WATER AND SANITATION PROGRAM: WORKING PAPER Nudging and Habit Change for Open Defecation: New Tactics from Behavioral Science David Neal, Ph.D. (Catalyst), Jelena Vujcic, M.P.H. (Catalyst), Rachel Burns Ph.D. (Catalyst), Wendy Wood, Ph.D. (University of Southern California) and Jacqueline Devine, MBA (World Bank, Water and Sanitation Program) March 2016 Authors: David Neal, Ph.D. (Catalyst), Jelena Vujcic, The Water and Sanitation Program is a multi-donor partnership, M.P.H. (Catalyst), Rachel Burns Ph.D. (Catalyst), Wendy part of the World Bank Group’s Water Global Practice, Wood, Ph.D. (University of Southern California) and supporting poor people in obtaining affordable, safe, and Jacqueline Devine, MBA (World Bank, Water and sustainable access to water and sanitation services. WSP’s Sanitation Program). donors include Australia, Austria, Denmark, Finland, France, The Bill & Melinda Gates Foundation, Luxembourg, The Acknowledgements Netherlands, Norway, Sweden, Switzerland, United Kingdom, This report was produced in collaboration between United States, and the World Bank. Catalyst Behavioral Sciences and the World Bank Water and Sanitation Program (WSP). Disclaimer The findings, interpretations, and conclusions expressed herein We gratefully acknowledge the expert input of Markus are entirely those of the author and should not be attributed to Brauer, Yolande Coombes, Craig Kullmann, Robert the World Bank or its affiliated organizations, or to members Dreibelbis, Francesca Gino, Steve Luby, Hans-Joachim of the Board of Executive Directors of the World Bank or Mosler, Mushfiq Mubarak, Nila Mukherjee, Mike the governments they represent. The World Bank does not Norton, Katherine O’Connell, Julia Rosenbaum, Jan guarantee the accuracy of the data included in this work. The Willemn Rosenboom, Kathleen Vohs, and Viengsamay boundaries, colors, denominations, and other information Vongkhamsao. shown on any map in this work do not imply any judgment on the part of the World Bank concerning the legal status of any The report was designed by Katy Martinez of Ruttle territory or the endorsement or acceptance of such boundaries. Design. Penny Scudder acted as project manager and editor. Copyright Statement The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to the work is given. © 2015 International Bank for Reconstruction and Development/The World Bank Contents Contents I. Introduction ................................................................................. 1 Global Prevalence and Persistence of OD .................................. 1 Current Behavior Change Strategies for OD .............................. 1 A Brief Primer on Habits and Nudges: System 1 Tactics for ........ Behavior Change ....................................................................... 2 How Can System 1 Derail Behavioral Interventions .................... 3 Way Forward: Augmenting OD Interventions with Nudges and Habit Change Tactics .......................................................... 4 Methodology for Arriving at the Principles .................................. 4 II. The 8 Principles ........................................................................... 6 Principle 1: STRATEGICALLY INCREASE/DECREASE THE PHYSICAL ......... AVAILABILITY OF KEY PRODUCTS AND INFRASTRUCTURE ... 6 Relevance to OD: ....................................................................... 6 Principle 2: LEVERAGE CONTEXT CHANGE ............................................... 8 Relevance to OD: ....................................................................... 8 Principle 3: PIGGYBACK ON EXISTING CUES AND BEHAVIORS .............. 10 Relevance to OD: ..................................................................... 10 Principle 4: REDUCE PERCEIVED AND ACTUAL FRICTION FOR NEW BEHAVIOR / ADD FRICTION TO THE OLD .............................. 12 Relevance to OD: ..................................................................... 12 Principle 5: SUPPORT CONTEXT-STABLE REPETITION ............................ 14 Relevance to OD: ..................................................................... 14 Principle 6: ADDING “RITUALIZED” ELEMENTS TO THE INTERVENTION CAN ENHANCE EFFECTIVENESS AND ADVOCACY............... 16 Relevance to OD: ..................................................................... 16 Principle 7: LEVERAGE REMINDERS AND INTENTIONAL CUES ............... 18 Relevance to OD: ..................................................................... 18 Principle 8: LEVERAGE DESCRIPTIVE AND “LOCALIZED” NORMS .......... 20 Relevance to OD: ..................................................................... 20 III. Concluding Remarks ................................................................. 22 REFERENCES ........................................................................... 24 www.wsp.org iii Executive Summary Executive Summary Open defecation (OD) remains a critical global health The 8 principles were developed through an iterative challenge, affecting almost 1 billion people around the process involving (a) thematic coding of field research world and contributing significantly to the estimated findings regarding OD, (b) identification of potential 842,000 people who die each year because of poor behavioral science principles matched to the themes sanitation, hygiene practices, and unsafe water supplies identified in the field research, (c) consultation with 9 (WHO, 2014). sanitation and hygiene experts with extensive experience in OD interventions around the world, and (d) consultation To date, most behavior change frameworks for addressing with 7 academic behavioral scientists with expertise related OD have focused on relatively conscious, “reflective” to each principle. The process culminated with the drivers of behavior, including people’s emotions (e.g., development of a simple summary and activation guidance pride, shame), rational knowledge (e.g., of germ theory), for each principle. social norms, and explicit action plans (e.g., commitments to change; see Sigler, 2014). The 8 System 1 Principles to support the initiation and maintenance of OD behavior change behavior change are Using the framework popularized by Kahneman (2011), as follows: these factors can be described as “System 2” drivers of behavior (i.e., relatively conscious and motivational 1. Ensure critical products and infrastructure are factors). It is now well established, however, that human immediately and consistently physically available for behavior can also be heavily influenced by “System 1” the end user. drivers (i.e. relatively automatic, cue-driven drivers; • Example: Promote latrine construction at secondary Marteau et al., 2012; Wood & Neal, 2015). System 1 locations (transit, markets), so that new latrine use factors of particular relevance to OD include people’s habits are not disrupted. hygiene habits (e.g., mindlessly repeated behaviors cued by context) and “nudges” (i.e., small changes to the 2. Create or capitalize on context change to drive new environment that can channel decision making and behavior of toilet use. behavior in new ways, Thaler & Sunstein, 2008). • Example: Capitalize on seasonal migration patterns or other events that disrupt existing behaviors – In this working paper, we draw on basic scientific findings time interventions to co-occur with these shifts. from psychology, cognitive science, and behavioral economics to propose a framework of 8 System 1 3. Piggyback on other existing behaviors and cues. Principles to support the initiation and maintenance of • Example: Build community latrines that piggyback OD behavior change. In doing so, we build from the on existing established behaviors in a community general framework advanced in the World Bank Group’s (e.g., washing clothes, water gathering). (2015) World Development Report: Mind, Society, and Behavior, which emphasized three core insights from 4. Strategically increase friction for the undesired behavioral science, namely that people think (a) behaviors and lessen it for desired ones. automatically, (b) socially and (c) using mental models • Example: Promote pre-packaged options that channel their decision-making. (e.g., “Easy Latrines” in Cambodia) that simplify the latrine construction process. iv Nudging and Habit Change for Open Defecation: New Tactics from Behavioral Science Executive Summary 5. Support context-stable repetition for latrine use. We emphasize that these 8 principles are meant to • Example: Reward context-stable use of community augment, not replace, approaches based on System 2 latrines (reward repeated use at the same place and thinking. A core insight from the behavioral sciences (see time, at least initially). Kahneman, 2011) is that human behavior is the product of both System 2 thinking (rational, motivated) and 6. Embed ritualized elements in the change process. System 1 thinking (automatic, cue driven habits). Thus, • Example: Integrate OD messaging into already the most challenging behavior change problems will ritualized cultural practices (e.g., “no loo, no bride” invariably require a set of targeted System 1 and System 2 campaign in India). tactics working in unison. 7. Leverage point-of-action reminders and cues. Finally, as field practitioners explore folding these ideas • Example: Create salient cues at typical OD sites to into OD interventions, we encourage the use, where act as reminders that these physical spaces have a practical, of randomized control trials (RCTs), the robust new meaning (e.g., use vermillion powder to measurement of outcome data, and the sharing of ritually cleanse the site). successes and failures alike. In particular, we encourage the sharing of new ways to translate, tailor, and “bring to life” 8. Highlight descriptive and “localized” norms that reduce these basic science principles as makes sense in specific cognitive demands. environments, cultures and sub-populations. • Example: Develop and frame incentive systems in ways that work at the level of a local group (e.g., local village or women’s group), rather than individuals or entire area. www.wsp.org v Foreword Foreword Rich and poor alike, people sometimes act in ways that In that spirit, Nudging and Habit Change for Open undermine their own health and well-being. After all, Defecation could not be more timely and valuable. This humans are creatures of habit, and many of our daily Working Paper tackles one of our most pressing sanitation actions run on auto-pilot with limited conscious thought. challenges—and one that has proven especially resistant to traditional, rational tactics such as information-based Until recently, this aspect of human behavior, often called interventions. By carefully marrying academic findings “System 1 thinking,” was largely off-limits for from behavioral science with field-based insights from development practitioners. We simply knew too little sanitation experts, the report creates a powerful blueprint about the cognitive biases and mechanisms that govern for new intervention tactics that are evidence-based yet System 1 thinking and we lacked practical tools for practical. I look forward to seeing how the eight System 1 applying them to real-world development challenges. principles described in the following pages inspire new progress in creating and maintaining change in open Fortunately, that is beginning to change. defecation practices. In 2015, the World Development Report: Mind, Society Varun Gauri, Ph.D. and Behavior, summarized a wide range of new scientific Head, Global Insights Initiative (GINI) findings on ways to “nudge” and trigger positive behavior The World Bank change by leveraging automatic thinking, social influence, and mental models. As the WDR showed, these approaches show high potential in shifting behaviors as diverse as corruption, parental caregiving practices, household savings, and conservation, among many others. vi Nudging and Habit Change for Open Defecation: New Tactics from Behavioral Science I. Introduction Global Persistence of described as “System 2” drivers of behavior (i.e., relatively Open Defecation (OD) conscious and motivational factors). Approximately 2.4 billion people around the world lack access to improved sanitation facilities and just under 1 As an example of a primarily System 2 approach to OD, billion people engage in open defecation (OD; WHO/ Community-Led Total Sanitation (CLTS) interventions UNICEF 2015). Poor sanitation, in turn, drives a range of engage the community in a process of conscious self- diseases, including diarrhea, trachoma, and soil-transmitted realization that leverages shame, pride, and concrete action helminth infections (Pruss-Unstun et. al 2008). Diarrhea planning to move people towards OD free (ODF) status alone has a devastating impact on child morbidity and (Sigler, 2014). Similarly, the SaniFOAM behavior change mortality, especially in low-income settings; it accounts for framework (Devine, 2009) focuses primarily on identifying approximately 800,000 deaths of children under 5 years of factors related to opportunity, ability and motivation age each year (Liu et al., 2012). Although the challenges that influence sanitation behaviors such as OD. These can associated with OD are global in nature, they are especially be regarded as principally, although not exclusively, prevalent in India, which accounts for roughly two-thirds of System 2 strategies. the global OD population and one-third of those without improved sanitation facilities (WHO/UNICEF 2015). These approaches have achieved some marked success and they incorporate the very best tactics currently known to In 2000, The United Nations Millennium Development promote OD change. However, even with these tools, Goals (MDGs) established a target of halving the global many attempts to alter OD in the field still fail, or achieve population that lacks access to safe water and improved only short-term success that does not “stick” or maintain sanitation. Between 1990 and 2015, approximately 2 over time (see Sigler, 2015). billion people gained access to improved sanitation, but OD practices have proven especially difficult to change In this report, we propose that further advances can be (Millennium Development Report, 2014). The sanitation made by incorporating new insights about the role that component of the MDG is currently expected to fall short habits, nudges and other “System 1” drivers (i.e., relatively by at least 500 million people (WHO, UNICEF 2014). automatic and non-conscious factors) play in supporting and sustaining behavior change (Marteau et al., 2012; Current Behavior Change Strategies for OD Thaler & Sunstein, 2008; Wood & Neal 2015). In doing The persistence of OD is complex and multi-determined, so, we build from the general framework advanced in encompassing a mix of supply-side issues (e.g., access to the World Bank Group’s World Development Report latrines, affordable building materials) and demand-side, (WDR, 2015), which emphasized three core insights or “user-centered,” issues (e.g., cultural and religious beliefs, from behavioral science, namely that people think (a) relative convenience and affordability of OD; for a recent automatically, (b) socially and (c) using mental models that review, see O’Connell, 2014). channel their decision making. To date, most of the demand-side interventions and In the following pages we propose 8 high potential System frameworks for OD have emphasized relatively conscious, 1 Principles to support the initiation and maintenance “reflective” drivers of behavior change, including people’s of OD behavior change. In brief, these principles were emotions (e.g., pride, shame), their rational knowledge generated through a multi-stage process beginning with a (e.g., awareness of germ/fecal matter transmission), social review of qualitative and quantitative research findings from norms, and explicit action plans (Sigler et al., 2014). the OD field literature. We then mapped recurring themes Popularized by Kahneman (2011), these factors are often from the field literature to evidence-based behavior change www.wsp.org 1 Introduction tactics from behavioral science, including social psychology, These processes are actively studied in multiple academic behavioral economics, and cognitive science. The principles fields, including social psychology (e.g., Kahneman, 2011; were then vetted, refined, and given executional detail Wood & Neal, 2007), behavioral economics (e.g., Thaler & through structured interviews with a team of 7 academic Sunstein, 2008), cognitive neuroscience (e.g., Marteau et experts and 8 sanitation and hygiene experts. Our process is al., 2012) and health psychology (Rothman et al., in press). described in more detail in the Methodology section below. Each of these disciplines has its own nuances and terminology, but all of them emphasize that drivers of In the following pages, we provide a brief overview of human behavior can be grouped into: System 1 versus reflective System 2 thinking, focusing on habit change and nudging. We then describe each of the 8 • A more conscious, goal-directed system that System 1 Principles, including the supporting basic science uses decision-making to direct behavior towards and examples of successful applications in real world emotionally and motivationally valued outcomes. settings. We also unpack the ways these principles (a) map This system is called System 2. to consistent patterns seen in OD behavior around the world, (b) inspire specific new OD intervention tactics, and • A more automatic, cue-driven system that uses (c) can be folded into existing program activities commonly familiar behavior patterns, signals from the used in OD behavior change. environment, and simple decision rules (e.g., heuristics) to guide behavior. This system is called A BRIEF PRIMER ON HABITS AND NUDGES: System 1. SYSTEM 1 TACTICS FOR BEHAVIOR CHANGE In recent decades, great advances have been made in the Most of the time, these two systems work together scientific study of behavior change. One key innovation harmoniously and efficiently (Kahneman, 2011). System 2 has been the discovery that many factors can significantly allows us to consciously monitor and carefully establish new impact people’s behavior, yet bypass their conscious behaviors, and ensures that these new behaviors meet our decision-making, attitudes, goals, and awareness (e.g., goals. Once a stable behavioral pattern is set up and Ariely, 2009; Thaler & Sunstein, 2008; Wood & Neal repeated, we spontaneously engage System 1, which allows 2007; WDR, 2015). us to redirect our limited attention, willpower, and goal setting elsewhere. BOX 1: HABITS AND NUDGES Sometimes, however, System 1 and System 2 come into conflict and push behavior in different directions. This • Habits Frequent, learned behavioral responses phenomenon is powerfully illustrated by behavior that are cued automatically by context cues, such prediction studies. Typically, these studies focus on a as physical settings and preceding actions in a se- specific behavior (e.g., seat belt use) and measure the quence (e.g., morning bathing sequence, food prep- strength of relevant System 2 factors (e.g., people’s goals/ aration habits, daily travel. intentions to wear a seat belt) and the strength of relevant System 1 drivers (i.e., their habits of wearing/not wearing • Nudges Environmental cues that signal a desired a belt). The critical question then becomes: which system response from the end user or channel their decision is the biggest predictor of what people actually do in the making (e.g., placing fruit at eye level to encourage future? In a meta-analysis of many such studies, Ouellette consumption, changing defaults so that people have and Wood (1998) found the striking pattern depicted to deliberately opt-out of healthy behaviors). in Box 2. 2 Nudging and Habit Change for Open Defecation: New Tactics from Behavioral Science Introduction BOX 2: IMPACT OF HABITS VS. INTENTIONS ON FUTURE BEHAVIOR Things we do rarely or in different Things we do often and in the environments… same environment… Intentions / Attitudes .62 Intentions / Attitudes .27 FUTURE FUTURE BEHAVIOR BEHAVIOR Habit Strength Habit Strength .45 .12 Box 2. Behavior prediction pattern reported in Ouellette and Wood’s (1998) meta-analysis. Numbers reflect correlation coefficients (r values). Habit strength reflects the frequency and context stability of the behavior in past performance. Intentions/attitudes reflect people’s stated preferences about what they wish to do in the future. Critically, for behaviors that people perform infrequently or then defecate), control of the behaviors generally will have in different settings, System 2 drivers, such as attitudes and shifted away from System 2 to System 1. For this reason, intentions, are stronger predictors of their future behavior. interventions that focus primarily on System 2 may often However, behaviors that are performed frequently in the have limited behavioral impact. The intervention will fail same setting are better predicted by habit strength, which to disrupt because System 1 is in charge of the behaviors belongs to System 1. This is because System 2 loses its and the intervention is targeting the wrong system. influence for behaviors that people have performed Demonstrating this, Webb and Sheeran’s (2006) meta- frequently and in the same setting and way each time. analysis of 47 studies found that interventions targeting intentions are generally effective at changing behaviors that As we explain next, these insights set the stage for a deeper people perform infrequently (e.g., blood donation) but are understanding of how System 1 can derail current generally ineffective at changing habits (e.g., seat belt use). intervention efforts to stop OD, and opportunities that exist to turn System 1 from a liability to an asset in OD System 1 can also cause relapse—or stickiness failures—for behavior change. interventions that initially succeed in changing behavior. Oftentimes, an intervention will temporarily change HOW CAN SYSTEM 1 DERAIL BEHAVIORAL people’s behavior, but this change does not last and people INTERVENTIONS? shortly return to their old behavior (Volpp et al., 2008). As practitioners and researchers know, it is difficult to get Why does this happen? Learning and memory research people to change their behavior (Webb & Sheeran, 2006), shows that System 1 habits, even when changed, tend not and even more difficult to maintain that change over to be forgotten. Instead, they become dormant in people’s time (Volpp et al., 2008). Unsuccessful interventions can memory and can be revived relatively easily even after generally be classed into those that fail to disrupt behavior significant time has passed (Bouton, 2000). Thus, people’s at all (i.e., they achieve no measurable behavior change) and short-term successes at changing their behavior can fail to those that initially change behavior but the changes fail to stick because habits re-exert themselves over time, causing stick (i.e., initial behavior change gives way to “relapse”). relapse to old ways of acting (Tobias, 2009). This does not mean that old habits never die. However, they are System 1 can play a strong role in both of these types of remarkably resilient and can re-emerge rapidly when cues failures. As we saw above, when people frequently repeat a associated with those habits are present. behavior in the same setting (e.g., defecation outside) and/ or in the same action sequence (e.g., wake up, then walk, www.wsp.org 3 A WAY FORWARD: AUGMENTING OD internal documents from the World Bank’s Water and INTERVENTIONS WITH NUDGES AND HABIT Sanitation Program) addressing OD behavior, and latrine CHANGE TACTICS construction, usage and maintenance. We reviewed studies To summarize, the System 2 tactics that work for new that reported on specific OD interventions (e.g., CLTS) or infrequently performed behaviors generally will not as well as studies that were purely descriptive (i.e., did not successfully disrupt and stick where System 1 is in charge. involve interventions). This process generated a database Therefore, these tactics need to be augmented with tactics of over 340 specific findings from the field. At Step 2, we that target System 1. In the remainder of this document we thematically coded the field insights to identify instances introduce 8 principles that have high potential to address where habits or other System 1 processes might plausibly System 1 and create disruptive and lasting change in OD underlie, or contribute to, an observed finding. Coding was behavior. Box 3 below summarizes these principles. performed independently by three expert coders and any inter-coder differences were resolved through discussion. METHODOLOGY FOR ARRIVING AT THE PRINCIPLES At Step 3, we used the behavioral science literature— including social psychology, health psychology, cognitive We arrived at the 8 principles via 5 steps. At Step 1, science, and behavioral economics—to identify high we collated the peer-reviewed literature (PubMed) potential strategies for influencing the common themes and grey literature (formative research reports and that emerged at Step 2. As part of Step 3, we also consulted BOX 3: 8 SYSTEM 1 PRINCIPLES FOR OD BEHAVIOR CHANGE Box 3. 8 Principles for Leveraging Nudges and Habit Change Tactics (System 1) to Support OD Behavior Change. 4 Nudging and Habit Change for Open Defecation: New Tactics from Behavioral Science a group of basic science experts from academia who study activation plan for each principle, by unpacking examples various System 1 tactics for behavior change (see Table of successful implementation in other domains as well as 1 below) This process generated a preliminary list of identifying potential links to existing OD change tactics approximately 40 principles. and hypotheses about genuinely new actions that could be taken. These activation ideas were refined in consultation At Step 4, we consulted nine field experts (see Table 2) with the academic and field experts. with extensive knowledge of OD behavior and intervention approaches in various regions (e.g., India, South East Asia and Africa) and settings (e.g., urban, rural, river). At this point, we refined the list of principles to 8 with the highest potential. Finally, at Step 5, we created an TABLE 1: ACADEMIC EXPERTS CONSULTED AT STEP 3 TABLE 2: OD, SANITATION AND HYGIENE EXPERTS CONSULTED AT STEP 4 Academic Experts Areas of Expertise OD, Sanitation Affiliation Prof. Markus Brauer Social psychology, attitude & Hygiene Experts University of Wisconsin change, social influence, Independent Yolande Coombes Consultant/World Prof. Robert Dreibelbis Sanitation and hygiene Bank Group University of Oklahoma behavior change Craig Kullmann World Bank Prof. Francesca Gino Psychology, behavioral Harvard University economics, ritual creation Steve Luby Stanford University Independent Prof. Mushfiq Mobarak Sanitation and hygiene Nila Mukherjee Consultant/World Bank Yale University behavior change Hans-Joachim Mosler EAWAG Prof. Mike Norton Social psychology, behavioral Harvard University economics, nudging Independent expert/ Katherine O’Connell World Bank Prof. Kathleen Vohs Social psychology, cognitive Julia Rosenbaum FHI360 University of Minnesota science, behavior change The Bill & Melinda Jan Willem Rosenboom Prof. Wendy Wood Gates Foundation Social psychology, health University of psychology, attitude change Southern California Viengsamay Vongkhamsao World Bank www.wsp.org 5 The 8 Principles II. The 8 Principles Principle 1: HOW TO EXECUTE/EXAMPLES OF APPLICATION Strategically increase/decrease the physical availability of key products and infrastructure Supporting new behavior: • The consistent physical availability of soap Principle Overview: within easy reach has been found to be critical to the formation of a new handwashing with soap As a necessary, but not sufficient condition for change practice (Luby, 2009). - increase the likelihood that supporting products/ • “Lucky iron fish” infrastructure for latrine use are consistently and (picture to right) used immediately physically available in the environment to treat iron defi- (without active searching/effort). Alternatively, or in ciency (http://www. addition, decrease the physical availability of products/ luckyironfish.com). infrastructure needed for OD. Fish is highly “physi- cally available” – can Basic Science: be kept in cooking pot and removed just before food Habits are automatically triggered by context cues, is added to automatically dose with iron. Also lever- including physical settings, preceding actions, and times of ages Principle 3 (piggybacking), 4 (friction) and 5 day. If these critical cues are not consistently present in the (context stable repetition). environment (without active seeking/effort), the desired Undermining existing (unhealthy) behavior: habit will not occur unless motivation is extremely high • Banning visual display of cigarettes (which reduced (Wood, Tam, & Witt, 2005). the physical availability of cigarettes) at point-of-pur- chase is effective at reducing impulse/habitual cigarette Relevance to OD: purchases (Wakefield, Germain, & Henriksen, 2008). Principle 1 captures the basic idea that certain physical structures and products (e.g. latrines near fields for farm Report, 2007). OD while working in agricultural fields workers) need to be consistently and easily available to is considered typical behavior since there is no access people if they are to have any chance of changing their to sanitation facilities and going home to use a toilet behavior. Importantly, these physical features will be was perceived as a waste of time (Qualitative Report for necessary but not sufficient to initiate behavior change. For Understanding Rural Sanitation, Bihar, 2012). example, findings from a global review of WSP initiatives in rural settings indicate that, in many countries, people Several opportunities to leverage Principle 1 emerged in who have easy physical access to a latrine at home still the literature we reviewed. First, OD often becomes engage in OD (O’Connell, 2014). However, maximizing difficult to practice (i.e., OD is physically less available) the consistent physical availability of enabling products during rainy season when rain is consistent, roads or remains a critical first step for new latrine-use habits to fields are flooded, dry space is constrained, and insects form. Where possible, latrines need to be available not are more prevalent (WSP reports from Kenya, Indonesia, only at home, but in other contexts and times of day that Meghalaya, Rajasthan, Bihar). Illness, disability and old age are a part of daily life, such as at work, near crop fields, in are also often cited as circumstances when OD is difficult public places like markets and in the homes of others. For or impossible to perform (WSP reports from Meghalaya, example, in rural Cambodia, 2% of adults with access to Rajasthan, Bihar). Safety at nighttime, and from wild latrines at home reported defecating in the open while at animals were also cited in a few reports as deterrents of home but when outside of the home, 43% of this same OD, especially among women (WSP reports from Kenya, group practices OD (Cambodia WSP Demand Assessment Indonesia, Meghalaya, Rajasthan, Bihar). 6 Nudging and Habit Change for Open Defecation: New Tactics from Behavioral Science The 8 Principles Physical Availability Challenges Physical Availability Opportunities • Lack of consistently available sanitation facil- • Rainy season reduces physical access to OD, ity (alternative not physically available) at home creating opportunity to shift people to latrines. and during work creates “gaps” in availability of • Illness temporarily reduces physical access latrines. to OD and could be leveraged too. • Lack of availability of masons and materials to • Old age reduces access to OD. build, maintain and improve latrines. • Safety (night time, wild animals) reduces • Structural soundness of latrines (risk of collapse access to OD. in rainy season), reducing physical availability of functional latrines. • When pit is full and not easily emptied, latrine becomes “unavailable”, causing relapse to OD. • In many contexts, plastic bags are highly physi- cally available, leading to use of “flying toilets”. PROGRAMMATIC IMPLICATIONS 1: MAPPINGS TO RECENT INNOVATIONS/TACTICS • New product innovations • Launching Micro-Finance • Examine the potential in a that ensure OD alternatives Initiatives to support latrine “total solid waste” strategy are consistently, easily readily purchase at the same time where all waste/trash is in a at hand (e.g., PeePoo bags as as OD interventions to ensure given setting is removed (not alternatives to “flying toilets”; alternatives to OD are consis- just sanitation). Thus, the http://www.peepoople.com). tently, immediately at hand. physical availability of all waste is radically altered in an • Ensuring latrines are environment at one time. See constructed in contexts experience in Rwanda, where beyond the home (e.g., a total solid waste strategy transit points, markets, may be demonstrating value schools, workplace). in shifting OD practices. PROGRAMMATIC IMPLICATIONS 2: NEW IDEAS AND EXTENSIONS • Micro-Finance Initiatives • During OD mapping altered and less physically that support keeping public activities in CLTS interventions, available. toilets open later at night and opportunities may exist to earlier in the morning (thus reduce the physical availability • Construct public toilets increasing physical availability), of common OD locations in before initiating CLTS to when operators would normally the village (e.g., repurposing ensure availability of toilets close them because they are common OD sites for an while households construct not sufficiently profitable. Thus, alternative use, barricading usual their toilets. removing disruptions in the OD sites) so they are physically physical availability of latrines. www.wsp.org 7 The 8 Principles Principle 2: LEVERAGE CONTEXT CHANGE Principle Overview: Relevance to OD: Disruptions to the physical environment and/or familiar One challenge for Principle 2 is that the physical context action sequences create a “window of opportunity” for a and other daily activities before and after an intervention new habit to form. like CLTS triggering remain mostly the same. However, we found several examples from the OD literature that may Basic Science: present windows of opportunity to leverage Principle 2. As When people undergo major shifts in context or life described under Principle 1, seasonality (especially rainy circumstances, their existing habits are temporarily season) presents a yearly context change were OD becomes vulnerable to change (see Rothman et al., in press). difficult to practice and presents an opportunity for more Context change can include major shifts in the external consistent latrine use. environment (e.g., moving to a new house or area), but even small context changes can sometimes be enough to The WSP reports highlight several life stage changes, such change behavior (Neal, Wood, Wu and Kurlander, 2011). as illness, birth, marriage or receiving a new daughter- Thus, a wide array of context shifts can be useful entry in-law, having elder or ill household members, as points to support the initiation of a new behavior. circumstances where having a latrine is advantageous and OD temporarily becomes a disadvantage. For example, those who adopted toilet use reported that marriage of a HOW TO EXECUTE/EXAMPLES OF APPLICATION son and having a new daughter-in-law was a trigger for adopting a toilet to avoid humiliation and shame (WSP • Providing free public transport vouchers is more report Bihar, 2012). Among this report and others, survey effective when the vouchers are sent to people respondents also indicated that having visitors to their who have recently moved house because their home, especially those that are accustomed to using toilets, car-use habits are temporarily disrupted was a motivator for toilet adoption to avoid shame or to (Verplanken, Walker, Davis, & Jurasek, 2008). improve social status (WSP reports Bihar, Meghalaya, Malawi). 8 Nudging and Habit Change for Open Defecation: New Tactics from Behavioral Science Context Change Challenges Context Change Opportunities • Physical context and other daily activities remain • Seasonal changes (physical context). mostly the same pre/post CLTS triggering. • Life stage changes, illness, birth (i.e., many new action sequences), marriage or receiving a new daughter-in-law, having visitors, having elders or those who are sick in the home. PROGRAMMATIC IMPLICATIONS 1: MAPPINGS TO RECENT INNOVATIONS/TACTICS • Capitalize on “circular migration” patterns to disrupt OD practices while people are temporarily away from their home environment (e.g., seasonal worker migration, major holiday migration such as Pchum Benh in Cambodia). They will be more likely to change when away from home and may bring their new “latrine habits” back to their home environment. See Chowdhury, Guiteras & Mobarak (2015). PROGRAMMATIC IMPLICATIONS 2: NEW IDEAS AND EXTENSIONS • Find mechanisms to deliver • Strategic timing of • Build new strategies that interventions when they will interventions so they occur can be deployed rapidly coincide with major changes during or immediately during disease outbreaks in physical context or life after large context changes, (e.g., Cholera) when people’s stage – e.g., promote OD especially changes that existing behaviors are altered change through trained birth impact cues directly involved (e.g., they stop shaking hands assistants, midwives, and in OD (e.g., rainy season and switch to the “Cholera antenatal clinics, etc. altering accessibility of handshake”) and they are habitual OD sites). actively thinking about hygiene. www.wsp.org 9 The 8 Principles Principle 3: PIGGYBACK ON EXISTING CUES AND BEHAVIORS Principle Overview: Relevance to OD: Cues and behaviors that are already well established in We found a number of challenges for Principle 3 in the people’s daily practices can be linked to the new, desired OD field research. In several states in India, OD is part of behavior. This increases the likelihood that the new a morning routine that includes a walk that is perceived behavior is performed and is often more effective than as pleasant and advantageous for overall well-being. OD trying to insert a new behavior in isolation from existing thus “piggybacks” on daily rituals of a time to walk, practices. check on crop fields, and socialize. For those working in agriculture, especially in rural settings, OD is perceived as Basic Science: typical behavior, natural or a part of the job since human Instead of creating a new, desired behavior “from nothing,” feces is commonly perceived as a fertilizer for crops. In it can be more effective to attach the behavior to an existing communities near rivers or bodies of water, OD may physical cue in the environment, or an existing behavior, piggyback on familiar or pleasant cues such as the feel or that is already established in people’s daily practice or sound of water or instant removal of waste. cultural understanding (Judah, Gardner, & Aunger, 2013). There are potential opportunities to leverage Principle 3 from existing behaviors or activities. For example, HOW TO EXECUTE/EXAMPLES OF APPLICATION existing community routines and daily practices around good hygiene could be leveraged to connect latrine use Piggybacking on existing behavior: to proper hygienic behavior. In Kenya, people who • It is more effective to teach children to floss after maintain good hygiene are perceived to be healthier, brushing (rather than before brushing) because this happier and confident, and are considered role models in sequence piggybacks the new behavior (flossing) some communities (WSP report Kenya, 2013). Religious onto an existing habit (brushing). The existing habit, or moral principles could also serve as piggybacking thus, becomes a cue to automatically perform the opportunities. Bundling latrine construction with other new behavior (Judah, highly desired improvements in house may also be an Gardner, & Aunger, 2013). opportunity to piggyback on decisions or actions that have already gained household momentum (Jenkins, 2005) or Piggybacking on are existing daily habits (see Programmatic Implications physical cues: below). • Mrembo wash station (picture to right) has a mirror installed on the front. The mirror is aspirational and triggers mirror-checking behavior, which causes people to engage with the wash station. 10 Nudging and Habit Change for Open Defecation: New Tactics from Behavioral Science The 8 Principles Piggybacking Challenges Piggybacking Opportunities • OD piggybacks on morning routine/walk. • Opportunity to piggyback on existing behavioral habits around good hygiene where those already • OD is inherent to some jobs such as working the exist (e.g., connect latrine use to other hygiene crops and using human feces as fertilizer. behaviors that may be more established in that community – washing clothes, bathing). • OD piggybacks on daily ritual of “time to talk, walk and socialize”. • Opportunity to piggyback latrine construction on other home improvements. • River OD involves many familiar/pleasant cues (feel of water, instant removal of waste etc). PROGRAMMATIC IMPLICATIONS 1: MAPPINGS TO RECENT INNOVATIONS/TACTICS • If ritualized socializing is part of the OD habit • Piggyback/bundle latrine construction in a community, explore construction of café/ and upgrades onto other, already established shaded area for socializing new public toilet construction activities (such as annual roof to maintain existing habit (see Principle 6 also). repairs or applications to have a water supply For example, see the “Bloc Sanitaire” connected to home). experience in Madagascar and Ethiopia. • WaterAid innitiative in Nepal, piggyback hygiene interventions onto successful immunization programs. PROGRAMMATIC IMPLICATIONS 2: NEW IDEAS AND EXTENSIONS • Build community latrines that also support • Piggyback voucher systems for latrine other, already established daily routines such construction onto antenatal care visits. as water gathering, washing of clothes, or socializing. Latrine use can then piggyback on these established, daily behaviors. www.wsp.org 11 Principle 4: REDUCE PERCEIVED AND ACTUAL FRICTION FOR NEW BEHAVIOR / ADD FRICTION TO THE OLD Principle Overview: Basic Science: Eliminate even minor amounts of choice, effort, and When a new behavior requires even small amounts of decision-making or “friction” required to perform the effort, decision-making or added steps (especially compared new (desired) behavior and/or add friction to the existing to the status quo), relapse to old ways of acting will be (undesired) behavior. much more likely (Murray & Häubl, 2007). Conversely, the likelihood of disrupting existing (undesired) behaviors can be increased by adding friction. HOW TO EXECUTE/EXAMPLES OF APPLICATION Relevance to OD: Removing friction from desired behaviors The convenience—or lack of friction—associated with OD • New medication management systems provide is a commonly cited factor that maintains the behavior. patients with their In India, for example, common OD locations, such as personal daily agricultural fields while working, are viewed as easily medications pre- accessible and natural places to practice OD. Rivers or sorted, thus removing bodies of water that are easy to access make OD simple, much of the friction convenient and easy to dispose of waste (O’Connell, (decision-making, 2014) and thus pose minimal friction for practicing OD. multiple steps) usually Perceived and actual maintenance requirements of latrines required to adhere to a can also generate friction to building or using latrines since medicate regimen. cleaning, repairing, and maintaining the superstructure and emptying the pit are commonly cited disadvantages Adding friction to undesired behaviors or barriers to latrine use or construction (Tyndale Biscoe, • When smoking 2013, WSP report for Kenya, Malawi, Meghalaya). bans were introduced in UK Despite these challenges, there are opportunities to pubs, people leverage Principle 4 in development of sanitation products. with strong habits Sanitation facilities that exhibit desirable qualities, such to smoke while as being easy to use, easy to maintain and clean, easy to drinking were no access, usable at all year and times of day, could decrease longer able to the perceived or actual friction to using a latrine. In effortlessly light a addition to physical or structural attributes, behavioral or cigarette when they emotional components can be leveraged to increase friction felt the urge. The for undesirable behavior, such as OD. For example, the resulting behavioral shame of a long walk to engage in OD can be used to cause friction (needing to friction for practicing OD. leave the pub to smoke) is thought to have disrupted the automated association between drinking and smoking and, in turn, helping to reduce smoking rates (Orbell & Verplanken, 2010). 12 Nudging and Habit Change for Open Defecation: New Tactics from Behavioral Science The 8 Principles Friction Challenges Friction Opportunities • The process of building a latrine is often • Current OD behavior sometimes involves multi-step, requiring a range of products sourced significant effort/friction (e.g., having to carry from different locations. This creates many water for anal cleansing). This friction could opportunities to abandon the process. Common be exploited to drive latrine use (e.g., see OD locations are easily accessible/lack friction Programmatic Implications below). (e.g., fields while working). • River OD is especially “frictionless” because the experience is pleasant/waste “disappears”. • Latrine maintenance generates friction through cleaning/pit emptying. This friction may trigger relapse to OD. PROGRAMMATIC IMPLICATIONS 1: MAPPINGS TO RECENT INNOVATIONS/TACTICS • Development of “Easy • Include desirable, “easy” exlcusive breastfeeding in Latrines” in Cambodia, product attributes (close Vietnam. Communications which pre-packages all proximity, usable all year and were then targeted narrowly required elements for a all times of day). and successfully to change working latrine in an easy- this specific SDA (see to-self-install design. • Identify the “smallest Jimerson, 2016). do-able action” (SDA) that is • Offering routine pit emptying easy to implement and will services that reduce the friction have the largest impact on associated with maintenance, the key outcome (e.g., Alive & thus reducing the likelihood Thrive/FHI 360 identified “do that users abandon the latrine not give water” as the highest and revert to OD. impact SDA to promote PROGRAMMATIC IMPLICATIONS 2: NEW IDEAS AND EXTENSIONS • Explore the potential of simple heuristics/ • In locations where anal cleansing with water rules that increase the distance people feel they is common, develop latrines that provide easy need to walk to engage in OD (e.g., a simple filling of water pots, making latrine use less “1000 step” rule would be easy to remember burdensome than carrying water for OD. and may add significant friction to OD). www.wsp.org 13 The 8 Principles Principle 5: SUPPORT CONTEXT-STABLE REPETITION Principle Overview: Relevance to OD: Where possible, encourage context-stable repetition. Several of the WSP reports indicated that seasonal changes, Context-stable repetition can include repeating the desired especially rainy season, affect the OD practice. Seasonal behavior in the same physical setting, at same time of day, changes present challenges for Principle 5 because the or in same action sequence. This can be enhanced through context in which sanitation behaviors are performed are implementation intentions. not stable (changes in access, availability, comfort, safety). Other challenges to a stable context for latrine use are Basic Science: poor latrine construction (Tyndale, Biscone, 2013) and Habits are more likely to form when people not only repeat degradation of the latrine structure. Both change the frequently, but repeat in the same setting, at the same time context for latrine use and disrupt consistent practice or of day and/or in the same action sequence (Neal, Wood, revert back to OD. For example, a market assessment for Labrecque, & Lally, 2012). This kind of repetition can WSP in rural Malawi identified latrine durability (through be promoted by implementation intentions instructions, seasonal changes and natural degradation) as an important which require people to visualize/specify the particular barrier to latrine construction and reconstruction (WSP context in which they will perform the new/desired Market Assessment for Rural Sanitation in Malawi, 2011). behavior (Gollwitzer & Shearan, 2006). To support context-stable repetition, the literature we HOW TO EXECUTE/EXAMPLES OF APPLICATION reviewed highlighted opportunities around sanitation product design and the context in which a sanitation facility would • If people are be placed. Since proximity of the latrine was a commonly asked to form 5 cited barrier to use (mainly to due inconvenience or a specific “if-then perceived waste of time to go to a latrine that was further links” to support away than defecate in the open), placement of latrines near fruit/vegetable the home but in an acceptable or culturally appropriate consumption, they place (e.g., away from sites of worship) could support will be significantly context stability. The WSP reports consistently indicated more likely to that convenience, cleanliness, comfort, privacy, easy carry out those maintenance, durability, and functionality through seasons intentions/change are key motivators for latrine adoption. While latrines their behavior designed with a user-centered approach support a desirable (see Harris et al., context for latrine adoption (Jenkins, 2005), they may also 2014). The if-then links must specify the particular support repetitive use in a desirable and stable context. context or situational triggers for the behavior (e.g., respondent writes down: If I eat out during the day, then I will have a banana after my food). 14 Nudging and Habit Change for Open Defecation: New Tactics from Behavioral Science The 8 Principles Context Stability Challenges Context Stability Opportunities • Poor latrine construction/need for frequent • Latrine construction near home with maintenance reduces the opportunity to practice desirable attributes. context-stable usage of latrines. • Seasonal changes can reduce context stability • Variable, limited opening times for public toilets of OD because people cannot engage in acts as barrier to context-stable usage. the behavior at all times or at typical places. PROGRAMMATIC IMPLICATIONS 1: MAPPINGS TO RECENT INNOVATIONS/TACTICS • Develop incentive systems that reward users for using latrines at the same time of day and same location (vs. incentivizing usage regardless of time and place). Reward systems of this kind are being trialed via membership cards that work via RFID technology (which allows people’s use of the latrine to be passively tracked without needing to swipe a card). PROGRAMMATIC IMPLICATIONS 2: NEW IDEAS AND EXTENSIONS • Examine the potential of “community toilets” (which may prioritize usage by many users, but few use with sufficient frequency to form lasting habits). We note that, although community toilets have high potential from a habit change perspective, they also carry unique management challenges. www.wsp.org 15 Principle 6: ADDING “RITUALIZED” ELEMENTS TO THE INTERVENTION CAN ENHANCE EFFECTIVENESS AND ADVOCACY Principle Overview: Relevance to OD: Where appropriate, promote inclusion of “ritualized From the qualitative reports in India, we know that certain elements” in the new behavior or intervention process to rituals can support OD. For example, the WSP reports promote adoption. from communities in Bihar and Rajasthan reported OD as a part of ritualized morning walks which were considered Basic Science: pleasant, beneficial and supportive of well-being and When behaviors become ritualized (i.e., linked with deep good bowel movement (WSP reports Bihar, Rajasthan). personal or culturally relevant meaning) they can (a) However, creating new rituals could support better attract more attention/psychological engagement, (b) be sanitation behaviors. For example, creating new rituals seen as more credible, (c) emotionally bind people to around purification or pride during public declarations of each other around group-based values, and (d) be more ODF status can be an opportunity to leverage Principle 6. likely to be socially transmitted (Lienard & Boyer, 2006; Rosano 2012). HOW TO EXECUTE/EXAMPLES OF APPLICATION In Bohol, Philippines, local fishermen were using dynamite and cyanide to fish, leading to the rapid Including a meaningful ceremonial component to an destruction of the aquatic ecosystem. To curb this action can elevate it to the status of a ritual, increasing behavior, statues of the Virgin Mary and other religious the likelihood that people perform the behavior. figures were submerged around the reef system, Opportunities can be found to create cues/visible instantly bringing the behavior into conflict with deeply signs of the ritual (e.g., ink on finger in elections). held, religious and ritualized cultural themes. 16 Nudging and Habit Change for Open Defecation: New Tactics from Behavioral Science The 8 Principles Ritual-Building Challenges Ritual-Building Opportunities • Ritualized morning walks, which include OD, are • Opportunity to create rituals around purification/ viewed as positive and pleasant. pride and public declarations around ODF. • “Filth” near home/close to religious or sacred sites considered is already considered morally wrong – this brings latrine use into conflict with religious rituals. PROGRAMMATIC IMPLICATIONS 1: MAPPINGS TO RECENT INNOVATIONS/TACTICS • If ritualized socializing is part of the OD habit • UNLI Rural Sanitation program in Phillipines in a community, explore construction of developed a simple hand gesture to café/shaded area for socializing new public toilet remind people/serve as a ritualized mnemonic to incorporate OD into an existing ritual. about the campaign. PROGRAMMATIC IMPLICATIONS 2: NEW IDEAS AND EXTENSIONS • In CLTS interventions, test ways to add a • Integrate ODF thinking into major cultural plausible ritual element at the end of the transect rituals including holidays, festivals (e.g. walk/ODF pledge. This ritual component could reinforce the “no loo, no bride” association symbolically capture the “end of OD” and the around weddings). transition point to ODF. Engage local religious/ traditional leaders to devise appropriate, context-specific rituals (e.g., Hindus throwing sindoor/Vermilion powder over common OD sites to signal ritualistic purification). www.wsp.org 17 The 8 Principles Principle 7: LEVERAGE REMINDERS AND INTENTIONAL CUES Principle Overview: If people intend to engage in a behavior, remind them of HOW TO EXECUTE/EXAMPLES OF APPLICATION the behavior periodically, especially when they are in the appropriate context. Telephone, email, text message, and postal reminders for numerous behaviors, including diet, Basic Science: physical activity, medication adherence and Following through on an intention requires a person smoking. Physical signs placed in locations where to remember the new behavior, but it is easy to forget the problem behavior is likely to occur. Signs can or neglect it. Reminders, especially in that appropriate leverage culturally powerful imagery. context, can mitigate forgetting (Cole-Lewis, & Kershaw, 2010; Elder, Ayala, & Harris, 1999; Fry, & Neff, 2009). Relevance to OD: The aforementioned rituals or existing routines can reinforce poor sanitation behaviors. For example, morning walks that include OD among communities in certain Indian states can serve as a cue or reminder to carry out typical behavior (OD). Other cues within that context, like the time of day or a sunrise, may also be cues for OD practice. Reminder & Cue Challenges Reminder & Cue Opportunities • Existing routines (e.g., the morning walk) • Opportunity to piggyback on religious cues that can serve as cue/reminder to carry out typical signal OD is not acceptable in immediate behavior (OD). environment (e.g,, use of Arabic on walls near common OD sites in Bangladesh). • Time of day (e.g., sunrise) can serve as cue/reminder to OD behavior. 18 Nudging and Habit Change for Open Defecation: New Tactics from Behavioral Science PROGRAMMATIC IMPLICATIONS 1: MAPPINGS TO RECENT INNOVATIONS/TACTICS • Incentive programs for latrine use can leverage text message reminders and can do so using messaging schedules that are timed and framed to promote latrine usage at the same time and place each day. PROGRAMMATIC IMPLICATIONS 2: NEW IDEAS AND EXTENSIONS • Clearly highlight traditional • To propote ODF status, • To propote ODF status, OD places with adverse cues adapt interventions based adapt handwashing (e.g., Hindus throwing sindoor/ on the RANAS model that interventions that place colored Vermilion powder over OD sites combine public commitments footsteps from latrines to wash - see also Principle 6). to wash with soap with stations (see Dreibelbis et al., distinctive head scarves that 2016), by painting colored act as a stable cue to remind footsteps from common OD people about adherence sites to public toilets. (Contzen, Meili & Mosler, 2015). www.wsp.org 19 The 8 Principles Principle 8: LEVERAGE DESCRIPTIVE AND “LOCALIZED” NORMS Principle Overview: So how can descriptive norms be optimally framed to Social norms are a powerful way to influence people’s influence behavior? New research shows that descriptive behavior. There are two major kinds of social norms (1) norms are most impactful when they are presented in a way Injunctive norms, which describe what people should do that matches the intended audience’s personal, immediate (e.g., “most people think that smoking is unhealthy and circumstances (Goldstein, Cialdini & Griskevicius, 2008). unattractive”) and (2) Descriptive norms, which describe Thus, social norms are generally more successful if they what most people actually do (e.g., “90% of people are refer not to “people in general” but rather to people who non-smokers”). Both can be effective at changing behavior, closely match the end user’s exact local circumstances (i.e., but evidence suggests that descriptive norms may work in a norms should be “localized”). more automatic, effortless “System 1” way. Also, descriptive norms can be made even more effective by framing them Relevance to OD: using “localized” language that links to the actual context Localized norms can pose challenges for stopping OD and immediate circumstances of the intended audience. and achieving consistent latrine use. We found a strong perception that OD is “normal practice” in certain Basic Science: communities among the literature we reviewed. In New research suggest that injunctive social norms are a number of communities OD was reported as very effective primarily for people who can engage in some common, practiced for years and through generations, degree of System 2 thinking (i.e., they can think about the normal and habitual (Coffey, Gupta, 2014; Patil, 2014; norm and engage in self-control). In contrast, descriptive WSP report Bihar, Indonesia, Kenya, Meghalaya, norms appear to work even when people cannot engage Rajasthan). System 2 thinking in this way. This is important, given recent evidence that poverty acts like a “cognitive tax,” However, there are several localized norms that might be limiting people’s ability to engage in effortful, System 2 leveraged to improve sanitation behaviors. For example, thinking (Mani, Mullainathan, Shafir, & Zhao, 2013). in Cambodia an advantage of owning a latrine is higher Thus, descriptive norms may be especially powerful and social status and prestige (WSP report Cambodia, 2007) under-leveraged as tools for driving behavior change among and in Malawi improved social status was a key motivator the poor. for latrine construction (WSP report Malawi, 2011). Communities in East Java who have pride in collective achievement were more likely than other communities to achieve ODF outcomes (Mukherjee, 2011). In Bihar, HOW TO EXECUTE/EXAMPLES OF APPLICATION 28% of cited pride and 45% cited honor as main reasons for opting for toilets (WSP report Bihar, 2012). In a global A sign in hotel rooms that stated, “75% of the review of influencers of OD in rural settings, shame and guests who stayed in this room participated in our humiliation were cited in Peru, India, Tanzania and Kenya new resource savings program by using their towels as drivers of latrine to own, construct or reconstruct a more than once” was more effective in promoting latrine (O’Connell, 2014). Each of these components re-use of hotel towels than a sign stating, “75% of (pride, honor, shame, humiliation) reflect and depend on the guests participated in our new resource local norms of respective communities and thus present as savings program by using their towels more than opportunities to utilize Principle 8. once” (Goldstein, Cialdini & Griskevicius, 2008). 20 Nudging and Habit Change for Open Defecation: New Tactics from Behavioral Science The 8 Principles Normative Challenges Normative Opportunities • Strong perception that OD is the “normal • Achieving social status from latrine ownership practice” in our community”. and reducing shame/increasing pride from OD are strong drivers. These drivers reflect and depend on • Awareness that OD has been practiced for the “local norms” of the community. generations, hence is accepted as local norm. PROGRAMMATIC IMPLICATIONS 1: MAPPINGS TO RECENT INNOVATIONS/TACTICS • Once OD behavior begins to change in an • Develop incentive systems that provide intervention (e.g., post-triggering in CLTS), rewards at the level of the local group (e.g., develop new intervention activities that village, women’s groups) rather than individual highlight how the majority behavior has shifted, households, to create and reinforce a new reinforcing the new descriptive norm. provincial norm around ODF status. PROGRAMMATIC IMPLICATIONS 2: NEW IDEAS AND EXTENSIONS • Construct public latrines at the “best schools” (creating an aspirational association) and fund them to open early so that parents and children can use them during school drop-off and pick up (see also Principle 1). www.wsp.org 21 Concluding What Remarks Influences Open Defecation and Latrine Ownership in Rural Households?: Findings from a Global Review References III. Concluding Remarks In this report, we have advanced 8 System 1 Principles In closing, we reiterate that these 8 principles are designed that can be used to promote the initiation and to augment, not replace, approaches based on System 2 maintenance of behavior change around OD and thinking. As we have emphasized throughout, human latrine use. Building on the World Bank’s 2015 World behavior is the product of both System 2 (rational, Development Report: Mind, Society, and Behavior, the motivated) and System 1 (automatic, cue driven habits). principles were derived from basic science in social Thus, the most powerful behavior change strategies are psychology, cognitive science, behavioral economics, and likely to come from combining different intervention health psychology. Critically, we focused specifically tactics that, collectively, address both systems. on “System 1” tactics that do not depend on the end users’ rational, effortful decision-making or their Finally, as field practitioners explore folding these ideas motivational systems (see Thaler & Sunstein, 2011; into OD interventions, we encourage the use, where Wood & Neal, 2015). Instead, we focused on ways to practical, of randomized control trials (RCTs), the robust disrupt existing OD habits and nudge people measurement of outcome data, and the sharing of automatically towards forming new, latrine use habits successes and failures alike. In particular, we encourage the that are maintained over time. sharing of new ways to translate, tailor, and “bring to life” these basic science principles as makes sense in specific The 8 System 1 Principles to support OD behavior change environments, cultures, and sub-populations. are as follows: 1. Ensure critical products and infrastructure are immediately and consistently physically available for the end user. 2. Create or capitalize on context change to drive new behavior. 3. Piggyback on other existing behaviors and cues. 4. Address friction for the old and new behaviors. 5. Support context-stable repetition for latrine use. 6. Embed ritualized elements in the change process. 7. Leverage point-of-action reminders and cues, and, 8. Highlight descriptive and “localized” norms that reduce cognitive demands. 22 Nudging and Habit Change for Open Defecation: New Tactics from Behavioral Science Concluding Remarks References Ariely, D. (2008). Predictably irrational. New York: Dreibelbis, R., Kroeger, A., Hossain, K., Venkatesh, M., & Harper Collins Ram, P. K. (2016). Behavior change without behavior change communication: Nudging handwashing among primary school students in Bangladesh. International Bouton, M. E. (2000). A learning theory perspective on Journal of Environmental Research and Public Health, 13, lapse, relapse, and the maintenance of behavior change. 129. Health Psychology, 19(Suppl.), 57-63. Elder, J. P., Ayala, G. X., & Harris, S. (1999). Theories and Cialdini, R. B. (2007). Descriptive social norms as under- intervention approaches to health-behavior change in appreciated sources of social control. Psychometrika, 72, primary care. American Journal of Preventive Medicine, 17, 263-268. 275-284. Cialdini, R. B., & James, L. (2009). Influence: Science and Fry, J. P., & Neff, R. A. (2009). Periodic prompts and practice (Vol. 4). Boston, MA: Pearson education. reminders in health promotion and health behavior interventions: Systematic review. Journal of Medical Coffey, D., Gupta, A., Hathi, P., Khurana, N., Spears, D., Internet Research, 11, e16. Srivastav, N., & Vyas, S. (2014). Revealed preference for open defecation. Economic & Political Weekly, 49, 43-55. Goldstein, N. J., Cialdini, R. B., & Griskevicius, V. (2008). A room with a viewpoint: Using social norms to motivate Cole-Lewis, H., & Kershaw, T. (2010). Text messaging as a environmental conservation in hotels. Journal of Consumer tool for behavior change in disease prevention and Research, 35, 472-482. management. Epidemiologic Reviews, 32, 56-69. Gollwitzer, P. M., & Sheeran, P. (2006). Implementation Contzen, N., Meili, I. H., & Mosler, H. J. (2015). Changing intentions and goal achievement: A meta-analysis of effects handwashing behaviour in southern Ethiopia: A and processes. Advances in Experimental Social Psychology, longitudinal study on infrastructural and commitment 38, 69-119. interventions. Social Science & Medicine, 124, 103-114. Harris, P. R., Brearley, I., Sheeran, P., Barker, M., Klein, W. Daw, N. D., Niv, Y., & Dayan, P. (2005). Uncertainty- M., Creswell, J. D., ... & Bond, R. (2014). Combining self- based competition between prefrontal and dorsolateral stri- affirmation with implementation intentions to promote atal systems for behavioral control. Nature Neuroscience, 8, fruit and vegetable consumption. Health Psychology, 33, 1704-1711. 729-736. Devine, J. (2009). Introducing Sanifoam: A Framework to Jacobson, R. P., Mortensen, C. R., Jacobson, K. J., Analyze Sanitation Behaviors to Design Effective Sanitation & Cialdini, R. B. (2015). Self-Control Moderates the Programs; Water and Sanitation Program, World Bank: Effectiveness of Influence Attempts Highlighting Injunc- Washington, DC, USA. tive Social Norms. Social Psychological and Personality Science. In press. Jenkins, M. W., & Curtis, V. (2005). Achieving the ‘good life’: Why some people want latrines in rural Benin. Social Science & Medicine, 61, 2446-2459. www.wsp.org 23 Concluding Remarks Jimerson, A (2016). A small doable action that achieved an Manning, M. (2009). The effects of subjective norms on increase in the “ideal” behavior: Alive & Thrive in Vietnam. behavior in the theory of planned behavior: A metaanalysis. FHI 360/Alive & Thrive. Presentation at the 2016 SBCC British Journal of Social Psychology, 48, 649-705. Summit, Addis Ababa, Ethiopia. Marteau, T. M., Hollands, G. J., & Fletcher, P. C. (2012). Judah, G., Gardner, B., & Aunger, R. (2013). Forming a Changing human behavior to prevent disease: the flossing habit: An exploratory study of the psychological importance of targeting automatic processes. Science, 337, determinants of habit formation. British Journal of Health 1492-1495. Psychology, 18, 338-353. Mukherjee, N. (2011). Factors associated with achievingand Kahneman, D. (2011). Thinking, fast and slow. New York: sustaining open defecation free communities: Learning from Farrar, Straus and Giroux. east Java. World Bank, Water and Sanitation Program Re- search Brief. Kelly, J. A. (2004). Popular opinion leaders and HIV pre- vention peer education: Resolving discrepant findings, and Murray, K. B., & Häubl, G. (2007). Explaining cognitive implications for the development of effective community lock-in: The role of skill-based habits of use in consumer programmes. AIDS Care, 16, 139-150. choice. Journal of Consumer Research, 34, 77-88. Kelly, J. A., St Lawrence, J. S., Diaz, Y. E., Stevenson, L. Y., National Institute of Mental Health Collaborative HIV/ Hauth, A. C., Brasfield, T. L., ... & Andrew, M. E. (1991). STD Prevention Trial Group. (2007). The community pop- HIV risk behavior reduction following intervention with ular opinion leader HIV prevention programme: Concep- key opinion leaders of population: an experimental analysis. tual basis and intervention procedures. AIDS, 21, American Journal of Public Health, 81, 168-171. S59-S68. Lienard, P., & Boyer, P. (2006). Whence collective rituals? Neal, D. T., Wood, W., Labrecque, J. S., & Lally, P. (2012). A cultural selection model of ritualized behavior. American How do habits guide behavior? Perceived and actual trig- Anthropologist, 108, 814-827. gers of habits in daily life. Journal of Experimental Social Psychology, 48, 492-498. Liu, L., Johnson, H. L., Cousens, S., Perin, J., Scott, S., Lawn, J. E. et al. (2012) Global, regional, and national Neal, D. T., Wood, W., Wu, M., & Kurlander, D. (2011). causes of child mortality: an updated systematic analysis for The pull of the past: When do habits persist despite conflict 2010 with time trends since 2000. Lancet, 379, with motives? Personality and Social Psychology Bulletin, 37, 2151–2161. 1428-1437. Luby, S. P., Halder, A. K., Tronchet, C., Akhter, S., Bhuiya, Nyer, P. U., & Dellande, S. (2010). Public commitment as A., & Johnston, R. B. (2009). Household characteristics a motivator for weight loss. Psychology & Marketing, 27, associated with handwashing with soap in rural Bangla- 1-12. desh. The American Journal of Tropical Medicine and Hy- giene, 81, 882-887. O’Connell, K. (2014). What Influences Open Defecation and Latrine Ownership in Rural Households?: Findings. Wash- Mani, A., Mullainathan, S., Shafir, E., & Zhao, J. (2013). ington, DC: World Bank. Poverty impedes cognitive function. Science, 341, 976-980. Oettingen, G., Mayer, D., & Thorpe, J. (2010). Self-regu- lation of commitment to reduce cigarette consumption: 24 Nudging and Habit Change for Open Defecation: New Tactics from Behavioral Science Concluding Remarks Mental contrasting of future with reality. Psychology and Rothman, A.J., Gollwitzer, P.M., Grant, A.M., Neal, D.T., Health, 25, 961-977. Shearan, P., & Wood., W. (in press). Hale and hearty poli- cies: How psychological science can create and maintain Oettingen, G., Pak, H., & Schnetter, K. (2001). Self-regu- healthy habits. Perspectives in Psychological Science. lation of goalsetting: Turning free fantasies about the future into binding goals. Journal of Personality and Social Psychol- Rozin, P. (1999). The process of moralization. Psychological ogy, 80, 736 –753. Science, 10, 218-221. Ouellette, J. A., & Wood, W. (1998). Habit and intention Rozin, P., & Singh, L. (1999). The moralization of cigarette in everyday life: the multiple processes by which past be- smoking in the United States. Journal of Consumer Psychol- havior predicts future behavior. Psychological Bulletin, 124, ogy, 8, 321–337. 54-74. Schultz, P. W., Nolan, J. M., Cialdini, R. B., Goldstein, N. Orbell, S., & Verplanken, B. (2010). The automatic com- J., & Griskevicius, V. (2007). The constructive, destructive, ponent of habit in health behavior: habit as cue-contingent and reconstructive power of social norms. Psychological Sci- automaticity. Health Psychology, 29, 374-383. ence, 18, 429-434. Payne, C. R., Niculescu, M., Just, D. R., & Kelly, M. P. Sheeran, P., Harris, P., Vaughan, J., Oettingen, G., & Goll- (2015). Shopper marketing nutrition interventions: Social witzer, P. M. (2013). Gone exercising: Mental contrasting norms on grocery carts increase produce spending without promotes physical activity among overweight, middle-aged, increasing shopper budgets. Preventive Medicine Reports, 2, low-SES fishermen. Health Psychology, 32, 802-809. 287-291. Sigler, R., Mahmoudi, L., & Graham, J. P. (2014). Analysis Patil SR, Arnold BF, Salvatore AL, Briceno B, Ganguly S, of behavioral change techniques in community-led total Colford JM Jr, et al. (2014) The Effect of India’s Total Sani- sanitation programs. Health Promotion International. tation Campaign on Defecation Behaviors and Child Health in Rural Madhya Pradesh: A Cluster Randomized Thaler, R., & Sunstein, C. (2008). Nudge: Improving deci- Controlled Trial. PLoS Med 11: e1001709. sions about health, wealth, and happiness. New Haven, Conn.: Yale. Prüss-Üstün, A., Bos, R., Gore, F., & Bartram, J. (2008). Safer water, better health: costs, benefits and sustainability of Tobias R. 2009. Changing behavior by memory aids: a so- interventions to protect and promote health. Geneva: World cial psychological model of prospective memory and habit Health Organization. development tested with dynamic field data. Psychogical. Review, 408–38. Qualitative Report for Understanding Rural Sanitation, Bihar, 2012. Tyndale-Biscoe, P., Bond, M., & Kidd, R. (2013). ODF Sustainability Study. FH Designs. Australia: PLAN Rossano, M. J. (2012). The essential role of ritual in the International, 1-181. transmission and reinforcement of social norms. Psychologi- cal bulletin, 138, 529-549. UN. (2014) The Millennium Development Goal Report 2014. United Nations, New York, NY. Rothman, A. J., Sheeran, P., & Wood, W. (2009). Reflec- tive and automatic processes in the initiation and mainte- UN. (2015) The Millennium Development Goal Report nance of dietary change. Annals of Behavioral Medicine, 2015. United Nations, New York, NY. 38(1), 4-17. www.wsp.org 25 Concluding Remarks Verplanken, B., Walker, I., Davis, A., & Jurasek, M. (2008). WSP Research Brief - Bihar (2013) World Bank, Water and Context change and travel mode choice: Combining the Sanitation Program. habit discontinuity and self-activation hypotheses. Journal of Environmental Psychology, 28, 121-127. WSP Research Brief - Kenya (2013) World Bank, Water and Sanitation Program. Volpp, K. G., John, L. K., Troxel, A. B., Norton, L., Fassbender, J., & Loewenstein, G. (2008). Financial WSP Research Brief - Indonesia (2008) World Bank, Water incentive–based approaches for weight loss: a randomized and Sanitation Program. trial. JAMA, 300, 2631-2637. WSP Research Brief - Meghalaya (2012) World Bank, Water Wakefield, M., Germain, D., & Henriksen, L. (2008). The and Sanitation Program. effect of retail cigarette pack displays on impulse purchase. Addiction, 103, 322-328. WSP Research Brief - Rajasthan (2012) World Bank, Water and Sanitation Program. World Bank. 2015. World Development Report 2015: Mind, Society, and Behavior. Washington, DC: World WSP Research Brief - Malawi (2011) World Bank, Water Bank. doi: 10.1596/978-1-4648-0342-0. and Sanitation Program. Webb, T. L., & Sheeran, P. (2006). Does changing behav- WSP Research Brief - Cambodia (2007) World Bank, Water ioral intentions engender behavior change? A meta-analysis and Sanitation Program. of the experimental evidence. Psychological Bulletin, 132, 249-268. WHO/UNICEF Joint Monitoring Program. 2013. Pro- gresson Sanitation and Drinking Water: 2013 update. New York: WHO/UNICEF. UN. (2015) The Millennium Development Goal Report 2015. United Nations, New York, NY. Wood, W., & Neal, D. T. (2007). A new look at habits and the habit-goal interface. Psychological Review, 114, 843-863. Wood, W., & Neal, D. T., (2015). Habit-based behavior change interventions. Working Paper. University of Southern California. Wood, W., Tam, L., & Witt, M. G. (2005). Changing circumstances, disrupting habits. Journal of Personality and Social Psychology, 88, 918-933. World Health Organization, Water Sanitation and Health (WSH) Burden of Disease and Cost-Effectiveness Estimates. 26 Nudging and Habit Change for Open Defecation: New Tactics from Behavioral Science