PERFORMANCE EVALUATION: HAITI WATER AND SANITATION PROJECT Baseline Survey | Draft Report Haiti Evaluation and Survey Services (ESS) November 2020 CROIX-DES-BOUQUETS CTE OFFICE PHOTO BY MARC LEE STEED PERFORMANCE EVALUATION: Haiti Water and Sanitation Project Baseline Survey | Report Date: November 2020 Submitted to USAID/Haiti Evaluation Mechanism Number: AID-521-C-17-00002 Haiti Evaluation and Survey Services for USAID/Haiti Contact: Jennifer Mandel, Chief of Party 2300 Clarendon Blvd, Suite 1000 Arlington, VA 22201, USA i | WATSAN FINAL REPORT (BASELINE) USAID.GOV CONTENTS Abstract............................................................................................................... Executive Summary..........................................................................................i 1. Introduction ............................................................................................1 1.1 Water and Sanitation Project ................................................................................................................ 1 1.2 WATSAN Performance Evaluation ...................................................................................................... 2 2. Desk Review Findings.............................................................................3 3. Methodology............................................................................................5 3.1 Sampling....................................................................................................................................................... 5 3.2 Fieldwork Preparations ........................................................................................................................... 6 3.3 Data Collection ......................................................................................................................................... 7 3.4 Data Analysis.............................................................................................................................................. 8 3.5 Indicator Definitions................................................................................................................................. 9 3.6 Disaggregations........................................................................................................................................11 3.7 Limitations................................................................................................................................................12 4. Findings ..................................................................................................13 4.1 Description of the Sample ....................................................................................................................13 4.2 Water.........................................................................................................................................................15 4.3 Sanitation...................................................................................................................................................31 4.4 Hygiene......................................................................................................................................................41 5. Conclusions ...........................................................................................45 6. References .............................................................................................47 7. Annexes .................................................................................................49 7.1 Annex A: Household Questionnaire..................................................................................................49 7.2 Annex B: Detailed Methodology .........................................................................................................79 7.3 Annex C. Detailed Project Context...................................................................................................86 7.4 Annex D. Evaluation Scope of Work.................................................................................................90 7.5 Annex E. Comparison of JMP vs. PIRS definitions...........................................................................99 USAID.GOV WATSAN BASELINE SURVEY REPORT | ii FIGURES Figure 1. USAID/Haiti WATSAN target communes................................................................................................. 1 Figure 2. At least basic drinking water service, by commune (IND 1)...............................................................15 Figure 3. Access to basic water service, by SES and commune............................................................................19 Figure 4. Access to piped source on premise, by SES and commune.................................................................19 Figure 5. Average round-trip collection time (minutes), by commune and water service ............................20 Figure 6. Average round-trip collection time (minutes), travel versus queuing, by commune.....................20 Figure 7. Households (%) reporting water “always” available from main source, by commune..................21 Figure 8. Households (%) reporting insufficiency last month, by commune......................................................21 Figure 9. Main reason for insufficiency last month, by water service and commune...............21 Figure 10. Main drinking water is “always” or “mostly” safe, by service level and commune......................23 Figure 11. Correlation between perception of safety and reported treatment of main drinking water....23 Figure 12. Plans to apply for a connection, by commune ......................................................................................30 Figure 13. Main reason household planning to apply for a connection, by commune....................................30 Figure 14. Water expenditures (monthly total and budget share), by commune and SES............................31 Figure 15. Main sanitation facility, percent (%) at least basic, by commune (IND 3)......................................32 Figure 16. Access to basic sanitation facility, by socioeconomic status..............................................................34 Figure 17. Level of access to hygiene facility, by commune...................................................................................42 Figure 18. Presence of handwashing facility at the household, by commune....................................................43 Figure 19. MDES for n=310, across possible baseline values (p1).......................................................................82 Figure 20. Haiti WATSAN results framework .........................................................................................................86 Figure 21. Households (%) with at least basic drinking water service, JMP vs. PIRS.................................... 100 TABLES Table 1. Basic water service: USAID PIRS versus JMP service ladder.................................................................. 9 Table 3. Basic sanitation service: USAID PIRS versus JMP service ladder.........................................................11 Table 4. Results of total number of survey attempts, by commune....................................................................13 Table 5. Sample characteristics (unweighted and weighted).................................................................................14 Table 6. Basic water service breakdown, by commune (IND 1) .........................................................................16 Table 7. Breakdown of water sources, by water service and commune ...........................................................17 Table 8. Problems faced with supply, by service level and commune.................................................................22 Table 9. Water treatment methods, by commune..................................................................................................24 Table 10. Water storage, by commune......................................................................................................................24 Table 11. Small container observation, by commune .............................................................................................25 Table 12. Other sources of water used for drinking, by commune....................................................................26 Table 13. Main source of water for other domestic purposes, by commune ..................................................27 Table 14. Other sources of water used for domestic purposes, by commune................................................28 Table 15. Access to, use of, and supply from tap on premises, by commune ..................................................29 Table 16. Breakdown of sanitation facilities, by sanitation service and commune...........................................33 Table 17. Facility location and sharing, by commune..............................................................................................35 Table 18. Risks faced when visiting sanitation facility, by commune ...................................................................35 Table 19. Any member of household trained in relation to open defecation, by commune ........................36 Table 20. Sanitation facility leaks or overflows seasonally, by commune ..........................................................37 Table 21. Access and barriers to accessibility, by commune ................................................................................37 Table 22. Cooking, laundry, and bathing water disposal, by commune..............................................................38 iii | WATSAN FINAL REPORT (BASELINE) USAID.GOV Table 23. Garbage disposal, by commune .................................................................................................................39 Table 24. Desire and willingness to pay for sanitation facility improvements, by commune ........................39 Table 25. Desired improvements to sanitation facility, by commune.................................................................40 Table 26. Septic tank emptying, by commune ..........................................................................................................41 Table 27. Hygiene facility by main water source and main sanitation facility....................................................43 Table 28. Reported handwashing among households without facilities or unobserved.................................44 Table 29. Sample sizes, p1 based on WATSAN data (water)...............................................................................83 Table 30. Sample sizes, p1 based on JMP data..........................................................................................................83 Table 31. Sample sizes, p1 based on WATSAN data (sanitation)........................................................................83 Table 32. Sample sizes, p1 based on JMP data (sanitation)....................................................................................84 Table 33. Population in each commune for post-stratification adjustment.......................................................85 Table 34. Households sampled from a single or multiple-household building, by commune........................85 Table 35. WATSAN activity progress to date .........................................................................................................87 Table 36. WATSAN activity planned activities for year 3 (2019-2020).............................................................88 Table 37. Basic water service breakdown, by commune (IND 1), PIRS ......................................................... 100 Table 38. Water sources used in Mirebalais, by SES............................................................................................ 101 USAID.GOV WATSAN BASELINE SURVEY REPORT | iv ACRONYMS ADS Automated Directives System AECID Agency for International Development Cooperation BCC Behavior Change Communications CFET Centre de Formation et d’Encadrement Technique CNIGS Centre National de l’Information Géo-Spatiale CTE Centre Technique d’Exploitation DAI DAI Global, LLC DDL Development Data Library DHS Demographic and Health Survey DINEPA Water and Sanitation National Direction EQ Evaluation Question ESS Evaluation and Survey Services ET Evaluation Team FY Fiscal Year HFC High-Frequency Check IDB Inter-American Development Bank IND Indicator IRB Institutional Review Board JMP Joint Monitoring Programme M&E Monitoring and Evaluation MEL Monitoring, Evaluation, Learning MTPTC Ministry of Publics Works, Transportation, and Communication PCA Principal Component Analysis PE Performance Evaluation PIRS Performance Indicator Reference Sheets SES Socioeconomic Status SI Social Impact SOW Scope of Work SRS Stratified Random Sampling TOC Theory of Change QIP Quick Impact Project UNICEF United Nations International Children’s Emergency Fund USAID United States Agency for International Development WASH Water, Sanitation, and Hygiene WATSAN Water and Sanitation Activity USAID.GOV WATSAN BASELINE SURVEY REPORT | ABSTRACT To improve access to basic water and sanitation in Haiti, USAID is supporting the Water and Sanitation activity (WATSAN), which involves infrastructure building and capacity building for service delivery. WATSAN works in select urban and peri-urban zones within five targeted communes, including Cap￾Haitian, Croix-des-Bouquets (Canaan), Mirebalais, Les Cayes, and Jérémie. USAID enlisted Social Impact’s (SI) Haiti Evaluation and Survey Services (ESS) activity to conduct a performance evaluation (PE) of WATSAN. This report details findings from the baseline household survey, which measured two key outcome indicators: percentages of households with access to basic drinking water (IND 1) and basic sanitation facilities (IND 3). Using Joint Monitoring Programme (JMP) definitions of basic water and sanitation services, 94 percent of households in the WATSAN zones have access to at least basic drinking water, and 63 percent of households in WATSAN zones have access to at least basic sanitation. FY19 HL8 PIRS definitions differ from JMP meaningfully (see Section 3.5.1) Apart from reporting on these two indicators, this survey provides valuable additional information. For water, this includes availability and burden of water collection, multiple concurrent source use, access and supply from piped households, and demand for new connections. For sanitation, this includes accessibility and safety, as well as demand and willingness to pay for improvements. Findings provide relevant context toward later evaluation of Water and Sanitation Project achievements as well as for context monitoring by USAID/Haiti. i | WATSAN FINAL REPORT (BASELINE) USAID.GOV EXECUTIVE SUMMARY BACKGROUND & WATSAN ACTIVITY Access to improved water services and sanitation facilities in Haiti remains among the lowest in the Western hemisphere. According to Joint Monitoring Programme (JMP) estimates, access to basic water services has declined over the past two decades, and progress toward basic sanitation has stagnated.1,2 To enhance access to improved water and sanitation in Haiti, the United States Agency for International Development (USAID) is supporting the Water and Sanitation project (WATSAN) implemented by DAI Global, LLC (DAI). This four-and-a-half-year intervention, which began in 2018 and will run until 2022, aims to increase access to sustainable water and sanitation services. Specifically, WATSAN aims to augment access to water and sanitation services in urban and peri-urban communities (referred to as “zones”) in targeted communes that are cholera hotspots or recovering from Hurricane Matthew. These communes include Cap-Haitian, Croix-des-Bouquets (Canaan), Mirebalais, Les Cayes, and Jérémie. The activity has three mutually reinforcing components: increased access to sustainable water supply services (Component 1), increased access to sustainable sanitation services (Component 2), and strengthened enabling environment for sustainable delivery, operation, and maintenance of water, sanitation, and hygiene (WASH) services (Component 3). WATSAN plans to realize these components through an integrated combination of infrastructure building, capacity building for service delivery management, and improving the enabling environment. WATSAN is designed to transition to longer-term, comprehensive interventions that build more sustainable capacity at local and national levels to reduce waterborne diseases in priority geographic zones. ESS PERFORMANCE EVALUATION In adherence with accountability and learning institutional objectives, USAID enlisted Social Impact’s (SI) Haiti Evaluation and Survey Services (ESS) project to conduct a performance evaluation (PE) of WATSAN.3 The WATSAN PE evaluation questions (EQs), as defined by USAID, are listed below. All three EQs will be answered in full when the performance evaluation is conducted later. This report details findings from the baseline survey which was conducted specifically to collect baseline values that will later be used to answer EQ 1, after subsequent surveys are conducted to measure changes over time. EQ 1) To what extent and in what ways has WATSAN increased access to new or improved basic drinking water and sanitation facilities? EQ 2) To what extent has WATSAN contributed to improved institutional capacity to better manage existing systems and increase revenue? EQ 3) To what extent does the WATSAN Theory of Change (TOC) remain valid and relevant to the activity’s current objectives? As per the ESS baseline survey scope of work (SOW), SI measured two indicators at baseline: (1) the percentage of households in the target communes using basic drinking water services, and (2) the 1 WHO/UNICEF, 2019a. 2 WHO/UNICEF, 2019b. 3 USAID enlisted SI to conduct this evaluation in May 2017. USAID.GOV WATSAN BASELINE SURVEY REPORT | ii percentage of households in the target communes using basic sanitation services. According to its Monitoring and Evaluation Plan, WATSAN aims to improve these indicators by 20 percent each from baseline over the life of the activity. The evaluation team (ET) measured additional indicators, described in this report, to provide context for these key indicators and the household-level WASH environment in WATSAN zones. Baseline findings will serve as a reference point for later PE data collection and allow for measurement of progress toward better access to improved water and sanitation. METHODOLOGY To measure the key indicators, the ET conducted baseline activities between November 2019 and March 2020. The ET worked with a local survey firm to administer a household survey across WATSAN zones in targeted communes (the household questionnaire is in Annex A). Before data collection, the ET led a one-week enumerator training. Field-team enumerators collected data using an ET-designed and electronically programmed survey in January and February 2020. The ET created the sampling frame using GPS data from Haiti’s Centre National de l’Information Géo-Spatiale (CNIGS) and enumeration area shapefiles from within the WATSAN zones in each of the five targeted communes. Based on sample size calculations, the team collected data from 310 households in each of the five targeted communes, for a total of 1,550 households. These households were drawn specifically from within the WATSAN zones in the targeted communes, rather than from the communes as a whole, to obtain representative estimates of the WATSAN zones. After completing data collection, the ET conducted a robust set of quality checks and finalized the dataset for analysis and reporting. The ET used Stata 15 statistical software to conduct sampling, monitor data quality, and clean and analyze data. “At least basic water service” is defined as water from an improved source that is either on-premises or not more than 30 minutes away by round-trip collection time, including queueing. “At least basic sanitation facility” is defined as an improved facility not shared with other households. This baseline report uses the JMP service ladder classifications and definitions to define water and sanitation services. This is important because it differs from the FY19 HL8 PIRS definition most recently established by USAID, in ways that substantially affect the interpretation of these baseline results. For example, the JMP post-2015 monitoring water service ladder considers carts with small tank or drum improved sources, along with tankers, bottled waters, and, critically, water kiosks. These are all differences compared to the USAID PIRS definition, which is not as generous in classifying such sources as improved (perhaps in line with older JMP definitions). Specifically, in this case, because of the widespread use of private kiosks, the decision to classify them as improved (as implied in the JMP definitions) or unimproved (as implied in the PIRS definition), makes an enormous difference on the resulting estimates of access to basic water service. See Section 3.4 for more on classifications and definitions. The ET disaggregated key outcomes by household head sex and socioeconomic status (SES). SES quintiles are calculated within each commune, rather than across all communes, an analytical decision made based on the priority interest in commune-specific results. Other indicators are disaggregated by the main outcomes (access to basic water and sanitation). Additional detail on the methodology is in Section 3 in the report and Annex B. iii | WATSAN FINAL REPORT (BASELINE) USAID.GOV FINDINGS WATER. Overall, the ET found that 94 percent of households across all WATSAN zones in the targeted communes have access to at least basic drinking water service (using the JMP definition – for results using the PIRS definition, see Annex E). Results show little variation across communes. The result is primarily driven by the widespread use of private kiosks within 30 minutes round-trip collection time. Private kiosks are the most common main drinking water source across all targeted communes. The widespread use of private kiosks, and its effect on the estimated value of this indicator, suggests several points about WATSAN. This estimate would imply that there is almost no room to improve the targeted 20 percent over baseline values. This is relevant to the extent that the project’s performance is tied to this indicator. However, this does not imply that there are no avenues through which the project could meaningfully improve households’ water access or use. Since the majority of households rely on private kiosks, there is still considerable room for source-switching over time toward better-quality water service, such as piped water. Actual experience in terms of water access could further improve if the project were successful in achieving its primary goals of improving service delivery and service quality, such that existing customers experience more reliable and better-quality water, and as additional households potentially switch from private kiosks to piped water. From this perspective, therefore, to understand the potential impact of the project on water access among households in the selected areas, we posit that it will be necessary to look beyond IND 1, even while it is necessary to measure it as a required indicator for this evaluation and as part of the Activity’s Monitoring, Evaluation, and Learning (MEL) Plan. EQ 1 asks not only to what extent access has increased, but also in what ways; further, it asks about “new” access to basic drinking water (understood as those who change from less than basic to at least basic water service) and “improved” basic drinking water (understood as those even within the “at least basic” classification that experienced improvements in their water service). In this way, we believe that EQ 1 is broad enough to encompass the critical nuance needed to understand not only whether the project is “moving the needle” on IND 1, but also what is happening “under the hood” of that indicator. In this report, we focus on IND 1, but in future phases (midline and endline), an important part of contextualizing any analysis of IND 1 will necessarily involve a close look at what households are experiencing in terms of specific water services used, reliability, etc. Household Head Sex. There were no significant differences in terms of access to basic water service between female- versus male-headed households, across any communes. Socioeconomic Status. As most households have access to basic water service, primarily through private kiosks within 30 minutes round-trip collection time, we do not observe any statistically significant differences in access to basic water service across socioeconomic status. However, there are substantial differences in access to an on-premises tap based on socioeconomic status in all target communes. The only exception is in Canaan, where there are no households reporting their own tap on premises. The most substantial disparity is in Mirebalais. Therefore, despite wide socioeconomic disparities in access to an on-premises tap, a widespread lack of use of those kinds of water services, perhaps because of failures of service delivery, leads to widespread reliance on private kiosks across SES quintiles, effectively “equalizing” access to at least basic water sources. Water Collection. Households’ average roundtrip time required to collect water, including travel and queue time, is about 14 minutes across communes. Households with less than basic water service spent substantially more time collecting water than those with basic access (note that this finding is partly USAID.GOV WATSAN BASELINE SURVEY REPORT | iv determined by the fact that basic water service is defined, in part, by collection time). In most cases, households spent more time traveling than queueing, regardless of commune or water service. Availability & Problems with Supply. Most households said that water from their main drinking water source was available “always” or “most of the time.” However, many said that they lacked sufficient drinking water at least once in the last month. Common problems reported with main water source include poor quality, time spent collecting water, and bad odor/taste, depending on the location. Most households with at least basic water service reported no problems with their supply. Perception of Safety. Across all areas, most households said they believed that water from their main drinking water source was safe to drink “always” or “most of the time.” This is especially true for those with at least basic service. In line with this, overall, a relatively small proportion report treating their water in any way before consuming. Water treatment does appear to be associated with the perception of safety, which is especially apparent in Jeremie and Les Cayes, as well as among those with at least basic service in Canaan and Mirebalais. Multiple Concurrent Source Use. The survey asked whether households regularly use any other drinking water sources besides their main source. In Cap-Haitian, sachet water was the most mentioned other source of drinking water, by 66 percent of households. Across all other communes, the most common response was “no other sources” (ranging from 37% in Canaan to 47% in Les Cayes), and sachet water was the second-most common response (ranging from 24% in Les Cayes to 39% in Mirebalais). Acknowledging that households may prioritize better-quality water as accessible for drinking, the survey also asked about households’ main water source for other purposes, such as cooking and handwashing. There was considerable variation across communes. Notably, however, 81 percent of households in Canaan and half the households in Jérémie use tanker truck water for non-drinking purposes. In Cap￾Haitian and Mirebalais, the use of a neighbor’s tap was also commonly reported. Access to Piped Water. Another important line of inquiry involved households with a piped connection, regardless of whether they used it as their main source of drinking water. Overall, 12.5 percent of households in WATSAN zones have an on-premises tap in their dwelling or plot; however, of these, only a small minority (11.7%) use it as their main drinking water source. Households in Mirebalais and Les Cayes were more likely to report using their tap as their main drinking water source, compared to Cap-Haitian and Jérémie. In Canaan, there were no households in the sample who reported an on-premises tap. Across communes, the most common reason cited for not using a tap as the household’s main drinking water source was that the tap does not supply good quality water. In Cap-Haitian, Jérémie, and Mirebalais, the second most common response was that the tap was inactive or disconnected, e.g., due to non-payment. In Jérémie and Les Cayes, many households also reported that their main source (usually private kiosks) already provides a regular supply. Intermittency of Piped Supply. All households with an on-premises tap were also asked about the number of days per week that water is supplied through their tap, and on those days, approximately how many hours per day water is supplied. These households reported having water an average of 2.4 days per week, and most (72.8%) reported having only about one to five hours of service on such days. Among communes, households in Les Cayes reported the highest average days of supply per week (3.2), and those in Mirebalais reported the lowest (1.9). Demand for Connections. The survey asked households without a piped connection about their plans to apply for one. The highest levels of intention to connect were in Mirebalais and Canaan, where the vast v | WATSAN FINAL REPORT (BASELINE) USAID.GOV majority (88 and 81 percent, respectively) expressed plans to apply for a connection. Results in Canaan are particularly notable, since no households had a piped connection at the time of the survey. SANITATION. The ET found that 63 percent of households across all WATSAN zones in the targeted communes have access to at least basic sanitation facilities. The highest rate of access is in Les Cayes (70%), and the lowest is in Jérémie and Mirebalais (each 58%). There is no single facility in any commune used by a majority of respondent households; however, the first and second most common latrine types in all communes are “improved,” per the report’s definition. Among common unimproved sources, pit latrines without slabs are used by 11.5 percent of households in Croix-des-Bouquets. Notably, open defecation is reported by more than 10 percent of households in Jérémie and Mirebalais. When breaking down main sanitation facility by service classification (at least basic versus less than basic), findings show that the majority of basic sources overall are made up of pit latrines with slabs, flush to pit latrines, and flush to septic tanks, all of which are not shared with other households. Overall, “less than basic” sanitation facilities are largely made up of pit latrines with slabs shared with other households (29.5%), open defecation (16.7%), and pit latrines without slabs or open pits (15.1%). Household Head Sex. Overall, there is no clear pattern across all communes regarding whether female￾or male-headed households are more likely to report a basic sanitation facility. Differences are not statistically significant, except in Cap-Haitian, where 72 percent of female-headed households reported access to at least basic sanitation, compared with 58 percent of male-headed households. Socioeconomic Status. Overall, a clear pattern emerged showing that wealthier households have substantially greater access to at least basic sanitation facilities, compared to poorer households. The widest disparity observed is in Canaan, where 90 percent of the wealthiest fifth have at least basic sanitation, compared to 37 percent of the poorest fifth. The smallest disparity is in Les Cayes, where 87 percent of the wealthiest have at least basic sanitation, compared to 52 percent of the poorest; this is the only commune where access to basic sanitation exceeds 50 percent among the poorest fifth of households. Facility Sharing, Accessibility, and Risks. Most facilities are on premises, either in the dwelling or on the plot. Households in Canaan report the highest frequency of off-site sanitation, at about 23 percent. Most households do not share sanitation facilities with other households. The highest rate of sharing is in Mirebalais, where about 29 percent of households share sanitation facilities, with the lowest in Cap-Haitian (19%). Overall, households reported that most men and women have equal access to toilet facilities—89 percent across communes. Similarly, most households, 89.3 percent, noted that everyone in their household can access and use their facility any time. Households with at least a basic sanitation facility were more likely to say that they did not face any risks when visiting their sanitation facility, compared to those with less than basic sanitation. Where those with at least basic sanitation mentioned risks, the most common were health related. These were most frequently mentioned in Jérémie and Cap-Haitian. Among those with less than basic sanitation, most also reported health risks, but risks of harassment was also commonly reported. This was especially the case in Mirebalais, Canaan, and Jérémie. Training in Open Defecation. Across communes, most households reported that they have not received training related to open defecation (79.8% total). Cap-Haitian households were most likely to report having received training related to open defecation (32.3%). Demand and Willingness to Pay for Improvements. A majority of households in all communes expressed a desire to make improvements to their sanitation facility. The highest levels of interest were USAID.GOV WATSAN BASELINE SURVEY REPORT | vi in Cap-Haitian (72%) and Jérémie (69%). Among those who said they would like to make improvements, most—at least 80 percent in all communes—expressed willingness to pay for these improvements. Link with Household Hygiene. While it was not one of the two main outcome indicators, the ESS baseline survey also measured household hygiene levels based on JMP definitions. “Basic hygiene” is defined as the presence of a handwashing facility on premises with soap and water. Basic hygiene is highest in Cap￾Haitian, where 83 percent of households have a handwashing facility with both soap and water. Looking at hygiene by main water source and main sanitation facility, data showed a stronger relationship between sanitation and hygiene compared to water and hygiene. In total, 66 percent of households with at least basic sanitation have at least basic hygiene, compared with 55 percent of households with less than basic sanitation. CONCLUSIONS This baseline survey measured key indicators IND 1 and IND 3, as related to the USAID-funded WATSAN project in urban Haiti. Overall, the ET found that the vast majority (94%) of households across all WATSAN zones in the targeted communes have access to at least basic drinking water (IND 1), while nearly two-thirds (63%) of households across all WATSAN zones in the targeted communes have access to at least basic sanitation facilities (IND 3). Given how it is defined and measured, IND 1 is already very high and, therefore, does not leave much room for WATSAN to improve it over time. However, beyond measuring these two outcome indicators, the baseline survey provided additional context showing that own-tap connectivity is relatively low, especially for the poorest households, leaving substantial room for improvements in expanding access to on-premises taps across all targeted communes. Likewise, there is room to improve service quality, which WATSAN emphasized for its planned interventions. 1 | WATSAN FINAL REPORT (BASELINE) USAID.GOV 1. INTRODUCTION In Haiti, overall access to improved water services has declined over the past two decades, and progress toward improving access to basic sanitation has stagnated.4,5 Though the number of people with improved access has risen, this has not kept pace with rapid population growth, and urban access to at least basic drinking water sources has declined.6,7 While access to basic sanitation has increased over time, only 37 percent of urban dwellers had access to basic sanitation by 2015.8 1.1 WATER AND SANITATION PROJECT In response, the United States Agency for International Development (USAID) is supporting the Water and Sanitation Project9, a four-and-half-year intervention (2018–2022), to support access to sustainable water supply and sanitation services and strengthen the enabling environment for sustainable delivery, operation, and maintenance of water and sanitation services. WATSAN focuses on increasing access to water and sanitation services in communes that are cholera hotspots or recovering from Hurricane Matthew. The five target communes include Cap-Haitian, Croix-des-Bouquets (Canaan), Mirebalais, Les Cayes, and Jérémie (Figure 1). WATSAN interventions are directed at specific zones within these five targeted communes, referred to as WATSAN zones. Figure 1. USAID/Haiti WATSAN target communes The activity is an integrated set of interventions organized into three mutually reinforcing components: increased access to sustainable water supply services (Component 1), increased access to sustainable 4 WHO/UNICEF, 2019a. 5 WHO/UNICEF, 2019b. 6 World Bank, 2018. 7 “At least basic” includes new JMP categories of basic and safely managed (see WHO/UNICEF, 2019c). 8 World Bank, 2018. 9 Previously referred to as the WATSAN Activity USAID.GOV WATSAN BASELINE SURVEY REPORT | 2 sanitation services (Component 2), and strengthened enabling environment for sustainable delivery, operation, and maintenance of water, sanitation, and hygiene (WASH) services (Component 3). A key principle of the activity is to transition to longer-term, comprehensive interventions that build more sustainable capacity at the local and national levels to reduce waterborne diseases in priority geographic zones. The WATSAN development hypothesis posits: “If access to improved water and sanitation services in Haiti is expanded through strategic technical assistance and infrastructure to improve the quality and availability of water and sanitation services, while the enabling environment for continued operation, maintenance, oversight, and use of those services is systemically strengthened, then a foundation will be established for sustained service delivery of safe and affordable water and sanitation services in Haiti, leading to reduced prevalence of cholera and other waterborne diseases and the improved health and prosperity of the Haitian people.” WATSAN’s results framework is based on the goal to provide 250,000 Haitians with access to new or improved water services over the life of the activity, along with 75,000 Haitians with access to basic or safely managed sanitation services over the same period. The activity results framework is included in the detailed project context in Annex C. Within its three components, WATSAN implements a variety of water and sanitation interventions, which differ from commune to commune, but are primarily in the form of water infrastructure engineering services, water infrastructure construction, technical assistance for water service providers, support to sanitation enterprises, wastewater treatment and fecal sludge management engineering services, wastewater treatment and fecal sludge management engineering construction, technical assistance to relevant national and sub-national government units, and knowledge dissemination and learning. 1.2 WATSAN PERFORMANCE EVALUATION In line with its institutional objectives of accountability and learning, USAID/Haiti enlisted Social Impact’s (SI) Haiti Evaluation and Survey Services (ESS) project to conduct a performance evaluation (PE) of WATSAN. The purpose of this PE is to measure whether WATSAN interventions have resulted in access to improved water and sanitation services in the targeted communes. The evaluation questions (EQs) are listed below. All three EQs will be answered in full when the performance evaluation is conducted later. This report details findings from the baseline survey which was conducted specifically to collect baseline values that will later be used to answer EQ 1, after subsequent surveys are conducted to measure changes over time. EQ 1) To what extent and in what ways has WATSAN increased access to new or improved basic drinking water and sanitation facilities? EQ 2) To what extent has WATSAN contributed to improved institutional capacity to better manage existing systems and increase revenue? EQ 3) To what extent does the WATSAN Theory of Change (TOC) remain valid and relevant to the activity’s current objectives? As per the ESS baseline survey scope of work, SI measured two indicators at baseline: IND 1: Percentage increase of households in the target communes using a basic drinking water service. WATSAN aims to increase this by 10 percent by midline and by 20 percent from baseline over the life of the activity. 3 | WATSAN FINAL REPORT (BASELINE) USAID.GOV IND 3: Percentage increase of households in the target communes using a basic sanitation facility. The project aims to increase this by 10 percent by midline and by 20 percent from baseline value over the life of the activity. The baseline survey will serve as a reference point to measure the performance of WATSAN interventions. EQs 2 and 3 will be addressed as part of the future PE. Following the conclusion of the baseline report approval, the evaluation team (ET) will de-identify the baseline dataset to submit to the Development Data Library (DDL) as per contract requirements and USAID Automated Directives System (ADS) 579 open data guidelines. See the full scope of work (SOW) in Annex E. The primary stakeholders for this evaluation include USAID/Haiti, WATSAN’s implementing partner DAI (and DAI’s partners Ayiti Nexus, Centre de Formation et d’Encadrement Technique [CFET], Zanmi Lasante, V3, and mWater), the Haiti Ministry of Publics Works, Transportation, and Communication (MTPTC), and the Haiti Water and Sanitation National Direction (DINEPA). Other stakeholders include other donors and entities active in Haiti’s WASH sector, such as the Inter-American Development Bank (IDB), the Spanish Agency for International Development Cooperation (AECID), and the World Bank. 2. DESK REVIEW FINDINGS Before launching the baseline survey, the ESS ET reviewed several documents, including the activity’s workplans, progress reports, monitoring, evaluation, and learning (MEL) plan, internal baseline report, and others.10 Findings from this desk review informed the baseline survey design and methodology and yielded important insights related to the interpretation of findings. First, USAID had identified the need to obtain baseline estimates of the key indicators for specific zones within each commune where WATSAN is being implemented. As a result of this explicitly articulated priority, the ESS baseline focuses specifically on the WATSAN zones within each targeted commune. The ESS baseline survey was, thus, designed to be representative of zones where the project is intervening (rather than representative at the commune level as a whole). In addition, interpretation of findings for both baseline and future rounds of data collection must consider the possibility that some benefits have accrued before the ESS baseline survey. For a variety of operational and logistical reasons, some WATSAN activities had already begun by the time this baseline was launched in early 2020.11 For example, in Les Cayes, WATSAN had begun providing fuel to power well generators in late 2019 and claimed to have observed a 20 percent increase in requests for new connections between then and January 2020. In Canaan, by the end of January 2020, WATSAN was expected to connect several reservoirs or water kiosks to the main distribution pipe before the baseline study. These reservoirs and kiosks had been unimproved water sources (supplied by trucks), but with a connection to the distribution pipe, they would now be considered improved. In Jérémie, the Centre Technique d’Exploitation (CTE) has begun to repair leaks with its own staff using equipment and supplies WATSAN provided. Similar work may have begun in Cap-Haitian in the first quarter of fiscal year 2020. In Mirebalais and Jérémie, starting in May 2019, the project may have been working to promote sanitation and end open defecation through community awareness, support sanitation marketing efforts, and developing community sanitation programs. A full list of documented project work to date is in Annex C. 10 See references for full list of reviewed documents. 11 By the time of this baseline, WATSAN was within Year 3 of project implementation. Years 1 and 2 of project implementation were largely dedicated to feasibility studies, assessments, and strategic planning for interventions. USAID.GOV WATSAN BASELINE SURVEY REPORT | 4 Further, even assuming that all WATSAN interventions are implemented successfully, trends in the key outcome indicators over time are unlikely to be strictly the result of any one activity or funding stream, but more likely the result of multiple donors working together to support CTEs in their objectives of expanding access to and improving the quality of key services. Some WATSAN interventions are interrelated with and interdependent on interventions implemented by other donors. As noted in a recent workplan, WATSAN is “leveraging the investments being made by the [IDB], AECID, and World Bank. The activity is counting on the World Bank to rehabilitate the Fonfred FSM site, AECID to rehabilitate the water distribution system in Jérémie and Cap-Haitian, and the [IDB] to expand the water distribution systems in Cap-Haitian and Les Cayes.”12 Lastly, WATSAN interventions largely focused on capacity-building for CTEs and the enabling environment. Associated interventions, in many cases, are expected to lead to improvements in service quality rather than increasing access to piped water. The ET’s review of activity documentation and discussions with WATSAN staff reinforced that WATSAN is mostly a technical capacity building intervention, focused on strengthening local CTEs and sludge management facilities and their ability to provide quality and regular water and sanitation services. Taking this into consideration, WATSAN does not expect many of its interventions to provide new access to basic water, but rather result in improvements in service quality (e.g., reliability, continuity of service, and water quality) among households already connected to piped water. For sanitation, the main focus is on improving treatment of fecal sludge waste. In other words, many WATSAN interventions are not likely to “move the needle” on IND 1 and IND 3, though that is the main focus of the ESS baseline survey, as per the evaluation SOW. Discussions with USAID, WATSAN, and ESS led to an agreement that ESS would include select questions in the baseline survey that measure service quality for those households already connected. These are explored in the Findings section of this report. WATSAN Communes The baseline survey obtained commune-specific results (for the WATSAN zones within each targeted commune), and the report presents commune-specific results throughout. Part of the reason for this is that the context in each commune is substantially different: not all are at the same “starting point” concerning WATSAN programming. The content below is taken directly from the WATSAN internal baseline report13 and provides an informative context for each commune, which aids in interpreting the different results observed between communes. Cap Haitian is Haiti’s second-largest city. It is a port city in the north of the country and was Haiti’s original capital. Cap Haitian has a rich aquifer that is easily reached by drilled wells, and, because the public water network has fallen into disrepair, most people get their water from wells. In the downtown area, these are drilled wells that provide reasonably safe water. In the outlying, poorer areas, the wells are shallow and dug by hand. Nearly everyone in Cap-Haitian relies on dry toilets, with a small percentage having flush toilets and a small percentage having no toilet. The only fecal sludge management is provided by a nonprofit that uses containerized toilets. Canaan (Croix de Bouquets commune), Haiti’s most rapidly growing area, began as an informal settlement after the earthquake and has grown to a population estimated at 216,000. 12 DAI, October 2019. 13 DAI, 2019a. 5 | WATSAN FINAL REPORT (BASELINE) USAID.GOV Canaan is part of the Croix de Bouquets commune and is further divided into several sections. The project is focused on the areas of Canaan III, IV, and V. The American Red Cross installed a deep well and built a primary distribution line up to a reservoir but stopped construction before building the distribution network. There are no drilled wells or other sources of safe water. The only main fecal sludge management site in Haiti is in Canaan. While it primarily serves Port-au￾Prince, it also draws clients from across the department and even the nation. Les Cayes is a mid-sized town on Haiti’s southern coast. The urban area has around 55,000 inhabitants and is organized in a neat grid of streets. Les Cayes’ public water system is the best organized in the five areas and has the most paid connections (2,500). Still, it is able to provide water only two times per week, and a number of private vendors sell water from deep wells. There is no fecal sludge management in Les Cayes, although a site was partially built many years ago. Most people use dry toilets, although the aquifer is so high in Les Cayes that this solution is not appropriate and contaminates the water that many people drink. Jérémie, on the southwestern tip of Haiti, was the city hardest hit by Hurricane Mathew in 2017. The water in the public system comes from natural springs in the Pic de la Selle and flows into town by means of gravity. Despite an abundance of water in the springs, the distribution network is in such bad shape that little reaches the public water system’s customers. Instead, people get their water from private kiosks or La Digue. There is no fecal sludge management in Jérémie. Mirebalais is in the center of Haiti, with 32,900 people in the urban area. The cholera epidemic began in Mirebalais, which was among the hardest hit areas. Mirebalais has grown tremendously in the past few years and has significant new infrastructure, including one of the largest hospitals in Haiti. The Spanish Cooperation has been working on improving the primary water transmission into Mirebalais, but the distribution remains very poor. Mirebalais has reasonable access to the fecal sludge management site in Canaan but rarely uses it. Instead, most people rely on dry toilets. 3. METHODOLOGY The baseline required an estimation of the current level of the two key outcome indicators, IND 1 and IND 3, the percentage of households with access to at least basic water and sanitation. The data collection method was a household survey, which the ET implemented electronically with a representative sample of households from the zones within each target commune where WATSAN is working. The ET selected households using a stratified random sampling (SRS) procedure. Targeted respondents within selected households were the adult household members most knowledgeable about the household’s water affairs. The household survey was administered in person with electronic data capture on handheld tablets. Interviews were conducted in January and February 2020. Baseline analysis involved tabulating descriptive statistics related to the sample characteristics, key outcome indicators (IND 1 and IND 3), and other related indicators, as measured through the household survey. 3.1 SAMPLING Sample Size Calculations. After reviewing project documentation, the ET conducted power calculations to determine a household sample size that would allow for detection of a 10 percent increase in access to basic water sources and sanitation (WATSAN targets 10 percent change by midline and 20 percent by endline; these calculations allow for detection of change at least as big as 10 percent). The ET conducted power calculations using Stata 15 statistical software. To account for and measure commune-level USAID.GOV WATSAN BASELINE SURVEY REPORT | 6 variation in both current and post-activity water source and sanitation access, the ET stratified these calculations by commune. Sample size calculations indicated a necessary sample size of 1,550 total households, with 310 from each of the five communes, to obtain estimates with a five percentage-point margin of error. Sample size calculations and details about parameters used are in Annex B. Sampling. To carry out the SRS approach, the ET used available spatial data. Specifically, the ET overlaid WATSAN target zones within each commune (shapefiles provided by the WATSAN Geospatial Information System Specialist) with building footprints obtained via the ET’s local survey firm from Haiti’s Centre National de l’Information Géo-Spatiale (CNIGS). The ET calculated the centroid (geographic center point) of each building within those target zones to extract the building’s GPS coordinates using ArcMap 10.2. After all buildings within each commune were identified and GPS coordinates calculated, data was imported into Stata 15.0 software. The ET assigned a randomly generated number to each building within the commune and then sorted by that random number, assigning a code from 1 through n, with n being the total number of building footprints within each commune’s WATSAN target zones. The ET drew a stratified random sample of 310 households from each commune’s WATSAN targeted zones. All zones within each targeted commune constituted a single stratum, thus resulting in five total strata. To achieve the desired sample size of 310 from each commune, the team attempted surveys with each household in the prescribed random order.14 For each building that was found to either be non-residential, abandoned, or where the household refused to participate, the next numbered building on the list was assigned to field interviewers. GPS points of all sampled buildings were preloaded into the electronic survey form, enabling field interviewers to navigate to each sampled building. In cases where a building constituted multiple dwellings, households, or apartments, the ET and local firm randomly selected one household from within each building using a combination of methods. For instance, if households lived on the same floor, enumerators spun a pen in a circle and first attempted to interview the household to which the pen pointed after coming to a stop. If households lived on different floors, enumerators marked small pieces of paper so that each represented a single household and then closed their eyes, shuffled the papers, and selected one at random. The enumerators interviewed the household indicated on the selected paper first. If enumerators discovered that a sampled building was not, in fact, residential, they recorded this on the survey form and move on to the next building on their assignment sheet. Intended Respondent. Within each sampled household, the targeted respondent for the baseline survey was the household head or their spouse, or the individual most familiar with the household’s water and sanitation affairs. The team ensured that enumerators were well-trained on the definition of the targeted respondent and could appropriately identify the right respondent in each household. 3.2 FIELDWORK PREPARATIONS Planning and Preparation. The ET oversaw fieldwork preparation before data collection. This included developing a baseline survey design document, developing a household questionnaire, calculating sample 14 The ET ensured the integrity of the random sampling while allowing fieldwork to be conducted feasibly. This means that the enumeration teams did not always conduct interviews chronologically (e.g., building 11 might have been attempted before building 8). Enumerators were assigned to conduct interviews in defined geographic areas, such that an interviewer could have a list of households in their area that could conceivably include buildings 1, 16, 310, and 345, while another enumerator had a list of households in their area that could conceivably include buildings 2, 48, 98, and 309. Records used for analysis were the first 310 valid submissions for each commune, in the order of randomization, starting with building 1 and proceeding until there were 310 completed interviews. In other words, if more than 310 valid records were ultimately submitted for any given commune, the first 310 valid submissions starting from building 1 were included for analysis. 7 | WATSAN FINAL REPORT (BASELINE) USAID.GOV size and sampling, contracting a local survey firm, reviewing translation and back-translation of the questionnaire, programming the electronic survey, developing quality assurance protocols, recruiting and onboarding a local monitoring specialist, reviewing and finalizing training manuals, submitting for institutional review board (IRB) review, mobilizing team travel, and supporting fieldwork oversight. Pre-Testing, Training, and Piloting. The questionnaire was pre-tested before enumerator training to identify any required adjustments in wording or survey logic. Enumerator training took place over a one￾week period in January 2020. With assistance from two field managers, the ET trained a total of 20 field interviewers and five supervisors. The enumerator training included a pilot activity, followed by a debrief to address final issues with questionnaire comprehension, translation, and programming before the official data collection start. The ET’s Assistant Team Leader traveled to Haiti with the team’s Research Assistant to oversee enumerator training and the data collection start. Ethical Considerations. The SI IRB reviewed the protocol for this baseline survey, which it deemed exempt based on the survey’s status as part of a program evaluation with no vulnerable populations or sensitive topics and involving no more than minimal risk. Respondents’ verbal consent was captured directly in the electronic questionnaire to confirm respondents’ understanding of the purpose, voluntary nature of the survey, and confidentiality measures alongside intended uses of the information provided. Timeline. The ET conducted preparations, data collection, and analysis between November 2019 and April 2020. This timeframe included approximately three weeks for finalizing the design and approval by USAID, three weeks for fieldwork preparations, one week for enumerator training, and three weeks for survey data collection (in January and February 2020). Following the completion of surveying, the ET conducted a robust set of quality checks and finalized the dataset for analysis. The ET then analyzed the baseline data and prepared this report. 3.3 DATA COLLECTION Household Questionnaire. The household questionnaire was the only data collection tool employed for this baseline survey. The household survey incorporated survey questions from the Joint Monitoring Programme (JMP) Core Survey Question Guide,15 including questions essential to constructing the main outcome variables of interest, along with several other supplementary indicators that provide additional useful context and information for USAID and WATSAN.16 The household questionnaire is in Annex A. Based on discussions with the WATSAN team and USAID in August 2019, the ET also noted an interest in understanding service improvements for those who already had connections (i.e., current CTE customers), as WATSAN focuses both on increasing access to basic water services and improving service for those who already have basic water. Thus, the ET agreed with USAID to include a limited set of questions in the baseline survey that measure current service levels and satisfaction among any household with a tap, including those who do not use it as their main drinking water source, intentions and barriers to connecting among households without a tap, among others. Electronic Data Capture. The survey was administered during in-person interviews in January and February 2020. Data was captured electronically using handheld tablets, with the questionnaire deployed 15 WHO/UNICEF, 2018b. 16 The ET’s survey included all core indicators other than W6, which involves water quality testing for E. coli. The ET did not incorporate this as part of the baseline survey. While SI has conducted such testing in the field for other evaluations, based on the logistical complexity and cost considerations involved, USAID explicitly requested to exclude E. coli testing from this survey. USAID.GOV WATSAN BASELINE SURVEY REPORT | 8 on SurveyCTO. Electronic surveying allowed for pre-programming advanced validations within the questionnaire to ensure high-quality data collection. Data Quality Monitoring. The ET conducted high-frequency checks (HFCs) on a nearly daily basis during data collection to check for any unexpected responses, patterns, or anomalies in the incoming data. HFCs, particularly at the beginning of data collection, are instrumental for identifying any potential issues and course-correcting to mitigate longer-term risks to data quality. The ET and local survey firm kept a detailed log of any questions arising from HFCs and their resolution. The ET further monitored data quality by conducting co-enumerations, accompaniments, and backchecks. These were administered by the data collection firm’s field team supervisors and SI’s local monitoring specialist. Co-enumerations involved a supervisor or the local monitoring specialist accompanying an enumerator to an interview, where both the enumerator and co-enumerator recorded respondent’s answers to main survey questions on separate forms. Accompaniments entailed a supervisor or the local monitoring specialist attending an interview alongside an enumerator and recording notes on interviewer performance and adherence to protocols. Backchecks were conducted both in-person and via telephone with individuals who had responded to the survey to verify the conduct and interview content. 3.4 DATA ANALYSIS Final Quality Checks. The ET then ran a series of data checks to look for errors, outliers, incomplete observations, duplicates, missing data, pilot data, and other erroneous components, and corrected, removed, or modified these data to a format that the team later used for analysis. Data Cleaning. The ET exported the raw dataset from the SurveyCTO server for cleaning. Data cleaning ensured that all data was labeled, formatted, and calculated for analysis. The data collection firm provided English translations of all text responses, especially for “other, specify” fields, which were recoded into existing or new responses categories or left in the “other” category as appropriate. Variable Construction. After data cleaning, the ET constructed variables required for analysis, including main outcome variables. Main outcome variables were calculated in accordance with JMP definitions, as per the stated intention of the USAID Performance Indicator Reference Sheets (PIRS) definition. The definitions used for this baseline analysis are described in further detail in Section 3.5 below. Weighting. The sample size analyzed from each commune was 310 households to create a representative sample at the commune level. However, since the total estimate (across communes) is of interest in addition to the commune-specific estimates, post-stratification weights were applied in the final analysis to ensure that the statistics related to the WATSAN zones as a whole remain representative of each of those zones. Details on the construction and application of weights are in Annex B. Data Analysis. Baseline analysis involved tabulating descriptive statistics related to the sample characteristics, key outcome indicators (IND 1 and IND 3), and other related indicators, as measured through the household survey. The main outcomes are also disaggregated by commune, sex of household head, and socioeconomic status (SES), as per the PIRS associated with IND 1 and IND 3 and in line with USAID’s Automated Directives System (ADS) 203, which requires integration of gender into evaluation design and implementation. The ET conducted all data cleaning and analysis using Stata 15.0 software and transparent and well-commented coding to ensure replicability and verification of results. 9 | WATSAN FINAL REPORT (BASELINE) USAID.GOV 3.5 INDICATOR DEFINITIONS Although the PIRS used by USAID and the Water and Sanitation Project team specify that USAID defines indicators IND 1 and IND 3 in alignment with the JMP service ladder for post-2015 monitoring, there are, in fact, differences between the USAID PIRS definition and JMP definitions, described in the two sub￾sections below. In this report, we use JMP definitions based on consultation with and concurrence by the USAID Program Office in Haiti. For the sake of transparency and to demonstrate the heavy influence of the definitional differences on the results, and upon request by USAID Technical Office and Water and Sanitation Project team, we have provided the main results for IND1 according to the PIRS definition in comparison with the JMP definition in Annex E. 3.5.1 IND 1: WATER The JMP defines basic water service as drinking water from an improved source, provided collection time is not more than 30 minutes for a round trip, including queuing. 17 The definition of basic water service provided by the most recent PIRS provided by USAID and the Water and Sanitation Project team (FY19 HL8 PIRS) is aligned in concept with this definition18 but uses different classifications of improved versus unimproved sources. The new JMP definitions categorize a greater number of sources as improved than the PIRS, meaning that all else equal, application of the new JMP definition will result in more optimistic estimates of access relative to the PIRS definition. By far the most consequential factor in determining the results of the baseline analysis is how private kiosks are classified, since the majority of respondents list private kiosks as their primary drinking water source. The JMP classifies water kiosks as improved without differentiating public versus private, while the PIRS does not list kiosks at all, implying that they are considered unimproved by USAID. In these areas of urban Haiti, there are public kiosks managed by CTEs that are similar to public standpipes, and two types of private kiosks: some that sell water from on-site wells, and others that sell treated water. Unfortunately, it is not possible to distinguish between the two types private kiosks types using the data collected in our survey – future survey waves will incorporate this into the questionnaire to ensure they can be distinguished. Since the JMP classified water kiosks as improved, we classify private kiosks as improved for the purpose of this report. As mentioned above, the main results for IND1 according to the PIRS definition/classifications are provided in Annex E for comparison. In Table 1 below, we compare the JMP post-2015 and USAID FY19 HL8 PIRS definitions, which demonstrates how the differences in classifications will affect analysis of water service data. Of course, at midline or endline, USAID prefers to use the PIRS definition, the same must be applied to this baseline data to allow for a fair and consistent comparison. Table 1. Basic water service: USAID PIRS versus JMP service ladder Red text indicates any instance in which PIRS and JMP are not perfectly aligned 17 See definitions in WHO/UNICEF JMP, 2018b and in Annex B 18 “Basic drinking water services are defined as improved sources or delivery points that by nature of their construction or through active intervention are protected from outside contamination, in particular from outside contamination with fecal matter, and where collection time is no more than 30 minutes for a roundtrip including queuing.” Note that the PIRS also includes the following language – note that at present this cannot be directly verified using our baseline measurements nor those of the Water and Sanitation project team: “The service must be able to consistently (i.e., year-round) produce 20 liters per day for each person.” Source: USAID FY19 HL8 PIRS USAID.GOV WATSAN BASELINE SURVEY REPORT | 10 Basic Water Service defined as: PIRS19 FY19 HL8 JMP16 Post-2015 service ladder (this report) (1) Source is classified as “improved” Piped into dwelling Improved Improved Piped into plot Improved Improved Piped to neighbor Not Listed->Improved a Improved Public tap/standpipe Improved Improved Borehole or tubewell Improved Improved Protected well Improved Improved Protected spring Improved Improved Rainwater collection Improved Improved Tanker truck Unimproved Improved Cart with small tank/drum Unimproved Improved Public water kiosk Not Listed->Unimproved a Improved c Private water kiosk Not Listed->Unimproved a Improved c Bottled water Improved w/conditions b Improved Sachet water Not Listed->Improved w/conditions a Improved Unprotected well Unimproved Unimproved Unprotected spring Unimproved Unimproved Surface water Unimproved Unimproved (2) AND within 30 minutes round-trip: Within 30 minutes round-trip collection time including queuing Yes Yes a These sources were not explicitly listed by the PIRS. Based on the description of improved sources in the PIRS, we classify piped to neighbor as improved, kiosks as unimproved, and treat sachet water similarly to bottled water (see footnote 19). b When another basic service is used for handwashing, cooking, or other basic personal hygiene purposes; otherwise considered unimproved. c The JMP lists only “Water Kiosk” without differentiating between public or private – a distinction which is relevant in this context in urban Haiti as described in the text. Our survey separated public and private as distinct response choices but was not sufficiently detailed as to be able to separate the two types of private kiosks described in the text. 3.5.2 IND 3: SANITATION The JMP defines basic sanitation service as the use of improved facilities not shared with other households.20 JMP describes improved facilities as “those designed to hygienically separate human excreta from human contact.” As with IND 1, the PIRS states its intention to align with the JMP definitions and is better aligned with the JMP in the case of IND 3. The PIRS is silent on some sources but contains language specifying that all other sources are considered unimproved; the PIRS specifies that a flush toilet without a sewer connection is unimproved while the JMP considers flush to pit latrine improved. Previous versions of the JMP service ladder included ventilated pit latrines (VIPs), which have now been superseded by pit and twin pit latrines and, thus, were erroneously included in the PIRS facility list. Beyond those differences, twin pit without slab is not listed in either source; we classify it as unimproved. The basic 19 “Drinking water sources meeting this criteria include: - piped drinking water supply on premises; - public tap/standpost; tube well/borehole; protected dug well; protected spring; - rainwater; and/or - bottled water (when another basic service is used for hand washing, cooking or other basic personal hygiene purposes. All other services are considered to be “unimproved”, including: unprotected dug well, unprotected spring, cart with small tank/drum, tanker truck, surface water (river, dam, lake, pond, stream, canal, irrigation channel), and bottled water (unless basic services are being used for hand washing, cooking and other basic personal hygiene purposes).” Source: USAID FY19 HL8 PIRS. 20 Round-trip collection time including any queueing. For definitions see: WHO/UNICEF JMP, 2018b. 11 | WATSAN FINAL REPORT (BASELINE) USAID.GOV sanitation service definition for this report uses the classifications shown in Table 3 below. The full JMP description of each type of facility is in Annex B. Table 2. Basic sanitation service: USAID PIRS versus JMP service ladder Red text indicates any instance in which PIRS and JMP are not perfectly aligned Basic Sanitation Service defined as PIRS FY19 HL8 JMP Post-2015 service ladder (this report) (1) Facility is classified as “improved” Flush to piped sewer system Improved Improved Flush to septic tank Improved Improved Flush to pit latrine Unimproved Improved Pit latrine with slab Improved Improved Twin pit with slab Improved Improved Other composting toilet Improved Improved Container-based sanitation Not Listed a Improved c Ventilated improved pit (VIP) latrine Improved Not Listed b Flush to open drain Unimproved Unimproved Flush, to don’t know where Unimproved Not Listed b Pit latrine without slab, open pit Unimproved Unimproved Twin pit without slab Not Listed a Not Listed b Bucket Unimproved Unimproved Hanging toilet / hanging latrine Unimproved Unimproved No facility / bush / field Unimproved No facility (Unimproved) (2) AND not shared with other households Not shared with other households Yes Yes a Not listed in PIRS. b Not listed in JMP. c JMP: “Container based sanitation refers to a system where toilets collect excreta directly in sealable, removable containers (also called cartridges), which are regularly collected by commercial service providers and delivered to treatment. Under “Bucket” in unimproved, JMP also says: “. . . or other container for the retention of feces” and refers to pans, trays, etc. d JMP lists pit and twin pit latrines in place of VIPs in past versions of the JMP service ladder. As with the water service ladder, the JMP sanitation service ladder has a category superior to basic, called “safely managed.” A “safely managed” sanitation facility is an improved sanitation facility that is not shared with other households and where excreta are safely disposed of in situ or transported and treated offsite. Therefore, just as with IND 1, this report refers to any sanitation facility meeting the definition for “basic” as “at least basic.” 3.6 DISAGGREGATIONS Commune. All indicators are disaggregated by commune to highlight specific differences among the areas where WATSAN is implementing, as described in Section 2. Household Head Sex. Key outcome indicators are also disaggregated by household head sex. Not all respondents were household heads. When the respondent was not the household head, the survey form prompted the respondent to specify the sex of the household head. During variable construction, the ET created a single variable to capture the sex of household head, whether that person was the respondent or not. Key outcomes are disaggregated by this variable. USAID.GOV WATSAN BASELINE SURVEY REPORT | 12 Socioeconomic Status. Key outcome indicators are also disaggregated by SES, which is quantified based on household characteristics, including dwelling materials (e.g., wall and roof), durable assets, and access to amenities such as electricity, among others. The ET combined these through a principal component analysis (PCA) that calculates an index partitioned into five quintiles. These five quintiles each signify a fifth of the population such that the first quintile (quintile 1) includes the poorest 20 percent of households and quintile five includes the wealthiest 20 percent of households. Importantly, SES is calculated within each commune (rather than across communes), so households are classified into relevant SES quintiles in comparison to counterparts within their commune. While calculating SES across versus within communes each has pros and cons, the decision to calculate within communes was based on a specific interest in commune-specific estimates of outcomes. This analytical choice maximizes the usefulness of comparing between SES quintiles, within communes. One drawback of the approach taken is that the characteristics of households in one SES quintile in a given commune may not align with the characteristics of households in the same SES quintile in another commune. For example, SES 5, the wealthiest, in Cap-Haitian are of higher wealth in comparison to SES 5 in Canaan, where households are considerably poorer, in general. Thus, some care must be taken in interpretation when comparing between communes. As a result of this analytical choice, “total” columns are not shown for breakdowns by SES, to emphasize the commune￾specific calculations. Water and sanitation access (IND 1 and IND 3). The ESS baseline survey also collected supplementary information about households’ water sources, use, and practices that serve to contextualize the main findings. In several cases, supplementary information is disaggregated by the main outcome variables, i.e., water and sanitation service (at least basic versus less than basic), to illustrate differences in the reported experiences of households with differing levels of access at baseline. 3.7 LIMITATIONS Apart from the definitional issues described above, there remain baseline survey limitations. First, the ET designed the survey to capture data representative of WATSAN target zones within each of the five communes. In this way, it provides useful and geographically precise information for USAID and WATSAN, with the tradeoff that these numbers cannot be directly compared to any that might exist in other sources that pertain to each commune as a whole. Also, the survey does not capture any potential variations to compare reliance on water sources between different seasons. Later rounds of data collection should be collected at the same time of year to ensure the highest level of comparability in this regard. Pre-post measurements will not measure impact or allow for strict attribution. The baseline survey is targeted in specific zones where WATSAN is present, which may help link WATSAN activities to observed trends; however, the evaluation design does not incorporate control or comparison zones that the ET can compare against WATSAN zones for more robust analysis or attribution. The ET will consider all available information sources to help explain and contextualize findings and any observed changes over time after later rounds of data collection, and the geographic precision may lend more confidence in linking any observed changes to activity interventions. However, the pre-post measurement without comparison zones or impact evaluation data does not allow us to attribute those changes solely to WATSAN activities. As described earlier, WATSAN activities had already started in some zones before this baseline. As of January 2020, just before the start of data collection, these activities included the provision of fuel for generator-powered well pumps, which WATSAN claimed had resulted in new connections or improved 13 | WATSAN FINAL REPORT (BASELINE) USAID.GOV service in some areas before baseline. The ET could not directly verify this information. However, the household questionnaire included questions about the households’ date of connection with a tap that may allow for some relevant contextualization of results. 4. FINDINGS This section begins with a summary of survey attempts and a description of the sample (section 4.1). The survey findings are then organized by major theme: water (section 4.2) and sanitation (section 4.3). Sub￾sections address findings related to the main outcomes of interest and additional supplementary data from the survey. 4.1 DESCRIPTION OF THE SAMPLE The targeted sample size for each commune was 310 households. Enumerators located 99.8 percent of listed buildings and found that 85.3 percent were residential. In total, 82.8 percent of buildings contained a single household. Most households sampled contained an eligible respondent available to respond to the survey. Among eligible respondents, 99.4 percent consented to the interview. Table 3. Results of total number of survey attempts, by commune Cap￾Haitian Canaan Jérémie Les Cayes Mirebalais Total Total attempted: 394 347 343 382 376 1842 Building located 394 (100.0%) 347 (100.0%) 343 (100.0%) 382 (100.0%) 372 (98.9%) 1838 (99.8%) Residential building 311 (78.9%) 312 (89.9%) 311 (90.7%) 318 (83.2%) 316 (84.9%) 1568 (85.3%) Of residential buildings: Single-household building 243 (78.1%) 302 (96.8%) 258 (83.0%) 239 (75.2%) 257 (81.3%) 1299 (82.8%) Of residential buildings: Multiple-household building 68 (21.9%) 10 (3.2%) 53 (17.0%) 79 (24.8%) 59 (18.7%) 269 (17.2%) Of multiple-household buildings: First selected household available 57 (86.4%) 10 (100.0%) 52 (98.1%) 76 (96.2%) 53 (89.8%) 248 (92.9%) Of selected households: Eligible respondent available 311 (100.0%) 310 (99.4%) 311 (100.0%) 314 (98.7%) 313 (99.1%) 1559 (99.4%) Of eligible respondents: Consent to participate 310 (99.7%) 310 (100.0%) 310 (99.7%) 310 (98.7%) 310 (99.0%) 1550 (99.4%) Overall response rate (Consents / total attempted) 77% 83% 90% 81% 82% 84% Table 5 presents both unweighted and weighted respondent and household characteristics of the sample population. The unweighted data represent the characteristics of the sample, while the weighted data (with post-stratification weights) are representative of the overall population in WATSAN zones, within the five targeted communes. Weighting is explained in section 3.4 above and detailed in Annex B. The report describes sample characteristics here based on the weighted data, since those represent the underlying population of interest. The average respondent age across communes was 39.0 years old. The majority of respondents (62.4%) were female; 37.6 percent were male. Among respondents, 67.8 percent were heads of their respective households. Participating households had five members on average. USAID.GOV WATSAN BASELINE SURVEY REPORT | 14 Table 4. Sample characteristics (unweighted and weighted) Unweighted Weighted (n = 1,550) (est. n=70,299) Commune Cap-Haitian 310 (20.0%) 37.6% Canaan 310 (20.0%) 26.8% Jérémie 310 (20.0%) 10.3% Les Cayes 310 (20.0%) 16.0% Mirebalais 310 (20.0%) 9.4% Head of household sex Female 794 (51.2%) 50.6% Male 756 (48.8%) 49.4% Household size (number members) Mean (SD) 5.0 (2.2) 5.0 (2.2) Median (Q1, Q3) 5.0 (3.0, 6.0) 5.0 (3.0, 6.0) Households with any children under five Yes (at least 1) 500 (32.3%) 34.3% No (none) 1,050 (67.7%) 65.7% Number of children under five (conditional on having any) Mean (SD) 1.2 (0.5) 1.2 (0.5) Median (Q1, Q3) 1.0 (1.0, 1.0) 1.0 (1.0, 1.0) Number of children under five (overall) Mean (SD) 0.4 (0.6) 0.4 (0.7) Median (Q1, Q3) 0.0 (0.0, 1.0) 0.0 (0.0, 1.0) Respondent age Mean (SD) 39.5 (13.7) 39.0 (13.3) Median (Q1, Q3) 38.0 (29.0, 48.0) 37.0 (29.0, 48.0) Respondent sex Male 594 (38.3%) 37.6% Female 956 (61.7%) 62.4% Respondent relationship to head of household Head of household 1,076 (69.4%) 67.8% Spouse of head of household 180 (11.6%) 13.4% Parent of head of household 29 (1.9%) 1.6% Child of head of household 145 (9.4%) 9.4% Sibling of head of household 43 (2.8%) 3.2% Other relative 74 (4.8%) 4.5% Other, specify 3 (0.2%) 0.2% Respondent highest level of education Did not go to school 141 (9.1%) 8.6% Some primary school 292 (18.8%) 16.9% Completed primary school 220 (14.2%) 14.2% Some secondary school 464 (29.9%) 30.8% 15 | WATSAN FINAL REPORT (BASELINE) USAID.GOV Unweighted Weighted Completed secondary school 233 (15.0%) 16.4% University/higher education 195 (12.6%) 12.7% Other, specify 2 (0.1%) 0.2% Refused 3 (0.2%) 0.3% Respondent marital status Single 461 (29.7%) 30.3% Married 514 (33.2%) 33.0% Co-habitation 485 (31.3%) 31.1% Divorced/separated 50 (3.2%) 2.9% Widow 38 (2.5%) 2.7% Other, specify 2 (0.1%) 0.1% Note: Weighted estimates with post-stratification weights. See Annex B for details. 4.2 WATER 4.2.1 BASIC WATER SERVICE Overall, the ET found that more than 90 percent of households in all WATSAN zones have at least basic drinking water service (Figure 2), using drinking water from an improved source that is within 30 minutes round-trip collection time. This is true in WATSAN zones across all target communes. Figure 2. At least basic drinking water service, by commune (IND 1) This result is almost entirely driven by the use of improved sources within 30 minutes round-trip collection time, rather than the use of on-premises taps, while households with less than basic service mostly use improved sources that are more than 30 minutes away, rather than unimproved sources (Table 6). 93% 96% 94% 94% 95% 94% 0% 20% 40% 60% 80% 100% Cap-Haitien Canaan Jeremie Les Cayes Mirebalais Total USAID.GOV WATSAN BASELINE SURVEY REPORT | 16 Table 5. Basic water service breakdown, by commune (IND 1) Cap￾Haitian Canaan Jeremie Les Cayes Mirebalais Total At Least Basic (IND 1) Own tap on premises 0% 0% 2% 10% 5% 2% Improved source, <=30 min RT 93% 96% 92% 84% 90% 92% Less than Basic Water Service Improved source, >30 min RT 7% 4% 5% 4% 5% 5% Unimproved source, <=30 min RT 0% 0% 1% 1% 0% 0% Unimproved source, >30 min RT 0% 0% 0% 1% 0% 0% Total 100% 100% 100% 100% 100% 100% Post-stratification weights applied. Analytical sample size is 301 for all communes. In terms of water sources used, in Cap-Haitian and Canaan, about 73 percent of households have basic water service, using water from a nearby private kiosk (Table 7). That same figure is 61 percent in Mirebalais, 53 percent in Jeremie, and 37 percent in Mirebalais. In Jeremie, 14 percent have basic water service from a tanker truck, and in Les Cayes, about 12 percent have basic water service through protected wells. As noted earlier, the use of a tap on premise is not common. About 10 percent of households in Les Cayes have basic service through their own on-premises tap (either in the dwelling or on their plot), along with about 4.5 percent of households in Mirebalais and about 2 percent of households in Jeremie. No households in Cap-Haitian and Canaan reported having an on-premises tap. 17 | WATSAN FINAL REPORT (BASELINE) USAID.GOV Table 6. Breakdown of water sources, by water service and commune Percentages shown in columns within each commune are out of the Commune Total Cap-Haitian Canaan Jeremie Les Cayes Mirebalais At Least Basic Less than Basic At Least Basic Less than Basic At Least Basic Less than Basic At Least Basic Less than Basic At Least Basic Less than Basic Improved Sources Piped into dwelling 1.6% 7.7% 2.6% Piped into plot 0.3% 2.3% 1.9% Piped to neighbor 0.3% 1.0% 0.6% 3.9% 0.6% 8.4% 0.3% Public tap/standpipe 1.0% 0.6% 1.0% 1.3% 2.9% 0.6% 1.6% 0.6% Borehole or tubewell 0.3% 6.5% 2.9% Protected well 0.3% 1.0% 0.3% 11.6% 0.6% Protected spring 0.3% 0.6% 0.3% 0.3% 1.0% 0.3% Rainwater collection 3.2% Tanker truck 2.3% 0.3% 14.2% 0.3% 0.3% Cart w/ small tank/drum 12.3% 6.1% 8.4% 0.3% 13.2% Public Water Kiosk 3.9% 1.3% 7.1% 4.5% 0.3% 6.8% 0.3% 0.6% Private Water Kiosk 72.6% 4.5% 73.2% 3.5% 52.9% 2.9% 36.8% 2.3% 61.0% 3.5% Bottled water 12.6% 6.8% 5.5% Sachet water 1.9% 1.0% 0.3% 0.3% 1.6% Unimproved Sources Unprotected well 0.3% 1.6% Unprotected spring 0.3% 0.3% 0.3% Surface water 0.3% Total 92.9% 7.1% 96.1% 3.9% 94.2% 5.8% 93.9% 6.1% 94.8% 5.2% Commune Total 100% 100% 100% 100% 100% Analytical sample size 288 22 298 12 292 18 291 19 294 16 Post-stratification weights applied. USAID.GOV WATSAN BASELINE SURVEY REPORT | 18 The widespread use of private kiosks, and its effect on the estimated value of this indicator, suggests several points about WATSAN. First, this estimate would imply that there is almost no room to improve the targeted 20 percent over baseline values. This is relevant to the extent that the performance of the project is tied to this indicator. However, this does not imply that there are no avenues through which the project could meaningfully improve households’ water access or use. As the majority of households rely on private kiosks, there is still considerable room for switching over time toward better-quality services, e.g., toward piped water. Actual experience in terms of water access could further improve if the project is successful in achieving its primary goals of improving service delivery and service quality, such that existing customers experience more reliable and better-quality water, and as additional households potentially switch from private kiosks to piped sources of water. From this perspective, therefore, to understand the potential impact of the project on water access among households in the selected areas, we posit that it will be necessary to look beyond IND 1, even while it is necessary to measure it as a required indicator for this evaluation and as part of the activity’s MEL Plan. EQ 1 asks not only to what extent access has increased but also in what ways; further, it asks about “new” access to basic drinking water (understood as those who change from less than basic to at least basic water service) and “improved” basic drinking water (understood as those even within the “at least basic” classification who experience improvements in their water supply). In this way, we believe that EQ 1 is broad enough to encompass the critical nuance needed to sufficiently understand not only whether the project is “moving the needle” on IND 1, but also what is happening “under the hood” of that indicator. In this report, given the scope of work, we focus on IND 1, but in future phases (e.g., at midline or endline), an important part of contextualizing any analysis of IND 1 will necessarily involve a close look at what households are experiencing in terms of specific water services used, reliability, etc. While IND1 may, therefore, not perfectly reflect the activity’s focus on technical assistance and service delivery, it remains part of the activity’s MEL Plan, and therefore must be measured and reported on; it is beyond the scope of this evaluation to suggest changes in the activity’s indicators. Water service, by head of household sex. There were no significant differences between basic water service on the basis of head of household sex. Overall, 95 percent of female-headed households had basic water service, compared with 94 percent of male-headed households. Across communes, there were no statistically significant differences between female- and male-headed households. Water service, by SES. Given that most households have access to basic water service, primarily through the use of private kiosks within 30 minutes round-trip collection time, we did not observe any statistically significant differences in access to basic water service across socioeconomic status (Figure 3).21 21 As described in section 3.6, socioeconomic status quintiles shown here are calculated within each commune. 19 | WATSAN FINAL REPORT (BASELINE) USAID.GOV Figure 3. Access to basic water service, by SES and commune However, given that access to basic drinking water service is driven primarily by the use of private kiosks, it is also informative to observe access to a piped source on premise by SES (Figure 4). In this case, there are substantial differences in access to a tap on premise based on SES, in all target communes. The only exception is in Canaan, where there are no households reporting their own tap on premise across the board. The most substantial disparity is in Mirebalais, where only 8 percent of households in the poorest SES quintile have access to a tap on premise, compared with 65 percent of households in the wealthiest SES quintile. Figure 4. Access to piped source on premise, by SES and commune 4.2.2 COLLECTION TIME Average round-trip collection times (including queuing) for households with at least basic water service ranges from 9 minutes in Canaan to 13 minutes in Jeremie and Les Cayes. Average round-trip times for households with less than basic water service ranges from 41 minutes in Mirebalais to 72 minutes in Les Cayes.22 Disaggregating travel and queue times by commune and water service (Figure 6), in most cases travel time exceeds that of queueing, except among households with at least basic service in Les Cayes and Jeremie. 22 Note the relatively small sample size of households with less than basic water service as shown in Table 7. 86 100 94 94 92 92 90 95 94 95 97 97 94 97 97 95 94 98 95 94 95 98 90 90 98 0% 20% 40% 60% 80% 100% Cap-Haitian Canaan Jeremie Les Cayes Mirebalais SES Quintile: 1 2 3 4 5 2% 0% 2% 8% 8% 7% 0% 3% 16% 19% 5% 0% 5% 33% 42% 8% 0% 8% 40% 40% 26% 0% 37% 47% 65% 0% 20% 40% 60% 80% 100% Cap-Haitian Canaan Jeremie Les Cayes Mirebalais SES Quintile: 1 2 3 4 5 USAID.GOV WATSAN BASELINE SURVEY REPORT | 20 Figure 5. Average round-trip collection time (minutes), by commune and water service Figure 6. Average round-trip collection time (minutes), travel versus queuing, by commune 4.2.3 AVAILABILITY & QUANTITY Overall, households with at least basic water service were more likely to report that water from their main drinking water source was always available, compared with those with less than basic access. This is especially the case in Jeremie, Mirebalais, and Canaan. Households were also asked about whether there was any time in the last month that they lacked a sufficient quantity of drinking water from their main drinking water source. Overall, households with less than basic access were more likely to report this, especially in Mirebalais (63%) and Canaan (50%). The only exception is in Cap-Haitian, where households with at least basic service were more likely to report insufficiency (39%) versus those with less than basic service (27%). Households in Les Cayes were overall the least likely to report insufficiency in the last month, regardless of water service. The most common reasons for insufficiency were that water was not available from the source, or that the source was not accessible ( Figure 9). 10 9 13 13 11 11 71 46 58 72 41 63 0 20 40 60 80 100 Cap-Haitien Canaan Jeremie Les Cayes Mirebalais Total ROUND-TRIP COLLECTION TIME (MINUTES) At Least Basic Less than Basic 6 5 6 6 7 4 3 8 7 4 0 20 40 60 80 Cap-Haitian Canaan Jeremie Les Cayes Mirebalais At least Basic Travel Queueing 51 27 33 46 31 20 19 25 26 10 0 20 40 60 80 Cap-Haitian Canaan Jeremie Les Cayes Mirebalais Less than Basic Travel Queueing 21 | WATSAN FINAL REPORT (BASELINE) USAID.GOV Figure 7. Households (%) reporting water “always” available from main source, by commune Figure 8. Households (%) reporting insufficiency last month, by commune Figure 9. Main reason for insufficiency last month, by water service and commune 4.2.4 PERCEIVED PROBLEMS WITH SUPPLY Reported problems with households’ main water supply differed between those with basic and less than basic water service (Table 8). Those with at least basic service were more likely to report “no problems” overall, except in Cap-Haitian. Those with less than basic service were more likely in all areas to report 62% 68% 84% 91% 65% 59% 58% 72% 90% 44% 0% 20% 40% 60% 80% 100% Cap-Haitian Canaan Jeremie Les Cayes Mirebalais At least Basic Less than Basic 39% 34% 14% 9% 31% 27% 50% 22% 11% 63% 0% 20% 40% 60% 80% 100% Cap-Haitian Canaan Jeremie Les Cayes Mirebalais At least Basic Less than Basic 67% 67% 50% 50% 90% 33% 33% 50% 50% 0% 20% 40% 60% 80% 100% Cap-Haitian Canaan Jeremie Les Cayes Mirebalais Less than basic Water not available Source Inaccessible Water too expensive Other, specify 63% 78% 51% 54% 87% 30% 17% 27% 31% 8% 0% 20% 40% 60% 80% 100% Cap-Haitian Canaan Jeremie Les Cayes Mirebalais At least basic Water not available Source Inaccessible Water too expensive Other, specify USAID.GOV WATSAN BASELINE SURVEY REPORT | 22 insufficient water from their main drinking water source. Poor quality of water was reported by at least 10 percent of households in all areas other than Canaan and up to 25 percent of households with basic service in Cap-Haitian and less than basic service in Mirebalais. Intermittent supply appears to be a particular problem in Canaan, reported by a quarter of households with less than basic service, and in Mirebalais, with 11 and 19 percent of households with basic and less than basic service reporting the same. A third and nearly half (47%) of households, respectively, in Jeremie and Les Cayes reported problems related to time spent collecting water. Nearly a third of households with at least basic service in Cap￾Haitian reported issues with bad taste, color, or odor, along with 13 percent of households in Canaan, plus about a fifth of those with less than basic service in Mirebalais. In Canaan, 10 percent of households with basic service reported problems with sediment and particles in the water. As shown earlier, those with at least basic service are mostly made up of households using private kiosks within 30 minutes round￾trip collection time, and those with less than basic service, private kiosks beyond 30 minutes. Table 7. Problems faced with supply, by service level and commune Cap-Haitian Canaan Jeremie Les Cayes Mirebalais Basic+ 8 gallons) (% yes) 41.6% 48.2% 13.5% 20.0% 27.1% Any time in last week not able to store sufficient water to meet needs (% yes at least once) 50.3% 48.4% 24.2% 20.6% 61.0% Household stores water in small containers (% yes) 97.4% 83.5% 81.3% 81.9% 84.2% How long water stored in small containers until consumed (days) Less than 1 day 88.0% 60.4% 74.8% 59.8% 70.7% 1 day 1.3% 16.3% 1.6% 2.4% 12.9% More than 1 day 10.6% 23.2% 23.6% 37.8% 16.4% Total 100.0% 100.0% 100.0% 100.0% 100.0% 25 | WATSAN FINAL REPORT (BASELINE) USAID.GOV The survey also involved small container observation. The vast majority (96.9%) of observed small water storage containers were covered. On average, 76.3 percent of households reported that the small containers are only used to store drinking water. Enumerators asked households to point out the container that they use most often for drinking water and asked additional follow up questions for that specific container to assess storage hygiene, indicating the likelihood of secondary contamination. Across communes, only about 12 percent mix water collected from different places in their container. An even smaller percentage, 4.9 percent, added that they mix treated and untreated water in their container. Jérémie households are noticeably the most likely to mix untreated and treated water (12.7%). Almost two-thirds (64.5%) of all households keep their container at ground level. During the observation, enumerators recorded that the majority (76.6%) of small containers had small openings of less than ten centimeters, and, on average, 61.5 percent use a utensil to scoop or draw water from the container. About one quarter of households had containers with unclean, cracked, or leaking exteriors, but most (96.8%) had containers with clean interiors and a lid or cover (91.5%). Notably, Cap￾Haitian households were most likely to use a utensil or scoop, least likely to have a container with an unclean, cracked, or leaking exterior, and most likely to have a container with a clean interior. Few households use containers with a spigot (6.7%). Table 10. Small container observation, by commune Cap￾Haitian Canaan Jeremie Les Cayes Mirebalai s Can you show me these small containers? No 0.0% 5.0% 0.4% 2.4% 1.9% COVERED containers 99.7% 91.9% 99.6% 97.2% 96.2% UNCOVERED containers 0.3% 3.1% 0.0% 0.4% 1.9% Are these containers used ONLY for DRINKING water? (% yes) 70.9% 76.8% 77.3% 88.7% 74.6% Do you ever mix water that you collected from different places in this container (% yes) 17.5% 5.7% 17.9% 10.1% 9.0% Do you ever mix treated and untreated water in this container? (% yes) 5.0% 1.6% 12.7% 5.2% 4.3% Is the container kept at ground level? (% yes) 64.6% 69.9% 61.8% 53.2% 70.7% Container has wide (>=10cm) or narrow ( < 10cm) mouth/opening? Wide (>=10 cm) 12.9% 27.6% 39.0% 36.7% 13.7% Narrow ( < 10cm) 87.1% 72.4% 61.0% 63.3% 86.3% Is a utensil used to scoop or draw water from the container? (% yes) 92.1% 41.5% 55.8% 45.2% 30.9% Is the outside of the container unclean, cracked, or leaking? (% yes) 3.3% 48.0% 33.9% 35.5% 33.6% Is the inside of the container clean? (% yes) 98.7% 95.1% 98.0% 97.6% 91.8% Container has a spigot? (% yes) 3.6% 11.0% 10.8% 6.5% 3.1% Container has a lid or fitted cover? (% yes) 97.4% 87.0% 86.1% 86.3% 96.1% 4.2.7 MULTIPLE CONCURRENT SOURCE USE 4.2.7.1 Other drinking water sources USAID.GOV WATSAN BASELINE SURVEY REPORT | 26 The survey asked households whether they regularly use any other drinking water sources aside from their main source. Respondents could select all that apply from a list of water sources. The results are presented below in Table 12. In Cap-Haitian, sachet water was the most commonly mentioned other source of drinking water, by 66 percent of households. Across all other communes, the most common response to this question was “no other sources” (ranging from 37% in Canaan to 47% in Les Cayes), with sachet water constituting the second-most common response (ranging from 24% in Les Cayes to 39% in Mirebalais). A few other sources were mentioned, especially in some communes. In Mirebalais, other common responses included bottled water (17%), piped to neighbor (8.1%), and private water kiosks (6.5%). In Cap-Haitian, 14.2 percent mentioned bottled water. In Canaan, 10 percent mentioned bottled water, and another 10 percent tanker trucks. In Jérémie, 7 percent mentioned tanker trucks, and another 7 percent carts with small tanks or drums. In Les Cayes, 8 percent cited protected wells, and 7 percent mentioned boreholes. Table 11. Other sources of water used for drinking, by commune Commune Cap￾Haitian Canaan Jérémie Les Cayes Mirebalais Total Piped into dwelling 0.0% 0.0% 0.3% 1.3% 1.6% 0.6% Piped into plot 0.3% 0.0% 0.0% 2.3% 0.6% 0.6% Piped to neighbor 2.9% 0.0% 1.0% 1.6% 8.1% 2.7% Public tap/standpipe 1.6% 0.6% 0.6% 1.3% 1.0% 1.0% Borehole or tubewell 0.6% 1.0% 0.0% 7.1% 3.2% 2.4% Protected well 0.0% 0.3% 0.0% 7.7% 0.6% 1.7% Unprotected well 0.0% 0.0% 0.0% 0.6% 0.0% 0.1% Protected spring 0.3% 0.0% 0.3% 0.6% 0.3% 0.3% Unprotected spring 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Rainwater collection 0.6% 3.2% 1.3% 0.0% 0.0% 1.0% Tanker truck 0.3% 10.0% 7.4% 0.0% 0.3% 3.6% Cart with small tank or drum 0.3% 0.0% 6.5% 0.6% 3.9% 2.3% Public water kiosk 0.0% 0.3% 4.8% 1.3% 0.0% 1.3% Private water kiosk 0.6% 1.0% 2.3% 2.9% 6.5% 2.6% Bottled water 14.2% 10.3% 2.6% 2.6% 17.1% 9.4% Sachet water 65.5% 36.5% 35.2% 23.5% 38.7% 39.9% Surface water 0.0% 0.0% 0.6% 0.3% 0.0% 0.2% No other sources 28.4% 37.1% 45.2% 47.1% 44.5% 40.5% Other, specify 0.3% 0.0% 0.6% 0.3% 0.3% 0.3% Don’t know 0.0% 6.5% 0.0% 1.0% 0.0% 1.5% Refused 0.3% 3.2% 0.0% 0.0% 0.0% 0.7% Analytical sample size 310 310 310 310 310 1,550 The survey asked respondents to select all that apply, so column percentages may exceed 100. Post-stratification weights applied. 4.2.7.2 Water used for other domestic purposes Anticipating that households may prioritize better-quality water for drinking, the survey also asked households about their main source of water for other purposes such as cooking and handwashing. The survey asked for their main and other sources of water for domestic purposes. In Cap-Haitian, the most 27 | WATSAN FINAL REPORT (BASELINE) USAID.GOV common sources of water used for other domestic purposes include protected wells (32%), piped to neighbor (20%), tanker truck (14%), and boreholes (13%). In Canaan, the large majority (81%) of households mentioned tanker trucks as their main source for other domestic purposes. Half of households in Jérémie also mentioned tanker trucks, with another 17 percent listing private water kiosks. In Les Cayes, about 30 percent of households use a protected well as their main water source for purposes other than drinking, with another 20 percent using their own tap, and 19 percent using boreholes. In Mirebalais, the most common source of water for domestic purposes other than drinking was neighbors’ taps (27%), followed by boreholes and their own tap (18% each). The survey did not ask about the round-trip collection time for water sources used for domestic purposes other than drinking; thus, these sources cannot be characterized by the same definitions as the main drinking water source. Table 12. Main source of water for other domestic purposes, by commune Main source for other domestic purposes Cap￾Haitian Canaan Jérémie Les Cayes Mirebalais Total Piped into dwelling 3.5% 0.0% 3.9% 20.3% 18.1% 6.7% Piped into plot 2.3% 0.0% 1.9% 5.2% 9.4% 2.7% Piped to neighbor 20.0% 0.3% 2.9% 6.5% 27.4% 11.5% Public tap/standpipe 6.1% 3.9% 2.3% 7.7% 5.2% 5.3% Borehole or tubewell 12.9% 7.1% 1.0% 19.4% 18.4% 11.7% Protected well 31.9% 1.0% 4.8% 29.7% 9.7% 18.4% Unprotected well 0.3% 0.3% 0.6% 2.6% 0.6% 0.7% Protected spring 3.5% 0.0% 0.6% 0.3% 0.6% 1.5% Unprotected spring 0.0% 0.0% 0.6% 0.3% 0.0% 0.1% Rainwater collection 0.6% 0.6% 6.8% 0.3% 0.0% 1.2% Tanker truck 13.5% 81.6% 50.3% 0.6% 5.5% 32.8% Cart with small tank or drum 4.8% 0.0% 0.0% 0.0% 0.0% 1.8% Public water kiosk 0.0% 0.3% 6.1% 5.8% 1.6% 1.8% Private water kiosk 0.0% 1.0% 17.4% 0.6% 0.0% 2.2% Surface water 0.3% 0.0% 0.6% 0.6% 3.2% 0.6% Other, specify 0.0% 3.9% 0.0% 0.0% 0.3% 1.1% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Analytical sample size 310 310 310 310 310 1,550 Post-stratification weights applied. Across communes, most households said that apart from their main source of water for other domestic purposes, they did not use others. The only exception was in Cap-Haitian, where only 47 percent of households answered this way; there, 26 percent of households mentioned rainwater collection, and 15 percent of households mentioned protected wells. Rainwater collection was also mentioned by 16 percent of households in Canaan. Boreholes were commonly mentioned in Canaan (6%), Les Cayes (7%), and Mirebalais (7%). Protected wells are commonly mentioned in Les Cayes (9%) and Mirebalais (6%). Households in Les Cayes also mentioned neighbors’ taps (10%). In Jérémie, another 10 percent of respondents mentioned tanker trucks, along with 5 percent mentioning carts with a small tank or drum. USAID.GOV WATSAN BASELINE SURVEY REPORT | 28 Table 13. Other sources of water used for domestic purposes, by commune Cap￾Haitian Canaan Jérémie Les Cayes Mirebalais Total Piped into dwelling 0.3% 0.0% 0.3% 1.3% 0.6% 0.5% Piped into plot 0.3% 0.0% 1.3% 9.7% 0.3% 2.3% Piped to neighbor 4.2% 0.0% 1.0% 1.0% 1.6% 1.5% Public tap/standpipe 0.6% 2.6% 2.6% 3.9% 3.9% 2.7% Borehole or tubewell 3.9% 6.1% 0.0% 6.5% 7.1% 4.7% Protected well 15.2% 3.9% 0.3% 9.4% 6.1% 7.0% Unprotected well 2.9% 0.0% 0.6% 2.6% 1.0% 1.4% Protected spring 1.3% 0.0% 0.3% 0.0% 1.9% 0.7% Unprotected spring 0.0% 0.0% 0.6% 1.0% 0.3% 0.4% Rainwater collection 25.5% 15.5% 4.5% 0.6% 1.0% 9.4% Tanker truck 3.9% 3.5% 10.3% 0.0% 1.6% 3.9% Cart with small tank or drum 6.5% 0.0% 5.2% 0.0% 0.3% 2.4% Public water kiosk 0.0% 1.6% 4.8% 0.6% 0.0% 1.4% Private water kiosk 0.0% 0.0% 1.6% 1.6% 0.0% 0.6% Bottled water 0.0% 0.0% 0.3% 0.0% 0.0% 0.1% Sachet water 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% Surface water 0.0% 0.6% 1.9% 1.6% 2.6% 1.4% No other sources 47.4% 57.7% 71.9% 60.6% 77.7% 63.1% Other, specify 0.0% 0.3% 0.0% 0.0% 0.0% 0.1% Don’t know 0.0% 7.1% 0.0% 0.0% 0.0% 1.4% Refused 0.0% 4.2% 0.0% 0.3% 0.0% 0.9% Analytical sample size 310 310 310 310 310 1,550 The survey asked respondents to select all that apply, so column percentages may exceed 100. Post-stratification weights applied. 4.2.8 HOUSEHOLDS WITH PIPED CONNECTIONS The information presented below pertains to any household with an on-premises tap. This information was collected to understand access to piped connections, even if those connections were not used as the main source of drinking water. This is especially relevant in the context of WATSAN, given its focus on building CTE capacity to improve service delivery. Overall, 12.5 percent of households in WATSAN zones have an on-premises tap, either in the dwelling or on the plot (see Table 15 below). Among all households that do have a tap, only a small minority (11.7%) actually use it as their main source of drinking water. Households in Les Cayes are the most likely to do so, but only about 35 percent with a tap use it as their main source. In Cap-Haitian, none of the 9 percent of connected households do so. The most common reason cited for not using a tap as the household’s main drinking water source was that the tap does not supply good quality water, followed in most cases by an inactive connection (either due to non-payment or another issue). All households with an on-premises tap were also asked about the number of days per week that water is supplied through their tap, and on those days, approximately how many hours per day water is supplied. Overall, among all households with a tap, households reported having water an average of 2.4 days per week, and most (72.8%) reported having only about one to five hours of service on such days. Reports of average days of service per week ranged from 1.9 days in Mirebalais to 3.2 in Les Cayes. 29 | WATSAN FINAL REPORT (BASELINE) USAID.GOV Table 14. Access to, use of, and supply from tap on premises, by commune Cap￾Haitian Canaan Jérémie Les Cayes Mirebalais Total Has tap on premises 9.4% 0.0% 11.0% 28.7% 34.8% 12.5% Analytical sample size 310 310 310 310 310 1,550 Of those with tap on premises: Uses it as main drinking water 0.0% - 17.6% 34.8% 13.0% 11.7% Analytical sample size 34 89 108 260 491 Why not main DW - Inactive/disconnected 31.0% - 21.4% 6.8% 17.1% 22.7% Insufficient amount of water 0.0% - 0.0% 6.9% 7.4% 2.5% Does not supply good quality 62.1% - 57.1% 51.7% 71.3% 60.3% Cannot afford to do so 0.0% - 3.6% 1.7% 2.1% 1.1% Other, specify * 0.0% - 17.9% 27.6% 2.1% 8.8% Do not know 6.9% - 0.0% 5.2% 0.0% 4.7% Total 100.0% - 100.0% 100.0% 100.0% 100.0% Analytical sample size 29 28 58 94 209 Avg. days/week of service 2.4 - 2.2 3.2 1.9 2.4 Analytical sample size 310 310 310 310 310 Avg. hours/day of service - Less than 1 hour per day 18.8% - 9.7% 8.8% 6.3% 13.7% 1-5 hours per day 81.3% - 67.7% 51.3% 81.0% 72.8% 7-17 hours per day 0.0% - 16.1% 35.0% 8.9% 11.0% 18-23 hours per day 0.0% - 0.0% 3.8% 2.5% 1.1% 24 hours per day 0.0% - 6.5% 1.3% 1.3% 1.3% Total 100.0% - 100.0% 100.0% 100.0% 100.0% Analytical sample size 16 31 80 79 206 Post-stratification weights applied. * In Jérémie and Les Cayes, many households explained in their “other” responses that their main source (usually private kiosks) already provides a regular supply. 4.2.9 UNCONNECTED HOUSEHOLDS Households without a piped connection were asked about their plans to apply (Figure 12). The highest levels of intention to connect were reported in Mirebalais and Canaan, where 88 and 81 percent of unconnected households, respectively, expressed plans to apply. Results in Canaan are particularly notable given the lack of any households with connections. In the rest of the communes, around a third of unconnected households expressed plans to connect. USAID.GOV WATSAN BASELINE SURVEY REPORT | 30 Figure 12. Plans to apply for a connection, by commune Analytical sample sizes: Cap-Haitian 281; Canaan 310; Jeremie 276; Les Cayes221; and Mirebalais 202. Households who planned to connect were asked about their main reason for doing so (Figure 13). Responses varied meaningfully between communes. The most commonly cited reason in most areas was to reduce time collecting water. Gaining a more reliable supply was also mentioned by a third of households who planned to connect in Canaan and Mirebalais. In Jeremie, another 35 percent of households planned to connect to obtain a better quality/safer water supply, along with 26 percent of households in Canaan, and around a fifth of households in Les Cayes and Mirebalais. Figure 13. Main reason household planning to apply for a connection, by commune 4.2.10 WATER EXPENDITURES Figure 14a presents water expenditures in absolute terms in terms of gourdes per household per month, as well as in relative terms as a percentage of the household’s total monthly budget. The highest monthly water expenditures are reported in Canaan, followed by Cap-Haitian, each with more than 1,000 gourdes (about 9 USD) per month per household. Households in Canaan also report spending more on water as a share of their total household monthly budget (13%) compared to other areas, which are all under 10 percent (Figure 14b). 38% 81% 39% 32% 88% 0% 20% 40% 60% 80% 100% Cap-Haitian Canaan Jeremie Les Cayes Mirebalais 16% 15% 28% 55% 23% 35% 34% 36% 13% 14% 9% 18% 10% 10% 33% 6% 32% 5% 26% 35% 18% 22% 0% 25% 50% 75% 100% Cap-Haitian Canaan Jeremie Les Cayes Mirebalais Better quality/safer water More reliable supply Now able to afford Reduce time collecting water Network has been extended here 31 | WATSAN FINAL REPORT (BASELINE) USAID.GOV The wealthiest quintile of households in all areas spend more in absolute terms than the poorest households, except in Cap-Haitian, where spending is similar. However, poor households’ water expenditures comprise a share of total monthly expenditures that is at least as large as those of wealthy households. Specifically, in Canaan, Les Cayes, and Cap-Haitian, poor households spend a greater share of their total monthly budget on water compared to the wealthiest households. Figure 14. Water expenditures (monthly total and budget share), by commune and SES 4.3 SANITATION 4.3.1 MAIN SANITATION FACILITY As discussed above, WATSAN outcome indicator three (IND 3) is “percentage increase of households in the target communes using a basic sanitation facility.” WATSAN aims to increase this by 20 percent from baseline over the life of the activity. This section outlines the findings for the baseline value of this outcome indicator. Within each commune, the ET presents this outcome indicator overall and disaggregated by household head sex and SES. In subsequent sections, supplemental data collected in the survey is provided to contextualize and detail additional information about sanitation practices at the household level. 1067.75 1446.938 832.152 368.886 804.273 1,263 1,031 643 327 544 1,268 1,871 1,495 485 1,056 0 200 400 600 800 1000 1200 1400 1600 1800 2000 Cap-Haitien Canaan Jeremie Les Cayes Mirebalais Gourde (a) Water expenditures (total, monthly) Total Poorest quintile (SES 1) Wealthiest quintile (SES 5) 7% 13% 7% 5% 6.4% 7% 15% 7% 6% 6.1% 6% 13% 8% 5% 7% 0% 5% 10% 15% 20% Cap-Haitien Croix-des-Bouquets Jeremie Les Cayes Mirebalais (b) Water expenditures (share of total monthly budget) USAID.GOV WATSAN BASELINE SURVEY REPORT | 32 A household’s main sanitation facility is defined as basic if it is classified as an “improved” facility that is not shared with other households. Since the new JMP sanitation service ladder includes a category that is superior to basic (“safely managed”), the ET uses the phrase “at least basic” to categorize any households meeting the definition for basic access described above. Because the definition of “basic” incorporates the element of sharing, it is possible for the same type of facility to be categorized as either “at least basic” or “less than basic” for any given household, depending on whether it is shared. This differs from the previous JMP classification, which was based on type alone. Overall, the ET found that 63 percent of households across all WATSAN zones in the targeted communes have access to at least basic sanitation facilities (Figure 15 below). The highest access is reported in Les Cayes (70%), with the lowest in Jérémie and Mirebalais (each 58%). Figure 15. Main sanitation facility, percent (%) at least basic, by commune (IND 3) In terms of the specific facilities reported, a pit latrine with a slab was the most common reported overall (37.6%) and across all communes (see below). However, there is no single facility in any commune used by a majority of respondents. Flush to pit latrine is the second-most common facility used in Cap-Haitian, Canaan, and Mirebalais, while flush to septic tank is the second-most common facility used in Jérémie and Les Cayes. All those sources are considered improved and defined as basic, as long as they are not shared by other households. Among common unimproved sources, pit latrines without slabs are reported by 11.5 percent of households in Canaan. Open defecation is reported by more than 10 percent of households in Jérémie and Mirebalais. In terms of sanitation facilities used (Table 16), most households in Cap-Haitian, Canaan, and Mirebalais have at least basic sanitation service using flush to pit latrines or pit latrines with slabs, and thirdly flush to septic tank. In Jeremie and Les Cayes, flush to septic tank is more commonly mentioned than flush to pit latrine. Open defecation is the most common facility used by those with less than basic sanitation service in Cap-Haitian, Jeremie, and Mirebalais. Pit latrines without slabs (open pits) are the most common less￾than-basic facility in Canaan, while flush to “do not know where” is the most common less-than-basic facility in Les Cayes. 65% 60% 58% 70% 58% 63% 0% 20% 40% 60% 80% 100% Cap-Haitien Canaan Jeremie Les Cayes Mirebalais Total 33 | WATSAN FINAL REPORT (BASELINE) USAID.GOV Table 15. Breakdown of sanitation facilities, by sanitation service and commune Percentages shown in columns within each commune are out of the Commune Total Cap-Haitian Canaan Jeremie Les Cayes Mirebalais At Least Basic Less than Basic At Least Basic Less than Basic At Least Basic Less than Basic At Least Basic Less than Basic At Least Basic Less than Basic Improved Facilities Flush to piped sewer system 0.3% 0.3% 2.3% 0.6% Flush to septic tank 10.6% 0.3% 7.4% 0.3% 15.8% 0.3% 29.7% 2.3% 9.7% 0.6% Flush to pit latrine 23.5% 3.2% 19.4% 5.8% 13.9% 1.6% 24.2% 1.9% Pit latrine with slab 28.1% 12.6% 29.0% 6.1% 32.6% 12.3% 22.6% 11.6% 16.1% 14.2% Twin pit with slab 1.3% 1.0% 3.2% 1.6% 1.9% 0.6% 1.3% 1.3% 7.1% 4.5% Other composting toilet 0.6% 2.3% 1.6% 1.3% 0.6% 0.3% 0.6% Container based sanitation 1.3% 0.6% Unimproved Sources Flush to open drain 0.3% 0.3% 1.0% Flush, to do not know where 3.9% 0.3% 3.5% 5.2% 1.9% Pit latrine without slab, open pit 1.6% 11.9% 7.4% 2.9% 5.5% Twin pit without slab 7.1% 0.3% 0.3% 1.3% Bucket 2.3% 1.0% 0.3% Hanging toilet / hanging latrine 0.3% 0.3% 1.9% 0.3% Do not know 0.3% Other, specify* 3.9% 3.5% 2.6% 0.3% No facility / Bush / Field 4.8% 6.5% 10.6% 3.2% 10.3% Total 65.2% 34.8% 60.0% 40.0% 57.7% 42.3% 69.7% 30.3% 57.4% 42.6% Commune Total 100.0% 100.0% 100.0% 100.0% 100.0% Analytical sample size 202 108 186 124 179 131 216 94 178 132 *Mostly neighbor's toilet and public facility. USAID.GOV WATSAN BASELINE SURVEY REPORT | 34 Basic sanitation, by head of household sex. Overall, there is no clear pattern across all communes regarding whether female- or male-headed households were more likely to report a basic sanitation facility. Differences were not statistically significant, except in Cap-Haitian, where 72 percent of female-headed households reported access to at least basic sanitation, compared with 58 percent of male-headed households in that commune (p=0.009). Basic sanitation, by SES. Access to basic sanitation was also broken down by SES quintile. Overall, a clear pattern emerged, showing that wealthier households have significantly greater access to at least basic sanitation facilities compared to poorer households (p<0.001 for each commune). The widest disparity observed is in Canaan, where 90 percent of the wealthiest fifth have at least basic sanitation, compared to 37 percent of the poorest fifth. The smallest disparity was observed in Les Cayes, where 87 percent of the wealthiest have at least basic sanitation compared to 52 percent of the poorest; Les Cayes is the only commune where access to basic sanitation exceeds 50 percent among the poorest fifth of households. Figure 16. Access to basic sanitation facility, by socioeconomic status 4.3.2 FACILITY LOCATION AND SHARING Most facilities are located on premises, either in the dwelling or on the plot (. Table 17). Households in Canaan report the highest frequency of off-site sanitation, at about 23 percent. Most households do not share sanitation facilities with other households. The highest rate of sharing is in Mirebalais, where about 29 percent of households share sanitation facilities, with the lowest in Cap-Haitian (19%). 40% 37% 33% 52% 37% 68% 40% 53% 61% 47% 63% 65% 56% 71% 57% 74% 68% 66% 80% 69% 82% 90% 82% 87% 81% 0% 25% 50% 75% 100% Cap-Haitien Canaan Jeremie Les Cayes Mirebalais SES Quintile: 1 2 3 4 5 35 | WATSAN FINAL REPORT (BASELINE) USAID.GOV Table 16. Facility location and sharing, by commune Cap￾Haitian Canaan Jérémie Les Cayes Mirebalais Total Where is this toilet facility located? In own dwelling 30.0% 26.8% 26.1% 47.1% 33.5% 31.8% In own yard/plot 58.7% 50.3% 57.7% 46.8% 52.6% 53.9% Elsewhere 10.6% 22.6% 15.8% 5.8% 13.5% 13.9% Refused 0.6% 0.3% 0.3% 0.3% 0.3% 0.4% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Analytical sample size 310 310 310 310 310 1,550 Share with non-HH members No 76.7% 82.1% 77.0% 77.7% 71.3% 77.8% Yes 19.3% 17.2% 23.0% 22.3% 28.7% 20.4% Don’t know 1.0% 0.7% 0.0% 0.0% 0.0% 0.6% Refused 3.0% 0.0% 0.0% 0.0% 0.0% 1.1% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Analytical sample size 296 291 278 301 279 1,445 Post-stratification weights applied. 4.3.3 RISKS WHEN VISITING FACILITY Households with at least a basic sanitation facility were more likely to say that they did not face any risks when visiting their sanitation facility, compared to those with less than basic sanitation (Table 18). Where those with at least basic sanitation mentioned risks, the most common were risks to health. These were most frequently mentioned in Jérémie and Cap-Haitian. Among those with less than basic sanitation, most also reported health risks, but it was also common to report risks of harassment. This was especially the case in Mirebalais, Canaan, and Jérémie. Table 17. Risks faced when visiting sanitation facility, by commune Cap￾Haitian Canaan Jérémie Les Cayes Mirebalais Total Overall No risks faced 42% 58% 31% 57% 49% 48% Yes, risk to health 51% 38% 67% 42% 48% 49% Yes, risk of harassment 12% 17% 15% 3% 22% 14% Yes, other, specify 3% 1% 2% 1% 1% 2% Do not know 2% 1% 0% 0% 0% 1% Refused 2% 0% 0% 0% 0% 0% Analytical sample size 310 310 310 310 310 1,550 At least basic No risks faced 53% 75% 42% 63% 66% 60% Yes, risk to health 45% 23% 58% 37% 34% 39% Yes, risk of harassment 13% 9% 4% 1% 10% 7% Yes, other, specify 0% 2% 0% 0% 0% 0% USAID.GOV WATSAN BASELINE SURVEY REPORT | 36 Cap￾Haitian Canaan Jérémie Les Cayes Mirebalais Total Do not know 1% 0% 0% 0% 0% 0% Refused 1% 0% 0% 0% 0% 0% Analytical sample size 202 186 179 216 178 961 Less than basic No risks faced 22% 33% 16% 43% 25% 27% Yes, risk to health 63% 61% 80% 54% 70% 66% Yes, risk of harassment 10% 31% 30% 7% 39% 25% Yes, other, specify 8% 1% 5% 4% 2% 4% Do not know 4% 2% 1% 0% 1% 2% Refused 4% 0% 0% 0% 0% 1% Analytical sample size 107 124 130 93 129 583 The survey asked respondents to select all that apply, so column percentages may exceed 100. Post-stratification weights applied. 4.3.4 TRAINING IN OPEN DEFECATION Across communes, the majority of households (about 80%) reported that they had not received training related to open defecation (Table 19). Cap-Haitian households were most likely to report having received training related to open defecation. There, about one third said they had. Table 18. Any member of household trained in relation to open defecation, by commune Cap￾Haitian Canaan Jérémie Les Cayes Mirebalais Total No 66.8% 92.3% 89.0% 81.0% 83.9% 79.8% Yes 32.3% 6.5% 10.6% 18.1% 12.6% 19.0% Don’t know 1.0% 1.3% 0.0% 1.0% 2.9% 1.1% Refused 0.0% 0.0% 0.3% 0.0% 0.6% 0.1% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Analytical sample size 310 310 310 310 310 1,550 Post-stratification weights applied. 4.3.5 FACILITY LEAKS The majority of households reported that their sanitation facility never leaks at any time of the year (81.2% across communes). Yet, Table 20 displays that among communes, Jérémie respondents were most likely to state that their facility experiences leaks; almost one-quarter of Jérémie respondents note that their facility “sometimes” leaks. Jérémie households are least likely (along with Mirebalais) to have at least basic sanitation. This may contribute to the relatively higher leak occurrence. 37 | WATSAN FINAL REPORT (BASELINE) USAID.GOV Table 19. Sanitation facility leaks or overflows seasonally, by commune Cap-Haitian Canaan Jérémie Les Cayes Mirebalais Total No, never 86.1% 84.5% 63.9% 76.1% 80.0% 81.2% Yes, sometimes 5.2% 7.4% 23.5% 15.2% 11.6% 9.9% Yes, frequently 1.0% 2.3% 6.5% 1.3% 2.9% 2.1% Don’t know 2.3% 5.2% 1.9% 6.5% 4.5% 3.9% Refused 5.5% 0.6% 4.2% 1.0% 1.0% 2.9% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Analytical sample size 310 310 310 310 310 1,550 Post-stratification weights applied. 4.3.6 ACCESS AND BARRIERS TO ACCESSIBILITY Overall, households reported that most men and women have equal access to toilet facilities—89 percent across communes. Similarly, the majority of households, 89.3 percent, noted that everyone in their household can access and use their facility at all times of day and night. In households where members are not always able to use the toilet facility, the causes of this inability varied across communes. As illustrated in Table 21, the highest percentages of households in Cap-Haitian and Jérémie stated that this was because they do not have a facility and defecate outdoors. In Canaan and Mirebalais, this was largely because their toilet is not always safe to use. In Les Cayes, households stated that the toilet is not always available to all members. Table 20. Access and barriers to accessibility, by commune Cap￾Haitian Canaan Jérémie Les Cayes Mirebalais Total Do men and women have equal access to the toilet facility? 94.2% 86.5% 81.3% 92.6% 78.1% 89.0% Analytical sample size 310 310 310 310 310 1,550 Everyone in household able to access/use facility at all times of day and night? 91.3% 86.8% 85.2% 94.8% 83.9% 89.3% Analytical sample size 310 310 310 310 310 1,550 What is the MAIN reason that household members are unable to use the toilet at all times of day or night? Limited mobility prevents members from using the toilet 4.8% 0.0% 2.2% 0.0% 6.0% 2.6% Distance/barriers prevent members from reaching the toilet 28.6% 13.2% 4.4% 0.0% 20.0% 16.6% Toilet is not always available to all household members 9.5% 0.0% 22.2% 42.9% 6.0% 13.3% Toilet is not always safe for all household members to use 4.8% 55.3% 13.3% 14.3% 50.0% 24.9% USAID.GOV WATSAN BASELINE SURVEY REPORT | 38 Cap￾Haitian Canaan Jérémie Les Cayes Mirebalais Total No latrine or defecates outdoors 52.4% 18.4% 55.6% 28.6% 14.0% 36.2% Other, specify 0.0% 10.5% 2.2% 7.1% 4.0% 4.6% Don’t know 0.0% 2.6% 0.0% 7.1% 0.0% 1.8% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Analytical sample size 21 38 45 14 50 168 Post-stratification weights applied. 4.3.7 DISPOSAL OF WASTEWATER AND SOLID WASTE Most households dispose of wastewater directly to open ground or a water body (Table 22). In Jérémie and Les Cayes, though this was still the most common response, households were also more likely to dispose of wastewater into a sink/drain connected to an open drain or open ground, or a sink/drain connected to a sewer. Table 21. Cooking, laundry, and bathing water disposal, by commune Cap￾Haitian Canaan Jérémie Les Cayes Mirebalais Total N/A (cooking/laundry/bathing is done away from the household) 0.0% 2.3% 0.0% 0.0% 0.0% 0.6% Other, specify 0.3% 0.3% 0.6% 0.0% 0.0% 0.3% Sink/drain connected to sewer 1.6% 0.0% 16.8% 18.7% 1.0% 5.4% Sink/drain connected to septic tank 1.0% 1.6% 3.2% 7.7% 1.3% 2.5% Sink/drain connected to pit 1.9% 0.6% 1.3% 3.9% 1.9% 1.8% Sink/drain connected to soak pit 8.4% 1.3% 9.4% 6.5% 2.6% 5.7% Sink/drain connected to open drain or open ground 24.5% 38.4% 27.1% 17.1% 58.7% 30.5% Disposed directly to open ground or water body 62.3% 55.2% 41.3% 45.2% 34.5% 52.9% Don’t know 0.0% 0.3% 0.3% 0.6% 0.0% 0.2% Refused 0.0% 0.0% 0.0% 0.3% 0.0% 0.1% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0 % Analytical sample size 310 310 310 310 310 1,550 Post-stratification weights applied. Garbage disposal varied somewhat by commune, but the largest percentage of households dispose garbage in a designated waste disposal area (29.7% overall). Households in Jérémie were the most likely to dispose of garbage in designated waste disposal areas (53% of households). A majority of Canaan households bury or burn garbage ( Table 23). 39 | WATSAN FINAL REPORT (BASELINE) USAID.GOV Table 22. Garbage disposal, by commune Cap-Haitian Canaan Jérémie Les Cayes Mirebalais Total Collected by public service provider 5.2% 0.0% 4.5% 12.3% 3.9% 4.7% Collected by private formal service provider 2.6% 0.0% 2.3% 3.5% 2.3% 2.0% Collected by private informal service provider 27.7% 0.3% 5.2% 11.6% 1.3% 13.0% Disposed of in designated waste disposal area 16.8% 28.7% 52.9% 41.9% 38.4% 29.7% Disposed of within household yard or plot 0.0% 1.0% 1.0% 3.5% 0.6% 1.0% Buried or burned 13.5% 58.1% 16.8% 11.9% 30.6% 27.2% Disposed of elsewhere 34.2% 11.9% 17.4% 15.2% 22.6% 22.4% Don’t know 0.0% 0.0% 0.0% 0.0% 0.3% 0.0% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0 % Analytical sample size 310 310 310 310 310 1,550 Post-stratification weights applied. 4.3.8 DESIRED IMPROVEMENTS A majority of households in all communes expressed a desire to make improvements to their sanitation facility. The highest level of interest was in Cap-Haitian (72%) and Jérémie (69%). Among those who said they would like to make improvements, the vast majority—at least 80 percent in all communes— expressed willingness to pay for these improvements. Table 23. Desire and willingness to pay for sanitation facility improvements, by commune Cap￾Haitian Canaan Jérémie Les Cayes Mirebalais Total Would like to make improvements 71.9% 63.9% 69.0% 57.7% 59.4% 66.0% Analytical sample size 310 310 310 310 310 1,550 Of those: Willing to pay for improvements 85.7% 89.9% 79.9% 78.2% 85.3% 85.0% Analytical sample size 223 198 214 179 184 998 Post-stratification weights applied. Among those who would like to make improvements, the most common response among those with less than basic water supply was to install a pour flush toilet (Table 25; the question asked respondents to select all that apply). Responses varied to a greater degree among those with at least basic sanitation; in general, however, among those households, installing pour flush toilets and improving ventilation were the most common responses, along with improving the water supply. An exception to this general pattern is in Cap-Haitian, where 91 percent mentioned improving the water supply and 68 percent mentioned installing a lockable door. Installing a lockable door was also mentioned by a sizable portion of households in Mirebalais and Canaan. USAID.GOV WATSAN BASELINE SURVEY REPORT | 40 Table 24. Desired improvements to sanitation facility, by commune Among those desiring improvements to sanitation facility Cap￾Haitian Canaan Jérémie Les Cayes Mirebalais Total Overall Install permanent superstructure 45% 48% 38% 30% 48% 42% Improve ventilation 66% 41% 51% 58% 59% 55% Install pour flush toilet 72% 74% 61% 68% 79% 71% Improve water supply 86% 49% 44% 35% 66% 57% Replace slab 45% 39% 19% 13% 43% 32% Install lockable door 70% 41% 21% 16% 54% 41% Other, specify 6% 2% 17% 7% 4% 7% Do not know 0% 0% 1% 0% 0% 0% Refused 0% 1% 0% 0% 0% 0% Analytical sample size 310 310 310 310 310 1,550 At least basic Install permanent superstructure 42% 54% 43% 28% 52% 44% Improve ventilation 64% 47% 57% 61% 63% 59% Install pour flush toilet 70% 71% 60% 64% 76% 68% Improve water supply 91% 55% 48% 31% 72% 61% Replace slab 44% 37% 21% 11% 47% 32% Install lockable door 68% 44% 28% 19% 50% 43% Other, specify 1% 1% 3% 5% 2% 3% Do not know 1% 0% 1% 0% 0% 0% Refused 0% 1% 0% 0% 0% 0% Analytical sample size 202 186 179 216 178 961 Less than basic Install permanent superstructure 51% 38% 31% 35% 44% 40% Improve ventilation 71% 31% 44% 52% 53% 50% Install pour flush toilet 76% 78% 63% 75% 85% 75% Improve water supply 76% 38% 40% 42% 59% 51% Replace slab 46% 42% 15% 17% 37% 32% Install lockable door 74% 36% 11% 10% 59% 39% Other, specify 14% 4% 35% 10% 6% 15% Do not know 0% 0% 0% 0% 0% 0% Refused 0% 0% 0% 0% 0% 0% Analytical sample size 107 124 130 93 129 583 The survey asked respondents to select all that apply, so column percentages may exceed 100. Post-stratification weights applied. 4.3.9 SEPTIC TANK EMPTYING In most cases, households with a septic tank reported that their tank empties to a leach field or soak pit (Table 26). In Cap-Haitian and Canaan, at least a third of households did not know. In Jérémie and Les Cayes, a plurality of households also reported discharging to an open drain. The vast majority (at least 85%) of households in Canaan, Jérémie, and Mirebalais with a septic tank reported that it had never been emptied. In Cap-Haitian, 21 percent said that it had been emptied, 21 percent did not know, and 56 41 | WATSAN FINAL REPORT (BASELINE) USAID.GOV percent said never. In Les Cayes, 20 percent reported that it had been emptied, 13 percent did not know, and 67 percent said never. Among those who reported that it had been emptied, most did not know the location to which it was emptied. The exception is Mirebalais, where half of the households with septic tanks that had been emptied reported that it was last buried in a covered pit. A majority of households said that they were willing to pay for a service provider to empty their pit latrine or septic tank. Table 25. Septic tank emptying, by commune Septic tank, discharges to Cap￾Haitian Canaan Jérémie Les Cayes Mirebalais Total Leach field, soak pit 50.0% 41.7% 47.1% 46.4% 66.7% 48.4% Sewer 5.9% 0.0% 7.8% 13.4% 21.2% 7.1% Open drain 2.9% 16.7% 37.3% 22.7% 3.0% 13.3% Open ground or watercourse 0.0% 4.2% 7.8% 6.2% 6.1% 3.5% Other, specify 2.9% 4.2% 0.0% 0.0% 0.0% 2.2% Don’t know 38.2% 33.3% 0.0% 11.3% 3.0% 25.4% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Analytical sample size 34 24 51 97 33 239 Septic tank, ever been emptied? Never emptied 56.1% 89.4% 91.9% 67.3% 84.9% 73.2% Yes emptied 21.3% 2.2% 6.5% 19.7% 7.5% 13.1% Don’t know 22.5% 8.0% 1.6% 13.0% 7.2% 13.5% Refused 0.0% 0.4% 0.0% 0.0% 0.4% 0.1% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Analytical sample size 253 274 246 269 265 1,307 Of those “yes, emptied”: Last time emptied, where to? Buried in a covered pit 3.8% 0.0% 0.0% 11.3% 65.0% 9.3% To don’t know where 96.3% 100.0% 93.8% 88.7% 35.0% 90.1% To uncovered pit, open ground, water body, or elsewhere 0.0% 0.0% 6.3% 0.0% 0.0% 0.6% Analytical sample size 54 6 16 53 20 149 Willing to pay for a service provide to empty pit latrine or septic tank All respondents 66.0% 63.1% 57.7% 58.4% 56.2% 62.3% Analytical sample size 253 274 246 269 265 1,307 Post-stratification weights applied. 4.4 HYGIENE While not one of the two main outcome indicators, the ESS baseline survey also measured household hygiene levels, based on the JMP definitions.23 “Basic hygiene” is defined as the presence of a 23 WHO/UNICEF, 2018a. USAID.GOV WATSAN BASELINE SURVEY REPORT | 42 handwashing facility on premises with soap and water. “Limited hygiene” is defined as the availability of a handwashing facility on premises without soap and/or water. “No facility” is defined as no handwashing facility on premises. Based on best practice in the JMP methods guide, the survey measured this indicator based on physical observation by enumerators of handwashing facilities and the presence of both water and soap at the handwashing facility. Overall, 62 percent of households practice or have access to basic hygiene (Figure 17). Basic hygiene is most common in Cap-Haitian, where 83 percent of households have a handwashing facility with both soap and water, i.e., basic hygiene. In all other communes aside from Canaan, more than half of households surveyed had basic hygiene. The lowest level of basic hygiene was in Canaan, where only about one-third (34%) of households surveyed had basic hygiene, and almost half (48%) had no handwashing facility at all. The percentage of households with no facility in most other communes was also relatively high, including 36 percent in Jérémie, 26 percent in Les Cayes, and 21 percent in Mirebalais. Only 8 percent of households in Cap-Haitian had no handwashing facility. Where they are present, most handwashing facilities observed in all areas include mobile objects such as buckets, jugs, or kettles (Figure 18). Fixed facilities such as a sink/tap were most common in Jeremie and Les Cayes, observed at about one fifth of households. There was a stronger observed relationship between sanitation service and hygiene, compared to water service and hygiene. Basic hygiene was reported by 66 percent of households with at least basic sanitation and by 55 percent of households with less than basic sanitation (Table 27). Figure 17. Level of access to hygiene facility, by commune 83% 34% 57% 64% 61% 62% 8% 17% 7% 9% 18% 12% 8% 48% 36% 26% 21% 26% 0% Cap-Haitien Canaan Jeremie Les Cayes Mirebalais Total No facility Limited Basic 43 | WATSAN FINAL REPORT (BASELINE) USAID.GOV Figure 18. Presence of handwashing facility at the household, by commune Table 26. Hygiene facility by main water source and main sanitation facility Main Drinking Water Source Main Sanitation Facility Hygiene At Least Basic Less than Basic At Least Basic Less than Basic Basic 62% 64% 66% 55% Limited 12% 8% 12% 10% No facility 26% 29% 21% 35% Total 100% 100% 100% 100% Analytical sample size 1,430 84 940 568 Post-stratification weights applied. Households that reported no facility (n=422) or whose facility could not be directly observed (n=36)24 were asked to self-report about the place where they usually wash their hands. Their responses are summarized in Table 28 below. Overall, most of these households confirmed that without a location in the household, there is no place to wash hands. In Cap-Haitian, 27 percent of these households said that they use a mobile object (such as a bucket, jug, or kettle). Depending on the commune, between about 12 to 20 percent of households said they do not have soap or water for handwashing. 24 Thirty-six cases could not be observed: 34 declined permission to observe; one was occupied, and another was originally recorded as “other” and recoded as “fixed facility in dwelling.” 8% 45% 36% 26% 20% 25% 6% 6% 18% 21% 7% 10% 84% 41% 46% 52% 67% 62% 7% 4% 2% 0% 25% 50% 75% 100% Cap-Haitien Canaan Jeremie Les Cayes Mirebalais Total No handwashing place Fixed facility (sink/tap): In dwelling Fixed facility (sink/tap): In yard / plot Mobile object: bucket, jug, kettle No permission or could not observe USAID.GOV WATSAN BASELINE SURVEY REPORT | 44 Table 27. Reported handwashing among households without facilities or unobserved Cap￾Haitian Canaan Jérémie Les Cayes Mirebalais Total Usual handwashing place Sink/tap in dwelling 0.0% 0.6% 0.9% 1.2% 10.8% 1.5% Sink/tap in yard/plot 0.0% 0.0% 0.0% 1.2% 1.4% 0.3% Mobile object 26.9% 10.6% 0.0% 2.4% 8.1% 14.1% No handwashing place 73.1% 88.2% 99.1% 95.2% 79.7% 84.0% Do not know 0.0% 0.6% 0.0% 0.0% 0.0% 0.2% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Analytical sample size 26 161 113 84 74 458 Has soap/water for handwashing No 11.5% 11.8% 19.5% 28.6% 12.2% 15.2% Yes, shown 26.9% 22.4% 30.1% 25.0% 18.9% 25.0% Reported but not shown 61.5% 65.8% 50.4% 46.4% 68.9% 59.8% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Analytical sample size 26 161 113 84 74 458 Post-stratification weights applied. 45 | WATSAN FINAL REPORT (BASELINE) USAID.GOV 5. CONCLUSIONS This baseline survey measured key indicators IND 1 and IND 3, as related to the USAID-funded WATSAN activity in urban Haiti. Aside from reporting on main outcome indicators, the survey collected information beyond JMP indicators that further contextualize findings for key stakeholders. Water. Overall, the ET found that the vast majority (94%) of households across all WATSAN zones in the targeted communes have access to at least basic drinking water service, with no significant differences between female- and male-headed households or by socioeconomic status. The high measured level of this indicator at baseline is almost entirely driven by the widespread use of private kiosks (about 67% of households overall), almost all of which are reported to be located within 30 minutes round-trip collection time, including queuing. There is, therefore, little room to improve upon this indicator as currently defined and measured. However, this does not imply that there are no avenues through which the project could meaningfully improve households’ water access or use. As WATSAN has stated, its interventions primarily target increasing service quality and, ultimately, connecting more customers to the network in the targeted communes. Other findings from the baseline survey suggest ample room for potential project impact. For example, there is still a clear disparity regarding access to a tap on premises. Across all target communes, wealthier households are substantially more likely to have a tap on premises, even if they do not use it as their main source of drinking water. Since the majority of households rely on private kiosks, there is still considerable room for switching over time toward better-quality services, such as piped water, assuming that there are ways to address potential affordability concerns or connection fee barriers among poorer households, which was beyond the scope of this baseline survey. In addition, there is considerable demand for new connections among unconnected households in Canaan and Mirebalais. Further, only a small minority of households with a tap rely on it as their primary source of water. One likely reason is poor service quality, that is, high levels of intermittency. Those with taps report having service for less than three days per week and less than six hours on those days. Actual experience in terms of water access could further improve if the project is successful in achieving its primary goals of improving service delivery and service quality, such that existing customers experience more reliable and better-quality water and, separately, as additional households potentially switch from private kiosks to piped sources of water. From this perspective, therefore, to understand the potential impact of the project on water access among households in the selected areas, we posit that it will be necessary to look beyond IND 1, even while it is necessary to measure it as a required indicator for this evaluation and as part of the activity’s MEL Plan. EQ 1 asks not only to what extent access has increased but also in what ways; further, it asks about “new” access to basic drinking water (understood as those who change from less than basic to at least basic water service) and “improved” basic drinking water (understood as those even within the “at least basic” classification who experience improvements in their water supply). In this way, we believe that EQ 1 is broad enough to encompass the critical nuance needed to understand not only whether or not the project is “moving the needle” on IND 1, but also what is happening “under the hood” of that indicator. In this report, given the scope of work, we focused analysis on IND 1, but in future phases (e.g., at midline or endline), an important part of contextualizing any analysis of IND 1 will necessarily involve a close look at what households are experiencing in terms of specific water services used, reliability, etc. USAID.GOV WATSAN BASELINE SURVEY REPORT | 46 Sanitation. Overall, the ET found that 63 percent of households across all WATSAN zones in the targeted communes have access to at least basic sanitation facilities. The highest access is reported in Les Cayes (70%) and the lowest in Jérémie and Mirebalais (each at 58%). Most households do not share sanitation facilities with other households. The highest rate of sharing is in Mirebalais (29%), with the lowest in Cap-Haitian (19%). Wealthier households are substantially more likely to have access to basic sanitation service. There may be at least three avenues for project impact as it relates to baseline findings presented in this report. First, there appears to be substantial demand for making improvements to sanitation facilities across all communes, as well as significant willingness to pay for those improvements. Second, among households with a septic tank, most say that it has never been emptied and do not know where it discharges to but are willing to pay for a service provider to empty their pit latrine or septic tank. The third relates to the activity’s behavior-change components, given that in Jeremie and Mirebalais, open defecation was reported as the main “facility” by more than 10 percent of households. Use of Baseline Findings. Results from this survey serve as a baseline for future time points and provide other information directly relevant to context monitoring for USAID/Haiti and WATSAN programming. Given the high estimated baseline values for IND 1 and IND 3, it is important to bear in mind that this survey covers areas within urban Haiti that may be the most likely to have access to improved facilities. While estimated baseline levels of the indicators IND 1 and IND 3 are high, based on the definitions and classifications used, there are potential pathways for project impact, especially concerning improved water service quality, expanded access to on-premise taps, and physical improvements to sanitation facilities. While IND 1 and IND 3 are required WATSAN indicators, and the primary focus is on EQ 1, it will be beneficial for USAID/Haiti in future rounds of data collection and analysis to understand the potential nuance behind the aggregate indicators, especially related to basic water service. 47 | WATSAN FINAL REPORT (BASELINE) USAID.GOV 6. REFERENCES Cayemittes, Michel, et al. 2013. Enquête Mortalité, Morbidité et Utilisation des Services, Haïti, 2012. Calverton, Maryland, USA : MSPP, IHE et ICF International. DAI Global, LLC. March 2018. “USAID Haiti Water and Sanitation (WATSAN) Project Activity Monitoring & Evaluation Plan.” USAID. DAI Global, LLC. April 2018. “USAID Haiti Water and Sanitation (WATSAN) Activity Quarterly Progress Report: First Quarter FY18 – December 4 to December 31, 2017, Second Quarter FY18 – January 1 to March 31, 2018.” DAI Global, LLC. July 2018. “USAID Haiti Water and Sanitation (WATSAN) Activity: Initial Assessment Report for Cap Haitian, July 31, 2018.” DAI Global, LLC. July 2018. “USAID Haiti Water and Sanitation (WATSAN) Activity: Liste des interventions envisages par le projet Eau et Assainissement de 7-20-18.” DAI Global, LLC. July 2018. “USAID Haiti Water and Sanitation (WATSAN) Activity Quarterly Performance Report, FY2018 Q3, April 2018 – June 2018.” DAI Global, LLC. October 2018. “USAID Haiti Water and Sanitation (WATSAN) Activity: Year 2 Work Plan, October 1, 2018 – September 30, 2019.” DAI Global, LLC. January 2019. “USAID Haiti Water and Sanitation (WATSAN) Activity Quarterly Performance Report, FY2019 Q1, October 2018 – December 2018.” DAI Global, LLC. October 2019. “USAID Haiti Water and Sanitation (WATSAN) Activity: Year 3 Work Plan, October 1, 2019 - September 30, 2020.” DAI Global, LLC. 2019a. “USAID Water and Sanitation Project Commune Specific Report.” DAI Global, LLC. 2019b. “USAID Water and Sanitation (WATSAN) Activity. Baseline Household Water and Sanitation Survey Sample Design and Questionnaire.” DAI Global LLC. 2019c. “USAID Water and Sanitation (WATSAN) Activity. Baseline Household Water and Sanitation Survey.” Targeted sub-commune level GIS data (.kml file). Institut Haitian de l’Enfance et ICF. 2018. Enquête Mortalité, Morbidité et Utilisation des Services (EMMUS-VI 2016-2017) Pétion-Ville, Haïti, et Rockville, Maryland, USA : IHE et ICF. Interamerican Development Bank. December 2018. “Haiti water supply, sanitation, and hygiene project in urban, periurban, and rural areas of Haiti’s Greater Northern Region.” Stukel, Diana Maria. 2018. “Feed the Future Population-Based Survey Sampling Guide.” Washington, DC: Food and Nutrition Technical Assistance Project, FHI 360. USAID. n.d. “USAID/Haiti Water and Sanitation (WATSAN). Section C - Statement of Work.” WHO/UNICEF. 2018a. “JMP Methodology. 2017 Update & SDG Baselines.” WHO/UNICEF Joint Monitoring Programme for Water Supply, Sanitation and Hygiene. 2018b. “Core questions on drinking water, sanitation and hygiene for household surveys: 2018 update.” New USAID.GOV WATSAN BASELINE SURVEY REPORT | 48 York: United Nations Children’s Fund (UNICEF) and World Health Organization. https://washdata.org/report/jmp-2018-core-questions-household-surveys-0. WHO/UNICEF Joint Monitoring Programme for Water Supply, Sanitation and Hygiene. 2019a. “Estimates on the use of water, sanitation, and hygiene in Haiti. JMP Haiti Country File.” https://washdata.org/data/country/HTI/download. WHO/UNICEF Joint Monitoring Programme for Water Supply, Sanitation and Hygiene. 2019b. “Drinking water, sanitation and hygiene service levels: Haiti – Urban.” https://washdata.org/data/household#!/dashboard/2042. WHO/UNICEF Joint Monitoring Programme for Water Supply, Sanitation and Hygiene. 2019c. “JMP basic water source definitions for post-2015 monitoring.” https://washdata.org/monitoring/drinking-water. WHO/UNICEF Joint Monitoring Programme for Water Supply, Sanitation and Hygiene. 2019d. “JMP basic sanitation definitions for post-2015 monitoring.” https://washdata.org/monitoring/sanitation. World Bank. 2018. “Looking Beyond Government-Led Delivery of Water Supply and Sanitation Services: The Market Choices and Practices of Haiti’s Most Vulnerable People.” WASH Poverty Diagnostic. World Bank, Washington, DC. http://documents.worldbank.org/curated/en/764651513150057693/pdf/122047-REVISED.pdf. World Bank DINEPA. 2017 “Renforcement de la capacité de la DINEPA dans la mise en œuvre de la stratégie de réforme du secteur de l’eau et de l’assainissement Évaluation de la durabilité financière du secteur de l’eau et de l’assainissement en Haïti.” World Bank Data. 2018. “Rural population (% of total population).” Data source: UN Population Prospects: United Nations Population Division’s World Urbanization Prospects: 2018 Revision. https://data.worldbank.org/indicator/SP.RUR.TOTL.ZS. 49 | WATSAN FINAL REPORT (BASELINE) USAID.GOV 7. ANNEXES 7.1 ANNEX A: HOUSEHOLD QUESTIONNAIRE NB: indicators that are directly in line with JMP Core Questions corresponding to its water, sanitation, and hygiene ladders are annotated with blue and bold formatting (e.g. JMP W2). Instructions to enumerators and skip patterns are indicated with red text. Other important notes and definitions are provided in gray and red. MODULE A: INTRODUCTION & CONSENT # QUESTION RESPONSES a1_0 Supervisor ID Will be a pre-populated drop-down list in the electronic form # # a1 Enumerator ID NB: Will be a pre-populated drop-down list in the electronic form # # a2 Commune 1 Cap-Haitian 2 Croix-des-Bouquets 3 Mirebalais 4 Les Cayes 5 Jeremie a3 WATSAN Target Zone (Sector) NB: Will be a pre-populated drop-down list in the electronic form # # a4 Enter Building ID Number # # # # # a5 Please re-enter Building ID Number # # # # # a6_date Date of Survey d d m m y y location Note: Below is the link to the household's location. Click here to open Google Maps navigation NB: Will be a link to Google Maps showing the location of the building in the electronic form sample1 Building/Dwelling was located? 1 Yes 0 No  ef1 (Module ENDFORM) sample2 If sample1= 1 Building/Dwelling is residential? 1 Yes, residential  sample3 2 No, commercial  ef1 (Module ENDFORM) 3 No, government ef1 (Module ENDFORM) -96 No, other _o Only if sample2= -96 Other, specify  ef1 (Module ENDFORM) open text: sample2 If sample2= 1 1 One household  sample4 2 Multiple households USAID.GOV WATSAN BASELINE SURVEY REPORT | 50 # QUESTION RESPONSES Building/Dwelling contains one household or multiple households? _1 If sample3= 2 Note: Randomly select a household in the building to interview ___________________________________ _2 If sample3= 2 Was the first randomly selected household there? 1 Yes 0 No sample3 If sample2= 1 Eligible respondent available at selected household? Eligible respondent must be 18+ years old; knowledgeable about water affairs in household. 1 Yes 0 No  ef1 (Module ENDFORM) _ If sample4= 1 Start Time Auto-recorded in electronic form h h m m consentscript If sample4= 1 Consent Script: My name is [INTERVIEWER NAME]. I am working with a survey firm called DDG. We are working with a firm called Social Impact, an independent evaluation company based in the US, conducting a survey about water and sanitation in some urban areas of Haiti as part of an evaluation for the US Agency for International Development (USAID). As part of this evaluation, we are surveying about 310 households in this urban area. Your household has been randomly selected to participate. As part of this survey, we would like to ask you to answer some questions related to your use of water sources, sanitation, hygiene practices, water-related illness, and some basic information about your household. We anticipate the survey will take between 30 to 40 minutes of your time. Your participation is voluntary and you may choose not to participate. You may also choose not to answer any specific question in the survey for any reason and you may request to end the survey at any time. There is no direct benefit for participating, other than knowing that you are providing information that can help organizations better plan for water and sanitation programs in your area. There is no risk to participating, other than the time you could spend doing something else. There is no consequence if you choose not to participate. You may also refuse to answer any specific questions within the survey, or to stop participating at any time with no consequence. Your personal identifying information will be kept confidential by the research team to the fullest extent possible permitted by local law, US law, and USAID policy. Results of the survey will be aggregated for analysis and published in a report at the end of the study. Your personal information will not be disclosed in any publication and will be stored securely by the research team. After publication of the report, we will remove all identifying information from the data so that no individual can be identified, and the de￾identified data will be submitted to USAID for publication. Your views are important and we hope that you will take the time to participate. If you have any questions about the study or your rights as a participant, you may contact M. Isnel PIERREVAL, 2817-1445 or Social Impact at irb@socialimpact.com or +1-703-465-1884. If you have any 51 | WATSAN FINAL REPORT (BASELINE) USAID.GOV # QUESTION RESPONSES questions for me, you are free to ask at any time during the interview. Do you have any questions before we begin? a9_consent Do you agree to participate? 1 Yes 0 No  ef1 (Module ENDFORM) If a9_consent=1 MODULE B: BASIC RESPONDENT & HOUSEHOLD INFORMATION First, I will ask you some basic information about yourself and your household. # QUESTION RESPONSES b1 Respondent name open text: b2 Record respondent sex 1 Female 0 Male b3 How old are you? In completed years - Don’t Know: -98 Refused: -99 years # # b4 What is your relationship to the head of household? 1 Head of household  b5 2 Spouse of head of household  b4_1 3 Parent of head of household b4_1 4 Child of head of household b4_1 5 Sibling of head of household b4_1 6 Other relative  b4_1 -96 Other, specify b4_o Only if b4 = -96 (Other, specify) Other, specify response - open text: b4_1 What is the head of household's sex? 1 Female 0 Male b4_2 How old is the head of household? years # # b5 What is the highest level of education you have completed? 1 Did not go to schoolb6 2 Some primary schoolb6 3 Completed primary schoolb6 4 Some secondary schoolb6 5 Completed secondary schoolb6 6 University/Higher Educationb6 -96 Other, specify -99 Refusedb6 b5_o Only if b5= -96 (Other, specify) Other, specify response - open text: b6 What is your marital status? 1 Single b7 2 Marriedb7 3 Co-habitationb7 4 Divorced/Separatedb7 USAID.GOV WATSAN BASELINE SURVEY REPORT | 52 # QUESTION RESPONSES -96 Other, specify -99 Refusedb7 b6_o Only if b6= -96 (Other, specify) Other, specify response - open text: b7 Including yourself, how many people are members of this household? A household consists of a single person or group of persons (related or unrelated) who contributes to or benefits from any pooled expenses for food and essential items for living. Members of the group may pool their incomes to some extent. A household is not the same as the dwelling (the physical structure or space). Note that it is possible that not everyone who lives in the dwelling is a member of the household. Likewise, it could be that a single household spans more than 1 dwelling (e.g. a family and their in-laws who live next door) if they meet the definition of household above. People # # b7_softcon If b7> 12 You have input a number of household members in the previous question that appears high. You have input: [cross-reference Question b7]. Confirm it is correct and then continue. 1 Yes, continue b8 If b7> 1 How many household members are under the age of five? Record the number of children that are members of the household who have not yet reached their fifth birthday. # # MODULE C: WATER Now we will talk about water. First, I will ask you a few questions about your household’s main source of drinking water. # QUESTION RESPONSES c1 What is the MAIN source of DRINKING water for members of your household? JMP W1 Piped water 1 Piped into dwelling  c7 2 Piped into compound, yard, or plot  c7 3 Piped to neighborc2 4 Public tap/standpipec2 5 Borehole or tubewellc2 Dug well 6 Protected wellc2 7 Unprotected wellc2 Water from spring 8 Protected springc2 9 Unprotected springc2 10 Rainwater collectionc2 Delivered water 11 Tanker truck c2 12 Cart with small tank / drum c2 Water kiosk 13 Public Water Kiosk c2 14 Private Water Kioskc2 53 | WATSAN FINAL REPORT (BASELINE) USAID.GOV # QUESTION RESPONSES Packaged water 15 Bottled water c2 16 Sachet water c2 Surface water 17 Surface water (river, dam, lake, pond, stream, canal, irrigation channel) c2 -96 Other, specify c1_o Only if c1= -96 (Other, specify) Other, specify response - open text: c2 Where is that water collected from? JMP W3 1 In own dwelling  c7 2 In own yard / plot  c7 3 Elsewhere c3 How long does it take to go there, get water, and come back? (Minutes) (Includes any queueing time.) If household members Do Not collect: -97, Do not know: -98, Refused: -99 JMP W4 minutes # # # -97 Members do not collect  c7 -98 Do not know  c7 c4 If c3 = some # of minutes (not -97 or -98): You mentioned that it takes {__|__|__} minutes to go there, get water, and come back. Of that total time, how many minutes out of {__|__|__} are spent traveling (not queueing or collecting)? TRAVELING: minutes # # # c4_softcon If c3 =c4 and c3>0 You mentioned travel time that is the same as the round-trip time. Is that correct? If not, swipe back and correct the response. 1 Yes, continue c5 If c3>0 Who usually goes to this source to fetch water for your household? Select the individual primarily responsible for water collection. JMP XW2 1 Adult woman (>=15 years old) 2 Adult man (>=15 years old) 3 Girl (<15 years) 4 Boy (<15 years) -98 Do not know c7 -99 Refusedc7 c6 If c3>0 and c5 is NOT = -98 or -99 How many trips did that person make in the last week? Do not Know: enter -98, Refused: enter -99 JMP XW3 Number of times # # -98 Do not know c7 -99 Refusedc7 c6_softcon If c6>21 The number of round trips entered in the previous question appears high: ${c6}. Please confirm this is correct to continue. Otherwise, swipe back and correct the entry. 1 Yes, continue c7 Is water always available from your MAIN DRINKING water source? If no, probe for most of the time, some of the time, or rarely. 1 Yes, water is always available  c9 2 No, water is available most of the time 3 No, water is available some of the time USAID.GOV WATSAN BASELINE SURVEY REPORT | 54 # QUESTION RESPONSES JMP XW5 4 No, water is rarely available -98 Do not know  c9 -99 Refused  c9 c8 In the past month, for how many days was water from your MAIN DRINKING water source unavailable when needed? If respondent hesitates or says they do not remember or ‘do not know’ at first, encourage respondent to make their best guess. If they cannot make a best guess, enter -98 for “Do not know”. Enter -99 for “Refused”. JMP XW8 Number of days # # -98 Do not know -99 Refused c9 In the past month, has there been any time when your household did not have sufficient quantities of drinking water when needed? JMP W5 1 Yes, at least once 0 No, always sufficient  c11 -98 Do not know  c11 -99 Refused  c11 c9_softcon If c7=1 AND c9=1 The respondent stated that water is "always" available from your main water source but also mentioned that in the past month, there was at least one time when their household did not have sufficient quantities of drinking water when needed. This appears contradictory. Please confirm this is correct to continue. Otherwise, swipe back and correct the entry. 1 Yes, continue c10 If c9=1 What was the main reason you were unable to access sufficient quantities of water when needed? JMP XW6 1 Water is not available from source  c11 2 Water is too expensive  c11 3 Source is not accessible (for example: source is locked, path to source is blocked, etc.)  c11 -96 Other, specify -98 Do not know c11 -99 Refused c11 c10_o Only if c10= -96 (Other, specify) Other, specify response - open text: c11 Do you or any other household members usually do anything to your MAIN source of DRINKING water to make it safer to drink? JMP XW13 (revised as ‘usually’ and ‘main source’) 1 Yes 0 No  c13 -98 Do not know  c13 -99 Refused  c13 c12 If c11=1 What do you usually do to the water to make it safer to drink? After response(s), probe: Anything else? Select all that apply. After response(s), probe: Anything else? JMP XW14 (Note: electronic form will disallow multiple selections if “do not know” or “refused” are selected) 1 Boil  c13 2 Add bleach/chlorine  c13 3 Strain it through a cloth c13 4 Use water filter (ceramic, sand, composite, reverse osmosis, etc.)  c13 5 Solar disinfection c13 6 Let it stand and settle c13 -96 Other, specify 55 | WATSAN FINAL REPORT (BASELINE) USAID.GOV # QUESTION RESPONSES -98 Do not know  c13 -99 Refused  c13 c12_ o Only if c12= -96 (Other, specify) Other, specify response - open text: c13 In your opinion, is the water from your MAIN source safe to drink? 1 Yes, always 2 Yes, most of the time  c14 3 Only sometimes  c14 4 Never safe to drink  c14 -98 Do not know  c14 -99 Refused  c14 c13_softcon1 If c13=1 AND c11=1 The respondent has said that their main source of drinking water is "always" safe to drink but has also noted that they or other household members usually do something to the source to make it safer to drink. Please confirm these are the correct entries then continue. Otherwise, please swipe back and correct entries as needed. 1 Yes, continue c14 What are the major problems, if any, that you face with regard to the water supply from your MAIN source of DRINKING water? You may say “no problems” if there are none. Select all that apply. (Note: electronic form will disallow multiple selections if “no problems”, “do not know”, or “refused” are selected) 0 No problems 1 Insufficient amount of water (cannot regularly access a sufficient QUANTITY water from this source for intended uses) 2 Poor quality / unsafe for drinking 3 Unreliable / intermittent / irregular supply (service is irregular or unpredictable, there are frequent planned or unplanned shortages or cuts, not possible to assume the water will be flowing when needed) 4 Not affordable 5 Time spent collecting water 6 Low pressure (water pressure is not high, takes household long time to get the amount of water needed from the tap) 7 Bad taste / color / odor 8 Contains sediment / materials / particles -96 Other, specify -98 Do not know -99 Refused c14_o Only if c16= -96 (Other, specify) Other, specify response - open text: c14_softcon1 If c13=1 AND c14=2 or 7 The respondent has said that their main source of drinking water is "always" safe to drink but has also listed water quality or bad taste/color/odor as a problem with their main source of drinking water. Please confirm these are the correct entries then continue. Otherwise, please swipe back and correct entries as needed. 1 Yes, continue c14_softcon2 If c13=4 AND c14 is NOT = 2 1 Yes, continue USAID.GOV WATSAN BASELINE SURVEY REPORT | 56 # QUESTION RESPONSES The respondent has said that their main source of drinking water is "never" safe to drink but has NOT listed water quality as a problem with their main source of drinking water. Please confirm these are the correct entries then continue. Otherwise, please swipe back and correct entries as needed. c14_softcon3 If c7=1 AND c14=3 The respondent has said that water is "always" available from their main source but has also listed "Unreliable / intermittent / irregular supply" as a problem with their main source of drinking water. Please confirm these are the correct entries then continue. Otherwise, please swipe back and correct entries as needed. 1 Yes, continue c14_softcon4 If c7=4 AND c14=3 The respondent has said that water is "never" available from their main source but has also NOT listed "Unreliable / intermittent / irregular supply" as a problem with their main source of drinking water. Please confirm these are the correct entries then continue. Otherwise, please swipe back and correct entries as needed. 1 Yes, continue if c1 is NOT=1 and c1 is NOT=2 Thank you. Now I will ask a few questions about piped water connections. if c1=1 or c1=2 Thank you. Now I will ask a few questions about piped water connections. # QUESTION RESPONSES c15 Only if c1 is NOT=1 or 2: Do you have a piped connection at your household? Note: select ‘yes’ if there is a piped connection to the household, even if the account is not currently active / is currently disconnected. 1 Yes 0 No  c20 c16 If c15=1 Why do you not use the piped connection as your main source of drinking water? Select the main reason. 1 Account inactive or disconnected due to non-payment c16_i 2 Account inactive or disconnected due to other issuec16_i 3 Does not supply a sufficient amount of water c16_i 4 Does not supply good quality water c16_i 5 Cannot afford to do so / using other sources is cheaperc16_i -96 Other, specify -98 Do not know c16_i -99 Refused c16_i c16_o Only if c16= -96 (Other, specify) Other, specify response - open text: c16_i If c15=1 or c1=1 or 2 When was your connection installed? 1 Last 3 months 2 Between 3 and 6 months ago 3 Between 6 months and 1 year ago 4 More than 1 year ago 57 | WATSAN FINAL REPORT (BASELINE) USAID.GOV # QUESTION RESPONSES -98 Don’t know -99 Refused c16_i2 If c15=1 or c1=1 or 2 Why did you decide to connect to the water network at that time? Please mention up to three reasons, in order of importance (the first mentioned is the most important). #1 1 Network was extended/now available in my area 2 Reliability of service 3 Quality/safety of water 4 Outreach/campaign 5 Was able to afford connection 6 Save time (e.g. collecting water) -96 Other, specify c16_i2_1_o -98 Don’t Know -99 Refused #2 1 Network was extended/now available in my area 2 Reliability of service 3 Quality/safety of water 4 Outreach/campaign 5 Was able to afford connection 6 Save time (e.g. collecting water) 0 No other reasons -96 Other, specifyc16_i2_2_o -98 Don’t know -99 Refused #3 If selected “No other reasons”, “don’t know”, or “refused” for #2, cannot select 1, 2, 3, 4, 5, 6, or -96 here. 1 Network was extended/now available in my area 2 Reliability of service 3 Quality/safety of water 4 Outreach/campaign 5 Was able to afford connection 6 Save time (e.g. collecting water) 0 No other reasons -96 Other, specifyc16_i2_3_o -98 Don’t know -99 Refused c16_i2_1_o Only if c16_i2 #1 = -96 (Other #1, specify why you decided to connect) Other, specify response - open text: c16_i2_2_o Only if c16_i2 #2= -96 (Other #2, specify why you decided to connect) Other, specify response - open text: USAID.GOV WATSAN BASELINE SURVEY REPORT | 58 # QUESTION RESPONSES c16_i2_3_o Only if c16_i2 #3= -96 (Other #3, specify why you decided to connect) Other, specify response - open text: c17_a Only if c1 =1 or 2 (piped connection is main source) OR c16=3, 4, 5, or -96 (has an active piped connection, is not main source): About how many days per week is water supplied, on average, from your piped connection? If respondent is unsure, encourage them to estimate as best they can. Do not know: - 98, Refused: -99. JMP XW7 Number of days # # c17_b Only if c1 =1 or 2 (piped connection is main source) OR c16=3, 4, 5, or -96 (has an active piped connection, is not main source) AND c17_a>0: On those days, about how many hours per day is water supplied, on average, from your piped connection? Read response choices to respondent. JMP XW7 0 Less than one hour per day 1 1-5 hours per day 2 7-17 hours per day 3 18-23 hours per day 4 24 hours per day -98 Do not know -99 Refused c18 Only if c1=1 or 2 (Main source is piped connection), OR c16=3, 4, 5, or -96 (Piped connection, not main source) How do you pay for water from your piped connection? Read response choices to respondent. Read response choices to respondent: 1 Post-paid: Billed monthly  c19 2 Post-paid: Flat amount monthly  c19 3 Pre-paid meter  c19 0 Does not pay for piped water c22 -96 Other, specify -98 Do not know  c19 -99 Refused  c19 c18_o Only if c18= -96 (Other, specify) Other, specify response - open text: c19 Only if (c1=1 or 2 (Main source is piped connection), OR c16=3, 4, 5, or -96 (Piped connection, not main source)) AND c18 is NOT=0 How much do you usually pay for water FROM YOUR PIPED CONNECTION per month? If respondent unsure, encourage them to estimate as best they can. Do not know: -98, Refused: -99. Gourde # # # # -98 Do not know -99 Refused c20 Only if c15= 0 (No piped connection on premises): Do you plan to apply for a piped connection from the CTE (Centre Technique d'Exploitation)/DINAT within the next six months? 1 Yes 0 No  c22 -98 Do not know  c22 -99 Refused c22 59 | WATSAN FINAL REPORT (BASELINE) USAID.GOV # QUESTION RESPONSES c21 Only if c15= 0 AND c20=1 What is the main reason you plan to apply for a piped connection from the CTE (Centre Technique d'Exploitation)/DINAT? 1 Network is/will be extended to my area c22 2 Reduce time collecting water c22 3 Now able to afford / was not previously able to afford c22 4 To have a more reliable supply of water c22 5 To have better quality/safer water c22 -96 Other, specify -98 Do not know c22 -99 Refused c22 c21_o Only if c22= -96 (Other, specify) Other, specify response - open text: Thank you. Now I will ask a few questions about other water sources used in the household. c22 Do members of your household regularly use ANY OTHER DRINKING water sources? Select all that apply. JMP XW1 (Note: electronic form will exclude from this list the one that was selected as the main source for drinking) 0 No, does not use any other sources Piped water 1 Piped into dwelling 2 Piped into compound, yard, or plot 3 Piped to neighbor 4 Public tap/standpipe 5 Borehole or tubewell Dug well 6 Protected well 7 Unprotected well Water from spring 8 Protected spring 9 Unprotected spring 10 Rainwater collection Delivered water 11 Tanker truck 12 Cart with small tank / drum Water kiosk 13 Public Water Kiosk 14 Private Water Kiosk Packaged water 15 Bottled water 16 Sachet water Surface water 17 Surface water (river, dam, lake, pond, stream, canal, irrigation channel) -96 Other, specify -98 Don’t knowc25 -99 Refused c25 c22_o Only if c22= -96 (Other, specify) Other, specify response - open text: USAID.GOV WATSAN BASELINE SURVEY REPORT | 60 c22_softcon If c1 is NOT=1 or 2 AND c15 is NOT =1 AND c22= 1 or 2 Earlier in the form, you have recorded that the household does not use or have a tap on premises. However, for the previous question, you have selected tap in the dwelling or tap on the plot. Please ensure your entries are correct, otherwise swipe back and resolve the entries as needed. 1 Yes, continue  c22_softcon_text c22_softcon_text If c22_softcon=1 You have marked that the entries were correct even though they appear to contradict. Please explain here. Other, specify response - open text: c23 FOR EACH SOURCE SELECTED IN c22 if c22 is NOT = 0, 1, 2, 097, -98, -99, the electronic form will repeat the following question: Is [cross-reference c22] located within a 30-minute round trip? Includes any queuing time. 1 Yes (round trip within 30 minutes or less) 0 No (round trip is more than 30 minutes) -98 Don’t know -99 Refused c24 FOR EACH SOURCE SELECTED IN c22 if c22 is NOT = 0, 1, 2, 097, -98, -99, the electronic form will repeat the following question: What are the major problems, if any, that you face with regard to the water supply from [cross-reference c22]? You may say “no problems” if there are none. Select all that apply. (Note: electronic form will disallow multiple selections “no problems” or “don’t know” is selected.) 0 No problems 1 Insufficient amount of water (cannot regularly access a sufficient QUANTITY of water from this source for intended uses) 2 Poor quality / unsafe for drinking 3 Unreliable / intermittent / irregular supply (service is irregular or unpredictable, there are frequent planned or unplanned shortages or cuts, not possible to assume the water will be flowing when needed) 4 Not affordable 5 Time spent collecting water 6 Low pressure (water pressure is not high, takes household long time to get the amount of water needed from the tap) 7 Bad taste / color / odor 8 Contains sediment / materials / particles -96 Other, specify -98 Do not know -99 Refused c24_o Only if c24=-96, (Other, specify) Other, specify response - open text: c25 What is the MAIN source of water used by members of your household for OTHER PURPOSES, such as cooking and handwashing? JMP W2 Piped water 1 Piped into dwelling c26 2 Piped into compound, yard, or plot c26 3 Piped to neighborc26 4 Public tap/standpipec26 5 Borehole or tubewell c26 Dug well 61 | WATSAN FINAL REPORT (BASELINE) USAID.GOV 6 Protected well c26 7 Unprotected well c26 Water from spring 8 Protected spring c26 9 Unprotected spring c26 10 Rainwater collection c26 Delivered water 11 Tanker truck c26 12 Cart with small tank / drum c26 Water kiosk 13 Public Water Kiosk c26 14 Private Water Kiosk c26 Packaged water 15 Bottled water c26 16 Sachet water c26 Surface water 17 Surface water (river, dam, lake, pond, stream, canal, irrigation channel) c26 -96 Other, specify c25_o Only if c25= -96 (Other, specify) Other, specify response - open text: c26 Do members of your household regularly use ANY OTHER water sources for OTHER PURPOSES, such as cooking and handwashing? Select all that apply. JMP XW1 (Note: electronic form will exclude from this list the one that was selected as the main source for other purposes. Electronic form will disallow multiple selections if “don’t know” is selected.) 0 No, does not use any other sources Piped water 1 Piped into dwelling 2 Piped into compound, yard, or plot 3 Piped to neighbor 4 Public tap/standpipe 5 Borehole or tubewell Dug well 6 Protected well 7 Unprotected well Water from spring 8 Protected spring 9 Unprotected spring 10 Rainwater collection Delivered water 11 Tanker truck 12 Cart with small tank / drum Water kiosk 13 Water Kiosk 14 Private Water Kiosk Packaged water 15 Bottled water 16 Sachet water Surface water USAID.GOV WATSAN BASELINE SURVEY REPORT | 62 17 Surface water (river, dam, lake, pond, stream, canal, irrigation channel) -96 Other, specify -98 Don’t Know -99 Refused c26_o Only if c26= -96 (Other, specify) Other, specify response - open text: c25_26_softcon Earlier in the form, you have entered that the household does not have a tap on premises. However, for one of the two previous questions, you have selected tap in the dwelling or tap on the plot. Please ensure your entries are correct, otherwise swipe back and resolve the entries as needed. 1 Yes, continue c25_26_softcon_text c25_26_softcon_t If c25_26_softcon = 1 You have marked that the entries were correct even though they appear to contradict. Please explain here. Other, specify response - open text: Thank you. Now I will ask a few questions about water expenditures, and water storage in the household. # QUESTION RESPONSES c27_a If c1=3 or c22=3 or c25=3 or c26=3 Does your household pay neighbors for the water from their tap? 1 Yes 0 No c27_1 -98 Do not knowc27_1 -99 Refused c27_1 c27_b If c27_a=1 How much does your household pay neighbors for the water from their tap? Select “PER MONTH” or “PER WEEK” Enter Amount (Gourde): Gourde # # # # 1 PER MONTH 2 PER WEEK -98 Do not know -99 Refused c27_1 Thinking about all the water sources you use for any purpose in your household, how much do you usually pay per month for WATER from all sources combined? Select “PER MONTH” or “PER WEEK” Enter Amount (Gourde): Gourde # # # # 1 PER MONTH 2 PER WEEK -98 Do not know c28 -99 Refused c28 c27_softcon1 If c27_b and c19>=0 and: (selected “Months” in c27_1 AND c27_1=0 and: (selected “Months” in c27_1 and “Months” in c27_b and c27_1=0 and: (selected “Months” in c27_1 and c280 Are these containers used ONLY for DRINKING water? 1 Yes 0 No -98 Do not know -99 Refused askmain If c34>0 Ask respondent to point out: Which is the container used most often for DRINKING water by members of the household? Swipe forward for questions about this container. Allow respondent to point it out. c36 If c34>0 Do you ever mix water that you collected from different places in this container? 1 Yes 0 No -98 Do not know -99 Refused c37 If c34>0 Normally, how long do you store drinking water in this container until you consume it? (Number of Days) If less than 1 day, enter 0. Do not know: -98, Refused: -99. Number of days # # -98 Do not know -99 Refused c38 If c34>0 Do you ever mix treated and untreated water in this container? 1 Yes 0 No -98 Do not know -99 Refused If c34>0: Note to enumerator: for the next group of questions, OBSERVE and RECORD, do not ask aloud. # QUESTION RESPONSES c39 If c34>0 Observe & record for the container used most often for drinking – do not ask aloud: Is the container kept at ground level? 1 Yes 0 No c40 If c34>0 Observe & record for the container used most often for drinking – do not ask aloud: Container has wide (>=10cm) or narrow (<10cm) mouth/opening? 1 Wide (>=10 cm) 0 Narrow (<10 cm) c41 If c34>0 Observe & record for the container used most often for drinking – do not ask aloud: 1 Yes 0 No USAID.GOV WATSAN BASELINE SURVEY REPORT | 66 # QUESTION RESPONSES Is a utensil used to scoop or draw water from the container? Can ask if needed. c42 If c34>0 Observe & record for the container used most often for drinking – do not ask aloud: Is the outside of the container unclean, cracked, or leaking? 1 Yes 0 No c43 If c34>0 Observe & record for the container used most often for drinking – do not ask aloud: Is the inside of the container clean? Can ask to see inside. 1 Yes 0 No c44 If c34>0 Observe & record for the container used most often for drinking – do not ask aloud: Container has a spigot? 1 Yes 0 No c45 If c34>0 Observe & record for the container used most often for drinking – do not ask aloud: Does the container have a lid or fitted cover? 1 Yes 0 No Thank you. The last few questions regarding water are about your experience with any water shortages. # QUESTION RESPONSES c46 In the last 30 days, how frequently have you had to use water from other sources you only use when you cannot obtain water from your main drinking water source? 0 Never 1 Rarely 2 Sometimes 3 Most or all of the time -98 Don’t Know -99 Refused c47 In the last 30 days, how frequently have you had water to drink that you thought might not be safe for health? 0 Never 1 Rarely 2 Sometimes 3 Most or all of the time c48 In the last 30 days, how frequently have you economized water by limiting the amount of water used for drinking or cooking? 0 Never 1 Rarely 2 Sometimes 3 Most or all of the time -98 Don’t Know -99 Refused c49 In the last 30 days, how frequently have you economized water by limiting the amount of water used for other purposes? 0 Never 1 Rarely 2 Sometimes 3 Most or all of the time -98 Don’t Know -99 Refused c50 In the last 30 days, how frequently have you been unable to complete all work/chores because of time needed for water collection? 0 Never 1 Rarely 67 | WATSAN FINAL REPORT (BASELINE) USAID.GOV # QUESTION RESPONSES 2 Sometimes 3 Most or all of the time -98 Don’t Know -99 Refused c51 In the last 30 days, how frequently have you worried that volume of water available would be inadequate for all needs? 0 Never 1 Rarely 2 Sometimes 3 Most or all of the time -98 Don’t Know -99 Refused MODULE D: SANITATION Thank you. We will now ask some questions about sanitation in the household. # QUESTION RESPONSES d1 What kind of toilet facility do members of your household usually use? If ‘flush’ or ‘pour flush’, probe: Where does it flush to? If not possible to determine the toilet facility, ask permission to observe the facility. JMP S1 Flush / Pour flush 1 Flush to piped sewer system d2 2 Flush to septic tank d2 3 Flush to pit latrine d2 4 Flush to open drain d2 5 Flush, to don’t know where d2 Dry pit latrines 6 Pit latrine with slabd2 7 Pit latrine without slab, open pitd2 Composting toilets 8 Twin pit with slab d2 9 Twin pit without slab d2 10 Other composting toilet d2 --------- 11 Bucket d2 12 Container based sanitation d2 13 Hanging toilet / hanging latrine d2 0 No facility / Bush / Field  d5 -96 Other, specify -98 Do not know d2 -99 Refused d2 d1_ o Only if d1= -96 Other, specify Other, specify response - open text: d2 If d1 is NOT=0 Do you share this facility with others who are not members of your household? JMP S2 1 Yes 0 No  d5 -98 Don’t Know  d5 -99 Refused  d5 d3 If d2=1 Do you share this facility only with members of other households that you know or is the facility open to the use of the general public? 1 Shared with known households (not public) 2 Shared with general public -98 Do not know -99 Refused USAID.GOV WATSAN BASELINE SURVEY REPORT | 68 # QUESTION RESPONSES d4 If d2=1 How many households in total use this toilet facility, including your own household? Enter number of households if less than 10. If 10 or more, ENTER the number 10. Encourage respondent to estimate if unsure. Do not know: enter -98 Refused: enter -99 JMP XS1 Number of households if less than 10 # # 10 10 or more -98 Do not know d5 Where is this toilet facility located? JMP S3 1 In own dwelling 2 In own yard / plot 3 Elsewhere  d7 -99 Refused  d7 d6 If d5=1 or 2 May I take a photo of this toilet facility? If respondent hesitates you may first try to assure them this is only for research purposes. However, respect any refusal. 1 Yes 0 No d6_photo If d6=1 Take a photo of toilet facility Electronic form will display button to take photo d7 Is everyone in the household able to access and use the toilet at all times of the day and night? JMP XS3 (adapted as binary) 1 Yes d9 0 No -98 Do not knowd9 -99 Refused d9 d8 If d7=0 What is the main reason that household members are unable to use the toilet at all times of the day or night? JMP XS4 1 Limited mobility prevents members from using the toilet d9 2 Distance/barriers prevent members from reaching the toiletd9 3 Toilet is not always available to all household membersd9 4 Toilet is not always safe for all household members to used9 -96 Other, specify -98 Do not knowd9 -99 Refusedd9 d8_o Only if d8= -96 Other, specify d Other, specify response - open text: d9 Do men and women have equal access to the facility? 1 Yes 0 No -98 Do not know -99 Refused d10 Do you or other household members face any risks when using the toilet? Select all that apply JMP XS8 (Note: Electronic form will disallow multiple selections if “No risks faced”, “Don’t know”, or “Refused” are selected.) 0 No risks faced d11 1 Yes, risk to health d11 2 Yes, risk of harassment d11 -96 Yes, other, specify 69 | WATSAN FINAL REPORT (BASELINE) USAID.GOV # QUESTION RESPONSES -98 Don’t Know d11 -99 Refused d11 d10 _o Only if d10=-96 Other, specify Other, specify response - open text: d11 If b8>0: The last time a child under five passed stools, what was done to dispose of the stools? JMP XS5 1 Child used toilet/latrine d12 2 Put/rinsed into toilet or latrine d12 3 Put/rinsed into drain or ditch d12 4 Thrown into garbage (solid waste) d12 5 Buried d12 6 Left in the open d12 7 Used as manure d12 8 Thrown into stream d12 -96 Other, specify -98 Do not know d12 -99 Refused d12 d11 _o Only if d11= -96 Other, specify Other, specify response - open text: d12 How does your household usually dispose of garbage? JMP XS14 1 Collected by public service provider d13 2 Collected by private formal service provider d13 3 Collected by private informal service provider d13 4 Disposed of in designated waste disposal area d13 5 Disposed of within household yard or plot 6 Buried or burned d13 7 Disposed of elsewhere d13 -96 Other, specify -98 Do not know d13 -99 Refused d13 d12 _o Only if d12= -96 Other, specify Other, specify response - open text: d13 How does your household usually dispose of water used for cooking, laundry, and bathing? JMP XS15 1 Sink/drain connected to sewer 2 Sink/drain connected to septic tank 3 Sink/drain connected to pit 4 Sink/drain connected to soak pit 5 Sink/drain connected to open drain or open ground 6 Disposed directly to open ground or water body -97 N/A (cooking/laundry/bathing is done away from the household) -96 Other, specify -98 Do not know -99 Refused USAID.GOV WATSAN BASELINE SURVEY REPORT | 70 # QUESTION RESPONSES d13 _o Only if d13= -96 Other, specify Other, specify response - open text: If d1=2, 3, 6, 7, 8, 9, or 10 or d13=2 You have mentioned that you have a flush toilet connected to a septic tank OR pit latrine, OR that you have a sink/drain that empties to a septic tank. Toilet type: [cross-reference d1] Disposal of wastewater: [cross reference d13] d14 If has septic tank (d1=2 or d13=2): Where does your septic tank discharge to? JMP XS10 1 To a leach field, soak pit d15 2 To a sewer d15 3 To an open draind15 4 To open ground or watercoursed15 -96 Other, specify -98 Don’t knowd15 -99 Refusedd15 Only if d14= -96 Other, specify Other, specify response - open text: d15 If has pit latrine or septic tank (d1=2, 3, 6, 7, 8, 9, or 10 or d13=2): How many years ago was your pit latrine/septic tank built? If in the past year, enter 0. Encourage respondent to estimate if unsure Do not know: -98 Refused: -99 JMP XS11 years (enter 0 if in the past year) # # -98 Do not know -99 Refused d16 If has pit latrine or septic tank (d1=2, 3, 6, 7, 8, 9, or 10 or d13=2): Has your pit latrine or septic tank ever been emptied? JMP S4 1 Yes emptied 0 Never emptied  d20 -98 Do not know  d20 -99 Refused  d20 d17 If has pit latrine or septic tank (d1=2, 3, 6, 7, 8, 9, or 10 or d13=2) AND d16=1 How many years ago was your pit latrine/septic tank last emptied? If in the past year, enter 0 Encourage respondent to estimate if unsure Do not know: -98 Refused: -99 JMP XS12 years (enter 0 if in the past year) # # -98 Do not know -99 Refused d18 If has pit latrine or septic tank (d1=2, 3, 6, 7, 8, 9, or 10 or d13=2) AND d16=1 The last time it was emptied, where were the contents emptied to? JMP S5 Removed by service provider 1 To a treatment plant  d19 2 Buried in a covered pit  d19 Emptied by household 3 To uncovered pit, open ground, water body, or elsewhere  d19 71 | WATSAN FINAL REPORT (BASELINE) USAID.GOV -96 Other, specify -98 Do not know  d19 -99 Refused  d19 Only if d18= -96 Other, specify Other, specify response - open text: d19 If has pit latrine or septic tank (d1=2, 3, 6, 7, 8, 9, or 10 or d13=2) AND d16=1 The last time your pit latrine/septic tank [D1] was emptied, who emptied it? e.g. Company or organization name, etc. Don't know: -98 Refused: -99 JMP XS13 open text: -98 Do not know -99 Refused d20 If has pit latrine or septic tank (d1=2, 3, 6, 7, 8, 9, or 10 or d13=2) Would you be willing to pay for a service provide to empty your pit latrine or septic tank? 1 Yes 0 No -98 Do not know -99 Refused d21 Does your sanitation facility [S1] leak or overflow wastes at any time of the year? Probe for problems during rainy season or floods. JMP XS9 0 No, never 1 Yes, sometimes 2 Yes, frequently -98 Do not know -99 Refused d22 Would you like to make improvements to your toilet? 1 Yes 0 No  d25 -98 Do not know  d25 -99 Refused  d25 d23 If d22=1 How would you like to improve it? Select all that apply. (Note that electronic system will only show relevant options based on answer to d1) 1 Install permanent superstructure  d24 2 Improve ventilation  d24 3 Install pour flush  d24 4 Improve water supply  d24 5 Replace slab  d24 6 Install lockable door  d24 -96 Other, specify -98 Do not know  d24 -99 Refused  d24 Only if d23= -96 (Other, specify) Other, specify response - open text: d24 If d22=1 Would you be willing to pay for these improvements? 1 Yes 0 No USAID.GOV WATSAN BASELINE SURVEY REPORT | 72 -98 Do not know -99 Refused d25 Have you or any member of your household received training in relation to open defecation? 1 Yes 0 No -98 Do not know -99 Refused MODULE E: HYGIENE Thank you. We will now ask some questions about hygiene in the household. # QUESTION RESPONSES e1 Can you please show me where members of your household most often wash their hands? Only applies to households using sanitation facilities with on-site storage (latrines, septic tanks, composting toilets and twin pits). JMP H1 Fixed facility observed (sink / tap) 1 Fixed facility (sink/tap): In dwelling  e2 2 Fixed facility (sink/tap): In yard/plot  e2 Mobile object observed 3 Mobile object: bucket, jug, kettle  e2 0 No handwashing place in dwelling, yard, plot e4 -99 No permission to observe e4 -96 Could not observe for another reason e1_o Only if e1= -96 (Not observed: Other reason) Other, specify response - open text: e4 If e1=1 or 2 or 3: Note to enumerator: for the next group of questions on the following screen, OBSERVE and RECORD, do not ask aloud. e2 If e1=1 or 2 or 3 Observe availability of water at the place for handwashing. Verify by checking the tap/pump, or basin, bucket, water container, or similar objects for presence of water. Note this is not read aloud to respondent. Verify by checking the tap/pump, or basin, bucket, water container, or similar objects for presence of water. Note this is not read aloud to respondent. JMP H2 1 Water is available 0 Water is not available e3 If e1=1 or 2 or 3 Observe availability of soap or detergent at the place for handwashing. Note this is not read aloud to respondent. JMP H3 1 Soap or detergent available 0 Soap or detergent not available e4 If e1<=0 For cases where observation of handwashing facility was not possible: Where do you and other members of your household most often wash your hands? JMP XH2 1 Sink/tap in dwelling 2 Sink/tap in yard/plot 3 Mobile object reported (bucket, jug, kettle) 4 No handwashing place in dwelling/yard/plot -96 Other, specify -98 Do not know -99 Refused e4_o Only if e4= -96 (Other, specify) Other, specify response - open text: e5 If e1<=0 1 Yes, shown 73 | WATSAN FINAL REPORT (BASELINE) USAID.GOV For cases where observation of handwashing facility was not possible: Do you have soap or detergent in your household for washing hands? Can you show it to me? JMP XH2 2 Reported yes, but not shown 0 No -96 Other, specify e5_o Only if e5= -96 (Not observed: Other reason) Other, specify response - open text: If b8>0 MODULE F: WATER RELATED ILLNESS Activated only if there is at least one child under five in the household based on the question from Module A. # QUESTION f1 I would like to ask you a few questions about each of the [#] children under five in this household. We will go one by one for each child. Note: Interviewers will ask questions f2 through f7 for each child one by one. The electronic data collection system will automate this process. The electronic system will have an additional validation programmed in to ensure the number of children was accurately reported earlier and reflected below. Child: f2 Child name f3 Child age (years) Age at child's last birthday. If less than 1, enter 0 Do not know: -98 Refused: -99 If f3=0 f4 Child age (months) If less than 1 month, enter 0 Do not know: -98 Refused: -99 f5 Has [name] experienced any episode of diarrhea (watery stools) in the last two weeks (14 days)? If f5=1 f6 Did any household member stay home from work or school, or miss any time away from other productive activities to take care of [name] over the last two weeks during this illness? If f6=1 f7 How many days did household member(s) miss from work or school or other normal activities to care for [name]? Entries can include partial days (e.g. 1.5 days) Do not know: -98 Refused: -99 (open text) years # months # 1 Yes 0 No  next child -98 Don’t Know -99 Refused 1 Yes 0 No  next child -98 Don’t Know -99 Refused days # # 2 (open text) years # months # 1 Yes 0 No  next child -98 Don’t Know -99 Refused 1 Yes 0 No  next child -98 Don’t Know -99 Refused days # # … (open text) years # months # 1 Yes 0 No  next child -98 Don’t Know -99 Refused 1 Yes 0 No  next child -98 Don’t Know -99 Refused days # # N (open text) years # months # 1 Yes 0 No  next child 1 Yes 0 No  next child days # # USAID.GOV WATSAN BASELINE SURVEY REPORT | 74 # QUESTION -98 Don’t Know -99 Refused -98 Don’t Know -99 Refused MODULE G: ADDITIONAL HOUSEHOLD INFORMATION # QUESTION RESPONSES g1 Observe & record: Main material of the floor of the dwelling. Do not read aloud. Natural Material Rudimentary Material g2 Elaborate Material 11 Earth/Sand g2 12 Dung g2 21 Wooden planks g2 31 Parquet or waxed wood g2 32 Cement/Concrete/Masonry g2 33 Mosaic/Ceramic g2 -96 Other, specify g1_o If g1=-96, (Other, specify) Other, specify response - open text: g2 Observe & record: Main material of the roof of the dwelling. Do not read aloud. Natural Material Rudimentary Material Elaborate Material 11 No roof g3 12 Thatch / Fins / Leaves g3 21 Fins / Bamboo g3 22 Wooden planks g3 23 Canvas / Tent g3 31 Metal g3 32 Wood g3 33 Cement g3 34 Shingles g3 -96 Other, specify g2_o If g2=-96, (Other, specify) Other, specify response - open text: g3 Observe & record: Main material of the exterior walls of the dwelling. Do not read aloud. Natural Material Rudimentary Material Elaborate Material 11 No walls g4 12 Bamboo / Cane / Palm / Trunk g4 21 Bamboo with mud g4 22 Stones with mud g4 23 Cardboard g4 24 Reclaimed wood g4 25 Canvas / Tent g4 31 Cement g4 32 Lime stones / Cement g4 75 | WATSAN FINAL REPORT (BASELINE) USAID.GOV # QUESTION RESPONSES 33 Cement blocks g4 34 Cement bricks g4 35 Metal g4 36 Wooden board / Shingle g4 -96 Other, specify g3_o If g3=-96, (Other, specify) Other, specify response - open text: g4 What type of fuel does your household mainly use for cooking? Do not read aloud. 1 Electricity g5 2 Liquefied propane gas (LPG) g5 3 Natural gas g5 4 Biogas g5 5 Kerosene g5 6 Coal, lignite g5 7 Charcoal g5 8 Wood g5 9 Straw / Branches / Herbs g5 10 Agricultural residues g5 11 Dung g5 -97 No meal prepared in the household g5 -96 Other, specify -98 Don’t know -99 Refused g5 g4_o If g4=-96, (Other, specify) Other, specify response - open text: g5 How many rooms in this household are used for sleeping? # rooms Do not know: -98 Refused: -99 # Rooms g5_ If g5>10 You have entered a number larger than 10 for the previous question. Is the response input for the previous question correct? If correct, press yes below and swipe forward to continue. 1 Yes, continue g6 Does this household own any livestock, herds, other farm animals, or poultry? If respondent unsure, encourage them to estimate as best they can. Divide livestock by the number of households if shared between households. Do not know: -98 Refused: -99 1 Yes 0 No  g13 -98 Don’t Know  g13 -99 Refused  g13 animals If g6=1 How many of the following animals does this household own? ------------------------------ g7 If g6=1 Milk cows or bulls? # | # USAID.GOV WATSAN BASELINE SURVEY REPORT | 76 # QUESTION RESPONSES g8 If g6=1 Other cattle? # | # g9 If g6=1 Horses, donkeys, or mules? # | # g10 If g6=1 Goats? # | # g11 If g6=1 Sheep? # | # g12 If g6=1 Chickens or other poultry? # | # g13 Does any member of this household own any agricultural land? 1 Yes 0 No  hhhave -98 Don’t know  hhhave -99 Refused  hhhave g14 If g13=1 Record the size of the agricultural land owned by the household. If respondent unsure, encourage them to estimate as best they can. Do not know: -98 Refused: -99 Units for recording land size: How many? # # 1 CENTIÈME (1=129 square meters) 2 CARREAU (1=1.29 hectares) -98 Do not know -99 Refused hhhave Does your household have: --------------------------------------------- g15 Electricity? 1 Yes 0 No -98 Don’t know -99 Refuse g16 A radio? 1 Yes 0 No -98 Don’t know -99 Refuse g17 A television? 1 Yes 0 No -98 Don’t know -99 Refuse g18 A non-mobile telephone? 1 Yes 0 No -98 Don’t know -99 Refuse g19 A computer? 1 Yes 0 No -98 Don’t know -99 Refuse 77 | WATSAN FINAL REPORT (BASELINE) USAID.GOV # QUESTION RESPONSES g20 A refrigerator? 1 Yes 0 No -98 Don’t know -99 Refuse hhown Does any member of your household own: g21 A watch? 1 Yes 0 No -98 Don’t know -99 Refuse g22 A mobile phone? 1 Yes 0 No -98 Don’t know -99 Refuse g23 A bicycle? 1 Yes 0 No -98 Don’t know -99 Refuse g24 A motorcycle or motor scooter? 1 Yes 0 No -98 Don’t know -99 Refuse g25 An animal-drawn cart? 1 Yes 0 No -98 Don’t know -99 Refuse g26 A car or truck? 1 Yes 0 No -98 Don’t know -99 Refuse MODULE Y: GPS & CONTACT # QUESTION RESPONSES y1 Respondent phone number (main) * if no phone, enter eight 7s * if refused, enter eight 9s # # # # # # # # y2 If y1is NOT=”77777777” or “99999999” Respondent phone number (alternative) * if no alternative, enter eight 7s * if refused, enter eight 9s # # # # # # # # y3 End Time NB: Will be recorded automatically on the electronic form h h m m y4 Did the respondent ask to stop the survey before reaching the end? 1 Yes 0 No -98 Don’t know USAID.GOV WATSAN BASELINE SURVEY REPORT | 78 # QUESTION RESPONSES -99 Refuse If a9_consent is NOT=1 MODULE ENDFORM: INCOMPLETE FORMS # QUESTION RESPONSES ef1 If sample1=0 or sample2 is NOT=1 or sample4=0 The form has ended because a residential household with an eligible respondent was not located. _______________________________ ef2 If a9_consent=0 The form has ended because the respondent has refused to participate. _______________________________ MODULE Z: INTERVIEWER FILLS OUT AFTER LEAVING THE HOUSEHOLD: # QUESTION RESPONSES gps_survey Record household GPS Coordinates Note: Should be done outside the dwelling where signal is better. Must be done before leaving premises. NB: Will be a push-button reading on the electronic form, which records latitude, longitude, altitude, and accuracy z1_comments_refusal If a9_consent=0 Notes on the interaction Leave blank if nothing to note Other, specify response - open text: z1_comments If a9_consent=1 Notes on the interaction/interview Leave blank if nothing to note Other, specify response - open text: z2_comments If a9_consent=1 Comments on specific questions Leave blank if nothing to note Other, specify response - open text: 79 | WATSAN FINAL REPORT (BASELINE) USAID.GOV 7.2 ANNEX B: DETAILED METHODOLOGY JMP WATER SOURCE CLASSIFICATION As additional context for the water source classifications described as part of the body of this report, we provide the following material from the JMP post-2015 monitoring service ladders. Source: WHO/UNICEF JMP 2018b. USAID.GOV WATSAN BASELINE SURVEY REPORT | 80 JMP SANITATION FACILITY CLASSIFICATION Source: WHO/UNICEF JMP 2018b. 81 | WATSAN FINAL REPORT (BASELINE) USAID.GOV SAMPLE SIZE CALCULATIONS WATSAN’s MEL Plan (March 2018) indicated that the activity intends to increase access to basic water sources and basic sanitation by 10 percent each over baseline by midline and 20 percent over baseline by endline. Based on ESS’s comprehensive review of the WATSAN internal baseline methodology, the ET believed that WATSAN’s intention was to state targets in terms of percentage point increases over baseline, rather than percentage (to illustrate the difference: 10 percent increase over a hypothetical baseline value of 50 percent is five percentage points, which would imply a target of 55 percent, whereas a ten-percentage point increase would imply a target of 60 percent). The ET conducted power calculations with the understanding that the activity intended targets as percentage point increases over baseline values. Using Stata 15’s power command, the ET calculated sample size requirements for a one-tailed test, as it is only interested in detecting increases over baseline. Calculations were stratified by commune to allow for measurement of changes in each commune individually. This was important given the different context in each commune and considering the nature of interventions in each commune may have different effects on the outcome indicators. Calculations assumed the team employs an SRS approach and draws independent repeated cross-sectional samples at each phase. The formula for the sample size required for each commune is, thus, given by the equation below:25 n (per commune) =� 𝑧𝑧1−𝛼𝛼�2 𝑃𝑃�(1− 𝑃𝑃�)+𝑧𝑧1−𝛽𝛽�𝑃𝑃1,𝑒𝑒𝑒𝑒𝑒𝑒�1−𝑃𝑃1,𝑒𝑒𝑒𝑒𝑒𝑒�+𝑃𝑃2,𝑒𝑒𝑒𝑒𝑒𝑒�1−𝑃𝑃2,𝑒𝑒𝑒𝑒𝑒𝑒� 𝛿𝛿 � 2 where: n is the required sample size for each timepoint where proportions will be compared δ = P2-P1, i.e., the difference in percentage points over baseline, set as 10 percentage points P1 = estimated baseline value P2 = estimated endline value, i.e. P1 + δ 𝑃𝑃� = average of P1 and P2 Z1-a = z-value corresponding to 95 percent confidence level for a one-tailed test, set as 1.64 Z∞ = z-value corresponding to 80 percent power, set as 0.84 No design effect (DEFF) needed because of the random sampling design (i.e. no clustering) Based on these calculations, the ET required a sample size of 310 households in each commune to detect a minimum effect size of a ten-percentage point increase over baseline values, regardless of the true baseline value (P1). If the true baseline value was near 50 percent, this sample size would have allowed the team to detect any difference that is ten percentage points or more. If the true baseline value was different than 50 percent, the sample size would have allowed for detection of even smaller increases in the key outcomes (see Figure 19).26 This applied to both outcome indicators, IND 1 (water) 25 Stukel, 2018. Feed the Future Population-Based Survey Sampling Guide. 26 In addition, P1 values are typically taken from existing data sources for the purpose of sample size calculations. There are large discrepancies between potential P1 values between numbers quoted in the USAID SOW, JMP/DHS figures, and the WATSAN USAID.GOV WATSAN BASELINE SURVEY REPORT | 82 and IND 3 (sanitation). The ability to detect a ten-percentage point increase regardless of the actual baseline value was important because actual baseline values were likely to differ between communes; this sample size allowed the ET to detect the activity’s targeted increases (if achieved) regardless of baseline value and between-commune differences. Figure 19. MDES for n=310, across possible baseline values (p1) ADDITIONAL SAMPLE SIZE CALCULATION TABLES To conduct sample size calculations, the ET conducted multiple iterations of the analysis using values of P1 obtained from different sources to understand sample size requirements for different values of P1. First, the ET calculated sample size requirements based on WATSAN’s internal baseline figures. While ESS had raised several methodological questions regarding the Water and Sanitation Project internal baseline as well as questions regarding how the final percentages shown in that report were calculated, it was the only source of data available to the ET at such a disaggregated level (commune), and thus these numbers were used as part of a broader sample size calculation exercise. The sample sizes calculated using WATSAN’s numbers are shown below in Table 29 and Table 31 for water and sanitation, respectively. One major caveat in terms of the WATSAN results was that they presented unlikely results of higher access to basic sanitation than basic water, which was unlikely the case in reality. The ET also calculated sample size requirements using the urban averages reported in the JMP Haiti Country Report from 2017 which uses Demographic and Health Survey (DHS) data, adjusted through regression. 27 The level of access for urban areas as a whole was likely higher than for the specific communes included in the Water and Sanitation Project. The sample size calculations with JMP urban average numbers are presented in Table 30 and Table 32 below. Regardless of the questions surrounding the relevance of the WATSAN or JMP numbers to the five target communes, conducting the sample size calculations with varying values for P1 had the benefit of demonstrating a range of sample size requirements across a wide range of potential P1 values. The value of this exercise was that across all calculations, the ET was able to obtain the maximum number from all of these – 308, keeping all other parameters at the specified levels – and used this figure rounded internal baseline report (See Annex D for more details). Further, those estimates may be different than the baseline values for our smaller zones within the five targeted communes. 27 See https://washdata.org/data/country/HTI/download. +5 +6 +7 +8 +9 +10 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% MDES (percentage points) Baseline Value (P1) 83 | WATSAN FINAL REPORT (BASELINE) USAID.GOV up to the nearest 10 – 310 – as the proposed sample size for the ESS baseline survey. Using that number (per commune), the ET then conducted minimum detectable effect size (MDES) calculations, as shown in the body of the report, which demonstrated that 310 surveys per commune would allow for detection of the intended target change in the key outcome indicators, regardless of the true value of P1. WATER Table 28. Sample sizes, p1 based on WATSAN data (water) Commune P1 – water P2 – water δ Z1-a Zβ Sample Size Cap-Haitian 0.197 0.297 0.10 1.64 0.84 229 Mirebalais 0.327 0.427 0.10 1.64 0.84 290 Cayes 0.480 0.580 0.10 1.64 0.84 307 Jérémie 0.178 0.278 0.10 1.64 0.84 217 Canaan 0.027 0.127 0.10 1.64 0.84 87 Total 1.130 Note that WATSAN’s internal estimates for access to basic water are provided in their internal baseline report but the calculation of the final percentages shown is not transparently documented in their baseline report; ESS was therefore not able to trace how these numbers were arrived at based on their survey data. Table 29. Sample sizes, p1 based on JMP data Commune P1 – water P2 – water δ Z1-a Zβ Sample Size Cap-Haitian 0.85 0.95 0.10 1.64 0.84 111 Mirebalais 0.85 0.95 0.10 1.64 0.84 111 Cayes 0.85 0.95 0.10 1.64 0.84 111 Jérémie 0.85 0.95 0.10 1.64 0.84 111 Canaan 0.85 0.95 0.10 1.64 0.84 111 Total 1.130 SANITATION Table 30. Sample sizes, p1 based on WATSAN data (sanitation) Commune P1 – water P2 – water δ Z1-a Zβ Sample Size Cap-Haitian 0.613 0.713 0.10 1.64 0.84 276 Mirebalais 0.510 0.610 0.10 1.64 0.84 304 Cayes 0.525 0.625 0.10 1.64 0.84 302 Jérémie 0.525 0.625 0.10 1.64 0.84 302 Canaan 0.473 0.573 0.10 1.64 0.84 308 Total 1,492 Note that WATSAN’s internal estimates for access to basic sanitation appear to be higher than their estimates for access to basic water sources, which does not align with any other published source of data from JMP or DHS. The results presented in the WATSAN internal baseline report do not provide sufficient levels of tabulated data to allow ESS to validate how these percentages were calculated. USAID.GOV WATSAN BASELINE SURVEY REPORT | 84 Table 31. Sample sizes, p1 based on JMP data (sanitation) Commune P1 – sanitation P2 – sanitation δ Z1-a Zβ Sample Size Cap-Haitian 0.44 0.54 0.10 1.96 0.84 308 Croix-des-Bouquets 0.44 0.54 0.10 1.96 0.84 308 Mirebalais 0.44 0.54 0.10 1.96 0.84 308 Les Cayes 0.44 0.54 0.10 1.96 0.84 308 Jérémie 0.44 0.54 0.10 1.96 0.84 308 Total 1,540 USE OF SPATIAL DATA FOR SAMPLING In ArcMap 10.2 software, building footprint data points were added using data provided by CNGIS. The ET used the “Select by Location” tool to select points from those shapefiles that were located within WATSAN zones in the five targeted communes; the latter of which were provided by the Water and Sanitation Project team. The ET calculated the centroid of each building footprint by using the “Add XY coordinates” tool from the data management toolbox. The selected points were exported to a new shapefile and added to the data frame. A “spatial join” was then conducted to extract other geographical information from the Water and Sanitation Project layers, including the commune and sector. Data from the Attribute Table of the resulting joined file was then exported. Minor data cleaning issues resolved (e.g. replacing commune for sectors or zones without commune names in the Activity shapefile). The exported data was then opened in Microsoft Excel, to save from text to comma delimited file type. That file was saved and then imported in Stata 15.0 software to conduct the sampling. APPLICATION OF POST-STRATIFICATION WEIGHTS For baseline data collection, the ET and field team carried out SRS. A sample of 310 was randomly selected from within each of the five communes to obtain representative estimates of statistics at the commune level. However, to ensure the total estimated statistics for WATSAN zones would be accurate, the ET applied post-stratification weights to the final analysis. The justification for application of these weights is that the actual population within the WATSAN zones of each targeted commune differ. For example, those with greater populations (such as Cap-Haitian) would otherwise be underrepresented by the 310 households in the sample, relative to those with smaller populations (such as Mirebalais) that also have 310 households in the sample. Thus, the post-stratification weights adjust the estimates of this specific target population of interest (all households within WATSAN zones) to ensure representativeness both at the commune level and overall. Stata’s -svyset- command was used to apply these weights. Commune was specified as the poststratum, and poststratum sizes (populations) were derived from the spatial sampling effort described in the previous section. The total number of buildings identified within all WATSAN zones in each commune represented the total “population.” These total population numbers excluded any buildings found to be non-residential during the survey attempts. One limitation to this approach was that the ET did know the total number of buildings that would not be attempted due to being non-residential. However, removing those that were known to be non-residential at least adjusted for known corrections to the sampling frame. 85 | WATSAN FINAL REPORT (BASELINE) USAID.GOV Table 32. Population in each commune for post-stratification adjustment Commune Sample Population a % of population sampled Cap-Haitian 310 26,427 2% Croix-des-Bouquets 310 18,845 4% Jérémie 310 7,224 3% Les Cayes 310 11,217 5% Mirebalais 310 6,586 2% Total 1,550 70,299 2% a Number of buildings in sampling frame with the number of non-residential buildings encountered subtracted. Weights can be considered household-level weights, except that in cases where there are multiple households per building, the ET unintentionally omitted a question about how many households lived in that same building, so those weights can only be interpreted as building-level weights. In other words, for 266 households in the sample, the ET assumed their weight is slightly underestimated because they were sampled from multiple-household buildings, as shown in Table 342. Table 33. Households sampled from a single or multiple-household building, by commune Commune Cap￾Haitian Canaan Jérémie Les Cayes Mirebalais Total Single household 242 200 257 234 251 1,284 Multiple households 68 10 53 76 59 266 Total 310 310 310 310 310 1,550 USAID.GOV WATSAN BASELINE SURVEY REPORT | 86 7.3 ANNEX C. DETAILED PROJECT CONTEXT PROJECT RESULTS FRAMEWORK & ASSUMPTIONS Figure 20. Haiti WATSAN results framework A series of critical assumptions about the context in which WATSAN will be implemented underpins the results framework. As listed in the WATSAN Monitoring, Evaluation, and Learning Plan,28 the activity makes the following key assumptions pertaining specifically to the stakeholders and institutions with which it partners: Haitians will pay for an improved water quality and service; Haitians will support improved water management and fecal sludge management practices; Service providers will implement water safety plans based on deteriorating water quality, rising treatment costs, climate change pressure, and customer demand; Service providers will be motivated to improve services when given incentives, training and tools; and Stakeholders have the willingness to participate in capacity building and knowledge sharing. These assumptions supplement those in the results framework, as well as others not detailed here, related to government commitments, and the general economic and financial environment, as detailed in the WATSAN MEL Plan. 28 DAI, March 2018. 87 | WATSAN FINAL REPORT (BASELINE) USAID.GOV WATSAN PROGRESS AS OF FEBRUARY 2020 Below is a summary of major activities WATSAN has completed as of February 2020, based on WATSAN quarterly reports and workplans. The list highlights major accomplishments especially as related to the baseline household survey and may not be a comprehensive list of activity interventions to date. Table 34. WATSAN activity progress to date Component 1 Increasing Access to Sustainable Water Services Task 1.1 Water Infrastructure Engineering Services WATSAN has made progress toward the completion of infrastructure assessment reports for target communes, assessments of potential sub-interventions in Cap-Haitian, and Quick Impact Projects (QIP) in Cap-Haitian. Installation of volumetric and household meters in Les Cayes and Jérémie to reduce non-revenue water losses. In Les Cayes, WATSAN has begun providing fuel to power well generators as of late 2019. Task 1.2 Water Infrastructure Construction As of FY19 Q1, WATSAN had received bids for an expression of interest for related construction and QIP work. Requests for proposals for QIP work in Cap-Haitian were planned for FY19 Q2. Begun in Q2 2019, the activity includes a variety of activities from repairing the chlorine pump and electrical installation to replacing broken doors and windows, the perimeter fence, and painting. Repair of leaks in the existing piped water infrastructure in Jérémie to reduce non-revenue water losses. By the end of January 2020, WATSAN was expected to connect several reservoirs or water kiosks to the main distribution pipe in Canaan. Task 1.3 Technical Assistance for Water Service Providers WATSAN has completed CTE (Centre Technique d’Exploitation) assessments in Cap￾Haitian and Les Cayes and has conducted trainings with DINEPA. WATSAN has selected two potential grantees to provide more detailed proposals for the Enterprise Acceleration Fund (EAF) grants. Component 2 Increasing Access to Sustainable Sanitation Services Task 2.1 Support to Sanitation Enterprises WATSAN has drafted a sanitation approach, which has been shared with DINEPA, and which describes the types of interventions that the activity can implement under this task. The sanitation approach was to be finalized in FY19 Q2. Convening of 15 artisan groups, including masons, latrine excavators, and carpenters, to form private companies focused on construction of toilets/latrines. Task 2.2 Waste Treatment & Fecal Sludge Management Engineering Services WATSAN has begun evaluating the Les Cayes fecal sludge management facility and the Jérémie proposed facility. Task 2.3 Waste Treatment & Fecal Sludge Management Construction None: This is still in the assessment phase. Component 3 Improving the Enabling Environment for Sustainable Implementation, Operations, and Maintenance of Water and Sanitation Services Task 3.1 Technical Assistance to National and Sub-National Government Structures WATSAN has completed field surveys for CTE assessments in Cap-Haitian and Les Cayes. Establishment of a management information system (MIS) for CTEs. Formulation of task forces in each commune that bring together municipal authorities and local leaders to focus on improving overall sanitation in their respective communities. Task 3.2 Knowledge Dissemination Learning WATSAN has developed a framework for collaboration, learning, and adaptation (CLA). USAID.GOV WATSAN BASELINE SURVEY REPORT | 88 Cross-Cutting Task CC1: Communications WATSAN has begun to develop a behavior change communication (BCC) strategy and staffed its team with relevant experts. Implementation and customization to each commune has not been listed in the latest available quarterly report. Though, as of May 2019, the project team may be working to promote sanitation and end open defecation through community awareness, support sanitation marketing efforts, and develop community sanitation programs in Mirebalais and Jérémie. Task CC2: Monitoring, Evaluation, and Learning (MEL) WATSAN has established baselines for activity indicators, collected baseline data on CTE-related indicators for Les Cayes and Cap-Haitian, prepared for a household baseline survey, used satellite imagery to analyze the physical landscape and population density, validated building footprints in Canaan for estimating activity indicators, and conducted flood risk analysis and mapped hydrology zones in Mirebalais. Task CC3: Enterprise Acceleration Fund WATSAN has solicited grant applications and has pre-selected three applicants to screen further. To date, all applicants have been pre-selected for Cap-Haitian. Task CC 4: Environmental Compliance WATSAN has developed a site-specific Environmental Mitigation and Monitoring Plans (EMMP) for Cap-Haitian and is in the process of developing an EMMP for the Les Cayes QIP. The activity has also begun work on its Environmental Management System (EMS). Task CC 5: Gender and Equity WATSAN has finalized and submitted to USAID two strategies: A gender strategy and a pro-poor and social inclusion strategy. PLANNED INTERVENTIONS IN YEAR 3 According to WATSAN’s Year 3 Workplan (October 2019-September 2020), the activity intends to implement the following work in the five target communes:29 Table 35. WATSAN activity planned activities for year 3 (2019-2020) Task Commune TASK 1.1: WATER INFRASTRUCTURE ENGINEERING SERVICES Design of the Canaan Water Distribution System Croix-des-Bouquets Drilling of a test well at the Balan in Cap-Haitian Cap-Haitian Design of a solar energy system for the Cap-Haitian wells Cap-Haitian Analysis of the Savanne Water Reservoir in Les Cayes Les Cayes Improvements to the Transmission Line from the Mahotière Spring Cap Mirebalais Expansion of the Non-Revenue Water Loss Reduction Program Mirebalais; Cap-Haitian Rehabilitation of kiosks All communes Design support during the construction process All communes TASK 1.2: WATER INFRASTRUCTURE CONSTRUCTION F10 Wellfield repairs in Cap-Haitian Cap-Haitian Installation of hybrid solar pumps at the Charpentier Wellfield in Les Cayes Les Cayes Repairs to the Mahotière Spring Cap in Jérémie Jérémie TASK 1.3: TECHNICAL ASSISTANCE FOR WATER SERVICE PROVIDERS Training to CTEs: Organization and Strategy 29 USAID Haiti Water and Sanitation (WATSAN) Activity: Year 3 Work Plan (October 1, 2019–September 30, 2020), 13–22. 89 | WATSAN FINAL REPORT (BASELINE) USAID.GOV Task Commune Training to CTEs: Human Resource Management All communes except Croix-des￾Bouquets Training to CTEs: Financial Management Training to CTEs: Technical Operations Training to CTEs: Commercial Operations TASK 2.1: SUPPORT TO SANITATION ENTERPRISES Support artisans in developing business plans Support artisans in marketing sanitation improvements and the use of the catalog All communes Support artisans in conducting lessons learned workshop TASK 2.2: WASTE TREATMENT & FECAL SLUDGE MANAGEMENT ENGINEERING SERVICES MORNE-À-CABRIT FECAL SLUDGE MANAGEMENT SITE Croix-des-Bouquets Developing an onsite well to provide water for the lagoons Rebuilding the perimeter fence Improvements to the sludge intake Creating on-site disposal of the garbage and sludge Site clean-up Development of Operations and Business Plans FONFRED FECAL SLUDGE MANAGEMENT SITE Les Cayes Develop a business plan for the site Develop an environmental assessment Develop an operations plan TASK 2.3: WASTEWATER TREATMENT & FECAL SLUDGE MANAGEMENT CONSTRUCTION No expected work in Year 3 TASK 3.1: TECHNICAL ASSISTANCE TO NATIONAL AND SUB-NATIONAL GOVERNMENT STRUCTURES Roll out MIS system nationwide for CTEs All communes Establish a Maison de l’assainissement Les Cayes; Cap-Haitian Develop Sanitation Zoning Maps for the urban areas Les Cayes; Cap-Haitian Ongoing awareness raising of best sanitation practices All communes TASK 3.2: KNOWLEDGE DISSEMINATION AND LEARNING Urban sanitation: national workshop to disseminate desk review findings All communes Management of hybrid pumps in Haiti: production of design and operation manual Technical Research Paper for the University of North Carolina’s Water Conference CROSS-CUTTING TECHNICAL ACTIVITIES Living Water’s proposal to strengthen private water suppliers in Cap-Haitian Cap-Haitian Le Levier’s proposal to provide microcredit for toilet construction Unknown Sustainable Organic Integrated Livelihoods’ (SOIL) proposal to manufacture sanitation hardware in Haiti Saniclean’s proposal to manufacture interlocking blocks for toilet pit construction Konbit Monde’s proposal to build a biodigester in Cap-Haitian Cap-Haitian The University of Quisqueya’s proposal to develop a system to sanitize biosolids through plant beds Unknown Supporting Sanitation SMEs All communes Strengthening the Pit Emptiers’ Association Creating a Water Testing Lab USAID.GOV WATSAN BASELINE SURVEY REPORT | 90 7.4 ANNEX D. EVALUATION SCOPE OF WORK BASELINE SURVEY OF USAID/HAITI’S WATER AND SANITATION ACTIVITY 1. Introduction The Water and Sanitation Activity (WATSAN) is a four- and half-year intervention to support access to sustainable water supply and sanitation services, and strengthen the enabling environment for sustainable delivery, operation, and maintenance of water and sanitation services. The activity is targeting the urban areas of five communes: Cap Haitian, Mirebalais, Croix-des-Bouquets (Canaan), Les Cayes and Jeremie, and will result in 250,000 people gaining access to new or improved water services and 75,000 people gaining access to basic or safely managed sanitation services This baseline study and the follow-up mid-term and final performance evaluations are intended to measure whether WATSAN interventions in these communes have resulted in households and communities in the target communes having improved and sustainable access to water and sanitation services. 2. Summary Information Strategy/Project/Activity Name Water Sanitation Activity (WATSAN) Implementer Development Alternatives Inc Cooperative Agreement/Contract # AID-OAA-I-14-00049/720521 Total Estimated Ceiling of the Evaluated Project/Activity (TEC) Life of Strategy, Project, or Activity 12/30/2017 to 06/30/2022 Active Geographic Regions N/A Development Objective(s) (DOs) [Insert number and name of the DO that this evaluation relates to] USAID Office …Office of Infrastructure and Engineering (OIEE) 3. Background 3.1 DESCRIPTION OF THE PROBLEM Diarrheal disease is a leading cause of mortality in children under age five in Haiti, responsible for 12 percent of all child deaths. Research indicates that almost 60 percent of diarrhea is caused by unsafe water, lack of sanitation, and poor hygiene behaviors. Poor water, sanitation and hygiene (WASH) is also a key contributing factor in malnutrition, as repeated bouts of diarrhea, parasitic infections, and environmental enteropathy all lead to an inability to absorb and utilize nutrients. Moreover, lack of clean water and sanitation services contributed substantially to the severity and rapid spread of the cholera epidemic that began in Haiti in October 2010. Notwithstanding the fact that the average number of cholera cases has decreased since 2010, there have been recurring spikes in 2015 and 2016, largely in a dozen identified “hot spots.” Cholera is now persistent in Haiti, and resources must be invested in improving access to drinking water and sanitation services if cholera is to be eliminated on a sustainable basis. Access to water and sanitation in Haiti is still the lowest in the Western Hemisphere. Only 58 percent of the population has access to improved water sources, a decrease since 1990, when the access rate was 62 percent, as increases in services were not able to keep pace with population growth. However, actual 91 | WATSAN FINAL REPORT (BASELINE) USAID.GOV access rates may be lower, as many water systems’ functionality is poor. Only 28 percent of the Haitian population has access to improved sanitation, a rate that has been relatively constant since 1990. Safe collection, transport, and treatment of fecal waste are practically non-existent throughout Haiti. The government has focused on awareness and promotion campaigns to encourage households to build their own latrines and addressing the need for sanitation facilities in public schools, health institutions, and other public spaces. These campaigns to reduce open defecation have been somewhat successful; open defecation rates have dropped from almost 50 percent in 1990 to about 20 percent in 2015. However, surveys indicate that the use of improved sanitation facilities has only increased from 18 to 28 percent, meaning that much of the population still relies on shared or unimproved facilities. While sanitation access rates in urban areas are higher than the national average of 28 percent, they are still below 50 percent. Water - and especially sanitation infrastructure - often lags housing. In informal or unplanned settlements, where the poorest and most vulnerable urban populations generally live, urbanization and high localized population density may be accompanied by an increased risk of infectious disease transmission, primarily impacting the poor. The combination of a high fecal-related disease burden and inadequate infrastructure suggests that investment in expanding sanitation access in densely populated urban slums can yield important public health gains. 3.2 SUMMARY STRATEGY ACTIVITY TO BE EVALUATED 3.2.1 GEOGRAPHICAL FOCUS Access to improved water services and sanitation facilities is low throughout Haiti. However, USAID is unable to address this challenge nationwide. Therefore, priority geographies were selected based on the need and opportunity for sustained impact through coordination with other donors and USAID-funded programs. The WATSAN targeted communes are as follow: Cap-Haitian, Mirebalais, Croix-des-Bouquets (Canaan), Les Cayes and Jeremie. 3.2.2 COMPONENTS AND ACTIVITIES Component 1: Increasing Access to Sustainable Water Services Task 1.1: Water Infrastructure Engineering Services Deliverables under Task 1.1: Initial assessment report for all water systems to be supported, with prioritized list of proposed construction activities; EMMP for all construction activities; 100% design specifications for all construction activities; Signed firm fixed price contract(s) with construction firm(s). Task 1.2: Water Infrastructure Construction Deliverables under Task 1.2 Certificate of completion for all newly constructed infrastructure, signed by the QA firm; Water quality test results from newly constructed water systems meet JMP standards. Task 1.3: Technical Assistance for Water Service Providers USAID.GOV WATSAN BASELINE SURVEY REPORT | 92 Deliverables under Task 1.3: 5 service providers demonstrating at least 10% increase in cost recovery over a baseline established in Year 1; 5 service providers demonstrating improved management capacity, according to a Contractor proposed index. Component 2: Increasing Access to Sustainable Sanitation Services Task 2.1: Support to Sanitation Enterprises Deliverables under Task 2.1: • FSMs entrepreneurs and/or SMEs supported; • 15 SMEs demonstrating increased sale of latrine products. Task 2.2: Waste Treatment & Fecal Sludge Management Engineering Services Deliverables under Task 2.2: Initial assessment report for all proposed wastewater treatment systems to be constructed, with prioritized list of proposed construction activities; 611(e) analysis for all construction activities; EMMP or EA for all construction activities; 100% design specifications for all construction activities; Signed firm fixed price contract(s) with construction firm. Task 2.3: Wastewater Treatment & Fecal Sludge Management Construction Deliverables under Sub-task 2.3: Certificate of completion for all newly constructed infrastructure, signed by the QA firm; Wastewater and/or byproduct quality test results from newly constructed treatment systems meet JMP standards; 3 communes have waste treatment capacity. Component 3: Improving the Enabling Environment for Sustainable Implementation, Operation, and Maintenance of Water and Sanitation Services Task 3.1: Technical Assistance to National and Sub-National Governments Deliverables under Task 3.1: 5 national and/or sub-national institutions successfully received training; Strategies/plans for improved water and sanitation service delivery in place for all supported geographies. Task 3.2: Knowledge Dissemination & Learning Deliverables under Task 3.2 Three national or regional events on activity learning involving government, civil society, private sector and donor partners; 93 | WATSAN FINAL REPORT (BASELINE) USAID.GOV Five published knowledge products spanning all three activity components. 3.3 DEVELOPMENT HYPOTHESIS AND THEORY OF CHANGE The WATSAN development hypothesis is that reducing waterborne diseases in priority geographic areas is achievable if the activity spends resources on design, engineering and construction services, providing technical assistance to government water services providers and private enterprises. Effective implementation of WATSAN programming leading to reduced waterborne diseases can be achieved if the following conditions are met: a) Water services are provided to households; b) Sanitation services are provided in households and communities; c) Water and sanitation infrastructure are built d) Water and sanitation enterprises are supported The Haiti WATSAN Activity proposes an integrated program approach that combines development of water and sanitation infrastructure, builds the corresponding capacity to sustainably manage service delivery and improves the enabling environment for sustainable implementation, operation, and maintenance of water and sanitation services. A key principle applied to the Haiti WATSAN Activity is to transition to longer-term, comprehensive interventions that build more sustainable capacity at the local and national levels to achieve the goal of reducing waterborne diseases in priority geographic areas. 4. Baseline Survey Purpose and Evaluation Questions 4.1 BASELINE PURPOSE The purpose of this Baseline Study is to serve as a reference point to assess the impact of the WATSAN activities at mid-point and end-line of the WATSAN Activity. The baseline data survey will also guide the implementation of the WATSAN Activity as it will provide a benchmark that will be used to help improve problem diagnosis, alert stakeholders to changes in underlying assumptions, uncover potential weaknesses in program implementation, and provide a firm foundation for monitoring and evaluation. The primary stakeholders for this evaluation include USAID/Haiti, WATSAN implementing partners including DAI and its partners Ayiti Nexus (communications/BCC), Centre de Formation et d’Encadrement Technique (CFET) (capacity development), Zanmi Lasante (health and advocacy), V3 (engineering) and mWater (digital monitoring, evaluation and reporting), the Ministry of Publics Works, Transportation and Communication (MTPTC) and the Water and Sanitation National Direction (DINEPA). Other key stakeholders include other donors and entities active in Haiti’s WASH sector (IDB, USAID.GOV WATSAN BASELINE SURVEY REPORT | 94 AECID, World Bank), and a variety of NGOs and religious organizations devoted to improving learning in Haitian primary schools. 4.2 EVALUATION QUESTIONS AND ILLUSTRATIVE INDICATORS The baseline will set the stage for follow-up mid-term and final performance evaluations which will allow measurement of progress toward better access on improved water and sanitation. The baseline data will be a reference point for future evaluations intended to help answer the three following questions: To what extent and in what ways has WATSAN increased access to new or improved basic drinking water and sanitation facilities? This question should focus on the effectiveness of the original activity design and implementation. To what extent has WATSAN contributed to improved institutional capacity to better manage existing systems and increase revenue? This question should focus on to what extent and in what ways relevant institutions are applying WATSAN training. The answer to this question will focus on the specific institutions, systems, and training on which WATSAN is working. To what extent does WATSAN Theory of Change (TOC) remain valid and relevant to the activity’s current objectives? The answer to this question will examine the assumptions underlying the TOC to see if they remain valid and will also assess whether the TOC remains relevant given the activity’s current goals and objectives. Illustrative indicators that will guide the design of the baseline study include: • Percent of people using improved water services (basic). • Percent of people using safely managed drinking water services. o Population using an improved drinking water source (piped water into dwellings, yards or plots; public taps or standpipes; boreholes or tube wells; protected dug wells; protected springs; rainwater; packaged or delivered water) that is located on the premises and available when needed, and free of faecal and priority chemical contamination. Drinking water from an improved source that does not fulfil the above-mentioned criteria is categorized as “basic” services, provided that the collection time is not more than a 30-minute round trip, including queuing. If the improved drinking water source is located further away, the service is categorized as “limited”. • Percent of people using an improved sanitation facility (basic). • Percent of people using safely managed sanitation services. o The population using safely managed sanitation services is defined as the population using an improved sanitation facility at the household level that is not shared with other households, and where excreta are treated and disposed of in situ or transported and treated off-site. Improved sanitation facilities include flush or pour-flush toilets to sewerage systems, septic tanks or pit latrines, improved pit latrines (pit latrines with a slab or ventilated pit latrines) and composting toilets. Improved sanitation facilities that do not fulfil the above-mentioned criteria for treatment are categorized as “basic” services. If the facility is shared with other households the service is categorized as “limited”. 5. Evaluation Design and Methodology The WATSAN performance evaluation will determine the effectiveness of the WATSAN program on the water and sanitation services outcomes and shall measure changes in those outcomes before and after 95 | WATSAN FINAL REPORT (BASELINE) USAID.GOV the interventions. ESS is encouraged to propose a pre-post evaluation design to estimate the changes observed in WATSAN interventions over the activity life. ESS is also expected to complete three (baseline mid-line and endline) rounds of data collection. While it is likely that the baseline will be implemented on a rolling basis as WATSAN identifies the specific locations in which it will work, options for data collection methodology will depend on the activity implementation strategy. The activity must provide information about implementation to elaborate the evaluation design document. It is expected that the baseline survey will produce baseline information on the highest result framework indicators and will set the stage to eventually test some of the key hypotheses behind the theory of change of the WATSAN Activity. 6. Deliverables and Reporting Requirements (Baseline) Work plan: Upon receipt of this SOW, ESS shall submit, within twenty (20) business days, a draft work plan to the Contracting Officer’s Representative (COR). The work plan will include: (1) the anticipated schedule and logistical arrangements; (2) a list of the members of the evaluation team, delineated by roles and responsibilities with their level of effort; (3) the identification of other required personnel and relevant local subcontractors, their LOE, roles and responsibilities and qualifications; and (4) the deliverable schedule. Inception Report: Within three weeks of approval of all evaluation team members, ESS must submit to Contracting Officer’s Representative (COR) an Inception Report, which will document what the team learned from its review of activity background documents and other relevant literature. The team will then draw upon what it has learned generally to identify information that will be useful in answering the evaluation questions. Finally, the team will then identify the information gaps that it will have to fill through primary data collection. This information will inform the evaluation design. Baseline Design: Within two weeks of submitting the Inception Report, ESS must submit to Contracting Officer’s Representative (COR) an Evaluation Design (which will become an annex to the evaluation report). The design will include: (1) a brief summary of key findings from the document review; (2) a brief summary of the gaps to be filled through the primary data collection; (3) a detailed baseline design, including key stakeholder groups to be participants, sampling approach, data collection and analysis methods, and a matrix linking each evaluation question in the SOW to the data sources, methods, and data analysis plan that will facilitate completely addressing the questions; (4) draft questionnaires and other data collection instruments or their main features; (4) the list of potential interviewees and sites to be visited; (5) known limitations to the design; and (6) a dissemination plan. USAID/Haiti will take up to 10 business days to review and consolidate comments through the COR. Once the evaluation team receives the consolidated comments on the Inception Report and initial evaluation design, they are expected to return a revised Inception Report and Evaluation Design within 5 business days. In-briefing/Presentation: Within two weeks of Baseline Design’s approval, the baseline team is expected to hold an In-Briefing Presentation, in person, to discuss their understanding of the design approach and proposed methodology. This is an opportunity to clarify assumptions and make sure that data collection tools will capture the information needed to answer evaluation questions. Out-briefing/Presentation: Within 20 business days after the fieldwork concludes, the baseline team is expected to hold two final Out-Briefing Presentations, in person to USAID, to discuss the evaluation findings, conclusions and recommendations. The first Out-Briefing will include the Program Office, USAID Technical Team and WATSAN Staff, while the second will be open to the entire Mission. USAID.GOV WATSAN BASELINE SURVEY REPORT | 96 Draft Report: The draft report should be consistent with the guidance provided in Section IX: Final Report Format. The report will address each of the questions identified in the SOW and any other issues the team considers bearing on the evaluation objectives. Any such issues can be included in the report only after consultation with USAID. The draft report will be submitted fifteen (15) business days after the second Out-Brief Presentation. Once the draft evaluation report is submitted, USAID will have 10 business days in which to review and comment on the draft, after which the ESS COR will submit the consolidated comments to the baseline team. The team will then be asked to submit a revised final report 10 business days hence, and again USAID will review and send comments within 10 business days of its submission. Final Report: The baseline team will be asked to take no more than 10 business days to respond to/incorporate USAID’s final comments. ESS will then submit the final report to its COR. All activity data and records shall be submitted in full and should be in electronic form in easily readable format, organized, and documented for use by those not fully familiar with the intervention or evaluation, and owned by USAID. 7. Team Composition The Baseline Team shall include a Team Leader (TL) and a Sector Specialist/Assistant Team Leader (ATL). The two positions are considered Key Personnel and essential to the work being performed. The Team Leader (TL) is ultimately responsible for the overall management of the evaluation team and final products, in conformity with this Scope of Work. The TL must be an experienced evaluation expert, with a documented track record of 10 years’ experience in evaluation. The TL shall also have demonstrated experience in evaluating WASH activities. The TL shall have a master’s degree in social science (population and demography), public health or other related WASH field, and will possess excellent writing and interpersonal skills. The Team Leader will be responsible for planning the evaluation, coordinating the evaluation implementation, assigning responsibilities and tasks, and authoring the final evaluation report. S/he must be fluent in English and French. Work experience with USAID will also be an asset. The Sector Specialist/ATL shall have a master degree in development rural and experience in coordinating large scale survey to ensure that all expertise areas required for the baseline study are effectively covered. S/he shall be an expert in WASH data collection and analysis, with at least 7 years’ experience in conducting evaluation and baseline studies. Work experience with USAID will be an asset and fluency in English and French is required. Haitian Creole is highly desirable. ESS is strongly encouraged to sub-partner with one local Haitian firm for data collection purposes. The selected Haitian firm should demonstrate proven capacity in collecting WASH data (quantitative data collection) and should have data analysis capabilities to perform all related data management functions. This approach is encouraged to build the local firm capacity and will also provide a Haitian perspective for the data collection and analysis. 8. Evaluation Schedule Task Estimated Due Date ESS receives revised SOW January 23, 2019 Revised Workplan January 31, 2019 ESS identifies local firm January 31, 2019 Workplan and full team approved February 8, 2019 97 | WATSAN FINAL REPORT (BASELINE) USAID.GOV Task Estimated Due Date Inception Report February 15, 2019 Evaluation Design February 22, 2019 Inception Report and Evaluation Design approved February 28, 2019 In Briefing January 28, 2018 Data collection and analysis March 2019 Out-Briefing Presentation TBD Baseline Draft Report submitted to USAID TBD USAID Feedback on Baseline Draft Report TBD Baseline Final Report submitted to USAID TBD Endline Data Collection and Analysis June – August 2021 9. Final Report Format The baseline final report should include an abstract of no more than 250 words; executive summary of no more than 5 pages; background on the local context and strategies/projects/activities being evaluated; the baseline purpose and evaluation questions; the methodology or methodologies; study limitations; findings, conclusions, and recommendations. For more detail, see “How-To Note: Preparing Evaluation Reports” and ADS 201mah, USAID Evaluation Report Requirements. An optional evaluation report template is available in the Evaluation Toolkit. The executive summary should be 2–5 pages in length and summarize the purpose, background of the activity being evaluated, main evaluation questions, methods, findings, conclusions, and recommendations and lessons learned (if applicable). The evaluation methodology shall be explained in detail in the report. Limitations to the evaluation shall be disclosed in the report, with attention to the limitations associated with the evaluation methodology (e.g., selection bias, recall bias, unobservable differences between comparator groups, etc.) The annexes to the report shall include: • The Evaluation SOW; • Any statements of difference regarding significant unresolved differences of opinion by funders, implementers, and/or members of the evaluation team; • All data collection and analysis tools used in conducting the evaluation, such as questionnaires, checklists, and discussion guides; • All sources of information properly identified and listed; and • Signed disclosure of conflict of interest forms for all evaluation team members, either attesting to a lack of conflicts of interest or describing existing conflicts of. • Any “statements of difference” regarding significant unresolved differences of opinion by funders, implementers, and/or members of the evaluation team. • Summary information about evaluation team members, including qualifications, experience, and role on the team. In accordance with ADS 201, the contractor will make the final evaluation reports publicly available through the Development Experience Clearinghouse within three months of the evaluation’s conclusion. 10. Other Requirements USAID.GOV WATSAN BASELINE SURVEY REPORT | 98 All quantitative data collected by the evaluation team must be provided in machine-readable, non￾proprietary formats as required by USAID’s Open Data policy (see ADS 579). The data should be organized and fully documented for use by those not fully familiar with the activity or the evaluation. USAID will retain ownership of the survey and all datasets developed. All modifications to the required elements of the SOW of the contract/agreement, whether in technical requirements, evaluation questions, team composition, methodology, or timeline, need to be agreed upon in writing by the COR. Any revisions should be updated in the SOW that is included as an annex to the Baseline Report. 99 | WATSAN FINAL REPORT (BASELINE) USAID.GOV 7.5 ANNEX E. COMPARISON OF JMP VS. PIRS DEFINITIONS As described in section 3.5.1., the different definitions used by JMP and PIRS lead to disagreement in the interpretation of baseline data, in that JMP allows for a greater number of sources to be categorized as improved relative to the PIRS definition. Moreover, the widespread use of private kiosks most heavily influences these differences, since the JMP and PIRS classify them differently – while the PIRS does not explicitly list private kiosks as an unimproved source, it includes language that “all other sources” other than those listed are considered unimproved, and this classification was confirmed via further consultations between our team, USAID Program and Technical offices, and the Water and Sanitation project team. Applying the PIRS definition and comparing with the JMP, we see that they agree in about 23% of cases – about 17% of cases where the household’s drinking water service is at least Basic, and nearly 6% of cases where the household’s drinking water service is less than basic. There is disagreement in about 77% of cases (i.e. 1200 households) where the JMP would classify a household’s drinking water service as “at least basic”, whereas PIRS would classify it as “less than basic”. As described earlier, this disagreement is driven heavily by the decision of whether to classify private kiosks as improved or not – among those 1200 cases of disagreement, nearly 77% of them are accounted for by private kiosks. Another approximately 10% are accounted for by the different classification of “cart with small tank/drum”. JMP vs. PIRS: Frequency and cell percentage (unweighted) Water | service | Water service (PIRS) (JMP) | Basic+ 30 min RT 1% 0% 1% 1% 1% 1% Unimproved source, <=30 min RT 80% 95% 85% 54% 76% 80% Unimproved source, >30 min RT 6% 4% 4% 4% 4% 5% Total 100% 100% 100% 100% 100% 100% Post-stratification weights applied. Analytical sample size is 301 for all communes. Compares with Table 6 in body of the report. By head of household sex and by socioeconomic status. Just as with the JMP definition, there are no statistically significant differences in basic water service for female- versus male-headed households in the WATSAN areas of the five communes. Likewise, using the PIRS definition of basic water service, there are no statistically significant differences in access based on socioeconomic status across the five communes, except in the case of Mirebalais, though the results are the opposite of what may be expected, with those in the highest socioeconomic quintile supposedly reporting substantially lower access (9% at least basic) 13% 1% 10% 41% 19% 15% 93% 96% 94% 94% 95% 94% 0% 20% 40% 60% 80% 100% Cap-Haitian Canaan Jeremie Les Cayes Mirebalais Total PIRS definition JMP definition 101 | WATSAN FINAL REPORT (BASELINE) USAID.GOV to improved sources compared with those in the lowest socioeconomic quintile (35% at least basic) – looking specifically at the sources of drinking water reportedly used by those in Mirebalais at different levels of socioeconomic status (see Table 38 below), this appears to be driven by households of lower SES being more likely to report getting piped water from a neighbor, which is classified as improved, in combination with those of higher SES being more likely to report using water from private kiosks and carts, which by PIRS definition are both considered unimproved. Table 37. Water sources used in Mirebalais, by SES Mirebalais, SES Quintile Water Source (%) 1 (lowest) 2 3 4 5 (highest) Piped into dwelling 1.6 1.6 1.6 4.8 3.2 Piped into plot 1.6 0 4.8 0 3.2 Piped to neighbor 22.6 8.1 11.3 1.6 0 Public tap/standpipe 0 9.7 1.6 0 0 Borehole or tubewell 6.5 1.6 3.2 1.6 1.6 Protected well 1.6 1.6 0 0 0 Protected spring 1.6 0 0 0 0 Unprotected spring 0 0 1.6 0 0 Cart with small tank/drum 9.7 16.1 12.9 14.5 12.9 Public Water Kiosk 1.6 0 0 1.6 0 Private Water Kiosk 48.4 59.7 62.9 74.2 77.4 Sachet water 3.2 1.6 0 1.6 1.6 Surface water 1.6 0 0 0 0 Total 100 100 100 100 100 Post-stratification weights applied. Analytical sample size is 301 in the commune. A full reproduction of all the tables and disaggregations in this report by the PIRS definition has not been undertaken for the purpose of this annex but is available upon request.