HAITIAN EMERGENCY RELIEF EFFORTS (HERE)
MID-TERM EVALUATION
Final Version 
March 31, 2009
Sergot Jacob MA PhD
Jean Armand Mondélis MBA 
Vanneur Pierre MSc 
GROUPE D’ÉTUDES, DE CONSULTATION ET D’AIDE À LA DÉCISION ÉCONOMIQUE ET STRATÉGIQUE SA®
PO Box 19188 ; Port-au-Prince, HAÏTI, HT6110 
Email: gecades@gecades.org • URL: http://www.gecades.org 
1 
HERE 
MID-TERM EVALUATION REPORT
Sergot Jacob MA PhD, Team Leader 
Jean Armand Mondélis MBA, Consultant 
Vanneur Pierre MSc, Consultant 
Acknowledgment 
The authors thank Claude Clodomir (M&E Manager, CHF Haiti) and Alberto Wilde 
(Chef of Party, CHF Haiti) for useful discussions and confidence at the early stage of the 
process, and comments on a the preliminary version of this report. We would also like to 
thank our surveyors: Kensky Princivil, Schedelyne Almonor, Lunès Ducarmel, Fritz￾Gerald Zéphirin and Alex Francois. We thank CHF staff in Gonaives and especially 
Aquel Gachette (the Community Mobilizer) and his driver, Wilnise Alabré (Employment 
Officer) and Patrick Louis (Administrator) for outstanding support during field work. The 
usual disclaimer applies. 
The views expressed in this report are those of the authors and do not necessarily 
represent the official position of Gecades SA. 
Coverage photo : Canal Drain Valira 
Source : Gecades SA, February 27, 2009 
2 
Executive summary 
The passage of 4 hurricanes and tropical storms in 2008 has caused the loss of productive 
assets of the Gonaives population. The cash injection through the Haitian Emergency 
Relief Efforts (HERE) program was believed to be an appropriate and necessary response 
helping in the clean-up phase and enabling beneficiaries to get access to the means of the 
recovery their assets and livelihoods. This mid-term evaluation aims most importantly at 
evaluating the impact of the Cash-for-Work (CFW) component of HERE on the quality 
of life of the targeted population. To do this, we use data both from the “treated” 
population and a non treated group. The sampling strategy insures that the treated group 
includes both individuals who are still under contract with CHF and those that have 
benefited since November 2008. First, we report data on the overall program and 
beneficiaries. Second, we use econometric models to measure the impacts of CFW. The 
results suggest that, all though the respondents’ participation in the CFW activities has 
the correct sign, it is not different from zero, statistically speaking. In the current 
predicament of Gonaives, CFW activities are doubtlessly essential, but the daily wage 
rate (3.75 USD) seems too low to significantly impact the quality of life of beneficiaries 
(which we proxied by the monthly expenditures) and the reconstruction efforts. A further 
possible interpretation is that the (officially) authorized period of participation in the 
program (12 days)—that leads to a 45 USD-payment for a round of participation—seems 
also too short to make any difference. We observe as well that the less wealthy strata of 
the population are the ones involved in this kind of job. The evidence also indicates that 
households that begin the rebuilding of their assets are those where member(s) benefited 
additional support from other humanitarian programs, and where another source of 
income and the possibility to pursue the same economic activity as before the hurricanes 
exist. 
We have also performed focus groups and key informants interviews. The qualitative 
data gathered reveal that the money received from the CFW activities was mostly used to 
feed the household members and in lesser extent to pay part of school fees and other 
expenses. In fact, it is worthwhile noting that the opportunity cost of time of the 
beneficiaries in the CFW is close to zero, given that participating in those activities was 
their best alternative after the hurricanes. Consequently, comparing with their after￾hurricanes situation without the HERE program, the beneficiaries are doubtlessly better 
off as a result of the program. However, this improvement should be viewed as short term 
and fragile given the characteristics of CFW component of the HERE program. 
As of the mid-term evaluation of the CFW component of the HERE program, the 
community is still waiting on program/projects with medium and long term potential 
impacts to rebuild their means of existence after the hurricanes. 
3 
Table of contents 
Acknowledgment 1
Executive summary 2
Acronyms and Abbreviations 4
1 The HERE program background 5
2 Methodology 7
2.1 Analysis of the documents pertaining the HERE program 7
2.2 Designing individual questionnaire, Focus groups protocol, and key informants guide 7
2.3 Performing the focus groups, key informants interviews, locating the beneficiaries, and 
testing the individual questionnaire 8
2.4 Sampling strategy, Collecting and analysing individual data 9
2.5 The econometric models 11
3 The HERE’s mid-term situation (by January 15, 2009) 12
4 Analysis of qualitative data from the focus group 16
4.1 After-hurricanes experiences of the affected population with the help of aid agencies 
and governmental authorities 16
4.2 Perception of the community about the aids/activities of CHF international Haiti after 
the passage of the hurricanes 17
4.3 Socio-economic Impacts of the CFW activities of CHF international in Gonaives 18
4.3.1 Opportunity cost of time using in CHF Cash-for-work activities 18
4.3.2 Standard of living after the hurricanes 19
4.3.3 Social capital 20
5 Descriptive analysis of the individual data 21
5.1 Socio-demographic characteristics of the beneficiaries 21
5.2 Socio-economic characteristics of the participants before and after the hurricanes 22
5.3 Usage of the money received by the beneficiaries from the Cash-for-work activities of 
CHF 24
5.4 Gender equity and other aspects in Cash-for-work activities recruitment 25
5.5 Reconstruction of the means of existence of the interviewees after the hurricanes 25
6 Understanding the micro-impacts the HERE CFW activities—Econometric evidence 
and discussions 26
6.1 The probability of participating in the CFW component of the HERE program 26
6.2 Participation in CFW and quality of life 27
7 Concluding remarks and recommendations 32
References 35
APPENDICES 36
Appendice I—Interview Guide with Local CHF Staff in Gonaives 36
Appendix II—Time Frame submitted by Gecades SA at the early phase of the process 38
Appendix III—The damages of the Hurricane Jeanne in photos (12 days later) 39
Appendix IV—Questionnaire for individual Survey 40
Appendix V—Focus Group Protocol 43
Appendix VI— HERE Project Status by January 15, 2009 45
Appendix VII—Gonaives in Photos after the Hurricane in September 2008 46
Appendix VIII— HERE Summary of Performance Results 47
Appendix IX—Buildings cleaned up by the CHF’s CFW activities 48
Appendix X—(A) CHF teams at work and (B) The need to find a consensus 53
Appendix XI—Statistical data 55
4 
Acronyms and Abbreviations 
ACF Action Contre la Faim 
CARE Cooperative for Assistance and Relief Everywhere 
CFW Cash for Work 
df degree of freedom 
HTG Haitian Gourdes 
HERE Haitian Emergency Relief Efforts 
MINUSTAH Mission des Nations Unies pour la Stabilisation d’Haïti 
OIM Organisation Internationale de la Migration 
USD United States Dollar 
5 
1 The HERE program background 
Extreme poverty, unemployment, non-schooling children, etc. continue to be the greatest 
challenges in Haiti, the poorest country in the hemisphere. Deforestation has reached 
such an extent that natural disasters are being pervasive and cause considerable damages 
in both human and material dimensions. The last years give significant examples. On 18th
and 19th of September 2004, the Hurricane Jeanne hit the northern part of the country. 
The city of Gonaives was completely devastated (see Appendix III). Nearly 2,000 dead 
and some 25,000 homeless people are registered. Homes, hospitals and schools were 
destroyed. Four years after, during the 2008 hurricane season, Haiti has been hit again by 
4 hurricanes and tropical storms (Fay, Gustav, Hanna and Ike) which have contributed to 
further deteriorate the precarious socio-economic situation and infrastructure of the 
country. 
The Haitian population as a whole was deeply affected by these natural disasters, but the 
city of Gonaives is known for being the most devastated area of the country. Located at 
150 kms from the Capital of Port-au-Prince, it was completely flooded and covered with 
mud. The agricultural production and the infrastructures were devastated. Besides the 466 
dead1
 , the latest statistics report 50,000 families affected, 65,000 families placed in 
temporary shelters and about 13,000 houses (administrative buildings, schools, hospital, 
etc.) destroyed or damaged by the passage of these four successive hurricanes in 
Gonaives.2
 Without shelter and being exposed to serious health problems, the victims are 
living in precarious conditions of power and total promiscuity. 
As of mid September 2008, after the initial distribution of foods and drinking water to the 
affected population in Gonaives, local assessments have identified intensive clean-up 
activities and rapid repair of key infrastructure as the priority response. This priority has 
laid the foundation for the HERE program sponsored by USAID through its Office of 
 
1
 Rapport d’évaluation des besoins après désastre. Cyclone Gustav, Fay, Hanne et Ike. Government of 
Haiti (2008). 
2
 Direction de la Protection Civile (cited in MINUSTHA). Haïti : Gonaïves, deux semaines après le déluge. 
September 19, 2008. 
6 
Foreign Disaster Assistance (OFDA) and implemented by CHF, an international 
development and humanitarian organization in Haiti. Its main objective is to provide 
livelihood assistance and support the recovery of critical social infrastructure in Gonaives 
and Petit-Goave. Through its approach of intervention, the program specifically aims at 
providing with immediate access to income to the flood and hurricanes affected 
population while performing basic clean-up and repair to key infrastructure. Through a 
community-based targeting approach, clean-up projects have been identified and labor 
teams were recruited to perform Cash-For-Work (CFW) activities. 
HERE is being implemented in Gonaives and Petit Goave. As mid of January 2009, CHF 
International Haiti was about to conduct a mid-term evaluation for HERE with a special 
focus on Gonaives. The main purpose of this mid-term evaluation is to assess the impact 
of HERE cash-for-work component on some important socio-economic outcomes of the 
beneficiaries. Specifically, the following objectives are being pursued: 
1. Assess to what degree the HERE cash-for-work activities have contributed to the 
improvement of quality of life of the beneficiaries; 
2. Gather the necessary demographic, social and economic data so as to give a good 
understanding of the living conditions in Gonaives; 
3. Provide data on the current type and level of employment; 
4. Describe the current level of community participation and involvement in disaster 
relief activities; and 
5. Provide ground-level analysis on the impact of HERE in Gonaives. 
The independent consulting firm Gecades SA has won the tender to conduct this work. 
The structure of this report is organized as follows. Section 2 outlines the methodology 
(the econometric models, the sampling strategy, the empirical data and the findings). 
Section 3 presents the background of the HERE program. Section 4 draws the situation of 
this program on January 15, 2009. Section 5 analyses data from the Focus Group. Section 
6 deals with individual impacts of the CFW component, and section 7 concludes and 
proposes recommendations. 
7 
2 Methodology 
As described below, several methods of data collection and analysis were used in order to 
better address the objectives of this mid-term evaluation. Each one complements the 
others. Gecades SA has fully completed the consultation—from the study design to the 
final report—in seven calendar weeks between January and March 2009. The different 
activities that have been performed according to the time table are shown in Appendix II. 
2.1 Analysis of the documents pertaining the HERE program 
In the first place, Gecades SA started to meet with the CHF contacts to discuss the 
relevant aspects of the evaluation, to deeper its understanding of the program to be 
evaluated and at the same time to identify and request the necessary documents for 
analysis. These first steps have allowed Gecades SA to better capture the objective of the 
program and consequently what is to be measured through the mid-term evaluation. CHF 
submitted to Gecades SA a set of documents pertaining the program (the HERE program 
basic document, project sites, payroll sheets, and the performance results until January 
15, 2009). We have analysed all these documents which provide us with precious 
information to understand the HERE program background and its situation trough the 
mid-term implementation. 
2.2 Designing individual questionnaire, Focus groups protocol, and key 
informants guide 
The second phase of the process consisted of devising the necessary tools for information 
collection devoted to achieve the first three objectives of this mid-term evaluation. Based 
on the analysis from the first phase, we have devised an individual questionnaire 
(Appendix IV) which has been pre-tested on the targeted and non targeted population of 
Gonaives (see below). This questionnaire contains questions about individual socio￾economic variables (age, sex, marital status, education level, formal and informal 
training, economic activities, employment, income), household variables (household size, 
number of dependents and employed, private transfers, expenditures, housing tenure, 
CFW participation) and neighborhood variables (distance, attendance associational life). 
8 
We developed as well a focus group protocol (Appendix V) in order to capture qualitative 
data aimed at describing the perception of the targeted community about the HERE 
program as well as its involvement in disaster relief activities after the natural disaster 
and at providing ground-level analysis on the impact of HERE in Gonaives. We 
developed also a key informants guide (Appendix VIII) that has been used to conduct 
interview with HERE’s stakeholders. That is, both quantitative and qualitative data can 
be combined to fully capture the (mid-term) impacts of the program on the community of 
beneficiaries. These tools were discussed with the key contacts of CHF to make sure they 
are designed according to the objectives of the study. 
2.3 Performing the focus groups, key informants interviews, locating the 
beneficiaries, and testing the individual questionnaire 
For adjustment purposes, we have proceeded by testing the data collection tools on the 
field of the study. With the help of CHF representatives in Gonaives, we have 
implemented a first field trip to locate the project sites selected by CHF, and the 
participants in CFW and non participants as well, test and calibrate the individual 
questionnaire and in the same time perform the focus groups and key informants’ 
interviews. All these interviews have been performed on February 13, 2008. We had 9 
questionnaires (6 participants and 3 non participants) pre-tested by our trained surveyors. 
Few changes have been necessary in the individual questionnaire. The final version is 
presented in Appendix IV. 
We conducted three focus groups (FG) with the direct beneficiaries who participated in 
the different clean-up sites as workers and team heads. Three different local served to 
perform these interviews : Hopital La Providence (14 individuals of whom 9 men and 5 
women), Ecole Nationale de Descahos (9 participants : 4 men and 5 women) and the 
home a team head nearby the Délégation de l’Artibonite (10 participants : 5 men and 5 
women). The information from the focus groups was recorded and then transcribed for 
content analysis. The FG protocol is presented in Appendix V. We also interviewed key 
informants (a controller and a program director) of the CHF in Gonaives (see the 
Interview guide in Appendix VII). These interviews have been transcribed and analyzed. 
9 
They give us the qualitative data needed to better understand the perceptions and 
experiences of the beneficiaries with the program. We report results from these 
interviews in section 4 and 5. 
2.4 Sampling strategy, Collecting and analysing individual data 
Two main types of evaluation exist (Human Resource Development Canada (1998) cited 
in Matchaya (2006)): (i) process evaluation that aims at determining if the program has 
been implemented as planned and (ii) summative evaluation that measure whether the 
program has succeeded in achieving its specific objectives. We will focus here mostly on 
the summative evaluation method which focuses on how individuals’ outcomes are 
altered by the HERE. The better way to do this is to compare what we observe with 
measures of what individuals’ outcomes would have been in the counterfactual situation, 
i.e. if they did not participate in the HERE. The difference in the main outcomes 
indicators is the impact of the program. But the counterfactual situation is not observed. 
Table 1 
Distribution of the sample 
Count %
Participants 94 75% 
Non participants 31 25% 
Whole sample 125 100% 
Source: The authors’ calculations
We then need to estimate proxies of counterfactual outcomes. To do this, we compare a 
treated group (TG) and a non treated group (NTG).3
 Generally, three pre-conditions are 
necessary for better results: size, experience and pre-disposal (Freeman et al. 1979). But, 
due to budget constraint, the whole sample was set to 125 individuals of whom 78% are 
direct beneficiaries—who are currently working (or had worked) with CHF only—of the 
project and 22% are non beneficiaries (Table 1). The latter constitute a control group. 
But, all the respondents are comparable on all other aspects. Furthermore, as HERE is an 
emergency response, a baseline was not available. It is then difficult to compare the 
current situation to the one that existed before HERE. However, we attempt to isolate 
 
3
 See Matchaya (2006) for a technical point on the evaluation problem. 
10 
effects from other similar programs currently implemented in Gonaives by focusing on 
people who have worked and are currently working in the CHF program only and by 
introducing other control variables. 
In order to perform individual data collection, we have used the list of the beneficiaries of 
the different project sites and from 6 payroll sheets submitted by CHF to randomly select 
a representative sample. Two methods are combined to construct the sample: the method 
of quotas and the probabilistic one. Once the number of beneficiaries from a site is 
defined with respect to the weight of the site in the overall 624 individuals being still 
under (or had a) CFW contract with CHF on the six project sites, the respondent were 
selected randomly. These are replicated in the selected sample (see table 2). We used 
payroll sheets over the period of November 12, 2008-January 27, 2009—as submitted by 
CHF—to indentify beneficiaries by the means of tokens. The rest of the sample is 
selected randomly as well from the non participants in CFW that we consider as a non 
treated group. The beneficiaries interviewed belong to 9 representative teams of the 41 
that exist (such clusters) so far and have been operating in different sites of Gonaives. 
The data collected were then entered into a data base in Excel and converted into SPSS4
for treatment (uni- and bivariate analysis) and into STATA10 for multivariate analysis in 
order to establish evidence of the HERE impact on its stakeholders. 
Table 2 
Composition of the sub-sample of CFW participants 
Project site Number of 
beneficiaries 
Percentage Sub-sample of 
participants per 
project site 
Hopital La Providence 152 24% 18 
Delegation de l'Artibonite 91 15% 11 
Université Indépendence 76 12% 9 
Université Independence 61 10% 7 
Bibliothèque J.S. Alexis 46 7% 6 
Loge Sympathie des coeurs 46 7% 6 
Collège Immaculée Conception 152 24% 18 
TOTAL 624 100% 75 
Source: The authors’ calculations
 
4
 Statistical package for the social Sciences. 
11 
2.5 The econometric models 
The main question that CHF practitioners may pose is whether the HERE program has 
any significant effect on the standard of living of the targeted population. In other terms, 
does the CFW component impact the households’ expenditures on foods and non food 
items? In the meantime, a concern may exist about how to isolate the CFW effects from 
other similar programs. The main problem we are facing here is to determine causality. 
In order to estimate the causal impact of CFW, we adopt three specifications. First, we 
model the participation in CFW activities (1). The dependent variable is response to the 
question (Q21): “Have you participated in the Cash-for-work activities of CHF 
international?” in Gonaives. As the answer is dichotomized (Yes=1 or No=0), model (1) 
is a binary probit. 
PartCFW=Xβ+μ (1) 
Second, we aim at measuring the causal impact of CFW on the quality of life of 
participants (2). We use average total consumption expenditures (ME) after the 
hurricanes (in natural log) as a proxy for income and we use these terms interchangeably. 
Pragmatically, we assume that in the current predicament of Gonaives, expenditures are a 
better measure than expressed income.5
 However, we need to deal with endogeneity 
concerns. As there are no official criteria to recruit workers (see below) who are then 
selected by a team chief representative of the site to be cleaned up, we then assume that 
participation in CFW (PartCFW) may suffer from endogeneity problem. So we need to 
find out excluded instruments (Z) responsible for participation (or not) in the program, 
that is, directly correlated to PartCFW and indirectly to the dependent variables 
(orthogonality condition).6
 We then adopt the following simultaneous model where X is a 
 
5 See Narayan and Pritchett (1999) for a point on this topic and similar hypothesis. In our data set, the 
monthly revenue expressed by the respondent before and after the hurricanes are respectively 6,006.66 and 
5,683.64 HTG in average. Whereas the monthly expenditures before and after the hurricanes are 
respectively 8,889.28 and 6,656.18 HTG in average. 
6
 To be valid, our instruments must satisfy two conditions: (i) Z is correlated with the endogenous 
participation in CFW activities, but (ii) uncorrelated with the error term in the income and reconstruction 
structural equations. 
12 
k-dimensional matrix that includes regressors at three levels: the individual, the 
household and the neighborhood variables (sex, education, number of employed in the 
household, the housing tenure status (whether owning=1 or renting=0), membership in 
associational life, etc.). We explore as well the effect of interacted terms. 
ME=PartCFWα+Xγ+η (2.1) 
PartCFW=Zδ+π (2.2) 
Alternatively, we also assume that participation in CFW may foster livelihood rebuilding. 
The dependent variable (Reconst) is the answer to the question Q30 : “Have you already 
started rebuilding your assets and other means of existence destroyed by the hurricanes?” 
(Yes=1 and No=0) Therefore, we adopt the following endogenous binary probit model 
(3): 
Reconst=PartCFWρ+Xθ+ε (3.1) 
PartCFW=Z +λ ω (3.2) 
The k-dimensional matrix X is—with few differences—the same as in (2). In dealing 
with the different causal mechanisms by which participation in CFW impact income, we 
use instrumental variable estimation. We then need valid instrument and theoretical 
arguments about the model. We select the following excluded instruments set: gender, 
number of years living in the locality where he/she is interviewed, age and marital status. 
ω, π and η are error terms that stand for non observed variables. μ and ε are random 
variable standing for unobservable household heterogeneity. We assume that μ and ε are
mean-zero and standard normally distributed. With this distributional assumption, we deal 
with binary probit models. α, θ, β, γ, δ, θ, λ, and ρ are parameters to be estimated.
3 The HERE’s mid-term situation (by January 15, 2009) 
HERE has a main component based on Cash-for-Work (CFW) activities which have two 
phases: (i) clean-up and (ii) rehabilitation. The former started on the 12th of November 
2008 and until 27th of January 2009, six sites (public and non public buildings mostly 
hospital, government administrative buildings and schools) have been cleaned up 
according to the 6 payroll sheets transmitted to Gecades SA. In sum, 624 temporary jobs 
13 
have been created to the benefit of the Gonaives population. But more globally, 30 
projects on different sites have been identified and planned pertaining to the clean-up and 
rehabilitation of public infrastructures (see the project list in Appendix VI). As of the 15th
of January 2009, 23 of these projects have been completed pertain to the Clean-Up phase. 
CHF aimed at employing temporarily 5400 workers by itself in CFW. This objective is 
achieved at 47% by the mid of January 2009 (see Appendix IX). The Infrastructure 
Rehabilitation phase has not started yet. 
The recruitment process is orchestrated by representatives of the CHF selected site to be 
cleaned. The workers are then set up in teams in which the criteria of selection seem to be 
the worker’s age, physical strength and the proximity to the site (see the coverage photo 
and Appendix X). One may distinguish the controllers, the team head and the workers per 
se. For the controllers, the selection is generally made by CHF local representatives with 
respect to years of experience with the individuals. A worker belongs to a team of 12-15 
persons supervised by a team head. The contract—which we consider as informal since a 
written document is not available for both parts—with a worker is normally valid for a 
“bi-weekly period” and is non renewable in order to offer the same opportunity to a 
maximum number of workers. At the end, the payment is done by Fonkozé on behalf of 
CHF. The daily wage rates are 380, 220, and 150 HTG respectively for controller, team 
leader and worker for six hours of work per day (from 7:00 AM to 1:00 PM or from 3:00 
PM to 9:00 PM) from Monday to Saturday. 
 
According to six payroll sheets for the period of November 12, 2008-January 27, 2009 
submitted to Gecades SA, a total amount of 1,169,080.00 HTG (29,227.00 US dollars)7
has been disbursed to create temporary jobs for 624 persons (of whom 41 team leader, 
574 workers, and 9 controllers). See table 3 for further details with regard to the sites and 
Appendix X for teams at work. We identified three categories of direct participants in 
CFW: the controller (who supervise 5-6 teams), the team head (who controls the work of 
about 12-15 workers), and the workers per se who perform clean-up jobs. The workers 
and their team chiefs are supposed to be hired for a two-week period while the controllers 
 
7
 Exchange rate : 1 USD=40 HTG. 
14 
may operate over a longer period.8
 Figure 1 illustrates this pattern that we observed on 
the field. The quality of the performed work does not seem to be a requirement since 
workers are not being paid on the basis of the quality of work done but on the basis of the 
number of work-days performed. 
As mentioned above, the program HERE has two main components: (A) Cash for Work 
and (B) Infrastructure Rehabilitation. As pointed out by a key informant, the program 
initially focused on public institutions but after meeting with local authorities and other 
aid agencies (CARE, CARITAS, OIM, etc.) involved in the same type of activities, CHF 
revised its plan of work. This fact per se suggests a good collaboration between CHF and 
other operators in the same area of intervention. By contrast, the community does not 
seem to have played any role in the selection of the sites to be cleaned. However, the 
population seems to be very committed and motivated in cleaning up its city. At this 
moment, it is however possible to identify the following impact of the CFW: 
1. for people participating in the CFW component of the HERE program, this 
alternative seems to be the unique one available to them after losing all their 
livelihoods; and 
2. the clean-up phase has facilitated the resumption of several activities (such as 
schools, hospital, library, etc.). 
 
8
 However, a controller expressed that the contract duration is the same for controllers and workers as well. 
15 
Figure 1 
Relationship between different entities of the HERE program 
Table 3 
Total Disbursement at the Mid-Term 
Project sites Number of 
beneficiairies 
Amount disbursed
(HTG)
% Period 
Hôpital la Providence 152 287,520.00 25% 12-27 Nov.08 
Délégation de l'Artibonite 91 158,200.00 14% 27 Nov-10dec08 
Délégation de l'Artibonite 
Université Indépendance 
76 143,760.00 12% 15-28 Nov.08 
61 115,920.00 10% 03-09 Déc.08 
Bibliothèque Jacques 
Stephen Alexis 
46 88,080.00 8% 12-24 janv.09 
Loge Sympathie des 
cœurs 
46 88,080.00 8% 10-23 janv.09 
Collège Immaculée 
Conception 
152 287,520.00 25% 14-27 janv.09 
Over all 624 1,169,080.00** 100% 12 nov08-
27janv.09 
Source : The authors’ calculations
Notes.—**Equivalent in USD : 29,227.00 USD (1 USD= 40 HTG). Daily wage rates : Controller : 380 HTG 
(9.5 USD) ; Team head : 220 HTG (5.5 USD) and worker : 150 HTG (3.75 USD). 
Controller 1 Controller 2 
Team Head Team Head Team Head 
Worker 1 
CHF 
Local staff 
Worker 2 Worker 3 
16 
4 Analysis of qualitative data from the focus group 
The data gathered from the focus group interviews can be categorized under three main 
points pertaining to the purpose of this mid-term evaluation: we investigated in the first 
place the general experiences of the beneficiaries with humanitarian aids after the 
hurricanes, then we put more focus on searching information related to the perception of 
the population about the activities of CHF as well as searching to identify areas of 
impacts of its activities on the beneficiaries. 
4.1 After-hurricanes experiences of the affected population with the help 
of aid agencies and governmental authorities 
From the three focus groups conducted during the evaluation, unanimously, the 
participants acknowledged the presence and the effort of different types of NGOs 
(CARE, OIM, PAM, OXFAM, ACF, CHF International) and the governmental 
authorities (through public local administration: the municipal administration and the 
Ministry of transport and telecommunication) providing humanitarian aids to alleviate the 
sufferings of the affected population of Gonaives. However, in general terms, when it 
comes to the efficacy and efficiency of the distribution of those many aids, one may raise 
some question marks depending on which aid agency or local authority implementing 
them. Firstly, from the information gathered, we noticed a lack of fairness or a certain 
degree of inequity in the providing of the aids to the people. Most of the participants 
stressed on the partisanship spirit that has prevailed particularly during the beginning of 
the process. Though the situation affected everyone, the women would be the first 
victims of this problem. Nevertheless, this information should be weighed cautiously as it 
is often heard of from population coping with such situations. Secondly, the highest level 
of dissatisfaction came from the experiences with the direct intervention of the 
governmental authorities through the municipal administration. The lack of control of 
their operational process has led to a high degree of inefficiency in their process of 
helping restore life in the city. Finally, despite the above-mentioned complaints, the 
participants on average have expressed a low degree of satisfaction (but within two 
17 
groups, it varies from 40% to 50% on average) with those many aids that have intended 
to helped them survive the after-hurricanes’ periods. 
4.2 Perception of the community about the aids/activities of CHF 
international Haiti after the passage of the hurricanes 
According to the participants of the focus group interviews, CHF seemed to differentiate 
itself from the low degree of satisfaction mentioned above. In many aspects, the 
interviewees have sorted out CHF from the negative remarks raised in the precedent 
section. Mainly, CHF has distinguished itself through: 
• Its approach of helping the population coping with the socio-economic difficulties 
after the hurricanes. Through its Cash-for-work activities to clean up social 
infrastructure, not only does it provide to the direct beneficiaries an economic way 
of going through the hard times, but also contributed to prevent the rapid 
dissemination of caused-infections and diseases through the whole community. In 
other words, it has helped contain the damages that could have been done in the 
area of public health in Gonaives. 
• The sense of fairness, coordination and control in the process of implementing its 
interventions on the field. The opinions voiced by most of the groups in this vein 
suggest that the payroll sheet reflects the reality of the work done. The 
“partisanship” spirit in reaching out to affected people was contained. In that way, 
more needed people are reached and the amount of work done is reasonable for 
the planned expenses. 
• Being perceived by the beneficiaries to be the first organization to bring about 
serious restoration in a reasonable timeframe to the social infrastructures in 
Gonaives, which was important in so many ways to keep this community alive 
after the hurricanes. The beneficiaries have expressed great satisfaction with the 
presence of CHF in Gonaives and the way it has been implementing the HERE 
program. They request that CHF stay for a long period of time in Gonaives as it is 
perceived to be an organization that can bring about real improvement in the life 
18 
of the community. For that matter, they recommend that CHF comes up with 
other types of projects having potential medium term or long term impacts, that is, 
projects capable of helping them rebuild their means of existence destroyed by the 
flood/hurricanes. 
However, the beneficiaries have voiced some complaints against CHF. They have 
reproached the organization with not being able to make the payroll on time and also with 
applying too low a wage rate for the CFW activities. For example, in one group, the 
participants were expressing their bitterness over the fact that they were recruited for a 12 
days-period of work but were stopped after 6 days. And it’s getting worse when 2 weeks 
after they have not been paid yet. 
4.3 Socio-economic Impacts of the CFW activities of CHF international in 
Gonaives 
The socio-economic impacts of the HERE program on the affected population of 
Gonaives are best measured by indicators of impact using statistical tools, but the 
analysis of opinions gathered from the focus groups may give corroborating thoughts or 
shed some light for better statistical interpretation of the results of the individual surveys. 
With that said, the purpose of this section is to expose the impacts of the program on the 
population according to their opinions pertaining to that matter. To this end, we have 
investigated through three broad areas. 
4.3.1 Opportunity cost of time using in CHF Cash-for-work 
activities 
The beneficiaries interviewed are comprised of diverse categories of people in terms of 
skills or activities for a living: tailor, seamstress, little farmer, painter, student, house 
builder, retailers, etc. After the passage of the hurricanes, as the whole city were 
devastated, there were no demands for such skills on one hand, and there was no money 
to replenish the stocks of goods or to replace working tools destroyed by the flood on the 
other hand. Most of the aids to the population on the first time consisted of foods, water, 
and other basic needs. Consequently, people spent their time in line waiting for foods (not 
19 
without difficulties), or doing nothing capable of producing economic benefit. According 
to their opinion, when CHF came to the city with the Cash-for-work activities, it brought 
hope to put their time back to work. For them, there was not another best alternative for 
their time. Without the CHF cash-for-work activities, it would have been almost 
impossible for them to find another job. Those testimonies suggest that CHF has created 
a major impact on the short term job market in Gonaives after the hurricanes. 
4.3.2 Standard of living after the hurricanes 
The CFW activities constituted the major source of income (not to say the only source) 
for the beneficiaries interviewed. The analysis in this section focuses on whether or not 
the generated income has contributed to improve the quality of life of the beneficiaries. 
For that purpose we have gathered information related to the use of the money earned for 
food items and for non food items. As for the food items expenses, the beneficiaries 
voiced that part of the money earned was used to feed the members of their household as 
it is considered their first priority. Given the many needs to satisfy after the hurricanes 
and the limited amount they receive from the CFW activities, they are not able to talk 
about real increase in their expenses for food consumption in comparison with before￾flood/hurricanes period. The best they can do is to optimize the use of this income to 
make sure basic food is provided in the family. However, considering that the cash-for￾work activities of CHF was the best alternative for the use of their time, one may say that 
those activities have greatly contributed to maintain the necessary level of food expenses 
in the households of the beneficiaries. 
Though the first priority of the beneficiaries is to feed their household or themselves with 
the money received, most of them have provided exciting items of information when it 
comes to talk about the use of the money for non food items. The most frequent expenses 
that came up were related to paying schools for their children. This was very important 
given that the disasters happened just before the beginning of the academic year. For the 
students, the money allowed them to pay (partially for some) to continue with their 
learning. Some participants are able to pay basic healthcare for their kids and for 
20 
themselves; some of them use part of the money to pay rent. One should not ignore 
however the fact that for some of them, they are just able to feed their household from the 
money they hope to receive. That said, it is worthwhile mentioning another interesting 
point raised by the participants: a current contract with CHF gives to the beneficiaries the 
capacity to borrow emergency money from a third party.9
 That is, the contract with CHF 
increases the credibility to pay back and consequently allows the beneficiaries to better 
cope with some urgent needs. 
The cash-for-work activities don’t allow the beneficiaries to make expenses related to 
making new acquisitions such as buying livestock, replacing tools to continue their 
profession, improving their dwelling. They are not able to accumulate savings either. 
However, individual survey reveals expenses made possible by CFW activities: food 
(52.1% of the respondent), commerce (10.4%) and other (37.5%) (see Graphic 3). 
4.3.3 Social capital 
In addition to the economic benefits to the beneficiaries, the presence of CHF has brought 
some other social advantages to the community. First of all, according to them, there is a 
sense of solidarity created among the members of a team working on the field; they 
support each other by sharing some time their food while working. In other words, they 
cultivate a fraternity spirit in the team-work. Second of all, they acquire the habit of 
getting together under CHF representative leadership to discuss about matters of interest. 
They left believe that CHF would be the first organization to perform that kind of activity 
in Gonaives, which has brought a sense of closeness among the members of the 
community. They would confess also that the approach used by CHF has allowed people 
from a given area to go to work to a different area without suffering any aggression by 
the residents of the latter. This kind of mobility and mutual acceptance are very difficult 
when it is about distributing food in the city. 
 
9 It is worthwhile to highlight the fact that this is a program that CHF is implementing in partnership with 
FONKOZE. The program is called “TiKredi”, and has been designed to revive micro enterprise activity in 
the city of Gonaives. The program is designed to re-establish and create 1,150 microenterprises in 
Gonaives. 
21 
As for the socio-economic impacts of the HERE program, the analysis of the information 
provided by the focus groups suggests the following conclusions: 
• Without the jobs created by the CFW activities in Gonaives after the passage of 
the hurricanes, the opportunity cost of time of the beneficiaries is close to zero. 
• Compared to the period before the hurricanes, it was hard for the beneficiaries to 
confess enjoying real increase in their quality of life after the hurricanes as a 
result of the CFW activities; this is explained by the many needs left behind by 
the hurricanes in comparison with the limited amount receive from the cash-for￾work activities. 
• However, the cash-for-work activities have allowed the beneficiaries to maintain 
the necessary level of expenses for food consumption in their household and 
mainly to keep up with school expenses and in a lesser degree to cope with some 
minor healthcare spending for their kids. In other words, seen in that way of 
comparison, the CFW activities has contributed to increase the purchasing power 
of the beneficiaries. 
• Compared to the after-hurricanes’ situation without CHF in Gonaives, one may 
argue in favor of a certain degree of improvement in the quality of life of the 
beneficiaries. This should be viewed as a short-term and very fragile improvement 
because it does not provide the beneficiaries with the capacity to rebuild their 
means of existence. 
Finally, these conclusions need to be validated by impact indicators measured via 
statistical tools based on the data collected from the individual surveys. This will be the 
purpose of the section 6 below. 
5 Descriptive analysis of the individual data 
We report in this section a rapid descriptive analysis of the data collected from the 
individual survey. The statistical tables are compiled in Appendix XI. 
5.1 Socio-demographic characteristics of the beneficiaries 
The analysis of the data collected shows that the interviewed sample is comprised of 
37.6% of women and 62.4% of men. In terms of age, 44.8% of them are between 25 and 
22 
35 years-old, 29.6% between 36 and 46 years-old and 19.2% are more than 46 years-old. 
(Table 12.1). 76% of the people surveyed are not married. As for the level of literacy, 
95.2% of them went to school but only 12% have reached beyond high school level and 
35.2% have reached high school level of education (Table 12.4). In terms of skills and 
training, the data indicates that 45.6% of the people interviewed have attended an 
informal training and 30.4% own a professional diploma in a given technical discipline 
(Tables 12.5 and 12.6). As for the size of the household of the participants to the survey, 
44% of them have between 4 and 6 members, 24% have declared having more than 6 
members in their household and 24.8% have between 1 and 3 members (Table 12.10). 
5.2 Socio-economic characteristics of the participants before and after the 
hurricanes 
Before the passage of the hurricanes, 87.2% of the participants had income-generating 
activities (Table 12.7) and 84.4% of them were independent workers or self-employed 
(Table 12.8). During this period of time, 24.8% have generated monthly revenue 
estimated to less than 2,500 HTG vs. 44.8% after the hurricanes; 27.2% had estimated 
monthly income between 2,500 HTG and 5000 HTG before the hurricanes compared to 
24% in that same range after the hurricanes. 39.2% of them generated more than 5000 
HTG estimated monthly income before compared to 29.6% in that interval after the 
disasters (Graphic 1). 
Graphic 1 
Source: The authors
23 
On the expense side, we notice from the data that 13.6% of the participants have 
estimated their household expenses to less than 2,500 HTG before the hurricanes 
compared to 20% declaring the same estimate after the hurricanes; an estimated monthly 
amount between 2,500 HTG and 5000 HTG had been spent in the household before the 
floods according to 28% of the participants; the percentage of participants went up to 
32% after the hurricanes for the same range of estimated expenses. However, the 
percentage of interviewees that declared more than 5000 HTG as estimated household 
expenses before the hurricanes decreases from 57.6% before to 45.6% after the natural 
disasters. 
Graphic 2 
Source: The authors 
The analysis of the data shows that for the less wealthy interviewees, there is a positive 
increase in the percentage of participants that enjoyed more expenses in their household 
after the hurricanes; however there is a decrease in the percentage of those who enjoyed 
more than 5 000 HTG of expenses per month in their household. On another vein, 72% of 
the interviewees did not receive any financial assistance from their relatives residing in 
Haiti after the hurricanes. Nonetheless, 76.8% of them have received money transfer 
from their relatives abroad (Tables 12.22 and 12.23). 
24 
5.3 Usage of the money received by the beneficiaries from the Cash-for￾work activities of CHF 
 
We observe from the data that 58.3% of the beneficiaries of the cash-for-work activities 
have worked for more than a biweekly period in the program. At the time of the survey, 
97.9% of the participants have finished their contract with CHF (Table 12.24 and 12.25). 
The members of the family of 76% of the interviewees did not participate in the cash-for￾work activities. Moreover, 86.4% of them declared that their family members were not 
direct beneficiaries of other humanitarian or reconstruction programs after the hurricanes. 
Among the Cash-for-work activities’ direct beneficiaries, only 37.6% have participated in 
other humanitarian or reconstruction programs after the floods (tables 12.27, 12.29 and 
12.30) 
Graphic 3 
Usage of money from cash-for-work activities
52.10% 37.50%
10.40% food consumption
school and other
expenses
retails activities
Source: The authors’ calculations
The salary received by the participants was used for different types of expenses. In fact, 
52.1% have used the money to buy food for their household; 37.5% of them have spent 
the money to pay school for their children, buy cloths, pay back their debts and satisfy 
other basic needs. Only 10.4% of them declared using the money for retail activities 
(little commerce). In spite of their participation in the Cash-for-work activities, more 
than three forth of the beneficiaries are not able to start rebuilding their means of 
25 
existence yet. Furthermore, there is no significant relation between the number of 
members in a household that enjoyed cash-for-activities and the reconstruction of the 
means of existence of that household (table 12.40). 
5.4 Gender equity and other aspects in Cash-for-work activities 
recruitment 
The analysis of the data revealed that the participation in the program is not conditioned 
by the sex of the participants: 70.2% of the women interviewed were direct beneficiaries 
of Cash-for-work activities vs. 80.8% of men (table 12.31: Chi2=1.834, df =1, p-value 
=0.176). The same observation is true for the level of education and the number of 
persons in the family of the beneficiaries (table 12.37: Chi2=7.403, df=7, p-value=0.192; 
table 12.38: Chi2=1.951, df=3, p-value= 0.583). 
However, being a member of a local association is significantly correlated to the 
participation in the program (table 12.48: Chi2=8.427, df=1, p-value= 0.004). Though 
participation in the program is not conditioned by sex, the length of participation and the 
sex of the participants are significantly associated according to the data: male 
beneficiaries have enjoyed more biweekly periods of contracts than female participants 
(table 12.33: chi2=5.235, df=1, p-value=0.022). 
5.5 Reconstruction of the means of existence of the interviewees after the 
hurricanes 
The analysis of the data revealed that the reconstruction of the means of existence of the 
interviewees is significantly linked to their participation in other humanitarian programs 
(table 12.41). Au contraire, for a household, the reconstruction of the means of existence 
destroyed by the hurricanes is independent of the quantity of its members who have a job, 
who have participated in the cash-for-work program or other programs (tables 12.40, 
12.42 &12.43). In the same vein, the reception of financial assistance from relatives 
living in other part of the country or abroad does not have a significant impact on the 
process of reconstruction of the means of existence of the interviewees (tables 12.45 
&12.46). Finally, the data show a significant relation between the reconstruction of the 
26 
means of existence of the participants and the fact that their spouse have a given 
generating income activities after the hurricanes (table 12.44, Chi2=7.895, df=2, p-value 
= 0.019). 
6 Understanding the micro-impacts the HERE CFW activities—Econometric 
evidence and discussions 
6.1 The probability of participating in the CFW component of the HERE 
program 
In table 4, we report the determinants of participation in CFW component. We find 
evidence that participation in the CFW component is most likely when the respondent is 
the household head (that is, he/she has the responsibility to bring the resources that is 
indispensable for the living of the household). This variable increases the probability of 
participation by 33 percentage points. We observe that the respondent age also matters. 
We interpret this as the opportunity given by the physical strength of the youths but 
marginal effect is low (1.19 percentage points). At the same time, participation is most 
likely when the spouse of the respondent has an economic activity after the Hurricanes.
We may explain this finding as a form social capital. As there are no specific criteria to 
select individuals to participate in the program and as friendship matters, it is then 
plausible that when the spouse has set up a small business, this may leads to informal 
networks that facilitate information on recruitment process. By contrast, participation is 
less likely with housing ownership. That is, as a house is a symbol of wealth, the possible 
explanation is that wealthier people do not participate in the CFW. They are supposed to 
have access to other assets. 
27 
Table 4 
Individual determinants of participation 
 in the CFW activities—Probit estimates (N=93) 
Explanatory variables Mean+
 Marginal 
effect 
(dF/dx) 
Robust 
Std. Err. 
Age (years) 36.7634 .0119617** .0057178 
Sex (male=1) .666667 -.002215 .1148812 
Marital status (Married =1) .27957 -.0233759 .1041587 
Years of education 8.97849 .0231027 .0186376 
The respondent attended in formal vocational 
training (Yes =1) 
.290323 -.2645916* .1586779 
Informal vocational training (Yes=1) .483871 .0767434 .0864085 
Number of dependents 5.34409 -.0273587 .0172222 
The spouse of the respondent has an economic 
activity after the Hurricanes (Yes=1) 
.290323 .1832286** .0744116 
The respondent owns the house where he/her lived 
before the hurricanes (Yes=1) 
.387097 -.2397565*** .0950878 
Other member of the household participate in the 
CFW project (Yes=1) 
.247312 .1108668 .0901165 
The respondent benefited from other humanitarian 
program after the Hurricanes (Yes=1) 
.322581 .0872015 .0919558 
Other member(s) of the household benefitted from 
other humanitarian program after the Hurricanes 
(Yes=1) 
.129032 -.0933407 .1815282 
Years of residency in the locality where the 
respondent is interviewed 
16.7913 -.0020808 .0032318 
The respondent is the Household Head .827957 .3258047* .1905681 
Number of employed in the household after the 
Hurricanes 
.172043 .0489311 .1444438 
Wald chi2(15) 25.72** 
Log pseudolikelihood -39.156351 
Pseudo R2 0.2473 
Observed probability .7526882 
Predicted Probability at the mean .8254971 
Source: The authors’ estimates
Notes.—The dependent variable=1 if the respondent participate in the CFW component of the 
HERE Program. Number of obs = 93 due to missing value from certain variables. Number of 
children in the household dropped because of collinearity. Stars denote the rejection of H0: 
The coefficient equals zero at 10% (*), 5% (**) or 1% (***) levels of significance. For dummy 
variables, dF/dx is for discrete change of dummy variable from 0 to 1. + The average value for 
dummies variables indicate the proportion of the farmers who answered « Yes » to a question. 
6.2 Participation in CFW and quality of life 
The preferred regression reported in table 5 suggests that the respondent’s monthly 
expenditures respond positively and significantly with years of schooling and 
28 
participation of other member of the household in the CFW component of the HERE.10
Meanwhile, even though the respondent’s participation in CFW has the correct sign, it is 
not different from zero. Having said that, one possible explanation to this finding is that 
CFW program is important, but the daily salary rate (3.75 USD) seems to be low and the 
(officially) authorized period of participation in the program (12 days) that lead to a 45 
USD payment seem not enough to significantly impact the workers’ quality of life. We 
observe at the same time that the participation of the respondents conditional to the 
participation of another member of the household in the CFW is negatively associated 
with the monthly expenditures. This seems to indicate that salary earned is too low to 
facilitate intra-household transfers. 
We paid attention to both the relevance and the weak instruments problem highlighted in 
the recent econometric literature. We confirm that our instruments are valid, since they 
do not have explanatory power in the structural model (Test of overidentifying 
restrictions: Score chi2(1) = 2.61403 (p=0.1059))11, but they remain weak as the F-test is 
below the level of 10 (Stock, Wright and Yogo 2002). We have however selected the best 
set of instruments available from our database. 
We also explored the probability to begin the reconstruction of assets. We ran both binary 
(Table 6) and IV-probit (Table 7) models. The latter allows us to deal with endogeneity 
concerns. Table 7 shows that the number of dependents decreases by 5 percentage points 
the probability of reconstruction. By contrast if the respondent benefited from other 
humanitarian program, the probability to start reconstruction is increased by 19 
percentage points. Two other factors have significant and bigger effects on this 
probability: The respondent has the same economic activity after the hurricanes (Yes=1) 
and the respondent’s spouse has an economic activity after the Hurricanes (Yes=1). All 
these findings illustrate that additional resources are needed at least to begin the 
reconstruction of families’ assets and livelihoods. 
 
10 The return from an extra year of schooling is high enough (9%) in the current predicament of Gonaives 
and is comparable to broadly overall levels (see an international update in Pscharopoulos and Patrinos 
2004). 
11 Overidentifying tests the Null that over-identification restrictions are valid. 
29 
Table 5 
Determinants of income (N=93) 
Explanatory variables Coef. Robust 
Std. Err.
The respondent participated in the CFW activities (Yes=1)--Instrumented .3322004 .9924017
Years of education .0907307*** .0258165
Sex (Male=1) -.0612628 .2238233
Marital status (Married=1) .2123671 .2006783
The respondent attended in formal vocational training (Yes =1) .064047 .1859255
The respondent attended in informal vocational training (Yes =1) .0850771 .1901465
The respondent had an economic activity before the hurricanes (Yes=1) .0275615 .3682519
Number of dependents in the household .0573027 .0391797
Number of employed in the household after the hurricanes .0394442 .1361995
The respondent’s spouse has an economic activity after the Hurricanes 
(Yes=1) -.5804057 .6530627
The respondent owns the house where he/her lived before the hurricane 
(Yes=1) .4810345 .3139856
The respondent received local transfers from relatives after the 
hurricanes (Yes=1) .1680777 .1922355
The respondent received international transfers from migrants after the 
hurricanes (Yes=1) .2049803 .2029563
Other member of the household participates(ed) in CFW activities 
(Yes=1) 1.474343*** .4298615
The respondent benefited from other humanitarian program (Yes=1) -.2890697 .1994012
Other member of the household benefitted from other Humanitarian 
Program (Yes=1) -.1342939 .271282 
The respondent is member of any Organization (Yes=1) -.4810588 .3890408
The respondent is the household head (Yes=1) .0269455 .3189838
The respondent participates(ed) in CFW * another member in the 
household participates(ed) in CFW (Yes=1) -1.441481* .7656141
The respondent participated in CHF * His spouse has an economic 
activity after the Hurricanes (Yes=1) .3731427 .7585342
Intercept 7.148332*** 1.096435
Wald chi2(20) 106.60*** 
R-squared 0.3760 
Root MSE .65918 
Diagnosis on the reduced form: 
R2 0.5096 
Adjusted R2 0.3646 
Partial R2 0.0376 
Robust F(2,71); [Prob > F] 1.53923; [0.22] 
Source: The authors’ estimates
Notes.—The dependent variable is the natural logarithm of the average monthly expenditures 
after the hurricanes. The instruments are : (1) Age of the respondent and (2) years of residency 
30 
in the same locality. Stars denote the rejection of H0: The coefficient equals zero at 10% (*), 
5% (**) or 1% (***) levels of significance.
In table 7, we deal with endogeneity concerns. We use the better set of excluded 
instruments available from the data base (the respondent’ age, sex, marital status, years of 
residency in the same locality, and attendance in the local associational life such as 
church, local development organization, etc.). We assume that these instruments are 
directly associated with participation in CFW—as the criteria for workers selection are in 
a certain extent physical strength and residency in the locality as mentioned above, while 
at the same time, an important role is left to the team leader, a representative of the 
project site, in the workers recruitment—and indirectly associated with the probability to 
begin reconstruction. Firstly, we observe that the appropriate Wald test accept the Null 
Hypothesis of the exogeneity of participation in CFW activities with regard to the 
dependent variable. Secondly, we observe that only two explanatory variables have 
explanatory power in the probability of having started rebuilding the mean of existence 
which is most likely when the respondent’s spouse has an economic activity after the 
Hurricanes (Yes=1), and less likely with Number of dependents in the household. The 
evidence seems then clear enough that with the low wage rates currently paid to the 
workers, it is difficult to boost reconstruction phase without additional resources to be 
brought by other members in the family. In other words, there is no doubt that CFW 
activities are necessary, but they are not sufficient to help the targeted population to move 
forward in terms of reconstruction. That is, longer term interventions are necessary. 
31 
Table 6 
The likelihood of beginning the reconstruction of families’ assets (N=86) 
Explanatory variables 
Marginal 
effects 
(dF/dx) 
Robust 
Std. Err. 
The respondent participated in the CFW activities (Yes=1)-- 
Instrumented 
-.1174379 .1335356 
Age -.0036462 .0050373 
Sex (Male=1) -.2354349 .1554226 
Marital status (Married=1) .0301508 .1327159 
Years of education -.0109685 .0157214 
The respondent attended in formal vocational training (Yes =1) -.0342044 .1123252 
The respondent attended in informal vocational training (Yes =1) .0917204 .0997736 
Number of dependents in the household -.0511803** .0243767 
The respondent owns the house where he/her lived before the 
hurricane (Yes=1) 
.059777 .1022134 
The respondent benefitted from other humanitarian program (Yes=1) .1950199* .1237549 
Other member of the household benefitted from other Humanitarian 
Program (Yes=1) 
-.1299686 .080908 
Years of residency in the locality where the respondent is 
interviewed. 
.0022012 .0033199 
Number of employed in the household after the hurricanes -.1227035 .0922153 
The respondent had an independent economic activity before the 
hurricanes (Yes=1) 
-.2608153 .1860345 
The respondent has the same economic activity after the hurricanes 
(Yes=1) 
.3302623*** .0948915 
The respondent’s spouse has an economic activity after the 
Hurricanes (Yes=1) 
.4264092*** .1405357 
The respondent received local transfers from relatives after the 
hurricanes (Yes=1) 
.0666121 .1277877 
The respondent received international transfers from migrants after 
the hurricanes (Yes=1) 
-.0715719 .090422 
The respondent is member of any Organization (Yes=1) -.0507773 .0940275 
The respondent is the household head (Yes=1) -.0095331 .1577107 
Observed probability .255814 
Predicted probability at the means .1614916 
Wald chi2(20); [Prob > chi2] 36.76 [0.0125] 
Log pseudolikelihood -32.64401 
Pseudo R2 0.3325 
Source: The authors’ estimates
Notes.—Dependent variable=1 if the respondent expressed having started to rebuild his assets 
and livelihoods. For dummy variables, dF/dx is for discrete change of dummy variable from 0 
to 1. 
32 
7 Concluding remarks and recommendations 
This report attempts to assess how well the HERE program works. For most of the people 
we interviewed, the program is the unique work opportunity, and for certain young 
workers, it is their first employment experience. We have no doubt that the hurricanes 
impoverished the local population. It is found that at this time —the one of data 
collection—, all the expected results have not been yet achieved according to the 
econometric evidence. However, the program gave access to cash which in turn helps 
families to deal with certain expenses (food, schooling, debt repayment, cloths, etc.) or 
set up a small business (retailed commerce), that would not be performed without the 
project. Econometric evidence shows that when the participant’s spouse (either husband 
or wife) is occupied in an economic activity, it is then most likely to start rebuilding 
livelihoods and assets. 
We recommend: 
1 the continuation of the program’s activities with—if the budget allows that—
a higher wage rate or otherwise by providing a hot meal to workers; 
2 CHF to make an effort to pay the workers on time in order to keep them 
motivated; 
3 that an appropriate place be negotiated with the local authorities in order to 
dump rubbish. The last pictures in Appendix X give evidence that the danger 
is still present as the streets are under the mud that has been extracted by 
workers of other operators from the drainage canal or a cleaned site. All 
tough this is not in CHF’s sites, the latter may need to call a consensus in 
order to avoid that the good work done to be compromised; 
4 that beneficiaries have their house cleaned by the program after the public 
building; 
5 that local authorities (Mairie, Délégation, etc.) and aid agencies find 
appropriate means (such as credit) to help the population rebuild their assets; 
33 
6 that people with the higher level of education be employed not only as 
workers but in a better position in order to take a greater role in the future of 
their city; 
7 that CHF local Staff in Gonaives attempt to have a closer look—may be by 
offering guidelines to the team leaders—on the recruitment process which so 
far needs to be better clarified. Some basic criteria (mostly age, physical 
strength) should be formally used in the selection process, since it is 
observed in the data set five participants with respectively 65, 70, 72 and 81 
years old ; 
8 that relationship between controller, team head and workers to be formalized. 
Years of education or vocational training should be used in giving more 
responsibilities and then greater wage. 
Finally, to allow the community to rebuild their livelihoods after the hurricanes, 
program/projects with medium and long term potential impacts should be designed and 
implemented. 
Table 7 
Probability to begin the reconstruction--Correcting for endogeneity (N=86) 
Explanatory variables Coef. Robust
Std. Err.
The respondent participated in the CFW activities (Yes=1)-- 
Instrumented .2978343 1.110701
Years of education -.0356597 .0547929
The respondent attended in formal vocational training (Yes =1) .0529173 .4941067
The respondent attended in informal vocational training (Yes =1) .5234379a .3312549
The respondent had an independent economic activities before the 
hurricanes (Yes=1) -.6590422 .5508609
Number of employed in the household after the hurricanes -.2409772 .3666541
The respondent’s spouse has an economic activity after the 
Hurricanes (Yes=1) .9221013* .514062
The respondent owns the house where he/her lived before the 
hurricane (Yes=1) .3430387 .3511965
The respondent received local transfers from relatives after the 
hurricanes (Yes=1) .4480245 .4755537
The respondent received international transfers from migrants after 
the hurricanes (Yes=1) -.1262637 .4070776
The respondent benefitted from other humanitarian program 
(Yes=1) .5220923 .4086488
Other member of the household benefitted from other .7548711 .9741961
34 
Humanitarian Program (Yes=1) 
The respondent is the household head (Yes=1) -.2497073 .4847565
Number of dependents in the household -.1711466** .0821636
Other member of the household participates(ed) in CFW activities 
(Yes=1) .8117825 .7537716
The respondent participated in the CFW activities (Yes=1) * 
Other member of the household participates(ed) in CFW activities 
(Yes=1) 
-.0206095 1.034549
The respondent participated in the CFW activities (Yes=1) * Other 
member of the household benefited from other Humanitarian 
Program (Yes=1) 
-1.902029a 1.182935
Constant -.0449505 1.356605
/athrho -.3373986 .4155691
/lnsigma -1.176386*** .0754994
rho -.325153 .3716332
sigma .3083913 .0232833
Log pseudolikelihood -55.778638 
Source: The authors’ estimates
Notes.—Dependent variable=1 if the respondent expressed having started to rebuild his assets 
and livelihoods. The a significant at the 11% level. The instruments are (Sex (Male=1)) : age, 
sex, marital status, years of residency in the same locality, and attendance in the local 
associational life (such as church, local development organization, etc.). Stars denote the 
rejection of H0: The coefficient equals zero at 10% (*), 5% (**) or 1% (***) levels of 
significance. a=significant at 11%. Wald test of exogeneity (/athrho = 0): chi2(1) = 0.66 Prob 
> chi2 = 0.4169. 
35 
References 
[1] Feeman H.E., Rossi P.H. & Wright S.R. (1979). Évaluer des projets sociaux dans les pays en 
développement. Paris : Centre de Développement de l’OCDE. 
[2] Matchaya, Greenwell Collins. 2005. The Impact of Rural Producers Organizations on People’s 
Livelihoods. The Case of National Small Holder Farmers’ Association of Malawi in Kasungu 
District: An Instrumental Variable Approach. Mimeographed. Leeds : Leeds University 
Business School (UK). 
[3] Narayan, Deepa and Lant Pritchett (1999). “Cents and Sociability: Household Income and 
Social Capital in Rural Tanzania”. Economic Development and Cultural Change. (April). 47(1) 
: 871-897. 
[4] Psacharopoulos, Goerges and Harry Anthony Patrinos (2004). « Returns to Investment in 
Education: A Further Update ». Education Economics 12(2): 112-134. (August).
[5] Stock, James H.; Jonathan H. Wright and Mothiro Yogo (2002). « A survey of weak 
instruments and weak identification in Generalized Method of Moments ». Journal of Business 
& Economic Affairs 20(4):518-529. 
36 
APPENDICES 
Appendice I—Interview Guide with Local CHF Staff in Gonaives 
 (Mission of February 13, 2009 in Gonaïves) 
Questions Responses 
What are the criteria to recruit the workers for CFW? 
 
Does the recruitment process target individuals or 
households? 
Can one worker benefit more than one contract over 
the execution period of the program? 
If not, why? 
If yes, according to what modalities? 
Have the people employed received any specific 
training from CHF? 
Does any mechanism exist to control the productivity 
of the individual workers or the teams? 
Does the remuneration of the workers base on the 
number of work-days or on the quality of the work 
performed? 
Does the community participate in the selection of the 
projects? 
If yes, which ones? 
 
If not, what has been the basis for their selection? 
And how does the community adhere to them? 
What is its responsability? 
Do the organizations operate according to a shared 
vision in Gonaives? 
What are their activities ? 
During what periods do they do their activities? 
Give a brief overview of their realizations. 
Quelles sont les relations du CHF avec les autorités 
locales ? 
37 
What are the relationships between CFH and the local 
authorities? 
What are the relationships between CHF and the other 
organizations? 
The name of Fonkozé appears on the payroll sheets, 
what role does it play in the process? 
Do you have any comments on the program? 
 Thank you very much for your time? 
38 
Appendix II—Time Frame submitted by Gecades SA at the early phase of the process 
Proposed work Calendar for the HERE mid-term Evaluation program 
January 2009 February 2009 March 2009 
L M M J V L M M J V L M M J V L M M J V L M M J V L M M J V L M M J V 
Activities 26 27 28 29 30 2 3 4 5 6 9 10 11 12 13 16 17 18 19 20 23 24 25 26 27 2 3 4 5 6 9 10 11 12 13 
Devising survey instruments, 
focus group protocols and 
template for data entry and 
treatment 
Sampling from the list 
submitted by CHF the 
beneficiaries to be surveyed 
according to a random 
sampling technique 
Training the surveyors on the 
survey and focus group 
instruments 
Pre-testing the 
instruments/finalizing the 
instruments and locating 
beneficiaries and non 
beneficiaries 
Performing data 
collection/validating collected 
data 
Entering and treating data 
Performing data analysis and 
writing preliminary report 
Reviewing the first draft and 
writing of final report (both in 
French and English) 
39 
Appendix III—The damages of the Hurricane Jeanne in photos (12 days later) 
Satellite picture of Gonaïves, before (to the left) and after (to the right) the passage of the hurricane Jeanne 
Source : Busselen Tony. 2004. L’ouragan Jeanne- 1 500 morts en Haïti. Comment est-ce possible? Web Site: 
www.ptb.be/scripts/article.phtml?lang=1&obid=24830. Processed on May 10, 2005. 
40 
Appendix IV—Questionnaire for individual Survey 
Sir, Madam, 
We seek to understand and to assess the direct and short-term impacts of the HERE 
program currently implemented by CHF international Haiti in Gonaives. We specifically 
aim at measuring the cash-for-work activities impacts on the standard of living of the 
beneficiaries who were affected by the recent hurricanes in Gonaives. The information 
that will be collected will be kept confidential. We inform you that we will write down on 
this questionnaire some of the information you will be providing during the survey. You 
may have access to the results of the survey if need be. These results should in many 
ways allow CHF to be more efficient and efficacious in implementing its activities in the 
community. Do you accept this interview? Thank you for your collaboration. 
Code (not to be filled) : Date:
Name of the 
surveyor :
First and last name 
of the surveyed 
person : 
No. Questions and screens Responses et codes Go to 
1. What is your age ? 
2. Sex 0 ( ) Female 
1 ( ) Male 
3. What is Your marital status? 1 ( ) Married ; 
0 ( ) Other 
4. What is your highest level of education? 
5. Do you have a diploma in any vocational field? 0 ( ) No 
1 ( ) Yes. explain : _______________ 
6. Have you attended any informal vocational training? 0 ( ) No 
1 ( ) Yes 
If Yes. explain : ______________ 
7. What economic activities did you use to perform before the 
passage of the hurricanes last summer? 
1( ) independent worker (little 
commercial activities, tailor/seamstress, 
etc.) ; 
2( ) public sector employee ; 
3( ) Private sector employee; 
4( ) Agriculture ; 
5( ) Other—explain : ____________ ; 
6 ( ) None. 
8. Do you perform the same activity (ies) after the passage of 
the hurricanes? 
0 ( ) No 
41 
1 ( ) Yes 
9. Do you currently have a job? 0 ( ) No 
1 ( ) Yes 
If yes, explain : 
10. How many people do you have under your care in your 
household? 
11. How many children (less than 15-years old) do you have in 
your household? 
12. What was your total revenue during last month (whether 
you were under a contract with CHF or not)? ____________ HTG 
13. What was the total monthly revenue of your household 
before the passage of the hurricanes? ____________ HTG 
14. Give an estimate of your weekly expenses to feed the 
members of your household during last month. 
1.before the hurricanes : 
____________HTG 
2.after the hurricanes : 
____________HTG 
15. Give an estimate of your other weekly expenses (toilet 
articles, transportation, healthcare, education, etc.) for your 
household during last month. 
1.before the hurricanes : 
____________HTG 
2.after the hurricanes : 
____________HTG 
16. How many members of your household have a job? 1.before the hurricanes : 
2.after the hurricanes : 
17. Does your spouse have a generating income activity? 1.before the hurricanes : 0( ) No 1( ) Yes 
2.after the hurricanes : 0( ) No 1( ) yes 
18. Did you own the house you inhabited before the hurricanes? 0 ( ) No 
1 ( ) Yes 
19. Have you received after the hurricanes any money transfer 
from your relatives living in other regions of Haiti ? 
0 ( ) No 
1 ( ) yes 
20. Have you received after the hurricanes any money transfer 
from your relatives living abroad? 
0 ( ) No 
1 ( )Yes 
21. Have you participated in the Cash-for-work activities of 
CHF international? 
0 ( ) No 
1 ( ) Yes 
If no, go 
to 25 
22. If yes, are you still under a contract with CHF ? 0 ( ) No 
1 ( ) Yes 
23. How many biweekly periods have you spent in the 
program? 
 
24. What have you done with the cash received from the 
program ? 
25. Have other members of your household participated in the 
program ? 
0 ( ) No 
42 
1 ( ) yes 
26. Are you beneficiary of any other humanitarian or 
reconstruction programs after the passage of the 
hurricanes ? 
0 ( ) No 
1 ( ) Yes 
27. Are other members of your household direct beneficiaries of 
the program ? 
0 ( ) No 
1 ( ) Yes 
28. What is the distance (walking time) between your dwelling 
place and the project site where you work for the Cash-for￾work activities program of CHF international Haiti in 
Gonaives? 
 
29. How long have you been living in the same locality? 
30. Have you already started rebuilding your assets and other 
means of existence destroyed by the hurricanes? 
0 ( ) No 
1 ( ) Yes 
If yes, give some examples : 
31. Are you a member of any network association (Church, 
political group, socio-professional association, youth 
groups, local development group, club, etc.) in your 
locality? 
0 ( ) No 
1 ( ) yes 
If yes, explain : 
43 
Appendix V—Focus Group Protocol 
A. General information 
Code (not to be filled) : Date : 
interviewer : Beginning time of the 
interview : 
Assistant : Ending time of the 
interview : 
B. Identification of the interviewed group 
Localities represented 
(local areas) : 
 Number of men: 
Projects (sites) 
represented : 
Number of women : 
Number of participants 
in the group : 
 
C. Generalities 
1. In a general manner, elaborate on your experience with the different types of 
huma- nitarian aids (aids from the central government, local authorities, NGOs, 
etc.) that have helped you to cope with socio-economic difficulties caused by the 
passage of the hurricanes in Gonaives. 
2. Express your degree of satisfaction from such humanitarian aids. 
D. Perception of the community about the aids/activities of CHF international Haiti 
after the passage of the hurricanes 
3. Do you think CHF (through its cash-for-work activities) has done something 
positive that the other aids providers have not done in seeking to improve the 
quality of life of the beneficiaries after the passage of the hurricanes? According to 
your answer, explain how? 
4. Express your opinion about the activities realized within the community by the 
HERE program of CHF International Haiti. 
E. Socio-economic Impacts of the Cash-for-work activities of CHF international in 
Gonaives 
5. Had you not been employed by the cash-for-work activities of CHF international 
Haiti would you have found other work opportunities after the passage of the 
hurricanes to be able to survive? 
44 
6. What do you do with the cash you have received from the Cash-for-Work 
activities of CHF? (Buying food, clothes, paying school for your children, 
acquisition of assets/livestock, commerce, savings, etc.). 
7. Make a comparison between the quality of life you are having after participating in 
the Cash-for-work activities of CHF international Haiti and that that you had 
before the losses brought to you by the passage of the hurricanes (for example, 
increase in the amount of expenses allocated to satisfying basic needs, new 
expenses that were not possible before, etc.). 
8. Talk about any other kind of benefits, apart from the economic ones, that the Cash￾for-work activities of CHF have created within the community (for example, 
creation of new association among the beneficiaries) ? 
9. In your opinion, what would be the greatest impact on the quality of your life as a 
result of participating in the Cash-for-work activities of CHF international Haiti 
after the passage of the hurricanes? 
45 
Appendix VI— HERE Project Status by January 15, 2009 
Item Project Site Building Type Clean-Up Phase Rehabilitation 
Phase 
Projected End 
Date 
1 Rehabilitation of surgery block Hospital Completed Started Jan 23rd 
2 
Rehabilitation of emergency services 
block Hospital Completed Started Jan 23rd 
3 Rehabilitation of Maternity Ward Hospital Completed Started Jan 23rd 
4 Rehabilitation of Pedriatic Unit Hospital Completed Started Jan 23rd 
5 
Rehabilitation of general medicine 
department Hospital Completed Started Jan 23rd 
6 Rehabilitation of ARV section Hospital Completed Started Jan 23rd 
7 
Providence Dépot Pharmaceutique 
(DSA) Hospital Completed Started Jan 23rd 
8 Ophtalmology Hospital Completed Started Jan 23rd 
9 
Foreign Doctor's Housing (Ministry of 
Heath Complex) Hospital Completed Started Jan 23rd 
10 K Soleil Paint Repair Hospital Completed Started Jan 23rd 
11 
K Soleil Rehabiliation of electrical and 
hydraulic systems Hospital Completed Started Jan 23rd 
12 
K Soleil Roof restructuring and cooling 
system repair Hospital Completed Started Jan 23rd 
13 Palais de justice Government Admin Completed Not started Jan 25th 
14 Loge Sympathie des Coeurs Public Utility 20/01/2009 Not started 5 work days 
15 Affaires Sociales Government Admin Completed Not started Feb 5th 
16 Universite de l'indépendance. School Completed Not started Jan 25th 
17 Mémorial de l'Ind. Public Monument Completed Not started Jan 25th 
18 Heureuse Independence Public Monument 20/01/2009 Not started February 9th 
19 Dépot départemental Government Admin Completed Not started Jan 25th 
20 Direction departementale Government Admin Completed Not started Jan 25th 
21 Résidences des Médecins Housing for Doctors Completed Not started Jan 25th 
22 DéLégation de l'ART. Government Admin Completed Not started Jan 25th 
23 Bibliotheque Jacques Stephen Alexis Government Admin Completed Not started Jan 25th 
25 Ecole Petit de Roseline School Completed Not started 26 work days 
26 Cesmed School Completed Not started 26 work days 
27 Ecole Nationale Descahos Education 19/01/2009 Not Started 26 work days 
28 Collège Eben-ezer Education 19/01/2009 Not Started 26 work days 
29 Ecole Pierre-Marcel Ste-Marie Education 19/01/2009 Not Started 26 work days 
30 
Assainissement routier et collecte 
d'alluvion dans les localités Nan M et 
Provence PG 
Road 
22/01/2009 Not Started 48 work days 
Notes.—This list has been submitted to Gecades SA by CHF on January 24. But payroll sheets have 
been made available only for the project number 14, 16, 22 and 23. By contrast, the College 
Immaculée Conception and Hopital La Providence for which we have payroll, are not listed above. 
46 
Appendix VII—Gonaives in Photos after the Hurricanes in September 2008 
1 2 
3 4 
Sources :1, 2, 3 :Alter presse, September 13, 2008 ; 4 : AFP/Thony Bélizaire, September 2008 
47 
Appendix VIII— HERE Summary of Performance Results
Cross￾cutting 
Theme Performance Indicator 
Suppo
rting 
Docum
ents 
LOP 
Targets 
Frequen
cy of 
Data 
Collectio
n 
Performance per reporting period 
Cumul
ative 
Total 
% 
Achiev
ed 
01-
Nov 
15-
Nov 
01-
Dec 
15-
Dec 
01-
Jan 
15-
Jan 01-Feb 
15-
Feb 
01-
Mar 
15-
Mar 
01-
Apr 
(A) Cash 
for Work 
Number of people 
employed by CHF in the 
CFW 
Payroll 
Sheets 5400 
Bi￾Weekly 744 459 456 424 442 0 2525 47% 
Number of people 
employed by CARE in the 
CFW 
Payroll 
Sheets 4500 
Bi￾Weekly 477 600 608 577 534 0 2796 62% 
(B) 
Infrastruct
ure 
Rehabilitat
ion 
Number of infrastructure 
type 
constructed/rehabilitated 
Last 
payroll 
sheets 35 
Bi￾Weekly 0 0 0 0 0 0 0% 
Source: CHF (Monitoring and Evaluation Unity), January 2009. 
48 
Appendix IX—Buildings cleaned up by the CHF’s CFW activities 
(Source : Gecades SA, February 13 and 26, 2008) 
Hopital La Providence 
49 
Hopital La Providence 
One of the Gecades SA’s surveyors showing the level reached by the water during the 
Hurricanes at the Hospital La Providence 
50 
Bibliothèque JS Alexis 
Université de l’indépendance 
51 
Université de l’indépendance 
Délégation départementale 
52 
Collège Immaculée Conception 
Loge Sympathies des Cœurs 
53 
Appendix X—(A) CHF teams at work and (B) The need to find a consensus 
Source : Gecades SA, February 26-27, 2009 
Panel A: CHF’s teams at work cleaning up the “Canal Ceinture Bien AC Gateau” 
nearby the national Road crossing Gonaives 
54 
Panel B: The need to find a consensus to evacuate the mud 
55 
Appendix XI—Statistical data 
 (df=degree of freedom) 
Table 12.1 
Distribution by age 
Age Frequency Percentage 
Less than 25 years-old 8 6.4 
Between 25 and 35 years -old 56 44.8 
Between 36 and 46 years -old 37 29.6 
More than 46 years -old 24 19.2 
Total 125 100.0 
Table 12.2 
Distribution by sex 
Sex Frequency Percentage 
Female 47 37.6 
Male 78 62.4 
Total 125 100.0 
Table 12.3 
Distribution by marital status 
Marital status Frequency Percentage 
Other 95 76.0 
Married 30 24.0 
Total 125 100.0 
Table 12.4 
Distribution by level of education attained 
Level of education Frequency Percentage 
None 6 4.8 
1st fundamental cycle 9 7.2 
2nd
 fundament cycle 27 21.6 
3rd
 fundamental cycle 24 19.2 
Secondary level 44 35.2 
University level 15 12.0 
Total 125 100.0 
56 
Table 12.5 
Distribution by attendance to any formal vocational training 
Formal training Frequency Percentage 
No 87 69.6 
Yes 38 30.4 
Total 125 100.0 
Table 12.6 
Distribution by attendance to any informal vocational training
Informal training Frequency Pour cent 
No 68 54.4 
Yes 57 45.6 
Total 125 100.0 
Table 12.7 
Distribution according to whether or not the respondent had an income generating activity 
before the hurricanes 
Activity Frequency Pour cent 
No 16 12.8 
Yes 109 87.2 
Total 125 100.0 
Table 12.8 
Distribution according the nature of the income generating activity before the hurricanes 
Nature of the activity Frequency Percentage 
Salary-based 17 13.6 
Self-employed 92 73.6 
Total 109 87.2 
Missing value 16 12.8 
Total 125 100.0 
Table 12.9 
Distribution according to whether or not, the respondent performs the same economic 
activity after the hurricanes 
Same activity Frequency Percentage 
No 64 51.2 
Yes 61 48.8 
Total 125 100.0 
57 
Table 12.10 
Distribution according to the number of dependents in the household 
Number of people Frequency Percentage 
None 9 7.2 
Between 1 and 3 31 24.8 
Between 4 and 6 55 44.0 
More than 6 30 24.0 
Total 125 100.0 
Table 12.11 
Distribution according to the number of children (less than 15 years old) in the household 
Number of children Frequency Percentage 
None 17 13.6 
Between 1 and 3 78 62.4 
Between 4 and 6 25 20.0 
More than 6 5 4.0 
Total 125 100.0 
Table 12.12 
Distribution according to the monthly income of the respondent after the hurricanes 
Income Frequency Percentage 
None 2 1.6 
Less than 2500 HTG 56 44.8 
Between 2500 and 5000 
HTG 
30 24.0 
More than 5000 HTG 37 29.6 
Total 125 100.0 
Table 12.13 
Distribution according to the monthly income of the respondent before the hurricanes 
Income Frequency Percentage 
None 11 8.8 
Less than 2500 HTG 31 24.8 
Between 2500 and 5000 HTG 34 27.2 
More than 5000 HTG 49 39.2 
Total 125 100.0 
Table 12.14 
Distribution according to the monthly expenditures of the respondent before the hurricanes 
Expenditures Frequency % 
I don’t know 1 0.8 
Less than 2500 HTG 17 13.6 
Between 2500 and 5000 HTG 35 28.0 
More than 5000 HTG 72 57.6 
Total 125 100.0 
58 
Table 12.15 
Distribution according to the monthly expenditures of the respondent after the hurricanes 
Expenditures Frequency Percentage 
I don’t know 3 2.4 
Less than 2500 HTG 25 20.0 
Entre 2500 et 5000 HTG 40 32.0 
Plus 5000 HTG 57 45.6 
Total 125 100.0 
Table 12.16 
Distribution according to the number of employed in the household before the hurricanes 
Number of people Frequency Percentage 
None 94 75.2 
Only one person 29 23.2 
Two people 1 0.8 
More than 2 people 1 0.8 
Total 125 100.0 
Table 12.17 
Distribution according to the number of employed in the household after the hurricanes 
Number of people Frequency Percentage 
None 106 84.8 
Only one person 16 12.8 
Two people 2 1.6 
More than 2 people 1 0.8 
Total 125 100.0 
Table 12.18 
Distribution according to whether or not the spouse has an income generating activity 
before the hurricanes 
Income activity before 
hurricanes 
Frequency Percentage 
No 50 40.0 
Yes 49 39.2 
No answer 26 20.8 
Total 125 100.0 
Table 12.19 
Distribution according to the number of employed in the household after the hurricanes 
Income activity after 
hurricanes 
Frequency Percentage 
No 66 52.8 
Yes 27 21.6 
No answer 32 25.6 
Total 125 100.0 
59 
Table 12.20 
Distribution according to the housing tenure before the hurricanes 
Frequency Percentage 
No 79 63.2 
Yes 46 36.8 
Total 125 100.0 
Table 12.21 
Distribution according to the reception of money transfers from relatives resided in Haiti 
after the hurricanes 
Transfer s from 
internal migrants 
Frequency Percentage 
No 96 76.8 
Yes 29 23.2 
Total 125 100.0 
Table 12.22 
Distribution according to the reception of remittances from external migrants after the 
hurricanes 
Transfer s from 
external migrants 
Frequency Pour cent 
No 90 72.0 
Yes 35 28.0 
Total 125 100.0 
Table 12.23 
Distribution according to the participation in CHF’s CFW activities 
Participation in CHF’s CFW Frequency Percentage 
No 29 23.2 
Yes 96 76.8 
Total 125 100.0 
Table 12.24 
Participants who are still under contract with CHF 
Under contract Frequency Percentage 
No 94 75.2 
Yes 2 1.6 
Total 96 76.8 
Missing value 29 23.2 
Total 125 100.0 
60 
Table 12.25 
Distribution according to the number of bi-weekly periods performed in the CFW 
Number of bi-weekly Frequency Percentage 
Only one 40 32.0 
More than one 56 44.8 
Total 96 76.8 
Missing value 29 23.2 
Total 125 100.0 
Table 12.26 
Distribution according to whether or not another member in the household participates in 
CHF’CFW activities 
Participation of other family 
 Members to CFW. 
Frequency Percentage 
No 95 76.0 
Yes 30 24.0 
Total 125 100.0 
Table 12.27 
Distribution according to whether or not the respondent benefits from other humanitarian 
programs after the hurricanes 
Beneficiary from other 
humanitarian programs 
Frequency Percentage 
No 78 62.4 
Yes 47 37.6 
Total 125 100.0 
Table 12.28 
Distribution according to whether or not other members of the household benefit from 
other humanitarian programs 
Beneficiary from other humanitarian 
programs 
Frequency Percentage 
No 108 86.4 
Yes 17 13.6 
Total 125 100.0 
Table 12.29 
Distribution according to the number of years of residency in the same locality 
Number of years of residency Frequency Percentage 
Less than 3 years 16 12.8 
Between 3 and 10 years 46 36.8 
More than 10 years 63 50.4 
Total 125 100.0 
61 
Table 12.30 
Distribution according to whether or not the respondent has already started rebuilding 
assets and livelihoods 
Start rebuilding Frequency Percentage 
No 95 76.0 
Yes 30 24.0 
Total 125 100.0 
Table 12.31 
Membership in local organization 
Frequency Percentage 
No 64 51.2 
Yes 61 48.8 
Total 125 100.0 
Table 12.32 
Sex and participation in CFW 
 Participation in CFW 
Sex No Yes Total 
Female 
Count 14 33 47 
% 29.8% 70.2% 100.0% 
Male Count 15 63 78 
% 19.2% 80.8% 100.0% 
Total 
% 
 29 96 125 
 23.2% 76.8% 100.0% 
Chi2=1.834 df=1 p-value=0.176
62 
Table 12.33 
Sex and number of bi-weekly periods performed in CFW activities 
 Number of bi-weekly periods 
Sex One More than 
one 
Total 
Female 
Count 
% 
19 14 33 
57.6% 42.4% 100.0% 
Male 
Count 
% 
21 42 63 
33.3% 66.7% 100.0% 
Total 
Count 
% 
40 56 96 
41.7% 58.3% 100.0% 
Chi2=5.235 df=1 p-value=0.022
Table 12.34 
Sex and participation in other humanitarian programs after the hurricanes 
 Participation in other programs 
Sex No Yes Total 
Female 
Count 
% 
25 22 47 
53.2% 46.8% 100.0% 
Male 
Count 
% 
53 25 78 
67.9% 32.1% 100.0% 
Total 
Count 
% 
78 47 125 
62.4% 37.6% 100.0% 
Chi2= 2.722 df=1 p-value=0.099
Table 12.35 
Sex and reconstruction of the livelihood 
 Rebuilding your livelihood 
Sex No Yes Total 
Female 
Count 
% 
36 11 47 
76.6% 23.4% 100.0% 
Male 
Count 
% 
59 19 78 
75.6% 24.4% 100.0% 
Total 
Count 
% 
95 30 125 
76.0% 24.0% 100.0% 
Chi2= 0.15 df=1 p-value=0.904
63 
Table 12.36 
Age and reconstruction of the livelihood
Age Reconstruction 
No Yes Total 
Less than 25 years old 
Count 
% 
7 1 8 
87.5% 12.5% 100.0% 
Between 25 and 35 years old 
Count 
% 
40 16 56 
71.4% 28.6% 100.0% 
Between 36 and 46 years old 
Count 
% 
27 10 37 
73.0% 27.0% 100.0% 
More than 46 years old 
Count 
% 
21 3 24 
87.5% 12.5% 100.0% 
Total 
Count 
% 
95 30 125 
76.0% 24.0% 100.0% 
Chi2= 3.148 df=3 p-value=0.359
Table 12.37 
Level of education and participation in CFW activities 
Level of education Participation in CFW 
No Yes Total 
None 
Count 
% 
0 6 6 
.0% 100.0% 100.0% 
1st fundamental cycle 
Count 
% 
3 6 9 
33.3% 66.7% 100.0% 
2nd
 fundamental cycle 
Count 
% 
9 18 27 
33.3% 66.7% 100.0% 
3rd fundamental cycle 
Count 
% 
8 16 24 
33.3% 66.7% 100.0% 
Secondary level 
Count 
% 
7 37 44 
15.9% 84.1% 100.0% 
University level 
Count 
% 
2 13 15 
13.3% 86.7% 100.0% 
Total 
Count 
% 
29 96 125 
23.2% 76.8% 100.0% 
Chi2= 7.403 df=5 p-value=0.192
64 
Table 12.38 
Number of dependents and participation in CFW activities 
Number of dependents Participation in CFW 
 No Yes Total 
None 
Count 
% 
1 8 9 
11.1% 88.9% 100.0% 
Between 1 and 3 
Count 
% 
8 23 31 
25.8% 74.2% 100.0% 
Between 4 and 6 
Count 
% 
11 44 55 
20.0% 80.0% 100.0% 
More than 6 
Count 
% 
9 21 30 
30.0% 70.0% 100.0% 
Total 
Count 
% 
29 96 125 
23.2% 76.8% 100.0% 
Chi2= 1.951 df=3 p-value=0.583
Table 12.39 
Reconstruction and number of bi-weekly periods performed in CFW activities 
Reconstruction of your livelihood Number of bi-weekly periods 
 One More than 
one 
Total 
No 
Count 
%
33 40 73 
45.2% 54.8% 100.0% 
Yes 
Count 
% 
7 16 23 
30.4% 69.6% 100.0% 
Total 
Count 
% 
40 56 96 
41.7% 58.3% 100.0% 
Chi2= 1.570 df=1 p-value=0.210
65 
Table 12.40 
Reconstruction and participation of other members of the household in CFW activities 
Reconstruction of your livelihood Participation of other members 
 of your family in CHF 
 No Yes Total 
No 
Count 
% 
77 18 95 
81.1% 18.9% 100.0% 
Yes 
Count 
% 
18 12 30 
60.0% 40.0% 100.0% 
Total 
Count 
% 
95 30 125 
76.0% 24.0% 100.0% 
Chi2= 5.540 df=1 p-value=0.19
Table 12.41 
Reconstruction and participation of the respondent in other humanitarian programs 
Reconstruction of Participation of the dependents in 
your livelihood other humanitarian programs 
No Yes Total 
No 
Count 
% 
66 29 95 
69.5% 30.5% 100.0% 
Yes 
Count 
% 
12 18 30 
40.0% 60.0% 100.0% 
Total 
Count 
% 
78 47 125 
62.4% 37.6% 100.0% 
Chi2= 8.442 df=1 p-value=0.004
Table 12.42 
Reconstruction and participation of other members of the household in other humanitarian 
programs 
Reconstruction of Participation members of your family 
your livelihood in other humanitarian program 
 No Yes Total 
No 
Count 
% 
83 12 95 
87.4% 12.6% 100.0% 
Yes 
Count 
% 
25 5 30 
83.3% 16.7% 100.0% 
 Total 
 Count 
 % 
 108 17 125 
 86.4% 13.6% 100.0% 
Chi2= 0.316 df=1 p-value=0.574
66 
Table 12.43 
Reconstruction and number of employed in the household after the hurricanes 
Reconstruction Number of employed in household after the hurricanes 
of your livelihood 
None One Two people More than two 
people 
Total 
No 
Count 
% 
81 12 1 1 95 
85.3% 12.6% 1.1% 1.1% 100.0% 
Yes 
Count 
% 
25 4 1 0 30 
83.3% 13.3% 3.3% .0% 100.0% 
Total 
Count 
% 
106 16 2 1 125 
84.8% 12.8% 1.6% .8% 100.0% 
Chi2= 1.076 df=3 p-value=0.783
Table 12.44 
Reconstruction and economic activity of the spouse 
Reconstruction of Economic activity of the spouse 
your livelihood 
 No Yes No answer Total 
 No 
Count 
 % 
54 15 26 95 
56.8% 15.8% 27.4% 100.0% 
 Yes 
 Count 
 % 
12 12 6 30 
40.0% 40.0% 20.0% 100.0% 
 Total 
 Count 
 % 
66 27 32 125 
52.8% 21.6% 25.6% 100.0% 
Chi2= 7.895 df=2 p-value=0.019
Table 12.45 
Local private transfers and reconstruction 
Reconstruction of livelihood Reception transfers from internal migrants Total 
 Non Yes 
 No 
Count 76 19 95 
% 80.0
% 
20.0% 100.0% 
 Yes 
Count 20 10 30 
 % 66.7
% 
33.3% 100.0% 
 Total 
Count 96 29 125 
 % 76.8 23.2% 100.0% 
Chi2= 2.275 df=1 p-value=0.131
67 
Table 12.46 
Reception of remittances and reconstruction 
Reconstruction of livelihood Reception transfers from external migrants 
 No Yes Total 
 No 
Count 
 % 
68 27 95 
71.6% 28.4% 100.0% 
Yes 
Count 
 % 
22 8 30 
73.3% 26.7% 100.0% 
 Total 
 Count 
 % 
90 35 125 
72.0% 28.0% 100.0% 
Chi2= 0.035 df=1 p-value=0.852
Table 12.47 
Housing tenure and reconstruction 
Reconstruction Housing tenure 
 No Yes Total 
 No 
Count 
% 
59 36 95 
62.1% 37.9% 100.0% 
Yes 
Count 
% 
20 10 30 
66.7% 33.3% 100.0% 
Total 
Count 
% 
79 46 125 
63.2% 36.8% 100.0% 
Chi2= 0.204 df=1 p-value=0.652
Table 12.48 
Participation in CFW and membership in associational life 
Participation in CFW Membership in associational life Total 
 No Yes 
No Count 8 21 29 
% 27.6% 72.4% 100.0% 
Yes Count 56 40 96 
% 58.3% 41.7% 100.0% 
Total 
Count 64 61 125 
% 51.2% 48.8% 100.0% 
Chi2=8.427, df=1, p-value=0.004