Information about Sub-Saharan Africa África subsahariana

1 A Status Report

Caroline Kende-Robb
Published Date:
January 2002
  • ShareShare
Information about Sub-Saharan Africa África subsahariana
Show Summary Details

Participatory poverty assessments (PPAs) are broadening our understanding of both poverty and the policy process. The limitations of quantitative measurements of well-being have long been recognized, and there is a rich tradition of anthropological and sociological work that uses a range of techniques to achieve an in-depth understanding of poverty for project work. In this tradition, PPAs use a systematic participatory research process that directly involves the poor in defining the nature of poverty, with the objective of influencing policy. This process usually addresses both traditional concerns, such as lack of income and public services, and other dimensions, such as vulnerability, isolation, lack of security and self-respect, and powerlessness.

PPAs are also highlighting the fact that policy change involves more than writing statements of intent in a policy document. It requires an understanding of the unpredictable situation within which agenda setting, formulation, and implementation continuously overlap and policy choices are made as outcomes of social processes. It also requires an understanding of how a broad-based dialogue with different people in society, including the poor, can help ensure that a policy will be implemented and sustained.

PPAs have demonstrated the value of the following:

  • Participatory policy research in the form of participatory problem identification, which includes the poor in the analysis of their own livelihoods using both qualitative and quantitative information
  • Participation in policymaking, which involves linking the information from participatory research into a broad policy dialogue among a cross-section of stakeholders, leading to increased awareness, attitude shifts, and changes in policy and the policy delivery framework.

PPAs are part of a trend within and beyond the World Bank that is challenging personal, professional, and institutional norms. On a personal level, the new approach is to learn from and listen to others; on a professional level, it is to appreciate that we are not the only experts and that many others can contribute to the debate on poverty and development; and on an institutional level, it is to change organizational culture, methods, and values from top-down practices to adaptable approaches that embrace risk-taking and error.


In the 1980s, the Bank’s poverty reduction objectives were often overshadowed by the focus on economic adjustment to achieve macroeconomic stability and structural change as foundations for long-term growth. Toward the end of the decade, however, the Bank and other development agencies began to act to mitigate the consequences of economic and structural adjustment for the poor. For example, the Social Dimensions of Adjustment program, funded by several multilateral and bilateral agencies, was launched in November 1987 in response to their concern about the position of the poor in the structural adjustment process in Africa. The program included a strong focus on strengthening national information systems, though with little use of participatory research.

The World Development Report 1990 (World Bank 1990), which focused on the issue of poverty, proposed a strategy for achieving more effective poverty reduction. That report was followed in 1991 by a policy paper, Assistance Strategies to Reduce Poverty (World Bank 1991), which laid out how the findings of the World Development Report could be used to strengthen poverty reduction efforts. The policy paper recommended that a poverty assessment be conducted for each country, with the objective of analyzing the nature and causes of poverty and developing a strategy for poverty reduction. In the World Bank’s process, the poverty assessment, which is done routinely for each country, feeds into the country assistance strategy, which lays out the Bank’s program of support for a country in relation to its development objectives and structural conditions (see World Bank 1992).

The World Development Report 2000/2001: Attacking Poverty (World Bank 2001) broadens the definition of poverty as presented in the World Development Report (WDR) 1990. The WDR 2000/2001 concludes that major reductions in poverty are possible but that achieving them will require a more comprehensive approach that directly addresses the needs of poor people in three important areas: opportunity, empowerment, and security. The WDR 2000/2001 drew on a large volume of research, including past and ongoing PPAs. (See Narayan and others [2000] for a summary of new PPAs undertaken in 23 countries for the WDR.)

World Bank poverty assessments use a variety of sources to diagnose the structural causes of poverty. Typically, a national household income or expenditure survey, or a multipurpose living standards measurement survey, is undertaken to provide basic information on the patterns of poverty. The early poverty assessments made little use of participatory techniques, and although they did employ a multidimensional concept of poverty, their principal criterion for defining who is poor was generally consumption or income. This approach, however, has changed over the past decade, with increasing attention being paid to information from participatory research sources. Such information is generally used to complement, enhance, modify, or interpret conclusions derived from household survey analyses and other quantitative sources.

Outside the Bank, there was also a growing realization of the importance of including the poor in diagnosis and policy work. A variety of sources led to this shift from projects to policy dialogue. PPAs developed in response to the broadened thinking on the multidimensional character of poverty associated with such publications as the Bulletin on Vulnerability (Institute of Development Studies 1989) and Putting the Last First (Chambers 1983). In the Bank, there was also ongoing project (as opposed to policy) work on understanding poverty and well-being through beneficiary assessments, participatory rural appraisals, developmental anthropology approaches, and similar methods.

The European donors (including Denmark, Norway, Sweden, and the United Kingdom, which support the PPAs through trust funds and operational funding) began to emphasize the social dimensions of poverty and provided funding for many of the Bank’s PPAs in Africa.1 In the Bank, development of the PPA was initially based on a series of papers by Clark (1992), Norton and Francis (1992), Salmen (1992a, 1992b), and Clark and Salmen (1993). In addition, the Bank’s Participation Learning Group (see World Bank 1994c) created a more receptive institutional environment for participatory approaches in both project and policy work.

Although to date more than 60 countries have undertaken PPAs with assistance from the World Bank, an equal number of PPAs have also been conducted with assistance from other agencies, including the United Nations Development Programme (UNDP), the Asian Development Bank (ADB), bilaterals, and nongovernmental organizations (NGOs). For example, as far back as 1990, The Gambian government and the UNDP formulated the Strategy for Poverty Alleviation through a process of dialogue with a cross-section of groups in society, including poor communities throughout the country2 The strategy provided an institutional framework whereby the poor could express their views on poverty3 And in Bangladesh, the UNDP undertook a national participatory poverty study (UNDP 1996).4 More recently, PPAs have been undertaken in partnership between and among various donors and NGOs. For example, the PPA in Mongolia (carried out in 2000) was supported by a partnership between the Government of Mongolia, the Centre for Social Development (local consulting firm), UNICEF, the World Bank, the ADB, and the Department for International Development—United Kingdom (DFID). The partnership for The Gambian PPA (1999 and 2000) was between the government the International Development Research Centre (IDRF), Canada, and Action Aid. The World Bank has adopted participatory research techniques on a broad basis in a variety of geographical regions and with a range of partners. This experience has enabled the Bank to understand the diverse causes and conditions of poverty and the processes that affect policy change. Appendixes A and B5 analyze the methodologies and impacts of participatory assessments on a country-specific basis. The objective is to learn from the organizations that have been our partners in this exercise and to reflect on the process.

What Is a Participatory Poverty Assessment?

A PPA is a method to include poor people in the analysis of poverty with the objective of influencing policy. The findings are transmitted to policymakers, thereby enabling the poor to influence public policy choices. PPAs have three key elements:

  • Field research. By directly consulting the poor at the community level, field research generates a better understanding of poverty from the perspective of the poor. The views of the poor contribute to the analysis of poverty and the formulation of public policy aimed at poverty reduction.
  • Policy influence. A cross-section of civil society (for example, NGOs, policymakers, administrators, civic groups, parliamentarians, and media) is included in the PPA process to promote wider ownership of the PPA results, thereby increasing the chance that the PPA will influence policy.
  • Country capacity. The results of PPAs are combined with other data sources, including quantitative household surveys, to better diagnose poverty.

Using PPAs to extract information just for research purposes, with limited participation and no link to policymaking, is considered bad practice. In the past, many of the World Bank’s PPAs were focused on the first element field research. Links to policymaking were weak or unsustainable. More recently, PPAs are being designed to include the second and third elements, resulting in a greater impact.

A PPA is typically one of many inputs into a World Bank poverty assessment (see box 1). Unlike household surveys, which collect statistical data on the extent of poverty through standardized methods and rules, PPAs focus on processes and explanations of poverty as defined by individuals and communities within an evolving, flexible, and open framework.

Box 1.Background to the World Bank’s Participatory Poverty Assessment

As a result of the World Development Report 1990 on poverty and the 1991 policy paper Assistance Strategies to Reduce Poverty, the Bank is committed to carrying out complete country-specific analyses of poverty in the form of poverty assessments. As of July 1998,99 poverty assessments had been completed (see appendix C). A majority (55 percent) of these were based on statistical assessments without participatory surveys. Each poverty assessment draws a poverty line based on the level of income or consumption associated with the minimum acceptable level of nutrition and other necessities of everyday life. People are considered poor if their income falls below this line (World Bank 1991). Poverty assessments generally include an analysis of the depth and severity of poverty and are increasingly using multiple poverty lines.

From 1994 to 1999, 43 poverty assessments included PPAs, which provided new dimensions in the analysis of poverty. Policy-focused research using participatory methods is undertaken to understand poverty from the perspective of the poor by focusing on their realities, needs, and priorities. Definitions of poverty, therefore, have moved beyond the conventional consumption and income indicators to broader issues, such as vulnerability, physical and social isolation, powerlessness, and lack of security and self-respect. The PPAs form part of the poverty assessment, which combines qualitative and quantitative data to achieve a better analysis of poverty.

The inclusion of other stakeholders at different levels in the country is required to link the information from the PPAs to policymaking. In many countries, this inclusion has led to the creation of partnerships between the Bank, government, and civil society with the objective of reducing poverty.

The World Development Report 2000/2001: Attacking Poverty (World Bank 2001) broadens the definition of poverty as presented in the World Development Report 1990. The WDR 2000/2001 concludes that major reductions in poverty are possible but that achieving them will require a more comprehensive approach that directly addresses the needs of poor people in three important areas: opportunity, empowerment, and security. The WDR also stresses the fundamental role of institutional and social change in strengthening development processes, and the importance of including poor people in development planning. The WDR drew on a large volume of research, including past and ongoing PPAs.

PPAs are sometimes referred to as qualitative surveys. This name can be confusing because there is a qualitative dimension to traditional survey work, and many PPAs contain quantified information and analysis. The terms “objective” for household surveys and “subjective” for PPAs may also be inaccurate. In household surveys, for example, interviewers and analysts will interpret informants’ answers subjectively. The use of these terms can create the appearance of a dichotomy, whereas in the best poverty analysis, the two merge into one integrated analysis (for example, the World Bank’s poverty assessments for Armenia and Zambia, and the Ugandan government’s 1999 Poverty Status Report). Traditional survey data can be used to count, compare, and predict. The strength of the PPA is not in counting but rather in understanding the hidden dimensions of poverty and analyzing the causality and processes by which people fall into and get out of poverty.

Participatory research is undertaken by facilitators using a diverse set of participatory tools determined by the research agenda and local context. Enabling the poor to participate leads to a reversal in the relationship between the community and the outsider that is implicit in traditional surveys. Facilitators of participatory research need different skills and behavior, including listening to and respecting the expertise of participants, building trust, handing over control, and allowing the community to define the poverty issues that matter. The poor are viewed as participants or partners in the research process, data are shared with them, and the analysis of research results takes place within the community. The poor thus have more control over the research process, and their capacity to appraise, analyze, plan, and act is recognized.

The extent and quality of participation have, however, varied extensively. Some PPAs have been criticized for limited participation, especially when interviews were done quickly (less than two weeks of field research in some countries) and the results were not fed back to the communities. In other PPAs, the quality of the participation has been questioned. Although participatory research methods may have been used, some research teams adopted a dominant role, undermining participation and resulting more in data extraction. For example, the manager of the PPA in Ecuador judged that genuine participation was limited and renamed it the Rural Qualitative Survey.

Secondary stakeholders (that is, those beyond the community) have also participated in PPAs. Such stakeholders can include, for example, other donors (bilaterals, UNICEF), national and international NGOs (Save the Children, Oxfam), academic institutions, religious groups and leaders, different levels of government, and local leaders. Even some poverty assessments that did not include direct consultations with the poor were participatory in the sense that they consulted a cross-section of secondary stakeholders (for example, Malawi).

Although PPAs and anthropological research have some similarities, there are three main distinctions. First PPAs provide a perspective from a cross-section of communities in different areas of a country, whereas anthropological research usually analyzes one or two communities in depth. Second, PPAs tend to focus on messages for policy. Third, PPAs provide a rapid overview of the current situation, which is quickly presented to the policymakers. Anthropological research usually takes longer and focuses more deeply on processes within communities, often without a policy focus.

In summary, PPAs have been used to provide clearer insight into the perceptions of the poor on the key issues related to poverty reduction (Norton and Stephens 1995). They are contributing to a greater understanding of the processes by which people fall into and get out of poverty, the complex coping and survival strategies adopted by the poor, and the major priorities and solutions identified by the poor, all within a local or regional context. By combining the PPA with the household survey information, the final poverty assessment is able to more fully analyze the various dimensions of poverty and make more informed and appropriate policy recommendations.

How Are PPAs Conducted?

Factors that influence the approach and consequent outcome of PPAs include political context, support, and commitment, both in country and within the Bank; relations between the Bank and the governments; and levels of expertise. Thus, there is a wide range of experiences among the PPAs undertaken to date (see appendix A for details of the timing, research teams, institutions involved, and methods used). Table 1 summarizes the typical characteristic of PPAs, and table 2 provides two case examples. In general, PPAs with the wider objectives of linking to the policymaking process and increasing a country’s capacity to analyze poverty tend to cost more.

Table 1.Summary of PPA Typical Characteristics
Number of communities selected for research40–60 communities
Time spent training2 weeks
Time spent on field research3–6 months
Time spent on analyzing data from field research2–3 months
Composition of research teamNationals of country, with men and women equally represented; ability to speak local languages; representatives from various ethnic and age groups
Agency conducting the fieldworkGovernment extension workers; local and international NGOs; academic institutions; independent consultants/firms
Donors that have contributed to government-led PPAsDFID, World Bank, Action Aid, Oxfam, UNDP, UNICEF, DANIDA, Asian Development Bank, Institute of Development Studies (United Kingdom)
Source: Robb 2000.
Source: Robb 2000.
Table 2.Case Examples of PFA Features at a Glance
FeatureMongoliaThe Gambia
Cost$108,100 (excluding World Bank staff weeks)$134,000 (cost of total project, which includes five PPA studies over a 3-year period)
Number of communities selected for research32 rural and urban communities29 rural and urban communities
Time spent on training2 weeks (March 2000). Provided by international consultant5 days. Provided by Action Aid The Cambia
Time spent on field research2 months (March–May 2000)1 month (August 1999)
Time spent in each community1 week5 days in rural communities; 6 days in urban communities
Time spent on analysis4 months3 months
Size of research team4 teams of 4 members7 teams of 4 members
Composition of research teamNationals with men and women equally represented; ability to speak local LanguagesNationals with men and women equally represented; ability to speak local languages
Agency conducting the fieldworkStaff of the Social Statistic Division, National Statistics Office, Government of Mongolia; Centre for Social Development (local consulting firm); UNICEF seconded staffAction Aid The Gambia; government extension workers and consultants
Donors who contributed to the PPAsWorld Bank, Asian Development Bank, DFIDInternational Development Research Centre (IDRC), Canada

The design of a PPA is determined by conditions in a given country, the research agenda, the size of the sample, the experience of the researchers, links to policymaking, and the extent to which capacity building for poverty analysis is included.

Tables 3a-3f summarize the experiences of some of the PPAs. The methodologies in table 3a are described in detail starting on page 13.

Table 3a.Methodologies Used
MethodologyNumber of PPAsaPercent
Rapid rural appraisal1327
Participatory rural appraisal1529
Beneficiary assessment919
Semistructured interviews and focus groups817

The numbers add up to more than 43 because some PPAs used more than one method.

Self-esteem, associative strength, resourcefulness, action planning, and responsibility.

The numbers add up to more than 43 because some PPAs used more than one method.

Self-esteem, associative strength, resourcefulness, action planning, and responsibility.

Table 3b.Time Spent in the Field
DurationNumber of PPAsaPercent
1–2 weeks38
2–4 weeks820
1–2 months38
2–4 months1537
4–5 months1127

Where data are available.

Where data are available.

Table 3c.Number of Communities Assessed
Number of communities in the PPANumber of PPAsaPercent

Where data are available.

Where data are available.

Table 3d.Agency Conducting the Fieldwork
AgencyNumber of PEASaPercent
Local NGCTb818
International NGO818
Academic institution1842
Government agency511
Independent consultants and firms511

The numbers add up to more than 43 because some PPAs used more than one type of agency.

Nongovernmental organization.

The numbers add up to more than 43 because some PPAs used more than one type of agency.

Nongovernmental organization.

Table 3e.Cost
Cost (US$)Number of PPMaPercent

Where data are available.

Where data are available.

Table 3f.Year Fieldwork Was Conducted
YearNumber of PPAsaPercent
Ongoing and planned1326

Where data are available.

Where data are available.

The discussion below focuses on three main issues to be considered when conducting participatory policy research: sequencing and duration, research teams, and methodologies.

Sequencing and Duration

Some PPAs have been conducted before the household survey and others afterward. Each data set can inform the other, so the sequencing will be determined by the context in country. If the PPA comes first, its results can help focus the research agenda for the quantitative survey and generate hypotheses. In Armenia, for example, the results of the PPA were used in designing the survey. The recent Mongolian PPA is linked to the Living Standards Measurement Study (LSMS)6 in three main ways: (a) research sites for the PPA were selected to correspond to the 1998 LSMS sites; (b) the results of the PPA will be used to determine the research agenda for the next LSMS; and (c) capacity built at the National Statistics Office may, in the future, promote better integration of data derived from both household surveys and PPAs. When PPAs have been conducted after the survey, they have been used to explain the results. For example, in Mali the household survey showed what seemed to be a disproportionate amount of money spent on clothing. The PPA found that clothing and cloth were considered investment items as well as status symbols. Conversely, the results of quantitative surveys can be used to identify the poorest geographical areas on which participatory research should focus. Emerging good practice suggests that the ideal situation is to have an iterative process, as is being developed in Zambia (see box 2 and figure 1).

Figure 1.Good Practice Research Cycle

In 1993 in Uganda, Togo, Benin, and Mali, short and rapid surveys were undertaken for three to four weeks.7 Methods were based on rapid rural appraisals (RRAs), so feedback to communities was limited. In Togo, time constraints were placed on the field workers by the World Bank’s internal deadlines. Some results were, consequently, not disaggregated by gender, and the final report was not written in a way that could be easily understood by policymakers. In some more recent PPAs, such as in Cameroon, the lack of time for community-level analysis meant that some results were too generic.

A balance needs to be achieved between quick fieldwork (which leads to less costly and more timely policy messages) and longer, more expensive fieldwork, such as household surveys (which can cost up to $1 million and take up to three years). PPA research teams have spent from one day to one week in a given community and have visited from 4 to 98 communities. Urban areas are more complex, and thus more time and flexibility are needed, since it is difficult to predict the nature of participation. Total time in the field for a PPA has ranged from one week to eight months, depending on the sample size and the number of research teams.

Research Teams

In Eastern Europe, most of the research was conducted by individuals from local universities. In other countries, NGOs undertook the field research (for example, Centre for Development of People [CEDEP] in Ghana, CARE in Cameroon, African Medical and Research Foundation [AMREF] in Kenya, Red Cross in Lesotho, Save the Children in Mali). International agencies have also been involved in the research process (UNDP in Togo, UNICEF in Lesotho). In South Africa, a local consulting company worked alongside a cross-section of NGOs, whereas in Mozambique and Zambia, local universities were involved. In Latin America, the community-level research was conducted by a cross-section of NGOs, universities, and government departments (for example, the government poverty agency in Mexico).

Box 2.Participatory Poverty Monitoring in Zambia


Using the same approach developed in the PPA, participatory poverty monitoring (PPM) has been undertaken in Zambia on a yearly basis (1995, 1996, and 1997) since the completion of the first PPA. The monitoring was conducted by the Participatory Assessment Group (PAG), the NGO involved in the PPA. The objective was to monitor changes in poverty over time.

Overall, it is evident that the PPA and the PPMs have made a considerable impact and contributed in a meaningful manner to the national policy agenda on poverty. The critical interest in the PPMs and their continuing contribution to policy dialogue lie in their empirical observation and elucidation of trends and changes in livelihood conditions in Zambia.

Two areas of PAG’s work will require continual reinforcement. Methodological skills need regular refreshing and upgrading through periodic training. The methodological approach requires repeated investigation of key policy areas using similar research techniques. What is required for successive PPMs to have additive value is consistent innovation in the use of research methods by the research team.

The second area that needs continual attention is the dissemination of findings, which involves identifying more precisely the clients for different types of PPM outputs and tailoring specific recommendations to those clients. An improved dissemination strategy is a priority. Initiatives might include local dissemination workshops, condensed reports for NGOs and other local institutions, and networking with other agencies and research institutes.

Linkages and impacts

The PPMs are not simply a tool for enriching the understanding of poverty in Zambia. They are also an important means of improving participatory planning in the provinces and districts by closing the information loops at those levels. PAG’s efforts (in dialogue, participation, and feedback) have been increasingly concentrated at the decentralized level and are well suited to ongoing decentralization efforts.

Link with the living conditions monitoring survey

There is still much informal discussion about linking the monitoring systems of the PPM and the living conditions monitoring survey (LCMS). The latest proposal suggests a quarterly meeting of a technical committee (comprising PAG and LCMS), with a rotating chair informing each institution of the other’s ongoing and planned work.

As far as harmonizing work programs, one problem identified was the difference in project cycles of the LCMS survey (at least one-and-a-half years) and the shorter cycle of the PPM. The timing of survey cycles appears to be the only major hurdle to partnership, since PAG and the Living Conditions Monitoring Unit are housed in the same complex at the Central Statistics Office, making it feasible, at least in practical terms, to harmonize their work programs.

Source: Based on a note prepared by D. Owen for field research for this study.

Some PPAs have used teams experienced in participatory research, as in Zambia, where the research team was given additional training in participatory rural appraisal (PRA) methods for the PPA exercise. Other PPAs have used local teams trained to conduct the research or have tapped into the country’s NGO and consulting firm networks (South Africa). In Ghana, the team was composed of a cross-section of individuals from NGOs, government line ministries, and academia.


There is a widening debate about the most appropriate methods to use when conducting participatory policy research. Below is a brief description of the main methodologies used8 (see table 4). In reality, these methodologies are complementary and can be used together. References are given for more in-depth information.

Table 4.Comparison of Participatory Methodologies
ItemRapid rural appraisals

Participatory rural appraisals

Beneficiary assessments

Participatory monitoring and evaluation

Participatory policy research

When1970Late 1980s198019901990
WhereUniversitiesNGOsWorld BankNGQsNGOs, universities. World Bank, governments, donors
ObjectiveData collection for projectsCommunity empowermentData collection for project managersUnderstanding impactData collection to influence policy
Main actorsOutsidersLocal peopleOutsidersLocal peopleLocal people and outsiders
Key techniquesVisualsVisualsConversational interviewsCombination of methods, e.g., RRA, PRA, BA, SARARCombination of methods, e.g., RRA, PRA, BA, SARAR
OutcomesFlans, projects, publicationsSustainable local action and institutionsBetter informed project managersAssessment of project processBetter informed policymakers
Main innovationMethodsBehaviorListening to the peopleLocal people’s contribution to determining indicators of successLinking local people to the national policy dialogue
Key resource earlier overlookedLocal people’s knowledgeLocal people’s capabilitiesLocal people’s knowledgeLocal people’s perceptions on impactLocal people’s knowledge for a better understanding of the problem and local people’s capability to analyze policy impact
Notes: NGO = nongovernmental organization; SARAR = self-esteem, associative strength, resourcefulness, action planning, and responsibility.Source: Adapted from Chambers (1997).
Notes: NGO = nongovernmental organization; SARAR = self-esteem, associative strength, resourcefulness, action planning, and responsibility.Source: Adapted from Chambers (1997).

What is a beneficiary assessment?

Many of the early PPAs were undertaken using a methodology called beneficiary assessment (BA), originally developed by the Bank in the early 1980s for use in the urban slums of Latin America. It was one of the methodologies that pioneered the inclusion of the voice of the poor in Bank operations. BAs draw from consumer research, traditional qualitative social science research, anthropological participant observation (observing people and interacting with them in their environments), conversational interviews, focus group interviews, institutional assessments, and investigative journalism.

A BA is designed in consultation with policymakers and others who will use the information. Teams of researchers collect information in selected communities through focus groups and individual interviews. A semistructured interview guide is drafted before the research begins. Information is collected mainly through dialogue between beneficiaries and researchers. The researchers then analyze the collected information—unlike a PRA, in which some of the analysis is done at the community level (see Salmen 1995a and 1995b for more details).

What are rapid and participatory rural appraisals?

Many PPAs have used the RRA methodology, which emerged in the 1970s. Its purpose was to develop an approach that would enable outsiders to learn about rural conditions and people’s realities quickly and cost-effectively. In the mid-1980s, RRAs evolved into the PRA approach, which placed greater emphasis on community participation.

RRAs and PRAs use such tools as mapping; diagrams of changes, trends, and linkages; matrices; and scoring. They also use group animation and exercises to facilitate information sharing, analysis, and action among stakeholders. The information is thereby made visible, which often creates ownership. The power of the PRA is frequently in “group-visual synergy” (Chambers 1997), with analysis being locally led. The main difference between BA and PRA is that PRA combines both verbal and visual techniques and emphasizes community-level analysis, whereas the BA emphasizes verbal techniques, and most of the analysis is done by the interviewer.

The PRA is also a set of principles that includes following up actions, embracing error, showing respect, being willing to unlearn assumptions and conditioned responses (reversals in learning), using methods or processes only if they make sense in the context (optimal ignorance), compensating for biases, and triangulating data. As Chambers (1997) has noted, “PRA stresses changes in the behavior and attitudes of outsiders, to become not teachers but facilitators, not lecturers but listeners and learners.” (Also see International Institute of Environment and Development 1991–2001.)

What is participation learning and action?

Participation learning and action (PLA) is an umbrella term for a wide range of similar approaches and methodologies, including RRAs and PRAs. The common theme in all these approaches is the full participation of people in the process of learning about their needs and opportunities and in the action required to address them (see International Institute of Environment and Development 1991–2001).

What is SARAR?

The methodology using self-esteem, associative strength, resourcefulness, action planning, and responsibility (SARAR) uses visual aids to stimulate discussions. These visuals are prepared in advance by the researchers (unlike the PRA, in which the visuals are created by the communities to express issues and concerns). The main objectives of the SARAR are to build local capacity to plan for community development or to raise awareness of health and sanitation issues. SARAR builds on local knowledge and strengthens local capacity through a variety of participatory methods. It has also been used by development agencies to increase participation and joint decisionmaking, although it is not often used in PPAs (see Srinivasan 1990).

Use of methodologies

There are many different participatory traditions from around the world: some provide the philosophy for participation, others provide the tools, and some provide both. PRA is one of the few that provides a broad philosophy in addition to distinctive tools. The selection of methodologies and tools depends on the context of the PPA (for example, capacity of in-country institutions, PPA manager’s knowledge of different methods, government approval, availability of skilled trainers, time available).

The tools and approaches can be very different, and all have advantages and drawbacks. For example, PRA enables some of the analysis to take place at the community level, leading to greater ownership of the results. A researcher from Zambia, where a PRA was undertaken, stated that community ownership meant that “problems would be thought about long after my departure,”9 To further promote this ownership in Zambia, charts and papers created by local people were left with the community. PRA places more emphasis on community-level interviewing, while BA concentrates on households or individuals (Norton and Stephens 1995) and involves less community ownership and control over the analysis and results.

Some have argued that the visual tools of PRA might not be suitable for all cultures. Although this statement might be true to some extent, the skill and sensitivity of the facilitator and the understanding that he or she has of the community usually determine the extent to which visual tools will be appropriate. PRAs have been conducted effectively in a diverse range of cultures in more than 100 countries.10

How these methodologies relate to policy work and methodological dilemmas

These methodologies were not originally designed to influence policy—they were developed specifically for communities and project work, BAs were traditionally used to seek the views of beneficiaries on the impact of projects and to feed this information back to project managers in an attempt to influence project design. SARARs and PRAs were used at the community level to develop community action plans with the wider objective of empowerment.

In the 1990s, participatory methods were used to achieve the broader objective of influencing policy. Sector assessments used participatory research to influence policy in the following areas: health and education in Zambia (work done by the NGO, Participatory Assessment Group; Milimo 1996); urban poverty and violence in Jamaica (Moser and Holland 1996); and wetlands management in India and Pakistan (Gujja, Pimbert, and Shah 1996). Whereas PPAs attempt to influence the broader policy framework, sector assessments attempt to influence specific policies.

In this new field of influencing policy through dialogue with the poor, ethical questions are being raised about the possible exploitation involved in using the poor to gain access to information without any benefit to them. When participatory methodologies were widely used at the project level, they comprised tools for gaining information and a set of principles, such as action follow-up, empowerment, and capacity building in the community. When participatory methodologies are used for policy work, however, these principles have often not been followed. It is suggested that when undertaking participatory research for policy work, the term participatory policy research (PPR) might be more appropriate. The debate has evolved because many PRA practitioners have questioned the process, principles, and ethics of working directly with communities for policy research when people’s expectations are raised and there is no direct follow-up at the community level—the result being more data extraction than community action. PPR uses tools from various methodologies but with a different overall objective: the creation of policy messages with communities contributing to the analysis, as opposed to direct action, community empowerment, and capacity building. But ethical questions remain about taking people’s time and raising their expectations when undertaking not only PPAs but any kind of poverty research, including household surveys.

PPR, therefore, is not generally a tool for empowerment (Chambers 1997), and while its research value is great, its value at the community level should not be overstated. For policy, the participatory research is meant to be imperfect, rapid, and restricted, and the principle of immediate action may not be feasible because the focus is on trends, not project identification. PPR is a way to inform policy rather than empower local people. In an attempt to respond to the principle of follow-up action, however, many PPAs have linked the information with action-oriented institutions. For example, in Argentina and Brazil the fieldwork has been linked with the work of country NGOs and government line ministries. As a result, the potential now exists for moving from information sharing to continuous dialogue with various stakeholders, including those at the community level.

What Is the Current Status of FFAs?

As of July 1998, 43 PPAs had been undertaken at the World Bank. The fraction of poverty assessments including a PPA has risen from one-fifth in fiscal 1994 to one-third in fiscal 1995 and one-half in fiscal 1996, fiscal 1997, and fiscal 1998.11 Of the 43 PPAs completed, 28 are in Africa, 6 in Latin America, 5 in Eastern Europe, and 4 in Asia.

Box 3 shows the distribution of the various participatory methodologies employed, by region, as of May 2001. Box 4 details some of the PPAs planned by the Bank and other organizations.

Box 3.World Bank Participatory Poverty Assessments: Status Report, March 2001

Burkina FasoFRAAlbaniaVarious
Central African RepublicRRABosnia and HerzegovinaPRA/various
EgyptFRAKyrgyz RepublicVarious
Equatorial GuineaRRALatviaVarious
The GambiaFRAUkraineVarious
MaliRRACosta RicaBA
South AfricaPRAASIA
Papua New GuineaVarious
Sri LankaPRA
Notes: RRA = rapid rural appraisal; PRA = participatory rural appraisal; BA = beneficiary assessment; SARAR = self-esteem, associative strength, resourcefulness, action planning, and responsibility; various = a variety of qualitative research methods were used, including open-ended interviews, focus groups, and semistructured interviews.
Notes: RRA = rapid rural appraisal; PRA = participatory rural appraisal; BA = beneficiary assessment; SARAR = self-esteem, associative strength, resourcefulness, action planning, and responsibility; various = a variety of qualitative research methods were used, including open-ended interviews, focus groups, and semistructured interviews.

Box 4.Examples of Planned PPAs

WhereDescriptionWhenDonor support
Burkina FasoNational Participatory Poverty Assessment as part of the implementation of the PRSP2001World Bank
ChadPerceptions of Poverty Study2000–01TBC
The GambiaPPA—3-year project2000–2002Government of The Gambia, IDRC
Guinea BissauRapid PPAMid-2001TBC
LaoPPAOctober 2000–May 2001TBC
MoldovaPPASeptember 2001TBC
MongoliaFollow-up to the first PPATBCTBC
NigerPerceptions of Poverty StudyNovember 2000–March 2001UNDP
PakistanPPAOngoingWorld Bank
PakistanPPA—managed at district level2001Government of Pakistan
Sierra LeonePPA2001–2002TBC
UgandaPPA—3-year projectOngoingDFID/UNDP/World Bank
Notes: TBC = to be confirmed; ADB = Asian Development Bank; DFID = Department for International Development—United Kingdom; IDRC = International Development Research Centre; UNDP = United Nations Development Programme.
Notes: TBC = to be confirmed; ADB = Asian Development Bank; DFID = Department for International Development—United Kingdom; IDRC = International Development Research Centre; UNDP = United Nations Development Programme.
1.The Department for International Development—United Kingdom seconded a Social Development Advisor (Andrew Norton) to the Bank to work on the development of participatory poverty assessments in Africa from 1992 to 1994,
2.See The Gambia (1994) for more details on this case,
3.In The Gambia, Action Aid and a nongovernmental organization (NGO) coordinating body assisted in organizing the participatory research on poverty using participatory rural appraisal techniques. In addition, a local team conducted research to gain an understanding of the informal networks within communities and throughout the country. Initially the policy environment was constrained, with the government unwilling to discuss poverty openly. As the dialogue gradually developed, more stakeholders were included until enough policy space was created to put poverty and related issues, such as decentralization and gender inequalities, on the political agenda. This process of consultation led to increased donor coordination and created an opportunity for the government and NGOs to redefine their heretofore controversial relationship.
4.Holland and Blackburn (1998) state that in the poverty study for Bangladesh, new issues were put on the policy agenda, such as the problem that demands for increasingly high dowry payments led to daughters’ being a burden to their parents and that wives were divorced or abused if the dowry was not paid. Furthermore, if daughters were educated and did not find a job, the demand for a dowry could increase. As a result, some parents were not sending their daughters to school. The study found that throughout Bangladesh, a priority for the poor was the enforcement of antidowry laws.
5.Appendix A analyzes the various methodological and organizational issues associated with each of the PPAs. Appendix B focuses on the value added of the PPAs and the impact on the Bank’s and borrower’s country programs and policies.
6.LSMSs use an integrated set of questionnaires and are designed to be repeated on a regular basis to track changing conditions over time. LSMSs can produce a comprehensive measure of household welfare, and evaluate its distribution across the population and over time; evaluate patterns in access to social services, such as schools, clinics, or welfare programs; identify the determinants of socioeconomic outcomes (for example, how women’s schooling affects fertility decisions, or how health status affects workers’ labor supply); and examine household responses to changes in economic conditions or government programs (for example, how price subsidies influence consumption patterns, or how clinic user fees affect health care choices). For further details, see World Bank 19961.
7.Although there were many limitations to these early PPAs, they are significant for having been the first Bank studies to use participatory research methods in poverty analysis.
8.The information in this section comes from a variety of sources but is based mainly on Rietbergen-McCracken and Narayan (1997).
9.Malako Nabanda, Participatory Assessment Group (NGO), Zambia.
10.Including developed countries. For example, PRA is now widespread throughout the United Kingdom. See Inglis and Guy (1996).
11.See appendix C for a detailed breakdown of all poverty assessments completed by the Bank to date

    Other Resources Citing This Publication