WATER AND SANITATION PROGRAM: WORKING PAPER 90044 Scaling Up Rural Sanitation What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Kathryn O’Connell August 2014 The Water and Sanitation Program is a multi-donor partnership administered by the World Bank to support poor people in obtaining affordable, safe, and sustainable access to water and sanitation services. The author would like to thank Jacqueline Devine for her guidance and inputs. Thanks also to Emily Christensen Rand, Susanna Smets, Jane Bevan, Eduardo Perez, and James Dumpert for their thoughtful comments and review. WSP’s Scaling Up Rural Sanitation is working with governments and the local private sector to develop the knowledge needed to scale up rural sanitation for the poor. The programmatic approach combines Community-Led Total Sanitation (CLTS), behavior change communication, and sanitation marketing to generate sanitation demand and build up the supply of sanitation products and services at scale. In addition, WSP works with local and national governments and the local private sector to strengthen the enabling environment—including institutional, regulatory, financial, service-delivery, and monitoring capacities—to achieve change that is sustainable. Starting in India, Indonesia, and Tanzania in 2006, Scaling Up Rural Sanitation is currently being implemented in more than a dozen countries. For more information, please visit www.wsp.org/scalingupsanitation. This Working Paper is one in a series of knowledge products designed to showcase project findings, assessments, and lessons learned through WSP’s Scaling Up Rural Sanitation initiatives. This paper is conceived as a work in progress to encourage the exchange of ideas about development issues. For more information please email Kathryn O’Connell at worldbankwater@worldbank.org or visit www.wsp.org. WSP is a multi-donor partnership created in 1978 and administered by the World Bank to support poor people in obtaining affordable, safe, and sustainable access to water and sanitation services. WSP’s donors include Australia, Austria, Denmark, Finland, France, the Bill & Melinda Gates Foundation, Luxembourg, Netherlands, Norway, Sweden, Switzerland, United Kingdom, United States, and the World Bank. WSP reports are published to communicate the results of WSP’s work to the development community. Some sources cited may be informal documents that are not readily available. The findings, interpretations, and conclusions expressed herein are entirely those of the author and should not be attributed to the World Bank or its affiliated organizations, or to members of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to worldbankwater@worldbank.org. WSP encourages the dissemination of its work and will normally grant permission promptly. For more information, please visit www.wsp.org. © 2014 International Bank for Reconstruction and Development/The World Bank 1818 H Street NW Washington, DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org Scaling Up Rural Sanitation What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Kathryn O’Connell August 2014 Executive Summary As part of its Scaling Up Rural Sanitation and Domestic cleanliness and durability); social norms around open defe- Private Sector Participation programs, the Water and Sani- cation; perceived latrine affordability; self-efficacy to build tation Program (WSP) of the World Bank has been com- latrines (respondent self-efficacy versus reliance on masons); missioning formative research studies among households. and competing priorities for other household expenditures. These studies have informed the development of behavior The review also identified a number of emotional, social, change communication (BCC) and other demand-creation and physical drivers. These include shame and embarrass- strategies and tools. WSP has used a conceptual framework, ment associated with open defecation, as well as perceptions called SaniFOAM (Sanitation Focus, Opportunity, Ability, of improved social status, privacy, and convenience associ- Motivation), to help program managers and implementers ated with latrine ownership and use. A number of back- analyze sanitation behaviors to inform effective sanitation ground characteristics influence sanitation behaviors. These programs (Devine 2009; PSI 2004). The SaniFOAM frame- include socioeconomic status, as well as contextual factors work has also been used to design formative surveys to un- that vary by region or country, such as perceptions of physi- derstand barriers and drivers of improved sanitation and cal and geographical conditions (e.g., access to water and monitor progress of the effectiveness of its behavior change soil profile), seasonal factors, and the time of year. program (Devine 2009). Figure 1 highlights the key factors that were found to influ- Since 2006, qualitative and quantitative market research ence rural sanitation behaviors based on the SaniFOAM studies have been carried out in multiple countries. To date, conceptual framework. It also includes an additional new no systematic comparison or summary of these studies has factor in the focus section of the framework (sociodemo- been conducted. As such, a desk review of existing WSP graphic and background characteristics), to demonstrate formative research studies was undertaken. The purpose of the relevance of a number of background and contextual the review was to identify commonalities and differences factors that were deemed relevant to rural sanitation across countries, and to determine factors that affect sanita- behaviors. tion behaviors, positively or negatively. Three specific sani- tation behaviors are covered in the review: open defecation, The review identified several other factors, including knowl- acquisition of toilets, and improvement of latrines. edge, enforcement of rules or regulations, values, intention to build latrines, roles and decision-making, and beliefs and Methodology attitudes. However, it is less certain how these aspects influ- This review collects the results from formative quantitative ence sanitation behaviors, thus limiting the strength of the and qualitative research reports and presentations from conclusions that can be made. This may be due to the dif- eight countries: Cambodia, India (Rajasthan, Meghalaya, ferent research objectives and interview guides of the stud- and Bihar), Indonesia (East Java), Kenya, Malawi, Peru, ies, quality of the data and reporting, regional differences, Tanzania, and Uganda. Studies were implemented from and/or their relevance to sanitation behaviors. 2006 until 2012. Most of the studies were conducted only in rural areas, with a few exceptions. Secondary data were Implications used for the review and analysis followed standard qualita- Given that this review identifies a number of factors that tive methodologies of thematic ordering and interpretation resonate with sanitation behaviors, there are opportunities to identify factors that could positively or negatively influ- to conduct “lighter” and more tailored formative research. ence the behaviors of interest. Measurement of the key determinants identified in this re- view will serve to monitor program impact and allow Key Findings for investigation of the barriers that are known to have The most salient factors influencing rural sanitation behav- the greatest influence on sanitation behaviors. In addition, iors that emerged from the review include access to and there may be opportunities to use more specific or less bur- availability of functioning latrines, sanitation products, and densome research methods, such as street intercept surveys services; latrine product attributes (e.g., perceptions of or supply-side assessments as a means to investigate pricing, iv Scaling Up Rural Sanitation What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Executive Summary affordability, and access barriers. Supply-side surveys, which the extent of this disparity through a meta-analysis of are already undertaken in most countries, would allow for existing datasets. Understanding the specific barriers an investigation into actual versus perceived affordability and drivers to improved sanitation among the rural and accessibility barriers. poorest will help improve programmers’ ability to de- sign effective behavior change interventions, particu- In countries or regions where new formative research stud- larly as the sector moves toward more equity-focused ies are planned, using standardized questions to ensure goals in the post-Millennium Development Goals greater comparability between studies and target groups (MDG) setting. will be an important methodological improvement. There are also opportunities for more specific behavioral questions Practical implications from the review include changing so- to delve deeper into self-reported latrine use, which will cial norms toward positive sanitation behaviors (i.e., “every- allow for further investigation into the barriers of latrine one uses a latrine”) and promoting awareness of actual usage among those who own latrines. For additional guid- latrine costs, coupled with messaging that underlines posi- ance, refer to the “Study Design and Questionnaire Tips” tive product attributes (improved latrines are safe, durable, document, available online in WSP’s Sanitation Marketing and hygienic). To ensure that messaging resonates with the Toolkit: http://wsp.org/toolkit/toolkit-home. target audience, communication campaigns could promote a number of positive emotional, social, and physical drivers, Given differences in coverage between wealthier and such as improved social status and pride associated with poorer households, there are opportunities to explore owning a latrine. FIGURE 1: KEY FACTORS FOUND TO INFLUENCE BEHAVIORS ACCORDING TO THE SANIFOAM CONCEPTUAL FRAMEWORK Focus Opportunity Ability Motivation Target population Access/availability Knowledge Attitudes and beliefs Skills and Desired behavior Product attributes Values self-efficacy Emotional/physical/ Sociodemographic and Social norms Social support social drivers background characteristics Sanctions/ Roles and decisions Competing priorities enforcement Affordability Intention Willingness to pay Key: ■ The most important SaniFOAM factors found to influence behavior ■ A new factor to emerge from the review www.wsp.org v Contents Executive Summary...................................................................iv Methodology ............................................................................iv Key Findings ............................................................................iv Implications ..............................................................................iv I. Introduction ............................................................................... 1 1.1 Methodology ..................................................................... 2 1.2 Current Rates of Latrine Ownership and Defecation Behavior ................................................... 3 II. Key Findings .............................................................................. 4 III. Opportunity ................................................................................ 5 3.1 Access and Availability ...................................................... 5 3.2 Product Attributes ............................................................. 6 3.3 Social Norms .................................................................... 7 IV. Ability ......................................................................................... 9 4.1 Skills and Self-efficacy ....................................................... 9 4.2 Affordability ....................................................................... 9 V. Motivation ................................................................................ 12 5.1 Emotional, Social, and Physical Drivers ........................... 12 5.2 Competing Priorities ........................................................ 13 VI. Socioeconomics Status and Other Contextual Factors ........ 16 6.1 Socioeconomics Status................................................... 16 6.2 Contextual Factors .......................................................... 16 VII. Other Findings ......................................................................... 19 VIII. Summary and Implications ..................................................... 23 References ............................................................................... 26 Appendix .................................................................................. 28 Figures 1: Key Factors Found to Influence Behaviors According to the SaniFOAM Conceptual Framework ..................................... v 2: SaniFOAM Conceptual Framework ..................................... 2 3: SaniFOAM Opportunity Factors Found to Influence Rural Sanitation Behaviors .................................................. 6 4: SaniFOAM Ability Factors Found to Influence Rural Sanitation Behaviors ............................................................ 9 5: Use of Masons to Build Unimproved or Improved Latrines, by Study (%) ....................................................... 10 vi Scaling Up Rural Sanitation What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Contents 6: Open Defecators Citing Cost or Affordability as a Key Barrier to Building Latrines or Making Improvements (%).............................................................. 10 7: SaniFOAM Motivational Factors Found to Influence Rural Sanitation Behaviors ................................................ 12 8: Common Uses of Extra Money in Bihar and Rajasthan (%) ............................................................. 14 9: Respondents Citing Health or Hygiene as a Reason to Build Latrines, by Study (%) .......................................... 19 10: Latrine-Owning Households Citing Convenience as a Main Reason to Build a Latrine (%) ........................... 21 11: Respondents Citing Satisfaction with Current Defecation Options, by Study (%)........................................................ 22 Tables 1: Estimates of Use of Sanitation Facilities in Rural Areas in 2011 as Reported by the Who/UNICEF Joint Monitoring Program (%) ......................................................................... 3 A: Details of the Formative Research Studies ....................... 28 Boxes 1: SaniFOAM Framework......................................................... 1 2: Study Limitations ................................................................. 3 3: Project Attributes: What Do Households Consider an Ideal Latrine? ................................................................................ 7 4: Do Households Have Accurate Price Perceptions of Latrines? ............................................................................ 11 5: The Role of Important Family Events as a Means to Raise Latrine Priority...................................................... 15 6: Examples of Contextual Factors Influencing Latrine Ownership and Open Defecation .......................... 17 7: The Influence of Sanctions and Enforcement, Beliefs and Attitudes, and Values on Rural Sanitation Behaviors......... 20 8: Regional Variations Regarding the Relevance of Different SaniFOAM Factors............................................................. 21 9: Examples of Practical Implications from the Review ......... 25 www.wsp.org vii I. Introduction As part of its Scaling Up Rural Sanitation and Domestic understand barriers and drivers of improved sanitation and Private Sector Participation programs, the Water and Sani- monitor progress of the effectiveness of its behavior change tation Program (WSP) of the World Bank has been com- program (Devine 2009). Study findings inform the design missioning formative research studies among households. of BCC, messaging, and sanitation marketing to stimulate These studies have informed the development of behavior improved sanitation behaviors. change communication (BCC) and other demand-creation strategies and tools. Since 2006, qualitative and quantitative market research studies have been carried out in multiple countries, includ- WSP has utilized a conceptual framework, called Sani- ing Cambodia, India, Indonesia, Kenya, Malawi, Peru, FOAM, to help program managers and implementers ana- Tanzania, and Uganda. To date, no systematic comparison lyze sanitation behaviors to inform effective sanitation or summary of these studies has been conducted. As such, a programs (Box 1). The SaniFOAM framework (Figure 2) desk review of existing WSP formative research studies was has also been used to design the formative surveys to conducted. The purpose of the review was to identify BOX 1: SANIFOAM FRAMEWORK The SaniFOAM framework uses a set of behavioral behavior, and includes health knowledge, awareness determinants organized under the domains of oppor- of different latrine types and price, perceived social tunity, ability, and motivation—factors with founda- support for latrine acquisition and use, household tions in the disciplines of consumer behavior, public roles and decisions regarding major expenditures, health, health psychology marketing, advertising, and and perceived affordability of latrines. Motivation is economics (Hallahan 2000; MacInnis et al. 1991; Moor- the individual’s desire to perform the promoted be- man and Matulich 1993; Rothschild 1999; Wiggins havior. Motivating factors include beliefs, attitudes, 2004) and expands on a behavior change framework and values surrounding health behaviors, as well as utilized by Population Services International (PSI 2004). emotional, physical, and social drivers; competing The “focus” concept of the framework also allows for priorities within the household; intention; and willing- determining what behavior should be improved and ness to pay. whose behavior needs to be changed. Examples of the drivers or inhibitors explored in the framework include Opportunity, ability, and motivation factors are con- beliefs about health and hygiene, access to suppliers, sidered to have the potential to influence sanitation perceived affordability of latrines, and awareness of behaviors, including use and acquisition of latrines. In sanitation options. practice, the framework is flexible; program managers can test and identify the unique set of factors that best Opportunity encompasses institutional or structural explain sanitation behaviors (such as latrine ownership factors that influence an individual’s chance to per- or use). The framework can be utilized across differ- form a behavior. Opportunity constructs include per- ent countries and with target groups (e.g., populations ceived availability of products and services, product living in rural versus urban areas). It is theorized that attributes, social norms regarding health behavior, targeting key factors through behavior change com- and sanctions and enforcement. Ability refers to in- munication (BCC) and other marketing activities will dividual skills or proficiencies needed to perform a facilitate behavior change. www.wsp.org 1 What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Introduction FIGURE 2: SANIFOAM CONCEPTUAL FRAMEWORK Focus Opportunity Ability Motivation Target population Access/availability Knowledge Attitudes and beliefs Skills and Desired behavior Product attributes Values self-efficacy Emotional/physical/ Social norms Social support social drivers Sanctions/ Roles and decisions Competing priorities enforcement Affordability Intention Willingness to pay commonalities and differences across studies, and to deter- interviews using questionnaires. The study population varied mine factors that affect sanitation behaviors, positively or according to the research method and study objectives, but negatively. Three specific sanitation behaviors are covered in generally included open defecators, latrine owners, and own- the review: open defecation, acquisition of toilets, and im- ers of improved latrines. In the quantitative surveys, most of provement of latrines. the respondents interviewed included male household heads. In the qualitative studies, the inclusion of female respon- 1.1 Methodology dents was more common. Questionnaire guidelines also var- This review collects the results from formative quantitative ied by study and were not standardized to allow for direct and qualitative research reports and presentations from cross-regional comparisons. In some instances, interviews eight countries: Cambodia, India (Rajasthan, Meghalaya, with suppliers were conducted, but this information is ex- and Bihar), Indonesia (East Java), Kenya, Malawi, Peru, cluded from the review, as a separate desktop review of sup- Tanzania, and Uganda. Studies were conducted from 2006 ply-chain studies is planned. Results were sometimes reported until 2012. Most of the studies were conducted only in according to the SaniFOAM framework, although in some rural areas, with a few exceptions where data were also col- countries (Cambodia, Malawi, Peru, Tanzania, and Uganda) lected from semiurban areas (Kenya, Peru, and East Java) the framework was not used as a means to structure and urban areas (Cambodia). Table A in the appendix lists the reports. the details of these studies. The review followed standard qualitative methodologies of The studies reviewed used a variety of methods, including thematic ordering and interpretation to identify factors that focus group discussions, in-depth interviews, and structured could positively or negatively influence the behaviors of 2 Scaling Up Rural Sanitation What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Introduction interest. Secondary data rather than primary data were used ings were routinely presented in the reports. These for the review. To conduct the analysis, a deductive list of limitations are summarized in Box 2. codes based on the SaniFOAM framework was used as a general guide for the analytic approach. Each report was 1.2 Current Rates of Latrine Ownership first read to investigate content and findings. Reports were and Defecation Behavior then coded using the broad SaniFOAM codes. After the To contextualize the findings of this review, defecation preliminary coding, themes were reviewed again and ar- practices and latrine ownership in 2011 are presented using ranged into a smaller set of themes to capture the emer- recent Joint Monitoring Program (JMP) prevalence figures gence of any subthemes. as shown in Table 1 (WHO/UNICEF 2013). The frequency with which the themes were mentioned in each report was noted in the analysis plan. This procedure TABLE 1: ESTIMATES OF USE OF SANITATION FACILITIES IN RURAL AREAS IN 2011 AS REPORTED BY THE WHO/UNICEF helped clarify which themes emerged consistently across all JOINT MONITORING PROGRAM (%) countries/regions for the different behaviors and which Improved Unimproved were idiosyncratic and specific to a country/region or re- Latrine Shared Latrine Open port. Comparisons were made across the different types of Ownership Latrine Ownership Defecation behaviors. Findings were verified as far as possible with tab- Malawi 53 31 9 7 ulated data in the reports or quotes in the case of qualitative Uganda 35 16 39 10 research. Tanzania 7 4 73 16 Kenya 29 19 35 17 Despite a number of commonalities that emerged from the Indonesia 44 11 10 35 analysis, there were some notable challenges given the vari- India 24 4 6 66 ability in study objectives and the extent to which all find- Cambodia 22 5 4 69 BOX 2: STUDY LIMITATIONS • Different survey methodologies and study instruments were used in the formative research studies, mak- ing comparison between countries more challenging. • Some research topics were explored in detail in some studies and not covered in others. • The definition of ownership of different latrine types varied by study. For example, in East Java, respon- dents were categorized as improved or unimproved latrine owners, sharers, or open defecators; whereas in Cambodia, respondents were classified by latrine ownership versus no ownership. • The review is based on summary reports, not the original data. Hence, findings have been filtered and interpreted by the authors of the reports. In particular, qualitative research has been translated, which may have led to lost insight or bias. • Tables of frequencies or means were commonly presented in reports, but statistical comparisons between target groups were rarer. • Although many of the surveys used Likert scales to address attitudes toward different concepts, the data were not analyzed in a consistent manner, making interpretations more challenging. • Not all of the studies used the SaniFOAM framework as a means to structure the survey questionnaires or present findings. Some of the reports retrofitted results to the framework, given that the studies predated the SaniFOAM framework. www.wsp.org 3 II. Key Findings Findings presented in this report illustrate themes that were of improved social status, privacy, and convenience associ- consistently found across regions/countries. The most sa- ated with latrine ownership and use. A number of back- lient factors influencing rural sanitation behaviors that ground characteristics influence sanitation behaviors. These emerged from the review include access to and availability include socioeconomic status, as well as factors that vary by of functioning latrines, sanitation products, and services; region or country, such as temporal and seasonal factors latrine product attributes (e.g., perceptions of cleanliness (such as the time of year), land ownership and access, and and durability); social norms around open defecation; per- perceptions of physical and geographical conditions (e.g., ceptions of latrine affordability; self-efficacy to build la- access to water and soil profile). The following sections dis- trines; and competing priorities for other household items. cuss these findings in detail according to the SaniFOAM The review also identified a number of emotional, social, opportunity, ability, and motivation factors, as well as so- and physical drivers. These include shame and embarrass- ciodemographic and other background characteristics. ment associated with open defecation, as well as perceptions 4 Scaling Up Rural Sanitation III. Opportunity Opportunity factors include institutional or structural as- 3.1.2 Functioning Latrines pects that influence whether an individual has the chance to Access to a functioning latrine is an important issue to ad- engage in the desired behavior. A summary of the key op- dress when unpacking reasons for open defecation. Defin- portunity factors that were found to influence behaviors is ing a household (or individual) as owning a latrine does not presented in Figure 3. ensure access to a working latrine. Latrines are commonly described as being full, overflowing, in need of repair, or 3.1 Access and Availability infested with maggots, posing a barrier to use. Thus, “own- Access and availability is the extent to which the promoted ing” a latrine is not necessarily a precursor to using the la- product or service can be found, or is perceived to be avail- trine. Example findings include: able, by target groups (e.g., Conteh and Hanson 2003). Ac- • In Tanzania, 20 percent of latrine owners stated that cess to, and availability of, latrines, products, and services there was a period in the last year when their latrine (such as masons to install latrines or shops selling sanitary was not usable. hardware) may influence whether or not latrines are pur- • Observations in Bihar show that 11 percent of la- chased or upgrades are made. This review focused on per- trines were not functioning on the day of survey. ceived availability as a potential driver for positive sanitation behaviors. Thus, assurances are needed to determine latrine functionality. Different dimensions of availability and access are found to be relevant among latrine owners and open defecators. 3.1.3 Perceived Supply-Side Access: Availability of Latrine Materials, Hardware, Suppliers, and Masons 3.1.1 Access to Latrines Knowing a supplier who stocks a variety of sanitation hard- Variance in latrine ownership by country or region is im- ware and a mason to assist with latrine construction are portant to note when exploring reasons for open defeca- central to upgrading and improving latrines. Often these tion. According to the JMP findings, latrine ownership (of factors serve as a barrier to moving up the sanitation ladder, any type) ranges from as little as 31 percent of households given that materials for improved latrines are perceived as in Cambodia to 93 percent in Malawi. If an individual does unavailable and costly. For example, in Bihar, two thirds of not have access to a latrine at work, or in the homestead, households that own latrines report that good-quality con- open defecation is the usual alternative. For example, in struction materials are not available. Cambodia, latrines are described as being far away and only found in towns, pagodas, or schools: Knowing a mason to assist with latrine construction is im- portant in contexts where labor is relied upon to build la- When farmers are in the field or when they go far from their trines, landscape is challenging, deeper pits are required, or villages they have no option other than using open fields. improved latrines/upgrades are desirable. In some countries/regions, up to 90 percent of households report Findings also demonstrate that owning a latrine, or having using masons to construct latrines. The importance of access to a latrine, does not ensure that it is used or used knowing where to find a supplier may be a determinant of consistently by household members. For example, among latrine ownership and upgrades. However, perceived avail- households with latrines, 18 percent of respondents in East ability of masons varies by study. For example, Java reported defecating in the open, and in Kenya, 89 per- • Perceived availability of suppliers or masons ranges cent of adults and only 66 percent of children consistently from 34 percent of latrine owners and 46 percent of use latrines. improved latrine owners in Meghalaya, 73 percent of www.wsp.org 5 What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Opportunity households in Rajasthan, 80 percent of households In general, most of the negative latrine attributes are in refer- in Bihar, and 85 percent of households in Tanzania. ence to unimproved latrines. Latrines are perceived as hav- • In Meghalaya, the second most common barrier ing low durability, requiring frequent maintenance and to making latrine improvements is noted as “diffi- constant relocation. They overflow, collapse, and/or become culty in finding a mason” (cited by 23 percent of full, and are perceived as unsustainable. Latrines are also per- households). ceived as unsafe and risky. There is fear that people, espe- cially children, will fall into the pit or the ground will cave Although perceived availability of masons may be an im- in, causing the user to sink into the defecation site. There is portant issue to address, these are not the only sanitation also the perception that latrines are unhygienic, difficult to suppliers that can provide latrines and make upgrades. clean, and emanate bad smells. For example, in Meghalaya, 56 percent of households believe that a key disadvantage of 3.2 Product Attributes using a latrine is the malodor. Given these negative percep- Sanitation products and services must not only be available tions regarding latrines, open defecators describe their be- and readily accessible, they must also have the level of qual- havior as a more pleasant and comfortable experience. ity and other positive attributes sought by the target popula- Example qualitative quotes from in-depth interviews are: tions. Product attributes are the subjective perceptions about the physical components of the latrine and perceptions of The logs can rot, and within three to four years it sinks. For the practical use of the product (e.g., Berkowitz et al. 2000; example what happened here recently, a woman sunk inside Rogers 2003). a latrine with a collapsing floor, the thing was so weak and she sunk inside. People went and rescued her. — Uganda The review found negative perceptions regarding the qual- ity, safety, comfort, and hygiene of latrines that reinforce If I defecate in the river, I feel more comfortable. I don’t have open defecation and hinder decisions to build or invest in to smell my own waste—unlike when I’m doing it in a latrines. (closed) latrine. — East Java FIGURE 3: SANIFOAM OPPORTUNITY FACTORS FOUND TO INFLUENCE RURAL SANITATION BEHAVIORS Access and availability: If an individual does not have access to a latrine at work or in the homestead, open defecation is the usual alternative, given that having access to a latrine is a precursor to being able to use a latrine. Furthermore, defining a household (or individual) as owning a latrine does not ensure that Opportunity there is access to a working latrine. Latrines are commonly described as being full or in need of repair, and this serves as a barrier to latrine use. Access and availability (supply-side perceptions): Knowing a supplier who stocks a variety of sanitation Access/availability hardware and a mason who can assist with latrine construction are central to making latrine upgrades and improvements. Often these factors serve as barriers to moving up the sanitation ladder, given that materi- als for improved latrines are perceived as unavailable and costly. Product attributes: There are negative perceptions regarding the quality, safety, comfort, durability, and Product attributes hygiene of latrines. Open defecators describe their behavior as a more pleasant and comfortable experi- ence. The negative product attributes, which are usually in reference to unimproved latrines, serve to rein- force open defecation and hinder decisions to build or invest in latrines. Social norms Social norms: Family members, peers, and others in the community defecate in the open, making this a common behavior that is rooted in culture and tradition and learned since childhood. These norms, which are held more strongly by open defecators, serve as a barrier to latrine acquisition and use. Sanctions/ enforcement 6 Scaling Up Rural Sanitation What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Opportunity Building a basic latrine each year is so much work. We had framework, social norms include behavioral standards that to collect grass, wood, and then dig the pit. And I had to exist in the community for an individual to follow, and are cook for the men and collect water. It was hard work but the presence or absence of traditions and cultures that gov- now we have a good latrine and I don’t have to worry about ern behavior (Andersen 1995; Fehr and Gaechter 2000; any of that. — Malawi Bettenhausen and Murnighan 1991).1 Increasing awareness around the durability of an improved Family members, peers, and others in the community latrine may be an important communication message. defecate in the open, making this a common behavior that Box 3 lists the qualities of an ideal latrine, such as durabil- is rooted in culture and tradition and learned since ity, along with other important attributes. childhood. In Peru, open defecation is described as “the 3.3 Social Norms 1 It is, however, noted that there are various definitions of social norms and the Social norms are the rules that govern how individuals in a precise definition of a social norm varies (for examples of definitions, see Elster 1989, Bettenhausen and Murnighan 1991, Fehr and Schmidt 1999, Lindbeck group or society behave. According to the SaniFOAM et al. 1999, and Bicchieri 2000). BOX 3: PRODUCT ATTRIBUTES: WHAT DO HOUSEHOLDS CONSIDER AN IDEAL LATRINE? Attributes of an ideal latrine include: • durability • privacy • cleanliness and being easy to clean • convenience • affordability • ease of construction For example, “privacy” is cited as an ideal attribute by 46 percent of households in Bihar, 67 percent in Rajasthan, and 80 percent in Meghalaya. Cleanliness is cited as the most important attribute in Kenya. The photos below show examples of improved latrines with slabs in Tanzania, which have many attributes of an ideal latrine. Photo credit: Water and Sanitation Program, World Bank. www.wsp.org 7 What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Opportunity most natural thing.” In East Java, a focus group participant to defecate in the open in their community.” In one noted, area surveyed, as many as 80 percent of respondents agreed or strongly agreed with this statement. Yeah, I am embarrassed if people pass by, but I think every- • In Rajasthan, 28 percent of open defecators state this body is used to it, everybody also does that. behavior is “practiced by generations” and 47 per- cent agree “we are used to defecating in the open.” And in Kenya, a participant described, • In Bihar, 49 percent of open defecators agree “we are used to defecating in the open.” Some people may have a toilet, but are not used to going to the toilet. It depends with how a person was brought up. If In certain circumstances (such as when traveling) or for cer- he is used to go to the bush, he will still go to the bush. tain target groups (such as children), the practice of open defecation is deemed more acceptable. Specific cultural Open defecation is described as traditional, habitual, and norms may also further influence open defecation, such as part of one’s daily routine, and these social norms are also the belief that females and male in-laws should not share held more strongly by open defecators. For example, the same latrine facilities, or in contexts where men are not • In Tanzania, 40 percent of all survey respondents meant to be seen going to a toilet. agree or strongly agree that “it is normal for people 8 Scaling Up Rural Sanitation IV. Ability Ability is an individual’s skills or proficiencies needed to do not know how to build an improved latrine. Other re- engage in a certain sanitation behavior. Key ability factors search shows that in East Java, 63 percent of latrine-owning that were found to influence behavior include skills and households report that it is “easy” to build a new latrine and self-efficacy to build latrines and perceived affordability. attribute this to the availability of masons and materials. In These are summarized in Figure 4. Peru, it is noted that suppliers are not always available to build latrines, and this serves as a barrier to latrine 4.1 Skills and Self-efficacy acquisition. Self-efficacy is an individual’s belief that he or she can per- form a promoted behavior effectively or successfully (e.g., As such, knowing a mason to assist with latrine construc- Bandura 1977; Becker 1990). Some households may build tion will be important in contexts where labor is relied their latrine themselves rather than hire a mason to do it. upon, such as when the landscape is challenging, deeper For these self-builders, the knowledge needed to go about pits are required, or improved latrines/upgrades are desir- this is referred to as skills. able. Knowing where to find a mason or services/supplier is an important driver for making latrine upgrades, but is less The extent to which masons are relied upon to build latrines important for building a simple pit latrine. varies by country/region, and by the type of latrine or struc- ture being built (see Figure 5). Unimproved latrine owners 4.2 Affordability typically believe they can build a basic pit latrine and have Affordability in the context of SaniFOAM is one’s ability to the skills and tools necessary to do this. However, self- pay for a sanitation product or service or to engage in a efficacy to build an improved latrine is lower and reliance sanitation behavior (e.g., Foreit and Foreit 2000). Afford- on masons is higher. One in three households in Tanzania ability can be influenced by many factors, including FIGURE 4: SANIFOAM ABILITY FACTORS FOUND TO INFLUENCE RURAL SANITATION BEHAVIORS Ability Knowledge Skills and self-efficacy: Unimproved latrine owners typically believe they can build a basic pit latrine and Skills and that they have the skills and tools necessary to do this. However, self-efficacy to build an improved latrine self-efficacy is lower and reliance on masons is higher. Knowing where to find a skilled mason is an important factor for making latrine upgrades, but is less important for building a simple pit latrine. Social support Affordability: Both open defecators and latrine owners consistently mentioned cost as a barrier to build- Roles and decisions ing and upgrading facilities. Open defecators overestimate the cost of latrines contributing to a perceived unaffordability. Latrines are perceived as expensive to construct, especially when associated with cement or deeper pits. Latrines are also perceived to be more expensive in certain seasons, such as during the rainy season when construction is perceived as more challenging, due to flooding. There are also chal- Affordability lenges accessing credit or loans to pay for latrines, which serve as a barrier to acquisition and upgrades. www.wsp.org 9 What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Ability FIGURE 5: USE OF MASONS TO BUILD UNIMPROVED OR IMPROVED LATRINES, BY STUDY (%) 100 90 80 70 60 50 40 30 20 10 0 Kenya Meghalaya Kenya Bihar Rajasthan Meghalaya Use of masons to build Use of masons to build improved latrines unimproved latrines Note: Graphs in this report are for illustrative purposes. Survey questions, target groups, and denominators may vary according to study. household income, availability of cash, time of year, access Open defecators cite lack of finances, insufficient funds, to credit, and availability of suitably priced sanitation op- “too expensive,” or “don’t have money” as key barriers to tions in the area. Affordability can be real or perceived. In building latrines or making improvements. Latrines are per- the latter case, knowledge of the true costs of a latrine may ceived as expensive to construct, especially when associated be an associated factor. with cement or deeper pits. Latrines are also perceived to be more expensive to build in certain seasons, such as during Assessments of wealth are estimated across all reports and the rainy seasons when construction is perceived as more indicate that those without latrines tend to be poorer than challenging, due to flooding. There are also challenges of those higher on the sanitation ladder. However, both open accessing credit or loans to pay for latrines. For example, in defecators and latrine owners consistently mentioned cost as Tanzania, 43 percent of non-latrine owners from the poor- a barrier to building and upgrading facilities (see Figure 6). est wealth quintile cite “inability to save” or “lack of access FIGURE 6: OPEN DEFECATORS CITING COST OR AFFORDABILITY AS A KEY BARRIER TO BUILDING LATRINES OR MAKING IMPROVEMENTS (%) 100 80 60 40 20 0 Cambodia East Java Kenya Tanzania Bihar Meghalaya Rajasthan Note: Graphs in this report are for illustrative purposes. Survey questions, target groups, and denominators may vary according to study. 10 Scaling Up Rural Sanitation What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Ability BOX 4: DO HOUSEHOLDS HAVE ACCURATE PRICE PERCEPTIONS OF LATRINES? The perception of the price of a latrine varies because open defecators have, in some cases, never owned, built, or even used a latrine, and owners of unimproved latrines have little experience with upgrading their facilities, although they might have looked at options for latrine upgrades. Notably, open defecators perceive latrines as much more expensive than do households that own latrines, but may own household items that cost as much as a latrine. For example, in Kenya, 90 percent of households own a radio, which costs approximately the same as building a latrine (WSP et al. 2013, 2). Generally, people are unaware of a range of affordable latrine options. Ensuring that households have accurate perceptions of costs associated with latrine purchases and upgrades may help to address the perceived afford- ability barrier. to credit” as a main impediment to building a latrine or car- Improved latrines are deemed expensive to install. Given rying out improvements. For open defecators, the perceived available income, which also varies by time of year, access to cost of a latrine may be so high that latrine acquisition is extra resources to build a latrine is a challenge. Constraints inconceivable (see Box 4). For example, a focus group are further exacerbated by the lack of formal credit mecha- member in Kenya stated, nisms for home improvements. These financial constraints are also associated with competing priorities (see “Compet- “This year I have no money to spend on anything but food. ing Priorities,” Section 5.2). It rained too much last year and our cassava crop has failed.” In summary, affordability barriers are linked with levels and For unimproved latrine owners, the high cost of materials fluctuation of income, lack of savings, lack of financing and and labor, coupled with lack of savings and access to credit, limited credit options for home improvement, and actual prevents improvements from being made. Given that pit versus perceived costs of building a latrine. latrines are notably lacking in durability, there is also need for repeated financial investments just to maintain or re- build them. www.wsp.org 11 V. Motivation For a behavior to take place, an individual must be moti- comfortable,” because it prevents individuals from getting vated to engage in it. Motivation refers to an individual’s scratched, stepping on thorns, or dirtying their clothes. Al- desire to perform a promoted behavior. Emotional, physi- though comfort was mentioned as a positive attribute across cal, and social drivers and competing priorities were found a number of countries, it was most notably important in to influence the behavior of interest. Key motivational find- Cambodia, where 66 percent of latrine owners cite comfort ings are summarized in Figure 7. as a key advantage of owning a latrine. 5.1 Emotional, Social, and Physical Drivers 5.1.2 Privacy Drivers are strong internal thoughts and feelings that moti- Privacy emerged as a motivator among latrine owners and vate behavior (e.g., Cole et al. 1993; Catania et al. 1990). open defecators to move up the sanitation ladder. It is im- They can be positive or negative, and can stem from unmet portant for people, especially women, to avoid being seen physical, emotional, or psychological needs. Such drivers exposing body parts. Improved privacy is a key reason for have been identified through research in several countries as latrine construction for around 45 percent of latrine owners motivators to engage in the adoption of positive sanitation in Bihar, Kenya, and Cambodia; 56 percent in Rajasthan; behaviors. The review found a number of relevant compo- and up to 90 percent in Meghalaya. This is also confirmed nents: comfort, privacy, shame and embarrassment, social by qualitative research, as illustrated by the following quote status, prestige, and honor. from a latrine owner in East Java: 5.1.1 Comfort We have to protect our body. If we have our own toilet, we Having one’s own latrine avoids exposure to the elements. can protect our body parts, so nobody else can see them. Being able to use a latrine is described as “more FIGURE 7: SANIFOAM MOTIVATIONAL FACTORS FOUND TO INFLUENCE RURAL SANITATION BEHAVIORS Motivation Attitudes and beliefs Drivers: Privacy, comfort, and improved social status emerged as key motivators to move up the sanitation Values ladder among the different behavioral groups, and were cited as common reasons for latrine acquisition. Improved social status was a particularly important driver to motivate open defecators to acquire latrines. Emotional/physical/ social drivers Embarrassment, shame, and humiliation also motivate individuals to use latrines. Although open defecation is noted as a common practice, this behavior may be a source of embarrassment, particularly for those who may have used facilities or own latrines that are no longer functional, and thus serves as a motivator for la- Competing priorities trine use. Competing priorities: Latrines are viewed as a household improvement, but one that has lower priority in Intention terms of family expenditure. School fees, food, transport, and healthcare are priorities for those with limited savings. Building, repairing, or improving a latrine are only considered if and when additional resources are available, and even then, other competing demands have priority. Willingness to pay 12 Scaling Up Rural Sanitation What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Motivation Although this factor is particularly important for women, it improved latrine owners are more likely than those with also resonates with men who want to protect a woman’s unimproved latrines to agree that having a latrine raises the honor and dignity, as illustrated by this unimproved latrine family’s status in the community. In Rajasthan and Bihar, owner from East Java: honor is the third most important reason for constructing latrines, as reported by 35 and 45 percent of male latrine My wife never goes to the river; she is not used to it. She feels owners. Improved status and prestige is also mentioned as a embarrassed and uncomfortable. So I thought I’d better key motivating factor for owning a latrine by 24 percent of build my own toilet. Cambodians. 5.1.3 Embarrassment, Shame, and Humiliation As reflected in the following statements, the notion of pres- Notions of embarrassment, shame, and humiliation moti- tige and pride is also important: vate latrine owners and open defecators to move up the sanitation ladder. Although it is noted as commonly prac- Now it is very easy for me to ease myself. Secondly when visi- ticed, open defecation is still a source of embarrassment, tors come I feel very comfortable and not scared. If you do particularly for those who may have used facilities or own not have a latrine and a visitor comes you will be very em- latrines that are no longer functional. Adjectives used to barrassed and look very small. So you feel very comfortable describe open defecation include “shy,” “shameful,” “un- and you feel that you are a man at home. Even outside there comfortable,” and “embarrassing.” For example, in Kenya you walk like other men walk. — Uganda 42 percent of households felt embarrassed when their la- trine was out of use. In Peru, persons living in households I feel proud because I have a well-maintained and clean without latrines report feeling embarrassed to receive visi- toilet. — Malawi tors. In Rajasthan and Bihar, the number one reason women are motivated to build a latrine includes notions of 5.2 Competing Priorities “feeling embarrassed to be seen uncovered” (66 and 56 per- Households and individuals face many competing demands cent, respectively). when it comes to spending. The lower the income, the more competing demands may influence behavior. Financial de- The notion of shame and humiliation is also prevalent, and mands can be for day-to-day necessities, occasional or peri- serves to promote latrine use and ownership. In Tanzania, odic expenses, or urgent or discretionary expenditures. for example, 42 percent of people who report openly defe- Households with strong financial pressures will often place cating because of collapsed latrines feel ashamed. In Kenya, a lower priority on sanitation and be less motivated to ac- 89 percent of households agree that people in the commu- quire a facility (e.g., Jenkins and Scott 2007). nity would feel ashamed if they did not have a latrine, and 37 percent of latrine owners report feeling ashamed when Latrines are viewed as a household improvement, but one their latrine was out of use. Women in particular feel a sense that has lower priority in terms of family expenditures. of humiliation, as illustrated by this female open defecator School fees, food, transport, and healthcare are a priority from Meghalaya: for those with limited savings. Building, repairing, or im- proving a latrine are only considered if and when additional We want to have a latrine for we face no more shame. It is resources are available, and even then, other competing de- difficult to live like this. We will try to build latrine using mands have priority (see Figure 8). For example, in East the available materials. Java, when households have extra money, the primary pri- ority is to pay debt, followed by purchasing items that can 5.1.4 Social Status, Prestige, and Honor be sold later (such as livestock or gold), and then purchas- Owning a latrine can positively influence one’s social status, ing luxury items (namely, electronics such as TVs or refrig- as owners are described as prestigious, well respected, and erators), which provide entertainment and improve the looked upon favorably by others. In East Java, for example, family status. www.wsp.org 13 What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Motivation FIGURE 8: COMMON USES OF EXTRA MONEY IN BIHAR AND RAJASTHAN (%) 100 80 60 40 20 0 Buying agricultural land Children’s education Repaying loans Bihar Rajasthan Note: Graphs in this report are for illustrative purposes. Survey questions, target groups, and denominators may vary according to study. In Tanzania, 43 percent of respondents agree that people in televisions, bicycles, and radios) over a latrine. Latrines are their community would rather buy animals than build a also not considered a “good” or “wise” investment, such as latrine, and 50 percent would rather buy a phone than a buying additional livestock or land, which is viewed as profit- latrine. In Cambodia, the second most common reason for able in the longer term. Rather, a (unimproved) latrine is not owning a latrine is “other priorities come first.” Least viewed as an asset that needs future investment for mainte- prioritized include home renovations, such as latrine acqui- nance, repairs, and possible reconstruction and/or is noted as sition, and this is noted as occurring in stages because of a household item that does not generate any revenue. How- lack of money. ever, in Peru, for example, although improved latrines may be more expensive than unimproved latrines, in the long run the However, although open defecators have fewer assets than overall costs are reduced given there is less need for reinvest- their latrine-owning counterparts, a significant proportion of ment, maintenance, and repairs associated with improved non-latrine owners prioritize ownership of other assets (e.g., latrines. 14 Scaling Up Rural Sanitation What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Motivation BOX 5: THE ROLE OF IMPORTANT FAMILY EVENTS AS A MEANS TO RAISE LATRINE PRIORITY Events such as a wedding, welcoming guests, and in-laws joining families may raise the priority of a home latrine above other demands on household resources. For example, in Bihar, latrine owners are more likely to agree that latrines are constructed when visiting children refuse to defecate in the open. Other events include sudden sick- ness, hosting an important social gathering, or an extended visit from a relative from the city or abroad. It may be that these events raise priority of latrines above other household demands and serve to promote a move up the sanitation ladder. As this screenshot from a 2013 World Bank video illustrates, an increasing number of brides in India are demand- ing the groom provide a household latrine before agreeing to marry. “Changing the Culture of Toilets in India” available on The World Bank’s YouTube channel: http://www.youtube.com/watch?v=liBPxiphF0U www.wsp.org 15 VI. Socioeconomic Status and Other Contextual Factors 6.1 Socioeconomic Status 6.2 Contextual Factors A clear finding to emerge from the review was the relation- Other contextual factors, such as the time of year and seasonal ship between household wealth and latrine ownership. factors, land ownership, and household members’ perceptions There is a positive relationship between a household’s socio- of the physical and geographical conditions (such as access to economic status and its position on the sanitation ladder. water and soil profile) were found to influence behaviors. Al- Improved latrine owners are wealthier than unimproved la- though these factors coalesced around common themes, find- trine owners or open defecators, are more educated, and ings varied by country and region, given variations in climate have higher literacy rates, which is consistent with findings and geography (see Box 6 for additional examples). Although from the JMP. For example, in Rajasthan, respondents from these components are not necessarily changeable through be- highest quintiles are more likely to own latrines than those havior change interventions, these sociodemographic and en- in the lowest quintiles. In contrast, those from the lowest vironmental characteristics are important because they serve socioeconomic quintiles are most likely to defecate in as facilitators or deterrents for positive sanitation behaviors. the open. This information is also valuable to situate some of the chal- lenges when aiming to promote better practices. 16 Scaling Up Rural Sanitation What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Socioeconomic Status and Other Contextual Factors BOX 6: EXAMPLES OF CONTEXTUAL FACTORS INFLUENCING LATRINE OWNERSHIP AND OPEN DEFECATION Temporal and Seasonal Factors Open defecation may be more frequent during certain times of the year: • At night, in cases where latrines are located further from the household and security is a concern or to allow for more privacy. • During work or when travelling, when access to latrines is limited, and where paddy fields are noted as a place for open defecation. Open defecation or latrine acquisition may vary by season, as seen in these examples: • In Sub-Saharan Africa countries, open defecation is more commonly practiced during the rainy season, as latrines are cited as overfull, washed away, or collapsed. Heavy rains prevent latrines from being built, rebuilt, or repaired. • During the dry season in East Java, rivers dry up and no longer provide an option to allow feces to float away, so alternatives to open defecation are sought. Land Access and Tenure Sufficient access to land and space to build latrines is noted as an important driver. In situations where latrines are overfull or have collapsed, it is impracticable or impossible to construct another latrine given insufficient land space. Some examples include: • In Meghalaya and East Java, 22 percent of open defecators report that they lack space to build a latrine. • In East Java, many report insufficient land to make improvements, such as space for a septic tank. Not owning one’s house or land can also act as a disincentive to construct a latrine, as mentioned by open defecators: • Meghalaya (16 percent) • Bihar (15 percent) Soil Profile Soil profiles, such as rocky or steep landscapes, pose a challenge to building latrines, as most commonly noted in East African countries and East Java. In some instances, water is too close to the surface, making it impossible to dig sufficiently deep pits. In cases where latrines have collapsed or are no longer functioning, households often report that they do not have the ability to rebuild latrines, due to flooding of land and soil quality, as illustrated by this participant from an in-depth interview: In this center we have one problem. We cannot dig pits because of the rocks we have here. You can only dig up to four feet and then you reach the rocks. If you don’t plan properly here you cannot dig a pit of 25 feet. — Uganda www.wsp.org 17 What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Socioeconomic Status and Other Contextual Factors Access to Water Sources A sufficient water supply for flushing away waste and latrine cleaning is seen is an integral part of the latrine decision-making process, and will also influence the type of latrine built. Scarce water supply, along with the perception that latrines consume a lot of water, can pose a barrier to building a latrine and prevent usage. For example, in Rajasthan, one in four households cited lack of water as a reason for open defecation. In contrast, in some areas where water supply is not an issue, having a river or water stream nearby is seen as an easy and inexpensive means to dispose of waste, such as in East Java (upper photo) and Peru (lower photo). Photo credit: Water and Sanitation Program, World Bank 18 Scaling Up Rural Sanitation VII. Other Findings The review identified several other factors, including meant or understood. Hygiene may be in reference to a la- knowledge, enforcement of rules or regulations, values, trine’s perceived cleanliness, rather than knowledge of the roles and decision-making, and beliefs and attitudes. How- fecal–oral contamination pathways. It may also reflect more ever, it is less certain how these aspects influence sanitation “top of the mind” reasons or social desirability. Thus, the behaviors, thus limiting the strength of the conclusions that importance of health and hygiene as a motivational concept can be made. This may be due to the different research ob- (and what this actually means to latrine owners and open jectives and interview guides of the studies, quality of the defecators) is inconclusive. data and reporting, regional differences, and/or their actual relevance to sanitation behaviors. Some examples are pro- In other cases, findings were only addressed in a handful of vided in this section to further illustrate the challenges in countries, such as sanctions and enforcement, beliefs and drawing conclusions. attitudes, and values (see Box 7). Although these concepts were explored in study questionnaires, findings were not In some instances, conclusions were difficult to draw given presented in reports, making it challenging to conclude the the measurement of the factors. An example of this was extent to which these concepts are consistent and relevant found for knowledge (see Figure 9). Knowledge questions across countries. However, in the reports where these con- were not routinely administered in all surveys and when cepts were included, findings were similar. they were, questions did not always address the disease pathways or transmission routes, making robust conclu- In other cases, such as roles and decision-making, inten- sions difficult to draw. However, for many households, a tion, and perceptions of latrine ownership as providing reason to build or upgrade a latrine is “good health and safety and security or convenience, there were differences hygiene,” and this is cited as a key advantage of owning a between countries (Box 8). For roles and decision-making, latrine. However, it is unclear what exactly respondents for example, although the review found a tendency for the FIGURE 9: RESPONDENTS CITING HEALTH OR HYGIENE AS A REASON TO BUILD LATRINES, BY STUDY (%) 100 80 60 40 20 0 Cambodia East Java Kenya Tanzania Bihar Rajasthan Note: Graphs in this report are for illustrative purposes. Survey questions, target groups, and denominators may vary according to study. www.wsp.org 19 What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Other Findings BOX 7: THE INFLUENCE OF SANCTIONS AND ENFORCEMENT, BELIEFS AND ATTITUDES, AND VALUES ON RURAL SANITATION BEHAVIORS Sanctions and Enforcement In certain situations, sanitation behaviors may be influenced by law enforcement and other regulations as men- tioned in Tanzania, and the three Indian states. For example, • In Tanzania, there is pressure from district officials to replace latrines that are full. • In Meghalaya, one of the main reasons for building a latrine is stringent village rules. • In Bihar, latrine owners are more likely to agree that sanctions are in place and that there are rules and penalties to stop open defecation. Beliefs and Attitudes There exist a number of beliefs and attitudes toward open defecation that serve to deter the extent to which feces are perceived as harmful to the environment and the concept of human waste as a source of pollution, but only found as relevant in Bihar, East Java and Kenya. For example, • In Bihar, farmers believe that feces are beneficial for farming, as it will increase fertility of the land and im- prove and increase crop production. • In East Java, respondents discuss how open defecation into a river is not harmful, given there is the belief that fish eat their waste or that feces can serve as fertilizer. These beliefs are more salient among open defecators, and may serve as psychological refuges that help to jus- tify their behavior. Values Only a few of the reports cite the importance of values: Peru, East Java, and Tanzania. In these countries, latrine ownership is associated with a number of positive family values: being clean, being health conscious, and being good parents and welcoming hosts. For example: • In Peru and East Java, having a sanitation facility at home represents modernity and progress. • In Tanzania, “modernity” is the most common reason for households to improve existing latrine. final decision to rest with the male head of the household, (see Figure 10), open defecators describe their behavior as there is country and regional variance, which makes gener- “easy” and practical. They report that if it is late at night or alizing results challenging. For example, in Meghalaya, raining heavily, they use flying toilets so there is no need to which is a matriarchal society, women have an important leave their household compound. The notion of conve- role in the decision-making process. nience may be related to a number of factors, such as whether or not there are open defecation sites allocated for Convenience is highlighted as both a barrier and a motiva- this behavior, how much available space and privacy there is tor to adopting better sanitation practices. Although latrine to defecate when one needs to, and the proximity of one’s owners cite convenience as an advantage to owning a latrine homestead to others. 20 Scaling Up Rural Sanitation What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Other Findings BOX 8: REGIONAL VARIATIONS REGARDING THE RELEVANCE OF DIFFERENT SANIFOAM FACTORS Drivers: Perceptions that latrines provide safety and security The concept of safety and security is both a motivator and a deterrent for improved sanitation, and there are dif- ferences between countries and regions in terms of where open-defecation behavior takes place and the type of latrine. For example, as a motivator for latrine acquisition: • There is fear of being dragged away by strong currents, especially for children, or that people can be bitten or attacked by wild animals. It is also perceived as dangerous when the ground is muddy. • A main reason for building a latrine is “improved security for women and children,” as mentioned by 69 percent of latrine owners in Rajasthan, 63 percent in Bihar, and 50 percent in Meghalaya. For example, as a barrier to latrine acquisition and use: • In Kenya, households report that they are afraid to use the latrine because of the distance/proximity of the latrine to their house, and would rather use flying toilets at nighttime. There is a fear that people will be bitten or attacked by wild animals. • Pit latrines are considered dangerous and in poor condition, especially after rains. Consequently, people report that they would prefer to defecate in the open rather than risk falling into a sinking hole or a latrine collapsing. The notion of latrines as being unsafe is in relation to pit latrines. A key trigger to upgrading latrines is around the notion that “improved latrines” are durable and safe. Improved safety is also described as one of the most com- mon reasons for improving an existing latrine. As such, the conflicting information regarding the role of safety in latrine adoption may be due to perceptions regarding the type of latrine being referenced by research participants. FIGURE 10: LATRINE-OWNING HOUSEHOLDS CITING CONVENIENCE AS A MAIN REASON TO BUILD A LATRINE (%) 100 80 60 40 20 0 Cambodia Bihar Meghalaya Rajasthan Note: Graphs in this report are for illustrative purposes. Survey questions, target groups, and denominators may vary according to study. www.wsp.org 21 What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Other Findings Finally, some other concepts emerged from the review, that in some cases, open defecators may be happy enough including satisfaction. The assumption is that the greater with their current practices (see Figure 11). Therefore, the degree of dissatisfaction, the higher the likelihood the relationship between satisfaction and an individual’s that a respondent will move up the sanitation ladder. position on the sanitation ladder is somewhat unclear, However, levels of satisfaction with current defecation although owners of improved latrines are clearly the practices vary by region and behavior group and suggest most satisfied. FIGURE 11: RESPONDENTS CITING SATISFACTION WITH CURRENT DEFECATION OPTIONS, BY STUDY (%) 100 80 60 40 20 0 Cambodia East Java Kenya Bihar Meghalaya Kenya Tanzania Bihar Meghalaya Rajasthan Open defecators Latrine owners Note: Graphs in this report are for illustrative purposes. Survey questions, target groups, and denominators may vary according to study. 22 Scaling Up Rural Sanitation VIII. Summary and Implications This global review of formative research studies identifies a • Using more specific or less burdensome research number of commonalities across countries and regions, as methods. Depending on the program needs and well as a number of emerging themes that positively and objectives, other methods could be used to answer negatively influence sanitation behaviors. Although some research questions that we know less about but are differences are found across countries, in general themes co- known to be important and specific to different alesced around facets of opportunity, ability, and motiva- study populations. For example, tion. Findings suggest that a number of factors serve to 0 street/village intercept surveys to address price promote positive sanitation behaviors. These factors include perceptions and willingness to pay for latrines, changing social norms, challenging perceptions of latrine or supply-side surveys to address actual product affordability, fostering positive latrine attributes, and in- availability and pricing. creasing consumer demand for latrines through emotional 0 qualitative research to explore determinants of hooks, such as associating latrine use and ownership with behavior when few people are actually “doing” improved social status. Ensuring that latrines are available the behavior. For example, if only 5 percent of a and functioning will also serve as a precursor to use. population is estimated to own an improved la- trine, investigating reasons for improved latrine The relationship between behavior and other concepts, ownership in a quantitative survey would require such as knowledge, sanctions, enforcement of rules or regu- a very large sample size. lations, and values and attitudes, is less clear. This may be • Using standardized research guidelines and ap- due to the different research objectives and questionnaires/ proaches to ensure greater comparability between guides of the studies, or it could reflect the actual relevancy countries and target groups. This has also been noted of these factors to sanitation behaviors. At any rate, making as important in other sanitation behavior change robust conclusions regarding their influence on sanitation frameworks, notably the RANAS model, which has behaviors is more challenging. underscored the need for standardized measurement of different theoretical factors through the use of sin- A number of research recommendations emerged from this gle questions in a survey (Inauen et al. 2013). There global review, given that it identifies a number of factors are also opportunities for more specific behavioral that resonate with sanitation behaviors. Most importantly, questions to delve deeper into self-reported latrine there are opportunities to conduct “lighter” and more tai- use. Namely, respondents should be asked about their lored formative research. These are summarized here: defecation behavior, and not what they think others • Including the most important determinants identi- in their households are doing. For example, specific fied in this review as a means to monitor program behavioral questions may address whether or not re- impact and assess the extent to which these factors spondents used a latrine the last time they defecated, may have changed over time. These determinants as well as clarifying defecation practices that occur are access and availability to functioning latrines, inside or outside the home (e.g., at work). This will sanitation products, and services; latrine product also allow for further investigation of barriers regard- attributes; social norms around open defecation; ing why people who own latrines may not use them. perceptions of latrine affordability; competing pri- For additional guidance, refer to the “Study Design orities for other household items; and a number of and Questionnaire Tips” document, available online emotional, social, and physical drivers. Addressing in WSP’s Sanitation Marketing Toolkit (http://wsp. wealth and contextual factors will also be important, org/toolkit/toolkit-home). as well as self-efficacy in contexts where improved • There may be value in conducting additional anal- latrines are being constructed. ysis on the primary data to allow for making more www.wsp.org 23 What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Summary and Implications robust conclusions and for exploring the importance • The “focus” component of the SaniFOAM frame- of SaniFOAM factors relative to each other. Analysis work ensures that program managers and imple- could also consider making statistical comparisons menters define the behavior to be changed and the between the different behavioral groups. The impact target group prior to research studies or interven- of wealth disparity could also be further investigated. tions, and this is noted as important in other sani- In particular, understanding the specific barriers and tation frameworks (Mosler 2012). However, the drivers to improved sanitation among the rural poor- SaniFOAM framework may also benefit from fur- est will help improve programmers’ ability to design ther clarification regarding the “focus” component, effective behavior change interventions, particularly namely to acknowledge more contextual factors as the sector moves toward more equity-focused goals that are known to affect sanitation behaviors (see in the post-MDG setting. Finally, the role of gender Figure 1). For example, perceptions of the physi- in decision-making should also be investigated. The cal environment such as available sources of water, findings presented in the quantitative research re- level of the water table, pattern of precipitation, ports generally reflect male perceptions, given that and available land space. These concepts have been the study respondents were usually male household suggested as an important component to address in heads or representatives. Future quantitative forma- sanitation behavior change frameworks (Dreibelbis tive research studies should ensure that women are et al. 2013). adequately represented in the sample. This will help to explore the role of gender and further facilitate Finally, to demonstrate how the results from this review can an understanding of the factors that may influence be utilized, Box 9 provides some thoughts regarding practi- sanitation behavior according to men and women. cal implications from this review. 24 Scaling Up Rural Sanitation What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Summary and Implications BOX 9: EXAMPLES OF PRACTICAL IMPLICATIONS FROM THE REVIEW • Open defecation is a common behavior and is sustained through local cultural norms. According to the SaniFOAM framework, it is a social norm that is traditional and practiced since childhood, and in some regions, such as in India and East Java, further fostered by designated sites allocated for open defecation. Although changing actual normal practice may be a long and slow process, a first step may be changing perceived norms. For example, regularly portraying latrine usage in mass media, in TV ads, or on enter- tainment shows can create the impression of normality. Campaigns should endeavor to give high visibility to latrine usage as a social norm by creating the illusion that “everyone’s doing it,” complementing efforts through Community-Led Total Sanitation (CLTS), and other triggering approaches. • There is a clear association between one’s position on the sanitation ladder and socioeconomic status. Wealthier people are more likely to own and use improved latrines. However, many believe that latrines are too expensive to purchase and install. Improving knowledge around the actual cost of latrines and com- paring the costs with other household items can serve to encourage latrine acquisition. • Similarly, messages around the affordability of latrines should be coupled with descriptions about positive latrine attributes that stress that improved latrines are safe, durable, and hygienic. This can also help to encourage use and acquisition. • Events such as a wedding or welcoming guests may raise the priority of a home latrine above other de- mands on household resources, and this can serve to promote latrine acquisition. As such, promoting latrine purchases or improvements prior to large annual holidays that involve visits from family members from other parts of the country, or other large social gatherings or ceremonies (such as religious festivals or weddings), may serve to promote latrine acquisition prior to these events. This could also be comple- mented with messaging that presents latrine-owning families as welcoming and good hosts. • Not owning one’s house or land can also act as a disincentive to construct a latrine. In areas where renting is more common, landlords may be considered as a target population for promoting latrines. Demand- creation strategies could stress the value that latrine ownership could add to their property. • Sociodemographic and other environmental characteristics, such as access to water, perceptions of soil quality, and seasonal and temporal factors, provide important contextual information for sanitation be- haviors. Changing physical factors on a large scale, such as the availability of water, requires long-term sustained investment, which is typically beyond the ability of a sanitation campaign to deliver. However, given variations between countries and regions, these factors should be noted as a means to help tailor and target behavior change interventions. www.wsp.org 25 What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review References References Andersen, R. M. 1995. “Revisiting the Behavioral Model Behavior Change Interventions in Infrastructure-re- and Access to Medical Care: Does It Matter?” Journal of stricted Settings.” BMC Public Health 13: 1015. http:// Health and Social Behavior 36 (1): 1–10. www.biomedcentral.com/1471-2458/13/1015. Bandura, A. 1977. Social Learning Theory. Englewood Devine, J. 2009. “Introducing SaniFOAM: A Framework Cliffs, NJ: Prentice-Hall. to Analyze Sanitation Behaviors to Design Effective Becker, M. H. 1990. “Theoretical Models of Adherence Sanitation Programs.” Working Paper. Washington, and Strategies for Improving Adherence.” In The Hand- DC: Water and Sanitation Program, World Bank. book of Health Behavior Change, edited by A.A. Shu- http://www.wsp.org/sites/wsp.org/files/publications/ maker, 5–43. New York: Springer. GSP_sanifoam.pdf. Berkowitz, E. N., R. A. Kerin, S. W. Hartley, and W. Elster, J. 1989. The Cement of Society: A Study of Social Rudelius. 2000. Marketing, 6th edition. Boston: Irwin Order. Cambridge, UK: Cambridge University Press. McGraw-Hill. Fehr, E., and S. Gachter. 2000. “Fairness and Retaliation: Bettenhausen, K., and J. Murnighan. 1991. “The Devel- The Economics of Reciprocity.” Journal of Economic opment of an Intragroup Norm and the Effects of In- Perspectives 14 (3): 159–181. terpersonal and Structural Challenges.” Administrative Fehr, E., and K. Schmidt. 1999. “A Theory of Fairness, Science Quarterly 36 (1): 20–35. Competition, and Cooperation.” Quarterly Journal of Bicchieri, C. 2000. “Words and Deeds: A Focus Theory Economics 114 (3): 817–868. of Norms.” Rationality, Rules and Structure, edited by J. Foreit, K. G., and J. R. Foreit. 2000. Willingness to Pay Nida-Rumelin and W. Spohn, 153–184. UK: Kluwer Surveys for Setting Prices for Reproductive Health Products Academic Publishers. and Services: A User’s Manual. New York: Population Catania, J. A., S. M. Kegeles, and T. J. Coates. 1990. “To- Council. http://www.popcouncil.org/uploads/pdfs/ wards an Understanding of Risk Behavior: An AIDS frontiers/Capacity_Bldg/WTP_Manual.pdf. Risk Reduction Model (ARRM).” Health Education Hallahan, K. 2000. “Enhancing Motivation, Ability and Quarterly 17 (1): 53–72. Opportunity to Process Public Relations Messages.” Cole, G. E., D. Holtgrave, and N. Rios. 1993. “Systematic Public Relations Review 26 (4): 463–480. Development of Trans-theoretically Based Behavioral Inauen J., M. M. Hossain, R. B. Johnston, and H.-J. Risk Management Programs.” Risk: Issues in Health, Mosler. 2013. “Acceptance and Use of Eight Arsenic- Safety and Environment 4 (1): 67–93. Safe Drinking Water Options in Bangladesh,” PLOS Conteh, L., and K. Hanson. 2003. “Methods for Study- One 8 (1): e53640. ing Private Sector Supply of Public Health Products in Jenkins, M. W., and B. Scott. 2007. “Behavioral Indica- Developing Countries: A Conceptual Framework and tors of Household Decision-Making and Demand for Review.” Social Science & Medicine 57 (7): 1147–1161. Sanitation and Potential Gains from Social Market- Dreibelbis, R., P. J. Winch, E. Leontsini, K. R. S Hul- ing in Ghana.” Social Science and Medicine 64 (12): land, P. K. Ram, L. U. Unicomb, and S. P. Luby. 2013. 2427–2442. “The Integrated Behavioural Model for Water, Sanita- Lindbeck, A., S. Nyberg, and J. W. Weibull. 1999. “Social tion, and Hygiene: A Systematic Review of Behavioural Norms and Economic Incentives in the Welfare State.” Models and a Framework for Designing and Evaluating Quarterly Journal of Economics 114 (1): 1–35. 26 Scaling Up Rural Sanitation What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review References MacInnis, D.J, C. Moorman, and B.J. Jaworski. 1991. Rothschild, M. 1999. “Carrots, Sticks, and Promises: A “Enhancing and Measuring Consumer’s Motivation, Conceptual Framework for the Management of Public Opportunity, and Ability to Process Brand Information Health and Social Issue Behaviors.” Journal of Market- from Ads.” Journal of Marketing 55: 32–53. ing 63 (4): 24–37. Moorman, C., and E. Matulich. 1993. “A Model of Sunstein, C. R. 1996. “Social Norms and Social Roles.” Consumers’ Preventive Health Behaviors: The Role of Columbia Law Review 96: 201–266. Health Motivation and Health Ability.” Journal of Con- WHO/UNICEF Joint Monitoring Program. 2013. Prog- sumer Research 20 (2): 209–228. ress on Sanitation and Drinking Water: 2013 update. Moser, H.J. 2012. “A Systematic Approach to Behavior New York: WHO/UNICEF. Change Interventions for the Water and Sanitation Sec- Wiggins, J. 2004. “Motivation, Ability and Opportunity tor in Developing Countries: A Conceptual Model, a to Participate: A Reconceptualization of the RAND Review, and a Guideline.” International Journal of Envi- Model of Audience Development.” International Jour- ronmental Health Research 22(5): 431–449; DOI:10. nal of Arts Management 7 (1): 22–33. 1080/09603123.2011.650156. WSP (Water and Sanitation Program), IFC (International PSI (Population Services International) Research Depart- Finance Corporation), and Ministry of Health, Kenya. ment. 2004. “PSI Behavior Change Framework ‘Bub- 2013. “Kenya Onsite Sanitation: Market Intelligence.” bles’: Proposed Revision.” Washington, DC: PSI. www. Nairobi, Kenya: Water and Sanitation Program, World psi.org. Bank. http://www.wsp.org/sites/wsp.org/files/ Rogers, E. 2003. Diffusion of Innovations, 5th edition. publications/WSP-Kenya-Market-Intelligence- New York: The Free Press. Brochure.pdf. www.wsp.org 27 What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review Appendix Appendix TABLE A: DETAILS OF THE FORMATIVE RESEARCH STUDIES Country/Regions Date Surveyed Population Methods* Report Names Cambodia: Kandal, 2006 Latrine owners and FGD (N = 6) Demand Assessment for Sanitary Latrines in Rural Siem Reap, Svay nonowners SI (N = 939) and Urban Areas of Cambodia (2007) http://www. Rien wsp.org/sites/wsp.org/files/userfiles/WSP- Demand-Assessment-Cambodia.pdf India: Meghalaya 2012 Household heads (chief IDI (N = 18) Consumer Research for Rural Sanitation in (3 districts) wage earners) FGD (N = 21) Meghalaya (unpublished) SI (N = 960) India: Rajasthan 2012 Chief wage earners from IDI & FGD Understanding Open Defecators Perceptions and (11 districts) rural areas (N = 56) Motivations to Toilet Usage and Communication SI (N = 3,301) Strategy to Handle Open Defecation in the State of Rajasthan (unpublished) India: Bihar 2012 Chief wage earners IDI & FGD Consumer Research for Understanding Rural (13 districts) (N = 45) Sanitation, Bihar (unpublished) SI (N = 3,971) Qualitative Report for Understanding Rural Sanitation, Bihar (unpublished) Indonesia: East 2008 Male and female heads FGD (N = 6) Total Sanitation and Sanitation Marketing Research Java of household, from rural IDI (N = 6) Report (2009) http://www.wsp.org/sites/wsp.org/ and semi-urban areas SI (N = 2,009) files/userfiles/WSP-Sanitation-Marketing- Nielsen-Report-Indonesia.pdf Kenya: Wangige, 2012 Household adults who FGD (N = 16) World Bank Formative Research Qualitative and Naivasha, Karatina, are decision-makers or SI (N = 2,000) Quantitative Report, 2012 (unpublished) Kitui, Garissa, involved in the purchase Bungoma, Ahero, of household items, from Kilifi rural or semi-urban areas Malawi: Dowa, 2011 Household heads, sup- IDI (N = 35) A Market Assessment of Rural Sanitation in Malawi: De- Mangochi, Nkhata pliers, and government FGD (N = 15+) mand, Supply and the Enabling Environment for Sanita- Bay representatives SI (N = 222) tion in Dowa, Mangochi and Nkhata Bay (unpublished) Peru: Callao, 2007 Household heads and FGD (N = 24) Qualitative Report: Water and Sanitation Program Cajamarcam An- mothers with no latrines, Alternative Sanitation Solutions (unpublished) cash, and Loreto unimproved latrines, and improved latrines, from rural or semi-urban areas Tanzania: Musoma, 2008 Household heads, sup- SI (N = 1,000 Market Research Assessment in Rural Tanzania for Kiteto, Rufiji, Iringa, pliers, government households and New Approaches to Scale Up Sanitation Demand and Sumbawanga representatives N = 200 and Supply (2009) local service http://www.wsp.org/sites/wsp.org/files/publications/ providers) TZ_TSSM_Research_Report.pdf Uganda: Tororo 2009 Household heads, IDI (N = 30) In-depth Consumer Assessment Report for Sanitation District adopters and nonadopt- Marketing Piloting in Tororo District, Uganda (2009) ers of latrines. http://pdf.usaid.gov/pdf_docs/Pnadw563.pdf *SI: Structured interviews; FGD: Focus group discussion; IDI: In-depth interview 28 Scaling Up Rural Sanitation