizations, NGOs, and community stakeholders.
CVA can also complement other programs that aim to strengthen formal lines of institutional
accountability (including electoral, political, and bureaucratic accountability). The three sectorsspecific to
government service units that were improved through CVA and NJP were community clinics, water and
sanitation, and agriculture extension.  
We conclude, based on the realities of the political economy of service provision in Bangladesh and the
CVA tool adopted and validated by WVI from other settings, that NJP succeeded at incorporating the CVA
approach into its programming. This resulted in the enhancement of service delivery in Bangladesh
through the improvement of the main channels/mechanisms of information, voice, dialogue, and
accountability.  
Whereas Section 6.11 discussed citizens’ and stakeholders’ perceptions of what could be used to make
CVA sustainable, in Section 7, we take a more holistic view of sustainability by defining what it is and how
to measure it. In doing so, we delineate a theory of change that transforms the more narrow scope of the
theory of change for CVA and NJP in Section 4 and broadens it to consider impacts on meso and national
levels, policy dialogue, and systems and institutions as outlined below. Because this theory of change was
not originally included as part of NJP, we leveraged comparative findings and analytic work from other
WVI programming in other countries adapted retrospectively for the present evaluation. We also leverage
USAID’s “Framework for Supporting Sustained Development” by engaging local systems. We then present
results from the evaluation that help to articulate this perspective of sustainability and systems
strengthening in NJP in Bangladesh based on CVA’s achievements.
Programming aspects of CVA present two important challenges regarding assessing impact on
sustainability. A first challenge is whether the evaluation can determine if CVA helped to promote
advocacy and influenced policy. NJP supported and facilitated community advocacy activities to Union,
Upazila, and Sub‐district levels. The action points of the action plans developed during the interface
meeting by service providers and service receivers were often followed up by the people mentioned in
the action plan. However, it is important to note that nothing in the CVA guidelines or NJP programming
specifically mentions or outlines explicit strategies regarding meso or national level policy dialogue or
action items, nor doesit provide specifics on “systemsstrengthening” and how it maps onto CVA activities
in the Bangladesh/NJP context by engaging local systems. WVB’s guidance was more likely finished with
the methodology and relied on the local national advocacy team to make the links.
Nonetheless, policy dialogue and systems strengthening are core goals of CVA. As noted in the Evaluation
Framework and Methodology, these realities were reinforced during the inception visit and recognition
of these goals helped to shape aspects of data collection for the fieldwork. Initial evidence generated
during the inception visit found incidents where NJP organized national level policy dialogues using
information provided by, and as the result of, the CVA process. This was actioned to improve services and
policy influence on an ad hoc basis (for example, this included actions by the Health Ministry on the
condition of community clinics in a WVB‐hosted national dialogue event in September 2019 and the
76
Second Coastal Water Conference in Khulna). But importantly, there was indication and linkages of
divisional and national level policy dialogue or advocacy workshops based on CVA field evidence in NJP's
CVA field implementation guideline as well asin the project proposal, even if the strategy was not detailed
or explicitly laid out.  
A second challenge is whether, beyond ad hoc events or isolated incidents, the evaluation in its
methodology is able to reveal systematic or systems strengthening findings that point to the
institutionalization of sustainability for social accountability in Bangladesh. This task is harder to perform
in a traditional sense given a lack of systems‐strengthening specific programming guidelines in CVA, the
timing of the evaluation, and a lack of a sustainability theory of change applied to CVA at the outset of
programming. Every development program hopes to sustain whatever progress it makes in the
implementation phase. This not only means narrowly sustaining whatever positive outcomes can be
tracked and measured, but also changing aspects of the contextual, institutional, or systems level inputs,
structures, and outputs that previously erected barriers but have now hopefully been reformed. Such
“systems strengthening” advances sustainability as the result of a catalytic process set into motion by the
program implementation. However, what “sustainability” and “systems strengthening” mean more
narrowly, analytically, and empirically in any given country or with respect to development programming
such as NJP is often minimally understood and lessfrequently discussed in program design and evaluation.   
Nonetheless, the evaluation attempted to assess evidence of systems strengthening as the CVA
components of NJP in Bangladesh explicitly mentioned sustainability as a desired outcome of the
programming. To do so, we in part relied on systems strengthening theories of change employed by WVI
in other contexts and adapted to Bangladesh. Moreover, social accountability more generally is by design
meant to enact more permanent, institutional, and systems‐level improvements even after the lifetime
of the programming, such that results here should similarly contribute to WVI’s global programming and
thinking on sustainability.
7.2 Policy Dialogue & Actions of CVA in NJP
7.2.1 Overview
Even if not clearly delineated, a core aspect of sustainability and systems strengthening extends beyond
what CVA’s narrow scope/mandate to ask whether its activities will persist. This involves examining
whether and how social accountability can either directly or indirectly lead to policy dialogue and actions
at meso and national levels that suggest real, lasting, and institutionalized changes and reforms. This
includes new policies, new legislation, new budgetary commitments, the adoption of new monitoring
tools, and additional actions undertaken by the government and other relevant institutional actors. If the
evaluation finds that CVA is working with some insights as to why it has been successful, the question
would then shift to determining whether CVA can be scaled.  
As such, the evaluation assessed the degree to which localized and local‐level results were capable of
“trickle‐up” as well as whether or not this process had already occured. Thisinvolved looking at the degree
to which advocacy and policy dialogue from CVA working groups had found their way up, both directly
and indirectly, the chain of delegation in public sector management, as well as how regional/meso and
national level stakeholders were working to influence policies relevant to CVA activities (in addition to
social accountability more broadly).  
In the following subsections, we first briefly review some aspects of the political economy of governance
in Bangladesh as it pertains specifically to avenues and challenges of policy dialogue. We then review the
policy dialogue and actions that we discovered along these lines during data collection for the evaluation.
77
To date some of these policies and actions have been implemented while others have discussed and
actioned, with a further subset still inchoate. Next, we attempt to measure their success thus far (where
appropriate) along with their potential future impact (where appropriate) by linking results from policy
dialogue to a retrospective theory of change on sustainability. The two year project extension provides
opportunity for learning, knowledge sharing, and improved policy advocacy and monitoring of policy
changes, which are subjects we will return to in Section 9.  
7.2.2 The Political Economy of Governance in Bangladesh  
Perthe theory of change as explained in Section 4 and itsrole in stressing the importance of understanding
the matrix of the political economy of governance in Bangladesh, CVA must work to both link and
strengthen the governance structures that exist at the local level, such as Union Parishads and Upazilas.
This follows USAID’s guidance for engaging local systems. But like any evaluation, it must do so with the
realities of the political economy of governance in a giving setting. This then helps in understanding how
to institutionalize the horizontal and vertical accountability mechanisms so that they do not solely arise
on an ad‐hoc basis, and continually blend local, meso, and national‐level actions and actors. Such actions
require communication with the government regarding not only the modalities and benefits of the CVA
approach, but also communication detailing these actions may operate alongside and in conjunction with
other development programming and financing through policy dialogue and advocacy. Linkages and
integration are necessary for the government to establish a system capable of feeding the results of local
CVA type approaches into meso and national level policy. However, the establishment of such a system is
often limited by the lack of a structure within the government to do so.   
Contextually, while the political economy of governance confronts numerous challenges across national,
regional, and local unitsin Bangladesh, there are many aspects of governance that informed an evaluation
of the potential for policy dialogue and advocacy for the evaluation, as well as the potential for
sustainability beyond the evaluation.  
The government has gained greater control of resources in the past 10 years due in part to its success in
parliamentary elections and majoritarian supremacy. Per the political economy mapping and theory of
change, these results pose the question as to whether national‐level politicians retain an  incentive to
pursue a reform agenda lacking any real electoral opposition or accountability. Importantly, this does not
preclude the possibility that the government remains interested in reform and is continuing to work to
improve service delivery  ‐‐  types, levels, and points of accountability are manifold in any regime.
Regardless, some divisions within government and problems with specific line departments continue to
pose challenges. The situation is such that there is nothing at the national level actively impeding the
success of CVA nor longer‐term progress promoting social accountability; however, there are also no
specific mechanisms in place to guarantee either short or long‐term success.  
In response to these realities and attempts to weigh their potential positive impact against many extant
challenges, the evaluation tried to uncover possible sites of pressure and institutional contact that may
facilitate the government’s own attempts to address critical development sectors via policy dialogue and
advocacy.
We note that the Prime Minister has recently publicly made many important strides on anti‐corruption
and system strengthening for rule of law. She has been at the helm of current anti‐corruption efforts in
an attempt to associate herself and the ruling party with an image of clean government. The government
has also recognized that despite sustained economic growth over the past five decades, they now must
turn to services, governance, and corruption to meet the critical welfare needs of the population.
78
Moreover, the government has begun to behave as though it has become more open to criticism than in
the past, with other evaluations also suggesting it has been successful in improving governance and
service provision across a variety of sectors.  
The media in Bangladesh can play an important role by supplementing the deficiencies of parliament to
provide a “check” on government action, with the recent RTI (Right to Information) Law helping to drive
reform alongside the efforts of the newly appointed Local Government Minister (in office from early
2019). Aiding in the expression of the need for “constructive dialogue” with the government (as opposed
to contentions interface meetings) could be a key aspect helping to leverage this reform movement from
the government to aid with the hopes and aims of the evaluation. And the evaluation found many
instances of media invites and attendance at divisional and national dialogues organized by NJP.
During the evaluation, officials at the meso and national levels reported their perceptions regarding how
service delivery was working. At the local level, it was believed that the education sector was performing
best, along with community clinics, which was unsurprising given that the clinics are the Prime Minister’s
brain child and that she remains personally invested in their success. It was expressed that WASH as a
sector was not working as well, with concerns including reference to the formation of WATSAN
committees. It was feared that perhaps too many of the standing committees, where they were
functional, were rubber stamps for government officials and duty bearers, with many remaining in
operation without actually exercising the necessary voice and monitoring that deliversreal accountability.
Moreover, assessing the performance of these institutions was hard given thatstanding committees were
non‐functional in many cases and that VDCs were not legal entities created by the government during the
present period.
7.2.3 Evaluation Results on Policy Dialogue
In thissubsection, we review the evaluation results on policy dialogue undertaken by the evaluation team.
While national and regional level NJP staff uniformly desired to have upazila, regional, and national level
policy dialogue and actions, they recognized that there were no specific monitoring or programming
aspects built into the design of NJP and CVA toward this end. Therefore, the evaluation relied on key
informant interviews, project documents, and other qualitative/anecdotal evidence involving policy
dialogue to understand, track, and summarize results on policy dialogue at meso and national levels.  
Summary of Results on Policy Dialogue:
● The CVA process has begun to “trickle‐up” to the Upazila level.
● CVA provided opportunities for policy dialogue at regional (division) and national levels
involving community people and government officials.
● CVA may have contributed indirectly to policy dialogue and government actions, in some cases
revealing internal tensions in the government’s approach to social accountability.
● Policy dialogue directly inspired by CVA led to changes in commitments, legislation, budgeting,
and policy by the government.  
79
Finding #1: The CVA process has begun to “trickle‐up” to the Upazila level.
Across the evaluation sample, we noted ways in which CVA was working to affect Upazila‐level officials
and policies in Section 6. Positive results from activities at the community and union‐level thus appear to
be “trickling‐up.” CVA working groups now report more access to Upazila‐level officials, including
increased dialogue with the chairperson and the council. Since the Upazila council is made up of Union
Chairpersons, this has helped create a conduit of information and feedback. In fact, respondents
expressed a belief that many Upazila officials were often not aware of local conditions at community
clinics and similar institutions prior to CVA. Accordingly, they noted that the CVA process had played an
important role by simply informing them of local realities. CVA working groups and their attending
activities have helped Unions and Upazilas to better mobilize resources to both improve service delivery
and respond to their own human resources problems among line ministry staff at the Upazila. These
information flows have also helped Upazilas seek improved assistance from line ministries.  
Moreover, the dialogue and interface meetings specifically engaged these officials, with access to the line
ministry having also improved. The CVA working group in Dacope (#1) reported that being a part of CVA
allowed them to pressure the Union Parishad to then pressure community clinics in their area (because
of participating in CVA to prepare them to put pressure on the Union Parishad), though they attribute this
change to the interface meeting with the Upazila officials attending, who the CVA group viewed as having
played a critical role in applying pressure on local actors (see case study box).  
Upazila officials reported that they appreciated the degree to which CVA helped to improve their own
monitoring tools, along with the oversight and accountability of officials within their purview. In fact,
Upazila dialogue sessions provided fora during which Upazila health officials and government officials
made new commitments for the monitoring and oversight of health care provision in their areas (see case
study box). A CHCP in Shatkira (#23) mentioned that she has noted a change in mindsets of Union Council
members since they began attending the Upazila dialogues. As a result of the pressure exerted by the
Upazila on Union members, she noted that the Union Chairperson in her area has begun to sit with the
community clinic committee in order to work with them to apply for infrastructure improvements for the
clinic. An NGO facilitator in Dacope (#3) replied that the Upazila dialogue was successful at improving
services for the Union Parishads within its area because of the force of “naming and shaming” that came
from Upazila leadership. Notably, this same respondent believed that phasing‐out CVA working groups or
Union‐level dialogues in favor of retaining only Upazila‐level dialogues would be a mistake as both
remained essential. The case studies of Upazila‐level dialogues from Dacope, Kalinganj, and Koyra
demonstrate how Union‐level interface and Upazila‐level dialogue were mutually supportive of one
another.
Case Study: Upazila‐Level Dialogues on Community Clinics
In May 2018, CVA held an Upazila Dialogue in Dacope on the topic of service provision at community
clinics. In Dacope, CVA had found that of the 22 clinics within the Upazila, many clinics fell short of
achieving monitoring standards ‐‐ 20 lacked safe drinking water, 9 did not have proper support service
providers, 13 lacked requisite medicinal supplies, and 12 lacked requisite medical equipment. The
Upazila dialogue provided the opportunity to relay these findings from CVA to local officials and service
providers to develop jointly agreed upon action plans to address the problem of health service
provision. During the interface meeting, Upazila officials had already agreed to provide additional
sanitation facilities and equipment. At the dialogue session, these findings and actions were relayed by
CVA members to a broader group of stakeholders, including government officials, line department
80
employees, and journalists. This larger audience ensured that community people were able to make
additional requests regarding infrastructure and supply equipment upgrades. In particular, Union
Parishad officials promised to make new commitmentsto address accessibility issues(around approach
roads) and pressured CHCPs to improve their attitudes toward community people. Health officials
worked with CVA to develop action plans to follow up on monitoring of infrastructure improvements,
and further budgeting commitments were made for 6 of the clinics. Participants at all levels were
pleased with the dialogue session and believed it helped them to better perform in their jobs.
Upazila dialogues were also held in Kalinganj and Koyra in July 2018. The dialogue meeting allowed CVA
working groups and Union officials to share the results of their activities to Upazila officials. It also
helped draw the attention of these officials to requests for action on maternal and child health
challenges at the community clinics. Health officers, politicians, and journalists were in attendance
along with community representatives from CVA. The participants agreed to institute and monitor the
government’s policies on the opening hours for clinics and to work with unions to provide further
budgeting to address patient comfort, sanitation, and equipment needs at clinics. Upazila leadership in
Koyra also provided motivation to health support staff, union members, and area residents by
reminding them that public health is one of the government’s key priority areas in development, and
that his office at the Upazila was also open to receive information and requests from within his area.
Similar sentiments were shared by the Upazila Health and Family Planning Officer at the conclusion of
the dialogue session in Satkhira.  
In fact, many CVA and Upazila officials expressed a desire to have an equivalent CVA working group at the
Upazila level, which may be an important consideration for future policy dialogue,systems strengthening,
and sustainability. As Upazilas have larger budgetary scopes, manpower, and oversight capabilities, many
respondents saw mobilization of Upazila into social accountability as an important aspect of future
improvements to service delivery. Moreover, the government line ministries often mentioned gearing
many of their activities, efforts, and reforms toward Upazilas, as opposed to unions, including the training
efforts guided by the Local Government Department.
The example of the Upazila Dialogue on community clinics in Shyamnagar is instructive in understanding
how the “bottom‐up” aspect of CVA coheres with the “top‐down” aspect of institutional structures and
government agencies in Bangladesh, which operate through the linkages created for interface and
dialogue as facilitated by NJP. The meeting occurred in August 2018. In attendance were CVA working
group and VDC members, Union and Upazila officials, and line department employees. The session was
facilitated by NJP, who presented on the conditions of the 41 community clinics in the Upazila using
information provided by the CVA working groups that came out of the interface meetings they had held
with local clinic providers and union officials. They were able to share recent progress on action items for
36 of the clinics, highlighting changes that came about through the actions of community people and
government officials that helped reallocate resources and attention to the most pressing needs. As a
result of those local interfaces, 88% of the action items were achieved through actions by local
government officials, with the remaining  12% mandated to the communities. The ability of the dialogue
session to relay these actions and subsequent progress to stakeholders was well‐received by those in
attendance, further giving government officials at the Upazila level and line department ministry officials
the opportunity to express their thanks and support while providing feedback. Community members and
Union officials were also given the floor to express their appreciation of CVA and directly field questions
and hear suggestions to improve interface and action items going forward. The dialogue meeting ended
with a speech from the chair of the meeting, a Union Health and Family Planning Official, who expressed
81
thanks to NJP, CVA working groups and community people, Union officials, Upazila officials, and other
relevant stakeholders for their participation in the meeting and their work to improve community clinics.
She also reaffirmed the government’s commitment to improving public health (mentioning the prime
minister specifically), with a promise from the Directorate of Health a budget allocation to reconstruct 3
community clinics in this area and further improve water and sanitation at health facilities.
Finding #2: CVA provided direct opportunities for policy dialogue at regional (division) and national
levels involving community people and government officials.
The evaluation team uncovered positive examples of policy dialogue and actions taken as a direct
component of CVA activities. Beyond the Upazila‐level, we consider these in terms of how CVA “trickled‐
up” to regional/meso (division) and national levels. While there are positive aspects and impact in this
regard, particularly the divisional and national dialogues, the realized and potential force and impact of
CVA has at times met pushback from some within the government. We try to highlight these tensions as
discovered in the evaluation, but none of these tensions suggest that gains from policy dialogue arising
from CVA and NJP will be unable to sustain or continue. Rather, we highlight these tensions to suggest
points of contact with certain institutional actors in order to specify what may be important
considerations required to continue policy dialogue and associated strategies in the future.
CVA directly contributed to policy dialogue and institutional engagement with divisional “dialogues” in
the project area on community clinics/public health, water and sanitation, and agriculture extension
services. Above the administrative unit of the Upazila lies the District and then Division, or region, in
Bangladesh. The divisional level provides important linkages between governance at the regional/meso
level and the national level, not only providing a site of development programming and financing by the
government, but also with representatives of relevant line department officials.
Case Study: Divisional Dialogue on Water and Sanitation
NJP hosted a divisional dialogue on water and sanitation  in Khulna in July 2019. Water and sanitation
are particular concerns in Khulna as a low‐lying coastal given its adjacency to the Bay of Bengal, which
has a tendency to flood, especially during monsoon season. This impacts not only the livelihoods of
farmers and their ability to grow crops, but also threatens the availability and quality of potable water
and proper sanitation. The divisional workshop was scheduled after the enactment of CVA activities
regarding 40 union WATSAN committees in Khulna and Satkhira districts. The purpose of the workshop
was to share these results prior to then discussing with the government and other stakeholders ways
to address the gaps in service provision revealed through CVA. These efforts were aimed at improving
the functionality of the WATSAN committees and helping assemble additional support from Union,
Upazila, Divisional, and national government officials to address water and sanitation needs. More than
100 people attended, drawn from upazila, district, and elected representatives; divisional and national
officers for DPHE; CVA members and community people; and journalists and NJP/NGO facilitators.   
The workshop first summarized how CVA activities, along with the Union WATSAN committees,
addressed the provision of water and sanitation services, allowing community people’s voice and
actions to be integrated into service demands at local levels. Various local government officials
expressed concern and made commitments to address water and sanitation needs. This included the
Upazila chairperson, who allocated a new budget that prioritized water and sanitation. They also
discussed various ways that officials and citizens at all levels could work to make Union WATSAN
82
committees more functional and better performing, directing Union Parishads to also prioritize water
and sanitation needs in their budget allocations over the next five years. In response to a question of a
Union Parishad Chairperson, the additional divisional commissioner replied that there would be a
decision that the Zila (District) Parishad will not lease any government ponds in the coastal belt as
community people use the water of those ponds for domestic and drinking purposes. He strongly
committed to the participants that they will take necessary initiative if there is any incident that Zila
Parishad have leased the government pond of these areas. Recommendations, action plans, and follow‐
up were developed and communicated at the end of the workshop for all stakeholders. The results of
the meeting were then reported by the media to various outlets.
Case Study: Divisional Dialogue on Improvements of Union Agricultural Service Unit
NJP hosted a divisional dialogue on the union agricultural service unit in January 2020 in Khulna. While
the Department of Agriculture Extension in the government of Bangladesh plays an important role
supporting the livelihoods of rural farmers, the CVA process in Khulna division had revealed a number
of shortcomings in the provision and quality of these services to local residents, including with respect
to the performance of the 40 union agricultural services units in the area. The divisional workshop was
scheduled after CVA activities related to the monitoring of 40 units, ensuring that these results were
shared with stakeholders along with a discussion of ways to address gaps. Close to 90 people attended,
drawn from upazila, district, and elected representatives; divisional and national officers for the
Department of Agriculture Extension; CVA members and community people; and journalists and
NJP/NGO facilitators.   
The workshop first summarized the challenges faced by farmers in the division and the quality of
services provided by agriculture support staff in the attending areas. Much of the feedback focused on
the results of the CVA scorecard sessions, which revealed a lack of adherence to governmentstandards,
including with respect to the staffing of Union agriculture officials, their attendance at work and at the
Union, the quality of service they are meant to provide, and monitoring of local service providers
regarding feed, seeds, and fertilizers. After hearing presentations from community people and area
farmers, various local government officials expressed concern and made commitments to address
agriculture extension services. A number of specific recommendations were made and summarized for
stakeholders, including the department, Upazilas, and unions, to make; with new commitments made
by Upazila Chairpersons and senior government officials. The results of the meeting were then reported
by the media to various outlets. During the fieldwork for the evaluation, one of the interviewed
agricultural officials had just attended this workshop and reported how much he had learned from it
and appreciated it.  
CVA has also played a role activating and engaging in national dialogues with government stakeholders,
including elected politicians and line ministry representatives. They did this by building on localized
versions of dialoguesto then create linkagesto higher unitsin a “bottom‐up” approach that leverages and
formslinkages at the meso and national levels. For example, the above case study on the Upazila dialogue
for community clinics did not end there. Indeed, NJP built on those successes in Dacope, Kalinganj, Koyra,
and Shyamnagar to host a divisional dialogue on community clinics in Khulna Division in May 2019. The
purpose of the meeting was to share the results from the “trickle‐up” approach of the 119 community
clinics in the division as well as how CVA was working to assist institutional actors at Union and Upazila
levels to better provide local services; and as a next step, they aimed to relay these findings to the
83
divisional level while seeking further support. Accordingly, relevant members from the community,
unions, and upazilas were in attendance, as were various departmental and divisional officials and
representatives from the media and NGOs. Given the attendance at this divisional meeting by community
people, service providers, and government officials at a local level on the one hand, and regional and
national leaders on the other, the “meso” aspect of bringing data, experiences, and voice from the
“bottom‐up” to articulate demands and requests for supply to the “top‐down” governance structure of
line department ministries found a site of common engagement and purpose at the divisional level.
After this divisional dialogue on community clinics, NJP then hosted a national dialogue on community
clinics in September 2019 in Dhaka. This provided a forum for duty bearers, service providers, citizens,
and government officials to engage in dialogue on community clinics. Members of CVA working groups
and other community people were able to attend to express demands and views from the local level to
national leaders. Public health officials, health NGOs and ministry members were in attendance, and were
willing to engage in dialogue. The evaluation concludes that the success of coordination, actions, and
mobilization of the National Dialogue was built on the success, rigor, and seriousness with which the
participants in the Divisional Dialogue undertook these activities, which in turn had been built on Upazila
dialogues.
Case Study: Second Coastal Water Convention in Khulna
Another example of a national dialogue occurred with the Khulna Water Conference in August 2019.
WVB and NJP in partnership with the government held a national conference in Khulna to discuss
“strengthening governance, enhancing private sector engagement, protecting ecosystems and adaption
to climate change impact.” The conference specifically focused on problems of safe drinking water and
sanitation in coastal areas of Bangladesh. The event drew 1,200 participants, including relevant
government ministries, NJP and WVB leadership, representatives of the community, and the media.
CVA working group members, and data gathered in the CVA process regarding water and sanitation,
informed manifold aspects of the conference agenda and activities. Members of the community met
with various government stakeholders to make demands of more aggressive government action on
service delivery, including a session with members of parliament. The presence of the meeting and the
opportunity for dialogue between citizens and officials was an important aspect to the conference’s
design and outcomes, and was reported on widely by the media. Numerous institutional actors and
government agencies made new commitments at the conference to improve service delivery, including
changes in allocative decisions on budgeting and priorities in the direction of supporting greater
demands and sectoral needs.
As a corollary to the information relayed about service provision during divisional and national dialogues,
NJP has produced a number of short “policy dialogue” memos on topics including community clinics,
water and sanitation, and agriculture. These short briefing memos, authored by NJP staff, convey basic
insights and findings from CVA activities in the three sectors. They can be sent to government officials,
NGOs, and other stakeholders, with many of the insights, results, data, and conclusions coming from CVA.
Most importantly, they can be shared with community people and community groupsto inform them and
motivate their actions. They are designed and written to provide further avenues for CVA to contribute
important information capable of informing policy dialogue between stakeholders, including the
government, by providing tools to support citizens’ voice and actions.
84
In a policy brief titled, “Community Clinics in Southwest Bangladesh: Challenges and Opportunities,”
published in July 2019, NJP first provides background to the community clinic sector as relevant to the
development challenges in Bangladesh. It describes CVA activities to monitor community clinics and
reports on initial baseline findings of the monitoring standards and scorecard session results from the 119
clinics in the project area. It next reflects on how community members ‐‐ through CVA working groups,
community clinic committees, and other community groups  ‐‐  can work to support and improve the
situation of clinics, which includes their guidance on advocacy and their efforts to hold duty bearers
accountable. Its conclusions on areas for further improvement and recommendations provide important
advice not only to service providers, donors, and government officials, but to the community people
themselves about how to raise their voices and seek action and accountability.  
A policy brief titled, “State of Water and Sanitation in Southwest Bangladesh: Pitfalls and Prospects,”
published in July 2019, and “Status of Agricultural Extension Services in Southwest Bangladesh:
Opportunities and Challenges,” published in December 2019, take a similar approach by utilizing simple
data visualizations, charts, and graphs to convey relevant background, CVA tracking, and other
information in ways that are easy‐to‐read and digest. This helps to inform citizens and citizen
organizations, who may face educational gaps, as well as policymakers and government officials, who
rarely have time to read and digest in‐depth policy reports.  
Moreover, the evaluation revealed that it is clear that NJP specifically, and WVB generally, have done a
good job using data from CVA and other programming in their policy and advocacy work, as was on display
prominently in these briefs. A WVB official reported his experiences being able to use systematic data as
well as quantitative indicators from NJP and CVA, to help make policy and inform advocacy (he mentioned
that some of this exists, but that they need more).  These policy briefings are one example of that, as are
the numerous presentations that NJP and CVA groups have made at Union, Upazila, Divisional, and
National dialogues. He also said results from the evaluation of NJP will help as well, but importantly, that
it will help them go to other Union, Upazila, and other government levels/officials to share best practices
and encourage them to use social accountability in their areas. With the quantitative tracking indicators
in the CVA database, the numerous reports issued by NJP on CVA and non‐CVA related activities, and
other publications like the reports, a variety of quantitative and qualitative data have been and can
continue to be used for policy dialogue and actions.
  
Finding #3: CVA may have contributed indirectly to policy dialogue and government actions, in some
cases revealing internal tensions in the government’s approach to social accountability.
As a result of these extensive data collection strategies by NJP, WVB has engaged in other national‐level
policy advocacy that less directly involves CVA specifically, but to which CVA and social accountability are
relevant and have contributed important insights to WVB programming and advocacy, which may have
indirectly influenced the government’s policy dialogue and actions.  
The evaluation revealed numerous occurrences of policy dialogue that one might consider as indirectly
inspired by CVA, or as a corollary to the type of activities conducted by CVA and to which social
accountability is relevant, that led to changes in government commitments, budgeting, and policy. Some
of these changes are evidence of policy dialogue and actions taken within the context of other NJP
programming to which social accountability may pertain. In these instances, it is hard to say whether
policy dialogue uniformly led to positive changes or improvements on the part of the government ‐‐ often,
85
it revealed many new promises, but was too soon to assess actual changes or the impacts of those
changes. In addition it also revealed points of tension within the government itself as well as how social
accountability tools may or may be received across different units and levels.  
For example, in an interview with a member of the CBHC‐Community Based Health Care under the
Ministry of Health and Family Welfare, the ministry’s NGO‐liaison officer for community clinics and rural
public health reported that he had attended the National Dialogue on Community Clinics sponsored by
NJP. Some of the other work in public health that he is helping to facilitate incorporates aspects of social
accountability. For example, the NGO CARE is working with 1,500 community clinics to involve community
people more. He noted that the government discourages NGOs from becoming direct service providers in
the field of public health as well as that their support for this community involvement precludes even the
provision of snacks nor other materials. Whether or not these efforts are working to improve the
functioning of clinics in those areas has not been assessed by the government, but it reflects important
aspects of social accountability. This officer replied that he was not aware of specific CVA activities or
modalities, yet he has worked with members of the NJP team and he could report familiarity with the
contribution of community input from the National Dialogue meeting.  
This official described other initiatives undertaken or planned by the government to which social
accountability modules and CVA‐like tools might pertain in public health. He said that in order for citizens
to become more active in local governance, the government was aiming to ensure the participation of 1
million community members by linking them with local governments nation‐wide (and that the
government is providing training and capacity building regarding how to link with Union Parishads). The
government invited NGOs and development partners to develop community support groups.
Encouragingly, where these community support groups were active, results on services and outcomes
were better. They expect the community clinic staff to mobilize resourcesin addition to engaging to create
awareness and mobilize the committees. However, unlike CVA, the government believes it is the
responsibility of the CHCP to organize these activities, rather having this come from citizen groups. As
reported by this official, these groups were not functioning, and they thus required the help of NGOs for
mobilization. Yet even so, he perceived this partnership as having achieved only limited success so far. A
government policy mandating the CHCP to mobilize never really took off in his area, with NGOs now trying
to work with Union officials toward implementation without a substantive direction, guidance, or
resources. The way he described it, itsounded to the evaluation team asthough tools and modalitiessuch
as CVA would help, however, he believed that any effort to mobilize officials and communities would
require support and government cooperation from the beginning of programming. We address
recommendations regarding government‐civil society linkages in Section 9.
Another official at the health ministry, who works specifically on community‐based health care and was
interviewed in Dhaka, had attended the National Dialogue on Community Clinics. In his role working with
the ministry on improving linkages with health care delivery and health access at the community level, he
provided some important background information on government initiatives that may prove fruitful in
thinking through how social accountability has been and will be incorporated into government
programming and policies in the future. Along with other stakeholders interviewed during the evaluation,
he credits the current prime minister for the government’s more recent aggressive action on improving
rural health. He mentioned that under the old government, community clinics were closed from 2001‐08.
Although the Awami League government had acted to establish community clinics when they were in
power in the 1990s, the BNP government stopped this after taking power in 2001, which Awami League
86
restarted again after winning elections in 2008. Specifically, in 2009, the government undertook an
aggressive revitalization program for community clinics themselves, which was supposed to provide voice
and action for health care recipients, 80% of whom are women (according to this official). When pressed
then why it was the case that fieldwork for the present evaluation revealed so many gaps in service
delivery at the clinics, including as reported by community people themselves, this official mentioned the
importance of timing. After the community clinics were re‐opened from 2009, he mentioned that public
campaigns to announce the services at clinics had to be undertaken, and that it took time to re‐open and
make the clinics functional again. Community support groups were created, and are the government’s
own mechanism to mobilize community members and receive feedback about the conditions of the clinic.
He reports that where these are active, health outcomes are better, and where they are not, outcomes
are worse, but he did not have a good explanation of why they were active in some places and not others
and believed that to work, they would require a lot more financing and incentives from the government.
Curiously, this stands in contrast to the mandate expressed by the health official above who said the
government refuses to provide any incentives to assist citizen‐led mobilization. Further, he relates gaps
in service delivery as also arising from problems in supply chains, stating that although local government
units like Unions and Upazilas were now more involved in community clinics, he is cynical that they have
any real incentive to involve themselves in improvement based projects. Instead, he believes they only
make a show of their investment by pretending to be engaged close to election time.
Given his experience in the ministry, his attendance at the National Dialogue, and the other information
he understood about CVA, this official did believe that CVA working groups could be one modality used
to overcome the problems and inertia that the government faces in its own attempts to mobilize citizens
in rural areas to monitor and improve health outcome at the clinic level. He believes CVA should be a
model for other areas, and believes NJP programming broadly and advocacy more specifically have been
key reasons for its success. Despite his cynicism of local government generally, he believes CVA‐type
programming could successfully work through local governments as a better way to manage health
service delivery, rather than having the ministry playing a centralizing role (at least in part because health
issues are local issues and often vary by location). Moreover, he favors knowledge‐sharing across
community clinics to make the gains from NJP sustainable, and the deployment of multipurpose health
volunteers as explained in the case study box demonstrates many of these linkages and knowledge‐
sharing.
Case Studies:  Possible Indirect Impact of CVA on Policy or Areas for Linkages and Knowledge‐
Sharing
During its interviews with line ministry officials, the evaluation team uncovered a variety of examples
of possible indirect impact of CVA on policy or fruitful areas for linkages and knowledge‐sharing
between CVA/NJP and the government of Bangladesh.
In a first example, the health official and NGO‐liaison in Dhaka mentioned that the government has
recently passed a bill in parliament to improve the monitoring modalities and tools of the officials in
local government responsible for service delivery. He mentioned that the bill allows for every upazila to
recruit a monitoring officer whose sole job will be monitoring; this legislation has apparently passed in
parliament and is now being implemented with a start in the recruiting for the monitoring officers in
2020 (energy for this bill came from government and NGO supervision, and various researchers). The
87
plan is to include 1 district coordinator and 1 monitoring officer per Upazila, but unlike in the past, this
will be detached from the actual doctors and health officials currently at the upazila level so that
monitoring does not bog down their other duties and vice versa with respect to actually providing health
care. That is, officialstrained in health would not be responsible for monitoring their own health officials
at the same time they must provide care, rather care would be decoupled from monitoring with the
creation of this new Upazila monitoring position. At Upazilas, there are an average of 35‐40 community
clinics. This monitoring official will check the community clinic once a month and attend the community
group committee meetings in the afternoon prior to reporting to the district.  District will then report
to division, and division will then report to national. He says there should be monthly reporting and
coordination meetings, and that community people are meant to be invited to these meetingsto discuss
successes and support needs (in ways analogous to the CVA interface and dialogue meetings). They will
also try to mobilize the union more in this process.
● If this legislation has indeed passed and the government has funded this position and started to
implement and train the monitoring modalities, CVA and NJP could think about waysto leverage
the creation of this monitoring position to refocus their own efforts on monitoring and linkages
with this officer as opposed to working with frontline health care providers.  Moreover, CVA
could think of ways to improve the citizen mobilization aspect of this program and lend
knowledge from their experiences, trainings, and results to that effect.
In a second example, officials at the health ministry mentioned that they have a new ICT‐based
monitoring system that works online and communicates results on monitoring standards and service
provision at the community clinic level.  
● The evaluation team inquired as to whether there would be scope for their tool to be
incorporated with the CVA database and vice versa, and he said yes and that the information
management systems people at the ministry and WVB should liaise in this regard.  
In a third example, the newly appointed line director in the health ministry mentioned how the
government is now working to prioritize tracking indicators and services and improving ways to
measure their existence, output, and effects. He mentioned ways in which the government is working
to prioritize communities and community health workers as a first line of defense along with
information‐gathering, and treatment for community clinics. For example, given a lack of treatment
options or medicines for psychiatric illnesses at the clinic level, they want to pursue an education
campaign to help communities become aware of and able to respond to psychiatric episodes that may
arise for area residents. The government is also conducting a pilot programme of community health
workers (“multi‐purpose health volunteers”) to collect data on families and households at the ward
level (it is currently being piloted in 19 Upazilas, four of which are the upazilas where NJP operate which
were included after the National Dialogue and NJP lobbied for it, even though CBHC had not previously
planned to include those four upazilas). These health workers can inform the CHCP to assist with
treatment and referrals, and, although these workers are volunteers, they can receive a per diem to
cover transport and associated expenses.  
● CVA and NJP could clearly leverage their strengths to partner with the ministry on the pilot and
roll‐out of the multi‐purpose health volunteers. First, this initiative arises in part from NJP‐
sponsored dialogues, and in fact, 4 upazilas were added to this pilot due to the direct lobbying
88
of NJP staff. Moreover, the synergies with CVA working groups and NJP’s extant work with
community clinics could provide tools and modalities to improve the work, monitoring, and
follow up of health volunteers, CHCPs, and area residents. These activities could also easily
inform interface and dialogue with Union and Upazila officials, and hope those levels reallocate
budgeting based on the information provided by the roll‐out of health volunteer treatment
modules. As of the evaluation, the pilot was planned to continue for the next year, having
started in late 2019, with the idea to scale beyond that if successful. (They plan to add 87
upazilas to the pilot in early 2020).
Despite the explicit or implicit synergies between the activities of the ministries and CVA, the evaluation
also found that many instances involving either a lack of knowledge among national‐level officials about
CVA specifically, or social accountability generally (even when they often knew about NJP and WVB), as
well as some tension regarding whether or not these officials believed citizen mobilization and social
accountability was necessary or helpful in improving the modalities of the work across various line
departments.
As noted, ministry workers often did not know about CVA activities. Some officials reported that they
knew what “CVA” was, but would get the acronym incorrect (e.g, “Citizen Voice and Accountability”), get
CVA confused with the work of the VDCs, confuse CVA with other NGOs or CSOs, or have a general
understanding of social accountability without real knowledge about how it has specifically worked in
Bangladesh. Despite the success of CVA, then, it has not permeated too deeply into the workings of
relevant line departments at the national level, even though the WVB’s reputation is well‐regarded and
its leadership well‐known across these departments.
In some instances, government/ministry officials in Dhaka rejected many of the premises of social
accountability when pressed as to whether they thought citizens should provide an important aspect in
improving government service delivery. For example, an official at the Local Government Department
rejected the notion that more needs to be done at a local level  ‐‐  instead, he believes Upazils are
sufficiently constituted to do development work because all 17 line ministries have representation at the
Upazila level. He recognized that Union and Upazila members need more training and capacity‐building,
supplied by his office, but when pressed as to why he did not action this, he responded that he does not
lack the budget to train but rather the manpower. So one can conclude that the government hasinstituted
Upazilas and designed them as such a way as to function as important sites of service delivery at the local
level, even funded these positions, but has not adequately planned how to train a cadre to work at
Upazilas or develop a robust plan for identifying, training, and properly remunerating such a group.
The evaluation team noted the irony that this local government official was rejecting social accountability
as a mechanism to spark local government action while at the same time describing his own inertia in
tackling the problem ‐‐ one that CVA could presumably work to fix. Moreover, he described new efforts
to start an open‐budget system and meeting at the Upazilas, but could not express the ways in which this
would do anything more than institute the budgets as mandated by the line ministries (e.g, this was not
participatory budgeting in a meaningful sense). He mentioned that there does not need to be a
constitutional amendment to allow a mandate for CVA working groups or VDCs, but he did not see a
reason to legally inscribe them beyond possibly their registration as NGOs, nor what purpose they would
89
serve in addition to the ward shava and open budget meetings. The evaluation could not determine
whether these viewsreflect more broad attitudes of government officials, but that is a subject where WVB
may continue to need dialogue and follow‐up for advocacy purposes.
Finding #3: Policy dialogue directly inspired by CVA led to changes in commitments, legislation,
budgeting, and policy by the government.  
The previous findings mention how CVA activities have “trickled‐up” to the Upazila level and informed
policy dialogues and action directly at divisional and national levels in community clinics, water and
sanitation, and agriculture extension, while at the same time CVA and other social accountability
modalities may have indirectly influenced policies acrossthe relevant line departments(and opportunities
forsynergies given the government’s own priorities and actions). Now we more firmly relate direct impact
from CVA to changes in the commitments, legislation, budgeting, and policies by duty bearers. These
commitments were often expressed during a variety of interface and dialogue venues with CVA members,
starting at the local level and making their way to Union and Upazila officials, line ministry officials, or
other elected and non‐elected government officials at meso and national levels. Policy dialogue has also
impacted budgetary allocation decisions ‐‐ which is easier to track at more local levels and harder to track
zooming out to regional and national levels. That many of the results here echo what has been said
demonstrates how the “bottom‐up” approach ofsocial accountability worksto activate meso and national
levels to actually translate voice and action into accountability and outcomes from the government.  
At the local level of the 119 community clinics, 40 Union WATSAN committees, and 40 Union agricultural
units, assessing the status of the action item monitoring and action plan that arose from the interface
meetings is a good place to evaluate and track “changes in commitments, legislation, budgeting, and
policy by the government.” While the action plan and follow‐up is an explicit programming aspect of CVA
that involves CVA working groups liaising with and working alongside officials and service providers at the
clinic and Union levels, it should be noted that these duty bearers at the local level that agreed to make
changes regarding service delivery quality while adhering to monitoring standards should indeed be
considered as “policy dialogue” (that is, CVA working groups expressing voice and demanding action), as
should subsequent changes in commitments, legislation, budgeting, and policy made by the government.
At this local level, these changes in commitments were mostly obtained by working to improve services;
changes in legislation are rarely actual changes in legislation as such but rather adherence to the
government’s previously legislated standards, circulars, and policies; whereas changes in budgeting arise
from the new allocation of funds (typically by the Union Chairperson and council, WATSAN committees,
or agricultural units) to support development projects. At thislocal level, we may not consider Union‐level
officials as simply performing in their previously prescribed roles and responsibilities as a “change in
policy” as such, but we may consider their actions to now overcome bureaucratic inertia as analogous to  
change in policy ‐‐ that change being that they now do their jobs. The hundreds of examples listed in the
action items and monitoring plans for community clinics (n=1,325), Union WATSAN committees (n=734),
and Union Agricultural services unit (n=515), as discussed in subsection 6.2 with the CVA database.
To get a sense of the importance and impact of CVA mobilization and action item follow‐ups as an
important aspect of changes in commitments, the figure below displays an example list of action items
(from the CVA database) for the water and sanitation sector. The CVA database does an excellent job
keeping track of action item monitoring, delineating the activity, as well as who is responsible for
monitoring it, by the date it should achieve completed status and then the current status indicated. The
figure shown below includes an example list of action items regarding water and sanitation, which are
also color coded to show progress (green=completed, yellow=pending, red=ongoing). As each item goes
from red to yellow to green, we can consider that as a new change to service delivery because of CVA and
90
NJP in Bangladesh ‐‐ each action item in some sense required a new policy, behavior, budgeted item, or
commitment to be realized.
Beyond the Union‐level changes as reported in the action plan items and monitoring, we document
numerous changes in commitments at the Upazila‐level. For example, the discussion of the Upazila
dialogue for community clinics in Shyamnagar included above mentioned that at the dialogue meeting,
new budgets and project commitments were made by representatives of the government and Directorate
of Health. The case study box below enumeratesthe specific commitmentsthat were made at the Upazila‐
level dialogues on community clinics in Dacope, Kalinganj, and Koyra (as mentioned as well in the case
study box above). These commitments, requiring new budgeting, often took the “hard” form of promising
new spending on direct improvements to infrastructure, sanitation, equipment, and services; but they
also took a “soft” form of promising to oversee the implementation of government policies and adherence
to standards.  
Case Study: Changes in Commitments, Budgeting, and Policies from Upazila Dialogues on
Community Clinics
From Upazila Dialogue in Dacope:
● action to repair and build a community clinic in a union
● promise to provide water ambulance service for remote areas/unions
● improvements to approach road and boundary of community clinics in a union
● direction for health providers to be more sincere
● new budgeting for 6 community clinics
● promise for community people to monitor construction work
● provision of 16 water tanks where potable water not available at clinics
91
From Upazila Dialogues in Kalinganj and Koyra:  
● actions to make sure that clinics are open per government guidelines and provide regular
services
● unions will reallocate funds to address clinic renovations, ensuring safe drinking water, and
managing safe latrines.
● recognition to land owners for the provision of land for the clinics
● promise by Upazila officials that their office for chairperson and council open to receive any
information regarding community clinics  
● promise by Upazila chairperson that he will recommend transferring community clinic staff to
other revenue streams to then be considered government permanent staff (and therefore
eligible to receive benefits of government staff).  
These positive changes made their ways to the meso levels via changes at the Divisional‐level. The
divisional dialogue on water and sanitation and agriculture extension services provided numerous
changes in commitments by relevant stakeholders, as per the case study box below.
Case Study: Changes in Commitments, Budgeting, and Policies from Divisional Dialogues on Water &
Sanitation and Agriculture Extension Services
From Divisional Dialogue in Khulna on Water & Sanitation:
● Union Parishad and the Department of Public Health Engineering commit to work closely to
make WATSAN committees more functional
● Union Parishad will prioritize WASH issues and allocate from the Union budget more
programming in this sector, including developing a 5 year plan that prioritizes and budgets for
WASH issues
● The Department of Public Health Engineering will work to capacity build with water
management committees and the caretakers of water sources
● The government will not lease ponds (and the Union Parishad officials will follow up on this)
● Pond excavation and maintenance must be pursued by the Union Parishads in partnership with
the Department of Health Engineering
● Union Parishads will be responsible for arranging regular coordination meetings with relevant
WASH actors (including the WATSAN committee)
From Divisional Dialogue in Khulna on Agricultural Service Unit:
● Commitment to recruit sub‐assistant agricultural officers in positions that are vacant at the
union level to improve services to farmers
● Union Parishad to provide a separate room for the office of the Union Agricultural service unit
(including furniture and essential equipment) where this is lacking
● Establish and keep functioning the digital labs at the Union Parishad  
● Devote improvements to maintain the duty roster and duration of services; with the sub‐
assistant agricultural officer ensuring the availability of mandated services through scheduled
field visits
● Strengthen the functioning of the Union Agriculture committee and the respective standing
committee
● Develop an agriculture club at each Union for women to receive training in homestead
gardening
● Circulate informational materials to farmers in the form of leaflets regarding the best methods
92
to cultivate pesticide‐free crops and vegetables
● Free the canalsfrom any extant leasesto store water for irrigation (with prodding from Upazilas
and Zila Parishad)
● Encourage gher ownersto take measuresto drain saline waterfrom their ghers without infusing
them into agricultural plots
● Strengthen feedback and monitoring mechanism for farmers to receive quality seeds and
fertilizers from dealers and retailers as per the official government rates
● Union‐level interventions on the production and use of organic fertilizers, and training and
sensitization on saline‐tolerant specifies to farmers
In a similar vein as to the changes in commitments, policies, and budgets as expressed in the case study
box for water and sanitation and agriculture extension, the Divisional Dialogue on Community Clinics also
provided a form for institutional actors ‐‐ including CHCPs, Union officials, Upazila officials, and divisional
officials ‐‐ to recommit to improving health services, making improvements to infrastructure and services
at the clinics, dialoguing and interfacing with community people, and working to access more funds from
the division and line department for health services in this division. The discussion and proposed action
items were made in reference to the government’s newly enacted legislation, the Community Clinic
Health Assistance Trust Bill (from 2018). Similar activities were undertaken in the National Dialogue on
community clinics, where national officials and ministry representatives committed to action points
regarding consultation with civil surgeons and the Upazila health and family welfare officer in Khulna and
Satkhira to provide the ministry a list of all community clinics requiring infrastructure improvements and
reconstructions; continuous consultation with and follow‐up with the CBHC to synthesize the NJP project
objectives on sustainability the ministry’s actions in public health; and a discussion of whether CVA
members could be used as health volunteers (as per the case study box above).  
The discussion of the divisional dialogue on water and sanitation in Khulna in 2019, as relayed above,
mentioned that at the meeting, not only were new budgetary allocations made by government officials,
including Upazila Chairpersons, but Unions, Upazilas, Divisions, and the national government all agreed
that water and sanitation had to become budgetary priorities for all levels of government over the next
five years (see Case Study Box above on the Second Annual Water Conference).  
At the national level, NJP and WVB has continued numerous policy advocacy activities with the
government regarding aspects of CVA, and many of these activities have had or will likely make an impact
on commitments, legislation, budgeting, and policies. For example, the Coastal Water Convention in 2019
produced numerous commitments and funding decisions regarding water and sanitation, including the
publication of the “Khulna Declaration 2019” after the convention, which lists the problems the
convection identified and 13 agreed upon action plans for stakeholders. Examples include:  
#3: “Existing laws and practices of leasing the natural and District Administration owned water
bodies for surface water should be amended and preserved as the sources of drinking water,
agricultural irrigation, household works etc. Relevant and necessary laws, acts, and policy should
be formed in this regard and implementation of those polic[ies] should be ensured under an
integrated management of respective ministry of government.”  
#6: “Effective measures should be taken by Local Government Division (LGD) for ensuring safe
drinking water supply in these rural areas. Considering the disasters these areas are continuously
affected by, more than adequate allocation should be ensured in the national budget.
[Department of Health Engineering] should increase the capacity of human resource through
increasing it. Union and village level WATSAN committee and capacity building of caretakers of
water sources should be under the direct supervision of DPHE.”
93
Moreover, as a result of one the recommendations from the national community clinic policy dialogue
and WVB’s ongoing advocacy supported through CVA evidence, the government of Bangladesh has
approved the recruitment and payment of 600 multipurpose health volunteers under the target areas of
the Nobo Jatra program in Khulna and Satkhira (as mentioned in the Case Study box above on linkages).
To address the shortage of health staff in the country, the Bangladesh Government introduced a
nationwide scheme to recruit multi‐purpose health volunteers to support the government’s nationwide
community clinic program. The primary responsibilities of the volunteers will be to monitor the growth of
children, ensure referrals of malnourished children and mothers to higher level health clinics and support
vaccination activities.  
This group of 600 volunteers was added to the national scheme after lobbying by Nobo Jatra via the
national policy dialogue on community clinics held in November last year. The 600 volunteers will be
recruited and paid by the government to support 122 community clinics through June 2022. Under a
performance‐based system, multi‐purpose health volunteers will be paid based on their performance (i.e.
how many children they monitor for growth and how many referrals they make where needed to higher
level health clinics). They may receive between 3000‐4000 taka a month for these activities. WVB and NJP
was successful on this front by advocating the government on behalf of the targeted communities by
showcasing their CVA work and lessons learned, as well as leveraging partnerships with the government,
government resources and funding, and relationships with line ministries for mutual support in health
sector programming.
Another policy activity involves the question of legalization of CVA working groups, perhaps in concert
with legalization of VDCs for the NJP areas, including possible budgets and funding provided for them by
the government. Many of the CVA working groups themselves and NJP field staff expressed the possibility
of legalizing CVA working groups as legally‐registered civil society organizations. This idea is expressed
with an eye toward the sustainability of improvements to the service delivery process because, as
expressed by one WVB official in Dhaka, “CVA is not about individual or household level sustainability, but
community level sustainability; this platform needs to therefore be recognized by the government.” This
official commented that the role of advocacy for WVB at the national level involves convincing the
government of the role of VDCs and CVA in working as a process from the ground up across institutional
levels.
In fact, one deputy commissioner asked WVB in a non‐NJP district to facilitate VDCs in his area to create
strongerlinkages with commissioners. Learning from this experience, the WVB official proposed field visits
with government officials as well as the sharing of case studies with local government officials outside of
NJP areas in order for them to learn about the positive effects of social accountability. He mentioned the
possibility of the development of joint 5 year plans between VDCs, CVA, and Union Parishads, which would
involve sharing plans and other ideas for sustainability. For example, in Myminshing District, he told the
example of a VDC MoU with the Deputy Commissions, which he wants to use this case study to convince
other government officials and others working to sensitize the National Human Rights Commission on
issues related to their lobbying the governments. He also presented an idea that involved sending
journalists to VDCs to publicize articles that focused on local level issues.  
Toward this end, WVB has developed a Memorandum of Understanding with senior government officials
and the National Institute on Local Government (NILG) (under the Local Government Division). This
involved embedding some of the NJP and WVB content into the Local Government Department training
curriculum of its ministerial cadres, Union Parishad members, Upazila Parishad members, and other
committees, including the use of VDCs and CVA working groups as groups to help identify beneficiaries,
social safety net programs, etc. This type of work not only helps government officials across institutional
levelsto become sensitized to the kind of work that VDCs and CVA do as an aspect ofsocial accountability,
94
but it also allows the government to “own” these structures. In particular, the MOU aims to review and
update capacity building manuals and training materials of NILG for local government institutions
regarding child rights and protection, strengthening local government institutions for participatory
planning and budgeting, and establishing evidence‐based advocacy mechanisms for child friendly local
government systems. VDCs and CVA could be one avenue toward sustaining social accountability and this
type of work with local units.  
It is important to note that during the evaluation not everyone agreed with trying to register CVA working
groups with the government. They instead expressed some hesitancy with regards to registration as they
believed that this could mean that the government would be able to “co‐opt” the working of CVA groups
down the line, as they have previously with other CSOs. The balance between sustaining social
accountability and the potential for capture should be considered now and in the future.
Moreover, progress on registration was hard to assess, with the government expressing mixed support.
At some meetings by government officials indicated either a lack of knowledge about progress on
registration, opposition to the idea of registration or legalization, or a lack of familiarity about VDCs and
CVA working groups. For example, a representative at the Local Government Division office seemed to
have little to no knowledge about CVA working groups, while downplaying the need to legally recognize
VDCs. Also, he expressed a desire not to make more institutions in parallel at the local level alongside
what other government‐mandated structures do, like the ward shava budget meeting. He also said that
he believes that two ward shava budget meetings were sufficient to decide what projects should be taken
for the Union Parishad for the year, which would be in contradiction to the frequency of activities and
meetings described with CVA. This type of sentiment was also suggested by some of the officials in the
health ministry with respect to whether it was necessary for VDCs or CVAs to exist if NGOs and other
citizen committees were already established, as explained above. An official at the Department of Health
Engineering did not really see the need to have continual CVA participation with Union WATSAN
committees because he believed that they were doing fine asis and that the Union Parishad could manage
the committee without citizen support.
While some health ministry officials agreed that some public health interventions and improvements
would necessarily require the participation of citizens and communities (even if they disagree with what
form this should take), officials in other ministries questioned the role of communities and social
accountability in development activities. For example, the Local Government Division expressed dismay
that citizens should be so involved in the Upazila programmes (beyond their contribution to ward shava
meetings) ‐‐ instead, he believed greeted investments and training were needed for Union and Upazila
officials to do their jobs, but better management and training from his line department were key to
making local government work better, not citizen activism.  
An official at the Department of Public Health Engineering was unable to say how many WATSAN
committees were in operation in Bangladesh, and when pressed, he was unable to say which, nor able to
consult one of many policy documents on his shelf to indicate where this information is collected and
reported. Officials in attendance at this meeting did not appear to know or understand the role of
WATSAN members nor the role of the committee other than to note their role in helping to implement
any programming mandated by the line department. At the same time, they were not aware of CVA
activities with respect to WATSAN, and rejected the notion that community‐activated mobilization around
water and sanitation was necessary for the committees to function or for services to be improved. This is
because the official at the Department of Public Health Engineering could not say specifically what role
WATSAN committees had played in recent interventions nor what the WATSAN committee’s role was vis‐
a‐vis the Union and Upazila structures. It was therefore unclear to him what citizens were contributing.
He also believed that development activitiesshould be directed and mandated by his department or other
95
relevant ministries and communicated to Unions and WATSAN committees from “top‐down” directives,
rather than entertaining “bottom‐up” development project requests from unions and the communities
they serve; he mentioned a role for members of parliament and divisional officials to be the linkage and
seek financing for these projects.
Moreover, he did not think citizen reporting on or monitoring of WASH services was necessary because
he said that they rely on the department’s own personnel at the Upazila level to oversee water and
sanitation projects within the upazila as directed by the ministry’s programming, circulars, and
employment policies. Even so, he admitted his department’s challenges in doing so. He replied that his
main mechanisms to identify issues related to WASH do not come from community people but rather
government statistics ‐‐ which by his own admission he often does not trust (and could not locate); he
understands the vulnerabilities of coastal areas and populations residing there without having to be told,
but also as confirmed by the mechanics they employ at each Upazila ‐‐ who by his own admission were
under‐staffed and under‐resourced.  
Summary of findings on policy dialogue and actions:  
We can summarize findings on policy dialogue and actions to say first that CVA has begun to “trickle‐up”
to the Upazila level. Further, CVA provided opportunitiesfor policy dialogue at regional and national levels
involving community people and government officials in certain instances. Last, policy dialogue directly
inspired by CVA led to changesin commitments, legislation, budgeting, and policy by the government. But
as expressed by WVB leadership, we need to think about how to balance sustainability of social
accountability more generally, not just CVA as such. Activation of the community groups is important for
sustainability, but so to is policy dialogue and actions.  
7.3 What Does Sustainability Mean for Social Accountability?
Having presented the evaluation of policy dialogue in the previous subsection, we now turn to a broad
discussion of how this and the findings in Section 6 suggest sustainability for social accountability in
Bangladesh.  
Variousscholars, practitioners, and donors have vastly different conceptions of whatsustainability means,
in particular for governance, and more generally, within the context of development programming.11 As
a result, the evaluation team was not able to find an operational definition of sustainability that was
mentioned in any of the original NJP documents nor the 2 year cost extension. In their “Local Systems: A
Framework for Supporting Sustained Development,” USAID defines sustainability thusly, “Sustainability
refers to the ability of a local system to produce desired outcomes over time. Discrete projects contribute
to sustainability when they strengthen the system’s ability to produced valued results and its ability to be
both resilient and adaptive in the face of changing circumstances” (5, 2014). But in exploring sustainability
more broadly across its programming and with respect to NJP specifically, the evaluation team found that
USAID does not have an internally consistent way it defines, measures, or evaluatessustainability through
which definitions, measures, and outcomes can be delineated enough to assess the long‐term impact of
NJP programming as outlined/thought‐out during project design nor implementation from which to track
progress.  
11 USAID, 2018, Sustainability of donor projects on decentralisation and governance reform in Indonesia, Assessment
Report.
96
For our purposes here, we have decided to focus specifically on sustainability with respect to social
accountability and CVA by trying to retrospectively fit evaluation findings with an appropriate (if post‐hoc)
theory of change regarding systems strengthening and the political economy mapping, borrowing from
this evaluation approach and those conducted by WVI in other settings, as well as USAID’s framework for
sustainability by focusing on engaging local systems.
How do we define systems strengthening, why does it matter, and how does it relate to sustainability?
Systems can either be strengthened in the short‐term or they can merely appearto be strengthening while
producing diminishing returns. Determining whether these results persist over the long‐term or instead
attenuate is often difficult to gauge with programming such as social accountability interventions like NJP
and within the narrow timeframes of typical project evaluations. Yet short‐term positive impacts can also
prove catalytic for long‐run impact. The figure below, drawn from a previous evaluation of CVA funded by
the World Bank in Indonesia (Westhorp, G. &Ball, D. 2018) maps how CVA potentially strengthenssystems
as the result of short‐term and long‐term impact and effectiveness, taking the reality of the Bangladesh
context into consideration and presenting findings with respect to the political economy mapping and
theory of change specified in Section 4.
97
To assess sustainability for social accountability, we considered: i) sustainability of CVA processes and ii)
sustainability of CVA outcomes.  
First, the sustainability of the CVA process: Our methodology in the evaluation first considered the
sustainability of CVA processes. Specifically, this involved examining various factors, including i) whether
local public service providers adopted a project methodology that they were using as a standard system
to improve quality ofservice delivery; ii) whether national policies supported by the project were adopted
and translated into local practices, internalized within the local system, and/or widely scaled up;   iii)
whether local governments created new initiatives to respond to public service challenges identified and
quantified through project support.
Second, the sustainability of CVA outcomes: Our evaluation methodology also considered the
sustainability of CVA outcomes. Specifically, this involved examining various factors, including i) whether
there was a formulation of a legal foundation for continued implementation of project‐supported
government initiatives (including CVA working groups), which have become the reference for activity
expansion by local stakeholders; ii) whether there was evidence of government continuation of project‐
introduced initiatives combined with organizational institutional changes undertaken in order to operate
those initiatives; iii) whether increased knowledge and skills of development partners provided greater
room for local agents to create additional initiatives and enhance impacts. Secondarily, we also consider
CVA itself as a path/mechanism toward the sustainability of other/larger NJP goals.
7.3.1 Evaluation Results of Systems Strengthening and Sustainability
With an eye toward applying this general framework of sustainability to the sustainability outlook of CVA
processes and outcomes, we delineate and summarize our assessment of sustainability for CVA/NJP
systems strengthening in Bangladesh by considering the quantitative and qualitative evidence presented
in Section 6 alongside the policy dialogue evidence from Section 7.2.3 to investigate: i) sustainability as
an aspect of the particular activities and modalities of CVA programming for citizens and institutions; ii)
sustainability of service improvements, dialogue, and voice that have already been working as well as how
98
they might continue to work; iii) sustainability of linkages, committees, and institutional roles activated by
CVA; and iv) sustainability of accountability.
Summary of Results on Systems Strengthening and Sustainability:
● Sustainability is an aspect of the particular activities and modalities of CVA programming, but
the previous successes of CVA regarding citizens and institutions may not sustain.  
● Sustainability of the service improvements, dialogue, and voice have been working as the result
of CVA, but they may not continue to work.
● Sustainability of linkages, committees, and institutional roles activated thus far by CVA, while
successful, may not sustain.  
● Sustainability of accountability from CVA is unlikely to persist.
Finding #1: Sustainability is an aspect of the particular activities and modalities of CVA programming,
but the previous successes of CVA regarding citizens and institutions may not sustain.
For citizens, sustainability must mean that the expression of “voice” and “action” via CVA is not a one‐
time event. CVA must continue voice and action to follow up on action plans, continue those action plans,
and begin to work in new areas of concern.  
One could argue that the initial investment that CVA provided to activate the dormant energy of civil
society and the empowerment of the citizens that participated in the CVA process could have lasting
effects on their behavior, actions, and feelings of efficacy. This may not only help them to become better
informed and engaged citizens with respect to social accountability, but also better informed and
mobilized citizens for electoral and political accountability. It could also improve citizen participation in
other parallel structures or agencies, such as the ward shava budget meetings, VDCs, other community‐
activated committees, or within other CSOs or NGOs. These are theoretical possibilities, but they are
impossible to measure during the timeline of the evaluation.
A more likely scenario, as informed by the evaluation, is that although CVA was very successful in
“activating” civil society and people’s social networks initially, citizens will require continued motivation
and a mandate to continue social accountability activities, as was the case even during the programming
period of the NJP project.  
As a result, despite the best wishes of the citizens involved and their expressed desires to continue CVA
activities without direct NJP support, the evaluation concluded that ending CVA too soon runs the risk
that this activation will merely flame out. Although the CVA process for citizens has worked well, the
uncertainty around whether and how it will continue beyond the two year extension, along with the
suggestion that it most likely will not, means that the CVA working groups and CVA participation in
committees is unlikely to be maintained.  
Despite the outcomes that we were able to detect and measure for this evaluation, these outcomes may
be fragile and as a result the citizen aspect of social accountability may not sustain. To definitely make a
99
determination in this regard would require follow‐up research for which it would take longer than the
2020 cycle to assess given the projection on sustainability as it pertains to citizens’ actions.  
Moreover, lacking a clear theory of change as to how the participation of citizens in CVA initially, per NJP’s
theory of change, is meant to sustain social accountability longer‐term as per the systems strengthening
theories of change noted above leaves to question whether NJP has a sense themselves of whether and
how citizen voice and actions are meant to persist beyond programming. While CVA and NJP programming
in previous years no doubt provided a catalytic moment to help overcome initial inertia on the part of
duty bearers, service providers, and institutional actors ‐‐ as the present evaluation clearly demonstrates
‐‐ these programmes must be an engine that keeps running, for which NJP would need to develop a plan.
The NJP and CVA theories of change do not adequately address this currently and the activities in the cost
extension do not provide an obvious avenue to sustain these successes, but they could with some
reformulation and possible reprogramming. We return to this discussion in Section 9.
For institutions, sustainability means that overcoming inertia is also not a one‐time event; institutional
actors need to continue to interface, dialogue, and produce joint action plans or else they are likely to
attenuate. Moreover, citizens and institutional actors must work together to follow up on action plans
and maintain activated committees.  
The evaluation found that the initial investment that CVA provided in activating the dormant energy of
civil society and institutional actors to overcome the bureaucratic inertia that has long been prevalent at
all level of local, meso, and national agencies could also have lasting effects on the behavior, actions, and
feelings of efficacy of institutional actors, duty bearers, and service providers. This may not only help them
to become better institutional managers and bureaucrats, but agents of oversight and monitoring and
service provision in the service delivery process.  
The evaluation uncovered a persistent, pervasive, and seemingly intractable level of bureaucratic inertia
as well as institutional pathologies of poor governance before CVA; but after CVA, there were numerous
instances of improvements to services processes and outcomes resulting from CVA. CVA does help
institutional actors to overcome inertia ‐‐ even by their own admission.
Yet it is impossible to measure whether and how these outcomes will persist outside of the evaluation
period, with the evidence in the evaluation period providing many reasons for skepticism. As easily as
inertia can be overcome, it can reappear. The evaluation summary on institutional actors found that they
needed motivation and a mandate to work in positive ways. Although the CVA process for institutions
worked well to provide the initial motivation to overcome inertia, it is not clear whether thisis guaranteed
to continue.
First, if the CVA working group were to disband, as the CVA working group is critical for social
accountability, the source of social accountability would disappear, therefore leaving institutional actors
with no incentive to continue to perform well. This is true whether one applies the CVA and NJP theory of
change to the present realities of social accountability, but is especially true if social accountability
disappearsfrom the NJP theory of change down the line. The systemsstrengthening figures above provide
theoretical guidance, but require specific programming aspects of social accountability modalities that do
not currently exist in the 2 year extension.  
At the very least, systems strengthening would require additional thinking and planning regarding what
investments are needed to continue to support the committees that are working well, even if they will
not receive any support in the future. Yet it is important to note that although CVA outcomes for
institutions have been improved, these gains are fragile and will require more time to fully assess. As a
100
result, the risk that this progress could easily be undone remains. While CVA initially helped to overcome
bureaucratic inertia, the activities in the cost extension were often unclear on exactly what mechanisms
were meant to support institutional strengthening through social accountability. We return to this
discussion in Section 9.
Finding #2: Sustainability of the service improvements, dialogue, and voice have been working as the
result of CVA, but they may not continue to work.
We also considered the sustainability of improvements in services, dialogue, and voice that worked (as
per findings in Section 6), as well as how they might continue to work to further social accountability and
improve systems strengthening.  
The evaluation found that in addition to the incorporation of citizen voice into the feedback loop of CVA,
CVA has ensured that local service providers, duty bearers, and government officials have the opportunity
to voice their ideas aboutservice provision and systemsstrengthening while enabling social accountability
to continue to help them do their jobs. Platforms for dialogue and interface, such as those provided by
CVA, were crucial to sustaining this momentum. This should be considered as “governance” effects of the
CVA process and the ways that CVA improves governance  ‐‐  in this instance, with officials reporting
themselves that CVA provides a forum that helps improve their own processes.  
Some aspects of what local service providers must do will likely continue as they required only a one‐time
fix, as was the case regarding informing providers of monitoring standards and the need for sincerity in
the provision of services. Other simple solutions that helped to improve actions in the short‐term may
also persist in the long‐run with limited effort, such as the sharing of phone numbers and contacts of
Union‐level officials in the WATSAN committee, Union chairperson and council, and the agriculture
extension officers. As discussed in Section 7.2.3, the operation of policy dialogues that work in tandem
with other social accountability or community‐activated programming, in areas where the government
has made an investment or desires change, could mean that CVA activities build on, create synergies with,
or evolve into those structures.  
Yet the evaluation found that more transformational and persistent change may not sustain in the
absence of regular interface and dialogue meetings between officials and community people. Such
meetings do not require a large amount of financing, but they do require points of contact both on the
citizen side and the institutional side to mobilize, organize, and facilitate those meetings.  
Given a problem of resources, the evaluation found that they will need to require minimum financing for
transportation, tools, and monitoring at least in the short‐term (we address ideas for phasing this out in
Section 9). As evidence of this, the government’s own efforts to support community mobilization for
community clinics were not operational or effective because they refused to provide them with even a
modicum of resources, such as tea or snacks to serve during interface meetings. The ability to fulfill the
various points of delegation, oversight, and contact per theories of change for systems strengthening
require at leastsome platformsfor continual engagement. The USAID cost extension frequently delegates
oversight of the actions ofservice providers, government officials, and duty bearersin varioussectors, but
it is not clear what the chain of delegation is, how it works with (or without) CVA working group advocacy
and mobilization, and where the impetus for its continued oversight and monitoring will come from in the
absence of NJP. However, a simple transposition of this aspect of the CVA and NJP theory of change to
the 2 year extension could, according to the evaluation, easily help to fulfill many aspects of systems
strengthening in low cost, low resource intensive ways. We return to this discussion in Section 9.
101
Finding #3: Sustainability of linkages, committees, and institutional roles activated thus far by CVA,
while successful, may not sustain.
Third, we considered the sustainability of linkages, committees, and institutional roles that have been
activated by CVA and whether they will continue to contribute positively toward systems strengthening.  
The evaluation concluded that beyond the activation of social accountability through CVA, the various
attending linkages, committees, and institutional roles were clarified and strengthened. Yet they also
require upkeep, maintenance, and vigilance.  
As a result of CVA, institutional actors now understand their roles and responsibilities, purview for action,
and the accountability mechanisms that ensure performance. As relayed in Section 7.2.3, these activities
have created, or have the potential to create, direct and indirect points of synergy, reform, and
improvements with many of the government’s other tools, activities, and programmes.  
Yet lacking CVA, it is not clear how this could be sustained or even how knowledge‐sharing and synergies
with the government or other institutional actors would continue. Some aspects of linkages and the
institutional roles being strengthened will likely continue as they required only a one‐time fix, but more
transformational change is unlikely to persist in the absence of further solidifying linkages, the
strengthening of parallel structures and committees, and the continued oversight of institutional actors
through social accountability.  
If sustainability of linkages, committees, and institutional roles is a goal that NJP desires to sustain, they
will need to do more thinking about how the positive results from CVA can be transposed into the theory
of change for the extension. Any partnership with or lobbying of government is not a one‐time event, as
NJP’s great success thus far demonstrates  ‐‐  it requires hand‐holding, coordination, follow‐up, and
evaluation over a number of years.  NJP could continue that in the future and moving into a more advisory
or technical role, but it will not happen on its own.
Moreover, the government, even if it agrees, will not be the change agent knocking on WVB’s door  ‐‐ 
rather WVB must take the lead. While the USAID cost extension frequently mentions government units,
actors, or agencies as primarily responsible for continuing this work, this dictation defeats the very
purpose of social accountability as something that occurs beyond government structures via community
activation, voice, and dialogue. Progress on social accountability in the short‐term does not suggest that
social accountability is necessary in the long‐run; in fact, evidence from Bangladesh and other contexts
suggests the opposite ‐‐ as social accountability grows it must be fostered as just that ‐‐ “social” ‐‐ and not
simply morphed into or incorporated into formal government structures.  
The evaluation team found at the time of fieldwork that this aspect of maintenance had been overlooked
by WVB, and instead, WVB understood that social accountability would be folded into parallel structures
as part of the phase‐out as opposed to existing as its own platform (or separate and apart from formal
parallel structures). We return to this discussion in Section 9.
Finding #4: Sustainability of accountability from CVA is unlikely to persist.
Sustainability should not be a one‐time process. With CVA having started but not finished the  
strengthening of formal linkages, these processes will need time to vest and grow. The considerations
102
undertaken in attempting to sustain accountability are complicated by many of the previously mentioned
factors.  
The result of the evaluation was inconclusive with regards to whether electoral accountability will serve
to provide enough of an incentive to motivate politicians’ behavior in Bangladesh. For voters, we might
consider the involvement of citizens in the CVA working groups, VDCs, scorecard sessions, and interface
meetings as improving the information, oversight, and efficacy of citizens, thus enabling them to become
more informed and mobilized citizens that are better able to vigilantly pursue political accountability
(including voting efforts). The evaluation also revealed a few instances of CVA working group members
desiring to run for ward representative (as communicated by a female CVA working group member in
Dacope during the inception visit), and becoming potentially more active in other parallel structures to
which their experiences in CVA could help to bolster the accountability of ward shava budget meetings,
Union committees, etc.
For institutional actors and politicians, even if CVA has provided such an incentive to perform better in
the face of political accountability, it would likely only be effective during campaign periods. Therefore,
we might expect increasesin performance and attention to service provision in the run‐up to election day,
but attenuating thereafter.  
Moreover, the short‐term benefits that have accrued as a result of CVA that induced improvements in
institutional accountability within governmentsystemsshould not replace non‐governmental/civilsociety
sources of social accountability, which remain necessary and will likely require both support and
facilitation, alongside formal political accountability. Though CVA processes also assist accountability in
larger ways, these gains may prove fragile as outcomes for accountability are still in their nacency and will
thus need significantly more time to vest and evaluate (e.g., process on legalization and national policy
dialogue will take time to negotiate, vest, follow up on, etc). It is unlikely that civil society will be able to
persistently demand accountability without mobilizing resources from CVA or similar social accountability
platforms, though the evaluation period did not allow the formation of definitive judgmentsin thisregard.  
Even where the evaluation found specific expressions of interest for strengthening partnerships between
the government/line departments and CVA/NJP, as well as general aspects ofsocial accountability ‐‐ these
expressions were often made with the understanding that NJP/WVB would need to take the lead in
developing and strengthening the partnership. Though as mentioned in Section 7.2.3, some opposition or
misunderstanding of CVA and the role of social accountability persists in line department offices.  
The evaluation concluded that while CVA provided ways to strengthen accountability from within
institutions and institutional actors, it is constructed around social accountability practicesthat also reside
beyond institutions. Therefore, even if NJP produces improved formal institutional accountability with
regards to government agency operations, and even if formal political accountability is improved as
democratic elections become more competitive, free, and fair ‐‐ allowing citizens to reward and punish
politicians through the ballot box ‐‐ social accountability and its importance will never be eclipsed even if
it changes.  
Even so, CVA is unlikely to provide long‐term and automatic systems strengthening to ensure
accountability, whether formal or informal  ‐‐ it is always likely to recede as the work of CVA working
groups declines. This is the lesson to draw from trying to map the CVA theory of change to a theory of
change aimed at systems strengthening and sustainability. As currently understood, USAID’s hope for
long‐term accountability through sustainability and systems strengthening fails to specify how this is
meant to be achieved; and instead, suggests it will not give a mandate for CVA. If that remains, we should
103
not expect the positive results on accountability from CVA to sustain, although we expand this section in
Section 9.
Summary of findings on systems strengthening:
Despite the many positive findings of CVA in the evaluation, the application of these results to systems
strengthening is harder to assess with any certainty. Even so, impressions suggest that many of the
positive results gained from CVA are unlikely to persist with regards to processes and outcomes. Citizens
and institutions have improved processes, but without further support, these gains will likely attenuate.
Service improvements, dialogue, and voice are working, yet none of these mechanisms are likely to
endure without furthersupport. Further, the evaluation questions whether the linkages, committees, and
institutional roles already activated by CVA, along with the accountability mechanisms, will sustain. We
build on these insights and conclusions in our list of recommendations enumerated in Section 9.
  
104
8. Discussion
8.1 Did the evaluation meet its objectives?
With the results as presented in mind and returning to the evaluation questions, we now summarize the
findings from the data collection in response to specific lines of inquiry and sub‐questions. While the data
collection did not provide the opportunity to alwaysfocus on each of these equally or to the same degree,
and not all are equally relevant toward assessing impact and effectiveness, we nonetheless believe the
evaluation yielded enough data to provide at least some summary conclusions for each objective.
1. To what extent have community health clinics met at least one or more additional minimum service
standards and did CVA contribute?
→ Both the quanƟtaƟve indicators and qualitaƟve evidence suggests consistently positive results across
all program areas, and that now community clinics are meeting the monitoring standards at a much higher
rate than before CVA; various CVA activities contributed to this.
1.1. What percentage of program health clinics met at least once or more additional minimum policy
standards?
→ Per the quanƟtaƟve indicators, nearly all of the community clinics met and achieved minimum policy
standards. This accords with the qualitative assessments provided in Section 6 by the CVA working groups
and confirmed by community clinic staff, and has been improved by Upazila, divisional, and national
dialogues on community clinics.
1.2. What is the average number of additional service standards met by health clinics, WASH, and
agriculture committees/extension workers at the evaluation period as compared to the baseline/initial
monitoring standards session?
→ Per the quanƟtaƟve indicators; community clinics, Union WATSAN commiƩees, and Union agriculture
services units all improved the average number of additional service standards, with the most robust gains
found consistently with community clinics and some additional challenges for some of the Union WATSAN
and Union Agriculture units in some unions.
1.3. Are there patterns in improvements and lack of improvements? Is there variation in change based on
geographic location or the domain targeted? (e.g., more success in obtaining additional agricultural
extension workers than in addressing stock outs in health facilities).
→  There are no discernible patterns of improvements in community clinics; water and sanitation;
agriculture extension; and no variation on geographic location or domain targeted.
1.4. Is there evidence of system strengthening, changing power dynamics or women’s empowerment?
→  There is posiƟve evidence of systems strengthening in numerous ways that have supported service
provision during the project period; the evaluation could not determine if these improvements are likely to
sustain, with some indications that much of this evidence is not likely to be maintained if CVA is phased
out.  
→ There is posiƟve evidence of changing power dynamics to make service providers, government officials,
and other relevant stakeholders more accessible, sincere, and less subject to capture as perceived by
community people.
105
→ Women's empowerment improved through parƟcipaƟon in CVA and both men and women (community
people and duty bearers) report becoming more sensitized to gender issues.
1.5. What role, if any, did government decision‐makers and other duty bearer groups (including
politicians) play in achieving service improvements, additional minimum policy standards or other
governance changes in clinics/WASH/agriculture?
→  Government decision‐makers and other duty bearers played pivotal roles in achieving service
improvements. Further, they achieved minimum policy standards (as per quantitative indicators) and
important other governance changes which can be described as more inclusive management for
community clinics with contributions from the health committee, for water and sanitation (including
constituting the WATSAN committee), and deploying adequate manpower for agriculture extension
services. Government participation in Upazila, divisional, and national dialogues was an important aspect
of the government’s contribution to improved services.
1.6. Is there evidence or to what extent could it be said that responsible government decision makers
responded to community lobbying/pressure through the activities?   
→  There is significant positive evidence that responsible government decision makers responded to
community lobbying and pressure brought about by CVA activities at all levels, from interface meetings up
to dialogue sessions at various administrative units.
2. In what ways, if at all, did CVA activities affect health, WASH, and agriculture service users and
providers?
2.1. What changes regarding health service quality and standards did users and staff observe? How has
this affected them?
→ Numerous posiƟve changes can be documented regarding health service quality and standards. This
affected them by helping users demand better services and staff supply better services, often as the result
of interface and dialogue.
→ Numerous posiƟve changes can be documented regarding water and sanitation service quality and
standards. This affected them by helping users demand better services and staff supply better services,
often as the result of interface and dialogue.
→ Numerous posiƟve changes can be documented regarding agriculture extension services and standards.
This affected them by helping users demand better services and staff supply better services, often as the
result of interface and dialogue.
2.2. Did the program produce any unexpected outcomes, positive or negative? Who was affected and
how?
→ A set of outcomes that could be perhaps considered somewhat unexpected was the level and quality of
participation by stakeholders at Upazila, divisional, and national dialogues where they occurred. From this,
the positive returns to systems strengthening and sustainability through the policy dialogue and actions
actually undertaken by NJP leveraging CVA processes and outcomes, while not proposed in the NJP theory
of change with much specificity, nonetheless did contribute positively in ways that were perhaps
unexpected.
→ The evaluaƟon did not reveal negaƟve outcomes.
106
3. To what extent were major program components achieved?
3.1. What role did local leaders (political and bureaucratic) play? Were there any differences in the
responses of local leaders? Were they receptive or hostile to the approach? What were some of the
political and social factors that might have affected their response? To what extent did they make
decisions in response to the project activities?   
→ Local poliƟcal leaders played an important role in the interface and dialogue sessions by first hearing
demand articulation by community people, and then working with CVA modalitiesto approve new funding
and devotion of resources to improvements of service delivery, including better monitoring and oversight
of officials at various levels. Local bureaucratic leaders played a role by more effectively monitoring
individual service providers, government officials, or others within their purview.  
→ Differences in responses by local leaders can be summarized as varying in the degree to which they
themselves became directly involved in all manner of interface and dialogue, or whether they delegated
this to others.  
→  In some cases there was some iniƟal uncertainty and passive resistance (not outright hostility) to
hearing community perceptions and participation, but this grew to become more sensitized over time.  
→ We idenƟfied poliƟcal factors that may have influenced their response as coming from the involvement
of officials across departments and levels at the Upazila, divisional, and national dialogues, and social
factors reflecting aspects of shaming, oversight, and accountability that arise from CVA  ‐‐  officials,
politicians, and bureaucratic agents wanted to overcome possible socialsanctioning if they did not perform
in their duties.. The evaluation demonstrated many instances where they made decisions in response to
the project activities.  
3.2. Did program teams, community or government stakeholders implement additional or innovative
strategies or activities to achieve the program goal?
→ Yes, parƟcularly at the level of CVA working groups and NJP field staff working on ways to acƟvate and
then sustain CVA working groups; as well as with numerous policy dialogues and commitments and
changes that came from Upazila, divisional, and national dialogues and policy actions worked on with NJP
and relevant government ministries and stakeholders.
3.3 Given the findings of the formative research, have the functioning of standing committees in
NJP target areas improved? If not, why not? If so, to what extent did NJP activities contribute to
improved functioning? Is there any evidence that standing committees, supported and trained by NJP,
played a role in improving services?  
→ CVA and NJP acƟviƟes played a very important role contribuƟng to improved funcƟoning of standing
committees as compared to the formative research. Evidence that standing committees supported and
trained by NJP playing a role in improving services was documented in numerous instances.
4. In what ways did the program contribute to the sustainability of project outcomes?
4.1. What evidence, if any, indicates that communities and local government will be able to maintain or
expand project benefits, especially improved service quality and standards in the program health
clinics/WASH/agriculture?
→ Evidence that indicates that communiƟes and local governments are able to maintain or expend project
benefitsis mixed ‐‐ certain gains are probably lasting and others could persist, but withoutsome continued
level of support for CVA working groups, local officials, and coordination with parallel structures, these
107
gains will likely decline; however, aspects of policy dialogue have allowed some gains to “trickle‐up” in
ways that are hopefully likely to persist longer term, while other aspects of this bottom‐up approach
require more direct support at meso and national levels to continue.  
4.2. Given the findings of the formative research on elite capture of participatory processes such
as ward shava, has this persisted or is there evidence of greater inclusion of citizens?  
→ The evaluaƟon found improvements to the idenƟficaƟon of and service delivery to beneficiaries, and
only a few instances of elite capture persisting (elite capture was not a significant problem identified in the
evaluation).
4.3. What else, if anything could have been done to strengthen the sustainability of improved service
quality and standards in program health clinics/WASH/agriculture?
→  The evaluaƟon found numerous instances of CVA working groups, NJP field staff, and government
officials confronting a number of perhaps unanticipated aspects of CVA programming that they often
addressed in creative and context‐specific way to help with sustainability and improving the quality of
services (Section 9 includes a longer discussion on recommendations).
4.4. Given existing weaknesses of current institutionalized social accountability spaces and entities (as per
the NJP formative research, e.g, standing committees, ward participatory budgeting, etc.), what do
national and local leaders think of the CVA approach? Do they see opportunities for elements of CVA to
be institutionalized by the government, and which elements do they deem feasible? Given a lack of clear
definition of sustainability or specified theory of change on systems strengthening, what avenues for
policy dialogue and action did CVA and NJP pursue (even if implicitly) that provides evidence on systems
strengthening?  
→  Local leaders expressed support for the CVA approach (parƟcularly at Unions and Upazilas), while
national leaders had a more mixed view. Some national leaders either supported CVA specifically or social
accountability generally, and avenues for partnership were identified to support government efforts on
civil society incorporation into service provision and monitoring. Others were not knowledgeable about
CVA, were more dismissive of the importance of incorporating community participation or social
accountability into the service delivery process, or were simply unaware of the local activities of members
of their own ministries in the NJP project areas.
→ Elements of CVA could be insƟtuƟonalized by the government through the legalizaƟon and registration
of VDCs, CVA working groups, or more explicit coordination with parallel structures and relevant
committees or other government‐mandated civil society participation.
→  NJP and WVB used parƟcipants and evidence from CVA in many Upazila, divisional, and national
interface and dialogue meetings, which resulted in many direct cases of changes in budgets, priorities, and
legislation. Other aspects of linkages, knowledge‐sharing, and advocacy directly and indirectly resulted in
behavior and policy changes on the part of institutional actors and government officials.
  
108
9. Conclusion and Recommendations
9.1 Summary
The evaluation concludes with a number of recommendations established with regards to the results,
analysis, and conclusions.
To summarize the results of the evaluation, we conclude that CVA in NJP has met many of its objectives
and indicators while producing a variety of positive results. These include improvements to citizens’ voice
and action, institutional responses, and accountability to improve services at the community clinic, in
water and sanitation, and with agriculture extension. While CVA has worked in many ways and provided
a net positive impact, the evaluation as designed cannot definitively determine and isolate the  aspects
most critical aspects to CVA’s success apart from the “whole package” of CVA programming in NJP. That
being said, numerous mechanisms and factors have been highlighted throughout that individually and
collectively are seen as having produced an effect. At the same time, while the combination of
quantitative and qualitative evidence from Section 6 along with the discussion of evidence of policy
dialogue and systems strengthening in Section 7 lend some hope that some aspects of CVA and social
accountability may prove sustainable beyond the project cycle, the evaluation also found that some of
the positive returns are unlikely to be maintained on their own.
The evaluation team’s interpretation of the 2 year extension, per USAID project documents, show that
CVA is not currently and explicitly mentioned in the 2 year extension. As a result, CVA would have to be
made sustainable by the end of the 2020 project year to be made sustainable (or otherwise through the
bridging period due to the Covid pandemic). Per the discussion in Section 7 on systems strengthening, the
evaluation concluded that evidence for sustainability at present is mixed  ‐‐ while policy dialogue and
action has been healthy and already resulted in many noticeable changes and revealed opportunities for
future linkages and advocacy work, these positive resultsrun the risk of attenuating without more thought
to NJP and WVB’s programming through the bridging period and extension phase. Moreover, per the
discussion in Section 7, USAID themselves have not been clear in regards to what they mean by
“sustainability” at a global level, nor at the project level involving NJP, including for the phase‐out now
and after the extension. With this in mind, we conclude the evaluation here in Section 9 by highlighting
some important considerations for plans of action for future sustainability followed by recommendations
to stakeholder groups that are proposed by the evaluation team and reflect their advice drawing on their
conclusions of the evidence in the evaluation.
9.2 Considerations for Plans of Action for Future Sustainability
As the evaluation’s design cannot definitively determine and isolate the most critical aspects of what
made CVA’s success apart from the “whole package” of CVA programming in NJP, the evaluation could
not isolate a single mechanism as the most important factor to consider in the development of plans of
action aimed at attaining future sustainability. Any change to one aspect of CVA runs the risk of initiating
a “jenga” game wherein it would not be clear to the player which pieces are most important to keeping
the structure standing, while the removal of any core piece would run the risk of toppling it over. CVA
seems to be a core piece of what has made NJP work and CVA is composed of many distinct activities and
processes that nonetheless exist concurrently, sequentially, and interactively. These realities complicate
efforts to easily identify the aspects of CVA that definitively worked and thus should be considered for
future sustainability and included in plans of action, and those that definitely did not and thus should not
be considered.
109
With those caveats in mind, we therefore first summarize general thoughts for future sustainability in this
subsection before turning to specific considerations for a plan of action for key stakeholders in the next
subsection.
Considerations for Plans of Action for Future Sustainability:
● Whether or not NJP’s theory of change should be amended after accounting for the results of
the present evaluation and CVA sustainability concerns, and if so, how?
● Timing, which is of the essence ‐‐ how to manage sustainability?
● Phase‐out versus continuation ‐‐ how should CVA be managed through the bridging, phase‐out,
and extension periods by NJP and WVB?
● Whether to seek input and buy‐in for sustainability from CVA participants, and if so, how?
Consideration #1: Whether or not NJP’s theory of change should be amended after accounting for the
results of the present evaluation and CVA sustainability concerns , and if so, how?
A first consideration involves whether or not NJP’s theory of change for the 2 year extension should be
amended after accounting for the results of the present evaluation and CVA sustainability concerns, and
if so, how?  
Taking this in cascade order, one might first consider whether there is a theory of change for NJP that
continues in which social accountability either does not or should not play a role, and if not, whether that
would work to still meet NJP objectives. That is, one option is to completely ignore the present
evaluation’s findings with respect to the new theory of change, or similarly, to consider them but refrain
from making any amendments to the theory of change.  
Next, should amendments be considered for the new theory of change, one might think about whether
that theory of change is one that social accountability can be made a part of, but which does not
necessarily involve CVA‐specific activities. That decision could leverage the results of the present
evaluation to edit the theory of change, but with either an expanded or differently scoped social
accountability approach that does not replicate CVA activities as they have been conducted in the past
and present. This could involve either an evolution turning the CVA tool into either a new tool, a different
tool altogether, or a different framework for social accountability and how it works in Bangladesh. This
decision would similarly need to be measured against the ability to maintain the objectives of NJP.
Finally, one might consider a theory of change that includes CVA as a component of the future of NJP. This
would require amendments to the current theory of change as specified in the 2 year extension. If this
decision were taken, the results of the present evaluation would likely be important to leverage with
respect to “lessons learned,” not only regarding what worked and did not work for CVA, but also for
reference when scaling and building further linkages with the government at meso and national levels.
Moreover, it would help to inform what CVA activities to keep, change, or add, and what policy advocacy,
knowledge sharing, linkages, and advocacy to pursue. Much of the evidence presented in this evaluation
suggests that to sustain the positive benefits from CVA for social accountability, the continuation of CVA
activities need not necessarily mirror what has happened since 2018 ‐‐ although that is one option. A more
likely scenario would instead require the continuation of specific forms of facilitation, training, and
110
checking‐in with NJP staff and partner NGOs rather than the maintenance of a more “heavy‐handed”
mobilization tool involving consistent oversight and monitoring. It could also include a heavier focus on
policy dialogue, advocacy, and linkages in a technical or advisory role at meso and national institutional
levels for NJP and WVB staff. The output from CVA working groups and the degree to which it has “tickled
up” to other stakeholders suggests these are possibilities, prompting a more thorough exploration of
these possibilities below. Similar to above, this plan would also need to be measured against the ability
to meet NJP objectives.
Consideration #2: Timing, which is of the essence ‐‐ how to manage sustainability?
Regardless of the decision made about whether or not to amend the theory of change, the evaluation
revealed very clearly that an important consideration is timing, which is of the essence, considering the
bridging period due to the Covid outbreak and recent cyclone activity in 2020. That is, the evaluation
found that regardless of whether NJP desiresto continue with CVA,social accountability, or neither, many
CVA working group members, government officials, partner NGOs, and WVB field staff expressed anxiety
about the lack of decision‐making on this front given the uncertainty as to the timing and process for the
decision. They expressed further uncertainty with regards to how it would or would not be brought into
the 2 year extension, as well as whether or not they should be empowered to go it alone. The evaluation
team did not find a clear phase‐out plan nor any mention of CVA for the extension, but the bridging period
would grantsome extension in the timeline to work some of this out. Moreover, NJP field staff themselves
were unsure while CVA working groupslacked guidance. As a result, this processrunsthe risk of becoming
demotivating given a perception that it is all talk but no action in addition to requiring a mandate from
NJP. Since the fieldwork was conducted in early 2020 and events for progress on CVA, NJP, and all manner
of development and donor‐related activities are on pause because of the global pandemic and cyclones,
WVB has some additional time for these considerations. Even so, the evaluation recommends these
decisions are made soon.
Consideration #3: Phase‐out versus continuation ‐‐ how should CVA be managed through the bridging,
phase‐out, and extension periods by NJP and WVB?
If the decision is to phase‐out CVA specifically and social accountability generally, a few actions will be
required. If the decision involves potentially allowing CVA, aspects of it, or aspects of social accountability
to continue, it is not just a question of whether CVA should continue, but also a question of in what ways
and how this would need to be realized.  
Accordingly, WVB will need to answer a number of important questions, such as: would this mean a
change to the program or would it require asking for additional money (is that even possible)? Or, does it
mean keeping the programming that is there, but incorporating CVA more holistically either without
changing, or reallocating, total funding? Furthermore, how can phase‐out be managed now (including
during bridging period) vs. during the 2 year extension if sustainability is the goal, and what could be done
with no additional funding?
It is important to note that the evaluation found that many of the activities included in the 2 year
extension as submitted to USAID echo CVA activities, though no explicit mention is made regarding the
specific mechanisms or actors that will be made responsible as change agent to achieve these objectives
(moreover, it often does not mention a responsible actor). However, CVA working groups could perform
many of the activities described either through explicit incorporation into that programming, or through
a variety of low‐cost, low‐tech “light touch” improvements following some additional technical guidance
from NJP, WVB, and partner NGOs. Additionally, the monitoring and evaluation plan for the 2 year
111
extension echoes many of the ways that CVA activities monitor and evaluate service provision, yet here
again, the planning documents themselves do not explicitly mention CVA or social accountability with
regards to monitoring and evaluation or specific CVA or social accountability indicators (such as those
gathered in the CVA database or as per the qualitative data collected for the present evaluation).
Lots of ideas on how to answer these questions were communicated to the evaluation team by CVA
working groups, partner NGOs, and WVB field staff during data collection. Rather than relay all of those
findings here, the evaluation team recommends that WVB considers this in their deliberations prior to
then seeking specific input from selected CVA working groups, partner NGOs, and their field staff. This
could be actioned through a variety of mechanisms and channels, including focus group discussions,
working group sessions in the field or Dhaka, consultation with other NJP staff or WVB staff, and
participation/guidance from relevant local, regional, and national stakeholders, including government
officials at Union Parishad, Upazila, divisional, and national levels. The evaluation team can provide
additional guidance and is willing to participate in any way deemed useful by WVB.  
  
Consideration #4: Whether to seek input and buy‐in for sustainability from CVA participants, and if so,
how?
Field staff at all levels and CVA groups have a number of good ideas on how best incorporate the findings
of this evaluation with an eye toward longer‐term sustainability. However, the evaluation found that they
are lacking a mandate with regards to both the plan for the future as well as their current plan (CVA  is of
course just one aspect of NJP). The evaluation team wantsto stressthat their perspective isthat that does
not mean that NJP or WVB need finalize everything for CVA participants, but it does suggest potential
benefits in streamlining the incorporation of their insights and participation in planning at both a very
level and day to day level now, in the near future, and in the long‐term, especially given the delays due to
Covid and cyclones. This would also provide room for them to more cogently expresstheir ideas and plans.
In addition, it would help organize them in a managed way that would aid in their understanding that if
CVA does continue over the next two years it would not simply be a continuance of past efforts. Rather,
this would ensure that they are explicitly told about operational changes and made aware of the bridging
period and that the 2 year extension’s top priorities would be focused on thinking through, experimenting
with, and testing different ideas to promote long‐term. However, to accomplish this, they first need a
mandate from WVB as well as parameters from which they must discuss and decide independently and
with NJP facilitation. NJP and WVB could then work with this feedback and these plans to incorporate
them into their broader goals and objectives during the bridging and 2 year extension, having already
outlined how CVA group sustainability fits into NJP’s own theory of change.  
9.3 Recommendations   We conclude in this subsection by listing a number of recommendations coming out of the evaluation as
summarized by the evaluation team for the various stakeholders for NJP and WVB.
9.3.1 To NJP/WVB  
To NJP and WVB, the evaluation team provides four principle recommendations.
Recommendation #1: Discuss and decide whether CVA or a social accountability tool should be in 2 year
extension and then how that decision affects the 2020 phase‐out (and bridging period).
Following on the “considerations” enumerated in subsection 9.2, the evaluation team recommends that
NJP and WVB discuss and decide whether CVA, or some social accountability tool, should be in the 2 year
112
extension (and in what form). Based on the outcome of this deliberation, it should then be determined
how this decision affects the 2020 bridging and phase‐out periods.  
If NJP/WVB take a decision of “no,” we recommend focusing on phase‐out now during bridging, which
still requires a clear mandate and communications with field staff and CVA working groups. This will need
to be integrated into the project timing that has now been affected by the Covid outbreak, which of course
affects NJP’s ability to reach and work alongside field staff and CVA participants.  
If NJP/WVB take a decision of “yes” or “maybe,” we recommend focusing on how the long‐term plan links
to the short‐term phase‐out period so that both provide credible plans of action now and in the future to
meet NJP goals (as elaborated above in subsection 9.2).12
Recommendation #2: If so, adapt the theory of change for the sustainability phase for CVA to remain
either explicitly or implicitly in NJP programming.
If the answer to recommendation #1 is yes or maybe, we recommend that NJP and WVB adapt their theory
of change for the sustainability phase to determine how this might incorporate CVA either explicitly or
implicitly, following on the discussion in subsection 9.2. The sustainability theory of change does not
include a Purpose Four, but NJP could nonetheless integrate CVA across aspects of the theory of change,
including specifying the correct indicators in the monitoring and evaluation plan, integrating these with
BHA indicators, and developing custom CVA or social accountability indicators. This requires figuring out
what needs to be changed at a programming level (and whether or not that requires new spending, or
rethinking/rephrasing on old spending), how this can be properly articulated to donors, and how the
extension monitoring and evaluation framework fits into the framework for the extension. This remains
particularly relevant as much of what is proposed in the latter is something that the former already does
and does well, and the bridging period allows some flexibility in timing to work this out. The evaluation
team would be happy to contribute further in this regard if deemed useful and appropriate by WVB.
Recommendation #3: better clarify CVA, and CVA indicators, in NJP programming.
While WVB has a monitoring and evaluation team dedicated to NJP at the national level, the evaluation
framework for NJP overall does not include anything specific for CVA programming or CVA custom
indicators. The only indicators that WVB was keeping track of are those agreed to by USAID, and then the
indicators recorded in the CVA database. While some of the other monitoring evaluations could plausibly
be linked to CVA, for example, the presence of monitoring standards at community clinics, these are not
currently linked to CVA tracking by that NJP team at the national level. Instead, in the field visit to the
regional offices, the evaluation team learned that the CVA coordinator has been keeping tracking tools
for CVA since they launched, including their presence and location, mobilization activities, programming,
outputs, and changes over time. These are then input into a CVA monitoring dashboard developed by WVI
and deployed in other contexts, but must be manually generated into reportsfor output by backend users,
nor does it appear to be consistently included (or systematically so) in NJP programming or reporting.
Integration of CVA database indicators with other NJP indicators could be a fruitful area for project
synthesis, especially regarding sustainability. Moreover, in its framework for sustainability, USAID
12 It is worth mentioning by way of background that it was not apparent to the evaluation team at the time of the
inception visit that WVB had precluded further financial or programmatic support for CVA in the 2 year program
extension in their application to USAID submitted in July 2019 for NJP. This did not become apparent to the
evaluation team until the field work visit in January 2020.
113
specifically mentions the importance of and recommends indicators for monitoring and evaluation for
sustainability.
The evaluation team had a discussion regarding what CVA custom indicators might look like qualitatively
with respect to monitoring whether or not people are demanding their entitlements (including changes
in whether people are becoming more informed about what those entitlements are); whether direct
influence of CVA working groups has occurred; whether and how the network around community action
has been built; and whether both demand and supply side actors share positive reports from interface
meetings. The evaluation team recommends centralizing and streamlining this process, but encouraging
NJP to work with the field staff to develop custom indicators from this report.
Moreover, the evaluation team recommends that WVB considers whether there is scope for NJP to
conduct a second‐round of perceptual surveys of the public regarding service delivery and outcomes in
the project areas to track changes to compare against baseline measures from subsection 6.2.  
Recommendation #4: After re‐evaluating the sustainability theory of change, develop ideas on scaling.
The evaluation team encourages the WVB leadership to leverage the results of this evaluation to develop
a guide on the best practices of CVA in order to scale‐up social accountability and CVA with NJP (as
discussed in subsection 9.2, assuming that decision is taken) and beyond NJP to social accountability in
Bangladesh or other contexts. The findings here may be relevant for multiple teams, including resource
development teams in country and WVI, who can leverage the evaluation findings to be more intentional
in how WVI integrates CVA as a core model across future grants. This could involve promoting the
understanding of CVA influences, enhancements, and contributions in development sectors as well as
how to translate these lessons learned into “something bigger” capable of involving government officials
and enabling them to gauge whether CVA can be of value to the work that they do. The hope could be to
propose CVA (and possibly other relevantsocial accountability models) at a higher level beyond the extant
project areas in NJP (this could coincide with the lessons learned from the WVB‐sponsored CVA approach
in non‐NJP areas, including setting up VDCs in other projects). All told, WVB leadership reported that they
are already taking CVA‐related activities to scale and thus view the evaluation as an important step in
providing evidence to support and inform this. WVB could therefore play an important role in technical
advising and capacity‐building for CVA and social accountability in Bangladesh, and devote some of the
human resources developed in NJP to contribute for that purpose.
At the same time, in its recommendation, the evaluation team stresses the importance of recognizing
numerous contextual, institutional, and program specific challenges towards the idea for scaling. CVA
must work to link extant governance structures capable of strengthening to understand how to
institutionalize both horizontal and vertical accountability mechanisms so that they do not solely arise on
an ad‐hoc basis (see also Section 4). This involves pressuring the government and developing an internal
(to NJP and WVB) action plan for recommendations to make at various levels (including and especially
government representatives at the national, district, and sub‐district level) to see the modalities and
benefits of the CVA approach alongside of, and in conjunction with, the rollout of the other development
programming and financing the government has already committed to; how to make the government
more open to social accountability models (the existence of the standing committees and the ward shava
and open budget meetings show that the government is at least somewhat open to social accountability,
but the extent of their willingness and their genuine political will moving forward are unclear); and how
to convince the government of the importance of inserting social accountability into governance and
governance structures beyond CVA/NJP. Linkages are necessary for the government to establish a system
capable of feeding the results of local CVA type approaches into policy at the national level. As reported
114
by WVB, right now, the central limitation remains that the government lacks a structure for it. In other
words, they do not know how to “think about it” properly. Thus the challenge for WVB becomes
determining how to articulate these lessons while mobilizing the government’s thinking.  
9.3.2 To CVA Working Groups
To CVA Working Groups, the evaluation team provides 3 principal recommendations.
Recommendation #1: develop a clear mandate and plan of action for phase‐out & 2 year extension.
To CVA working groups, we recommend that they work to develop a clear mandate and plan of action for
phase‐out and 2 year extension regardless of the decisions made by NJP, WVB, and USAID. Importantly,
they should consider what more support they need during phase‐out, what they would request for the
extension if possible, and what their plans are for post‐extension sustainability. Rather than wait, we
recommend they do so by producing some short‐term (until end of 2020), medium‐term (through 2022),
and long‐term planning. Moreover, rather than wait for a series of directives from WVB or USAID, we
recommend that CVA working groups start to work on this themselves under the direction of partner
NGOs and NJP field staff to the best of their abilities. The inability of community gatherings to occur in
Bangladesh because of Covid‐19 has no doubt put these actions on hold, but CVA working groups could
begin to prepare for this once travel restrictions are lifted.
Recommendation 2: general advice in terms of how to think through CVA working group sustainability.
If the decision involves potentially continuing with CVA, continuing certain aspects of it, or maintaining
other aspects of social accountability that CVA working groups might morph into, it is not just a question
of whether CVA should continue but also a question regarding what ways it should be maintained and
how this would need to be realized. By now, CVA working groups should be empowered to plan along
these lines while deciding for themselves what they want to do and how they would continue in the
future. While we do not have strong recommendations for the form or function CVA working groups
should take, we do provide some general advice to follow in terms of how they might think through it.
For example, let us assume that CVA Working Groups will continue to be supported in the 2 year
extension, but at a reduced financial and organizational level in comparison to the past 3 years. NJP could
direct their field coordinators to work with staff and CVA groups with a much leaner structure during the
cost extension on the following to sustain in the two years and beyond:
i) first, finish outlining goals for 2020 + develop goals for 2020‐22; that is, develop an “organizational
plan”;
ii) second, figure out what financial resources are absolutely required from NJP in 2020‐22 and determine
how the CVA working group can begin to save and develop sustainable financial resources; that is, develop
a “business plan”;
iii) third, figure out what training and capacity building they need now as well as what they desire to have
between 2020‐22 to sustain beyond that, including determining what they can do now on their own vs.
figuring out what additional support they will need from NJP to reach the new capacities they desire; that
is, develop an “operations plan”;
115
iv) fourth, determine what monitoring, coaching, and follow‐up they desire from the partner NGO and
NJP in 2020‐22 that is lighter in its footprint and limited to capacity‐building, training, and coaching; that
is, develop a “monitoring and evaluation plan.”
With this in mind, NJP and WVB could then work with this feedback, provided by the CVA working groups,
along with these plans. This may also include deputizing the role of “Systems Strengthening Specialists”
to facilitate CVA working group meetings (remotely as possible similar to digital social behavior change
[SBC] sessions). This would enable them to incorporate these provisions into their broader goals and
objectives with the 2 year extension while having already outlined how CVA group sustainability fits into
NJP’s own theory of change, organizational, business, operations, and monitoring and evaluation plans.
In any event, focusing CVA working groups on developing organizational, business, operations, and
monitoring and evaluations plans and planning should be fruitful in efforts to plan for sustainability.
Recommendation 3: consider creating CVA working groups at the Upazila level.
The evaluation found that many CVA and Upazila officials expressed a desire to have an equivalent CVA
working group at the Upazila level (building on the strengths of Upazila dialogues), which may be an
important consideration for future policy dialogue, systems strengthening, and sustainability and
possibility for CVA working groups at the Union level to think about to replicate their model at the Upazila
level. As Upazilas have larger budgetary scopes, manpower, and oversight capabilities, many respondents
saw mobilization of Upazila into social accountability as an important aspect of future improvements to
service delivery. Moreover, the government line ministries often mentioned gearing many of their
activities, efforts, and reformstoward Upazilas, as opposed to unions, including the training efforts guided
by the Local Government Department.
9.3.3 To VDCs, NGOs, other committees, and community people
The evaluation uncovered policy dialogue between WVB arising from CVA that lent recommendations on
some actions to consider for VDCs, NGOs, other committees, and community people.
Recommendation #1: regarding VDCs, consider sustainability of social accountability and other aspects
of CVA as an important dimension.
The evaluation found that WVB had instituted policy dialogue on VDCs with the central government. For
example, from 2020, VDCs were highly engaged in supporting responsesto the Covid crisis. VDC members
are part of a task force at sub‐district and ward levels to support the government’s response to Covid,
which demonstrates greater acceptance of VDCs by other stakeholders as well as indication of greater
ownership of VDCs by community people. While the evaluation noted both support and pushback of
legalization of VDCs within some line departments, we recommend that if VDCs are legalized, that they
are done so from WVB’s perspective as an aspect of making social accountability sustainable and
incorporating insights from the present evaluation and learning from CVA as an aspect of future VDC
activities. If they are not officially legalized, they still may form an important aspect of sustainability of
social accountability (and could be scaled beyond NJP project areas).
Recommendation #2: partner NGOs and other aspects of civil society can play an important role in
sustainability.
The partner NGOs will need a mandate from NJP and WVB about whether and in what form CVA and social
accountability may continue with NGO participation in the future. Even so, partner NGOs could be
leveraged to help train, capacity build, and mobilize CVA working groups or associated structures to help
116
with sustainability in the future. Moreover, these NGOs could attempt to work as liaisons between the
civil society activation coming from CVA and NJP with the civil society actions undertaken by the
government (for example, local health committees).
Recommendation #3: forging better linkages between CVA working groups and parallel structures for
sustainability.
The evaluation found numerous instances of CVA working groups supporting the activities of parallel
structures, such as Union WATSAN committees, ward shava and budget committees, health committees,
union parishad standing committees and the like. Because these parallel structures are mandated by the
government, but were often inactive until CVA mobilized community people to press for their activation,
now that these committees are up and running, they should work to forge strong linkages that will last
with the CVA working groups, or continue to develop their own modalities for better incorporating the
participation of local communities in their activities. And they should not wait for mandates from the
government, but rather use the lessons learned from CVA to begin to work on that now, perhaps with
coordination of partner NGOs and NJP.  
Recommendation #4: community people should continue to express “voice and action” in whatever
manner that takes to sustain oversight of government service delivery and maintain accountability.
Whether through CVA working groups, ward shava and budget meetings, scorecard sessions, interface
and dialogue meetings, or the like, CVA and NJP have helped to activate civil society in project areas. This
activation helped to improve the service delivery process and maintained better accountability from
service suppliers to service seekers. Citizens and communities in Bangladesh should continue to
participate in local governance to the best of their ability, and build on the spark created by CVA. They
should not wait for NJP, the government, or other donors to inspire this action, but instead, CVA working
groups should spread the word and distribute the positive results of their work to their communities and
other communities so that social accountability grows in Bangladesh.
9.3.4 To WVB and the Government of Bangladesh
To WVB and the Government of Bangladesh, the evaluation team provides three principal
recommendations.
Recommendation #1: continue to build on the successful work of policy actions and dialogue.
The evaluation documented numerous evidence of positive policy actions and dialogue across
institutional levels, offices, and actors directly and indirectly from CVA and the government as reviewed
in Section 7. The evaluation team recommends to all stakeholders to continue in these actions, and
develop further linkages, knowledge‐sharing, and scope for partnerships.  
With the 2 year extension, WVB could shift into a more technical and/or “advisory” role to the government
now that Upazila, divisional, and national dialogues have taken place and provide the government a model
to follow. WVB should be very clear on what precisely their “ask” is of the government to help define the
role that WVB can continue to play in supporting the government, and then how CVA formally or social
accountability informally may play a role in this (or not, as the case may be). Part of this conversation
should include topics like whether CVA working groups should be made legal, whether the government
should and could fund them (or where else WVB and the government may work to secure funding), how
CVA and social accountability can continue to work to influence government budgeting in relevant
117
development sectors. WVB staff are in a unique position to help with this in the advisory role of “match‐
maker” joining CVA and government agencies and line departments. Building on the success of the
Upazila, divisional, and national dialogues is a robust place to start from.
Recommendation #2: leverage learning from CVA, NJP and WVB to contribute to other civil society
efforts and programming supported by the government.
The evaluation found numerous instances of the government either explicitly or implicitly trying to
leverage aspects of civil society to improve service delivery and monitoring on accountability in public
health, water and sanitation, and agriculture extension services. Even where line department and
ministerial officials were not aware of CVA or even NJP, they have worked to leverage community
participation in the work they are doing, whether it is from citizens’ reporting on flooding in coastal areas
to the deployment of multipurpose health volunteers. The government should consider ways in which
CVA working groups are a model to assist in those efforts. A good example of knowledge‐sharing, linkages,
and partnerships to follow in this regard could be case of the government working with NJP and WVB on
the deployment of multipurpose health volunteers, or the inclusion of civil society organizations at
divisional and national dialogues on water and sanitation, community clinics, and agriculture extension.
Recommendation #3: continue to improve the interagency process by clarifying roles and
responsibilities for institutional actors, as well as delegation and accountability chains within
institutions.
A consistent finding that came out in the evaluation was there was pervasive “inertia” on the part of
institutional actors ‐‐ from local service providers to Union Parishad members ‐‐ to perform well in their
mandated jobs, or to “pass the buck” when it came to delegation and accountability chains within
institutions. This was not only reported by CVA working group members and community people, it was
also reported by institutional actors themselves at all levels. CVA helped to improve the interagency
process by clarifying roles and responsibilitiesfor institutional actors, but the governmentshould do more
to address the apparent and actionable human resource constraints that local and meso level officials
face in understanding their institutional roles and responsibilities, as well as devote more time and
attention to not simply mandating accountability, but actually supporting accountability and delegation
chains through programming within government institutions. This means coordination with Union
Parishad and Upazila officials, as well as between divisional and national line department officialsrelevant
for the three development sectors. Working with WVB, the government can learn how  CVA has helped
to foster incentives and platforms for interagency coordination to assist in this regard in other areas
beyond NJP project areas and for scaling. But the government should also focus on improving
accountability, reporting, and monitoring of institutional actors by higher‐level officials ‐‐ such as through
the new program described by the Ministry of Health and Family Welfare regarding the decoupling of
front‐line medical care with monitoring of health officials in Union Parishads and Upazilas. Advances in
reporting and monitoring systems that leverage simple‐to‐use, cheap, and scalable information and
communications technology that other development partners and WVB have experience with could be
very useful.
9.3.5 To Donor
To the Donor, the evaluation team provides two principal recommendations, especially given that NJP
and WVB requested a 12 month bridge phase to help communities recover from Covid and Cyclone
118
Amphan in 2020 (hopefully allowing some of the flexibility requested in this evaluation to continue
supporting and strengthening CVA working groups).
Recommendation #1: allow more flexibility in amendments to the 2 year cost extension, even keeping
the same budget and scope of work.
The evaluation team recommends to the donor that they allow some flexibility in amendments to the 2
year cost extension as per the considerations and recommendations above. This need not require a
change in budget depending on how WVB makes changes or not, but the donor should consider whether
there is scope to add or change the budget, especially given programming delays from Covid‐19.  
Recommendation #2: Operationally define “sustainability” to consistently apply across USAID
programming and with social accountability.
As discussed in the present evaluation in Section 7, despite the importance of USAID’s framework for
supporting sustained development and the principlesit lays out there for engaging localsystemsto impact
and assess long‐term development programming, USAID still has no consistent definition of
“sustainability” to apply to the entire portfolio of its development programming, with social
accountability, or even with prior NJP project cycle or the two year extension. The evaluation recommends
that the donor addresses this moving forward, and to leverage insights from its own framework for doing
so. Working from reverse order, the evaluation team encourages USAID to begin by including an
operational definition of sustainability in the 2 year cost extension. They could use the discussion and
guidance/evidence provided here if they want, but they could also go in a different direction. Next, while
sustainability was never defined or tracked for NJP or CVA previously, we encourage USAID to work with
WVB to retroactively apply the results of the theory of change, political economy mapping, and results on
policy dialogue in Section 7 to NJP. After that, it is for USAID to decide whether to use the definitions and
standards here or not in their other social accountability and development programming globally.
9.3.6 To WVI
Although the present evaluation covers CVA ‐‐ a social accountability tool previously developed by WVI
and deployed in other settings ‐‐ whether and how the findings pertain to WVI’s global programming or
use of CVA in other settings is a matter for WVI to discuss. Although the evaluation team could assist with
knowledge‐sharing and learning for how these results may pertain to extant or future CVA activities by
WVB, the present evaluation is not well positioned to comment specifically on other WVI programming
or the use of CVA in othersettings. Nonetheless, the evaluation team can comfortably recommend to WVI
that they take the entirety of the findings here and the recommendations to other stakeholders into
account.  
119
Annex A 
Coding Sheet 
Code Brief Identification Detailed about FGD/KII 
1. FGD-CVA-Dacope FGD with CVA (Citizen Voice and Action) Working Group 
Members, Dacope Union Parishad, Dacope (12.01.2020) 
2. FGD-Scorecard-Dacope FGD with Scorecard Session and Women’s Group 
 Vanue: Dacope Union Parishad, Dacope (12.01.2020) 
3. KII-Partner staff KII-NGO-Facilitator, Dacope 
(12.01.2020) 
4. FGD-CVA-Kailashgonj Focus Group Discussion 
CVA Working Group, Koilashgonj 
Place: Koilashgonj UP, Dacope 
12.01.2020 
5. FDG-VDC-Kailashganj Focus Group Discussion 
Dhanopoti-VCD, Koilashgonj 
Dacope 
12.01.2020 
6. KII-UHFPO-Dacope KII- UHFPO, Dacope Upazila 
12.01.12 
120
7. FGD-CG-CC-Dacope FGD (CG-CC) 
Venue: Kakrabunia, Rashkhola, Dacope Union, Dacope 
(13.01.2020) 
8. KII-AHI-Dacope KII: AHI, Robiul Islam 
 Venue: World Vision Office, Dacope 
(13.01.2020) 
9. KII-CHCP-CC-Dacope KII- CHCP, Tanusree Roy 
Venue: Saheber Abad Community Clinic, Dacope Union, Dacope 
(13.01.2020) 
10. FGD-CG-CC-Dacope FGD-CG 
Danopoti Community Clinic, 
13.01.2020 
11. KII-CHCP-CC KII: Khukumoni Mondol 
CHCP, Dhanopoti Community Clinic, Dacope 
(13.01.2020) 
12. KII-HI-Dacope KII- Health Inspector, Dacope (13.01.2020) 
13. KII-UAO-Dacope KII with UAO, Dacope (13.01.2020) 
14. FGD-WATSAN FGD (Union WATSAN Committee) 
Venue: Dacope Union Parishad, Dacope (14.01.2020) 
121
15. KII-UPChair KII: Chairman, Dacope Union Parishad 
 Venue: Dacope Union Parishad, Dacope (14.01.2020) 
16. KII-SAAO KII: SAAO, Dacope Union 
Venue: Dacope Union Parishad, Dacope (14.01.2020) 
17. FGD-WATSAN FGD: WATSAN Committee 
Koilashgonj UnionParishad 
(14.01.2020) 
18. KII-SAAO KII: SAAO, Kailashganj UP 
UP Office (14.01.2020) 
19. KII-UP Chair KII: Tapos Kumar Mistri 
Panel Chairman, Koilasgonj UP, Dacope 
Pannel UP Chairman 
(14.01.2020) 
20 FGD-NGO-Dacope FGD-NGO Staffs, Dacope, 14.01.2020 
21. FGD-CG-CC FGD (CG-CC) 
Venue: South Sreepur Community Clinic, Sreepur Union, 
Kaligonj, Sathkhira (15.01.2020) 
22. KII-HI KII- Health Inspector (HI), Proshanto Sarkar 
Venue: Kaligonj Upazila Health Complex 
Kaligonj, Sathkhira (15.01.2020) 
122
23. KII-CHCP KII- CHCP, Kakoli Mondal 
Venue: South Sreepur Community Clinic, Sreepur Union, 
Kaligonj, Sathkhira (15.01.2020) 
24. FGD-CG-CC FGD: CG Group, Thalna, Kaliginj 
15.01.2020 
25. KII-CHCP KII: CHCP Thalna, Kaligonj 
15.01.2020 
26. KII-FASI KII-Food and Sanitation Inspector, Kaligonj Shatkhira 
15.01.2020 
27 FGD-WATSAN FGD (Union WATSAN Committee) 
Venue: South Sreepur Union Parishad, Sathkhira 
(16.01.2020) 
28. KII-UZPChair KII: Chairman, Upazila Parishad 
Venue: Kaligonj UZP Complex, Satkhira (16.01.2020) 
29. KII-SAAO KII: SAAO, South Sreepur Union 
Venue: South Sreepur Union Parishad, Sarhkhira (16.01.2020) 
123
30. KII-UPChair KII: Chairman, South Sreepur Union Parishad 
Venue: South Sreepur Union Parishad, Sathkhira 
16.01.2020 
 
31. FGD-WATSAN FGD: WATSAN Committee 
Chompaful Union Parishad 
16.01.2020 
32. KII-UPChair KII- Chairman 
Chompaful Union Parishad, 16.01.16 
 
33. KII-SAAO KII-SAAO, Chompaful Union, Kaligonj 
16.01.2020 
34. KII-UAO KII: UAO, Kaligonj 
16.01. 2020 
35. FGD-CVA FGD with CVA (Citizen Voice and Action) Working Group 
Members 
 Vanue: Champafhul Union Parishad, Dacope 
(17.01.2020) 
36. FGD-SCORECARD FGD with Scorecard and Women Group 
Vanue: Thalna Cyclone Centre, Champaful UP 
(17.01.2020) 
124
37. KII-Partner Staff KII-Partner Staff 
Venue: NJP Office, Sathkira 
Dated: 17.01.2020 
38. FGD-CVA-SS FGD-CVA Working Group South Sreepur, Kaligonj 
17.01.2020 
 
39. FGD-Service Beneficiary FGD- Service Beneficiary (Maternity Health) 
Date: 17.01.2020 
40. FGD-VDC-Sonatala FGD: Village Development Committee
Place: Sonatola VDC‐1
Kaligonj 
Date: 17.01.2020