Rethinking impact: Understanding the complexity of poverty and change

Group finds traditional measures such as ‘rate of return studies’ are not suitable for evaluating research impacts.
complexity of povertySixty people from 33 organizations worldwide, almost half of them women and from outside the 16 centres supported by the Consultative Group on International Agricultural Research (CGIAR), convened this March in Cali, Colombia, to rethink the way agricultural researchers go about assessing their impacts on reducing poverty and economic, social and gender inequities.

Traditional assessment methods unsuitable
This group thinks that traditional economic impact assessment methods such as ‘rate of return studies’ are unsuited for evaluating research activities aimed at sustainable poverty reduction. Indeed, for the last 5-10 years, many CGIAR centres have been widening their range of methods to assess their diverse outcomes and impacts. The impact experts and local partners gathered at this meeting recommended that CGIAR management take immediate steps to acknowledge the legitimacy of this diversity and the broad range of impact assessment methods needed to evaluate it.

New ‘linking’ role for researchers
The group also stressed the central, rather than peripheral, role that today’s researchers must play in linking researchers, academics, farmers, marketers, policymakers and representatives of civil society in creating and sharing knowledge. To do this, the group warned, will require that research organizations change the way they do business. Specifically, these organizations will have to recognize the legitimacy and challenges of such boundary-spanning work by dedicating substantial time and resources to it and rewarding those who do it well.

Learning through participatory research
Learning organizations that are effective at innovation are also likely to be effective in engaging end-users. Thus, participants at this meeting highlighted the need to find more thoughtful ways of assessing who to involve in a given research project and how to involve them. Participatory tools demonstrably effective at this kind of ‘action research’ were discussed and evaluated. What was clear to all was a continuing need to better engage farmers and other end-users of research for development, or the civil society organizations that represent them, in meaningful ways at appropriate points throughout the research process. They also recommend that scientists managing complex research projects spanning the policy, civil society, agricultural, local development and private sectors be recognized and rewarded for taking on such complexity.

Excerpts from Workshop Brief No 2 states:

It is time for the CGIAR to present a clear strategy and code of conduct for engaging users (including farmers, the poor and the civil society organizations that represent them) in on-the-ground research processes. . . . The CGIAR guidelines for impact assessment currently being finalized (based on a rate of return methodology alone) are not adequate for much of the research it conducts.

We urge management to support the rapid development of another set of impact assessment guidelines specifically for evaluating complex collaborative research, and to adapt the performance measurement and other systems to reflect these new approaches. Without them, we risk inappropriately assessing the work we are doing that is most likely to lead to sustainable solutions to poverty, and possibly even driving it out of the CGIAR research portfolio.

More realism needs to be applied to the concept of attribution and causation within complex collaborative research, where impacts are not likely to be attributable to the CGIAR or single causes. Knowing that different collaborators play different roles over time and multiple causal strands contribute to impact, we should focus assessments on contribution rather than attribution. Over-emphasis on attribution may damage the trust needed for effective collaboration. In addition, greater emphasis needs to be placed on understanding adaptation processes rather than adoption per se of finished technology.

Principles for linking knowledge with action
ILRI agricultural economist, impact assessment expert and innovations leader Patti Kristjanson is committed to developing a set of principles for linking knowledge with action and to further linking the principles identified with tools, methods, approaches and strategies. Kristjanson says ILRI is collaborating with Harvard University’s Sustainability Science Program in development of a training course on this for CGIAR research managers and their partners.

The workshop participants agreed on the following four key messages.

Mission-oriented scientists need to rethink how they do research to have sustainable impacts on reducing poverty as well as how to evaluate those research impacts.
How scientists do research is key to achieving pro-poor, gender-sensitive and socially
inclusive results. Working more thoughtfully with, and helping to bridge boundaries
between, strategically chosen partners can help increase the probability of linking the
knowledge generated by the research to actions that lead to sustainable poverty

Mission-driven researchers need to continue to bring other (existing) evaluation methods and approaches into more regular practice.
A wide array of evaluation methods and approaches already exists that is not fully
used by the agricultural and natural-resource management R&D communities. Members of those communities should review the available options and try out methodologies they are not familiar with.

Many scientists still view non-economic assessment methods as ‘illigitimate’.
There is still a high degree of skepticism among agricultural and natural resource management researchers, particularly within the CGIAR, about using non-economic and non-statistical data in evaluations. More empirical evidence of the validity and value of approaches other than economic (e.g. ex-post assessments) is needed.

Methodology gaps still exist.
It can be argued that the CGIAR are employing inadequate evaluation methods for the 75% of its research unrelated to germplasm improvement (e.g. policies, institutions, natural resource management, gender and social inclusion). CGIAR research evaluators need to refine existing, as well as often employ multiple, methods to fill this important gap.

For more information, see

“Workshop on Rethinking Impact: Understanding the complexity of poverty and change: Summary”, 26–28 March 2008, Cali, Colombia, ILRI Innovation Works Discussion Paper 4, ILAC (Institutional Learning and Change) Working Paper 7 and PRGA (CGIAR Systemwide Program on Participatory Research and Gender Analysis) Working Document 26, September 2008.

The Challenge Paper, Initial Synthesis of Feedback, and Workshop Workbook

Further Information contact:
Patti Kristjanson
Innovation Works Leader
International Livestock Research Institute (ILRI)

Women and livestock: Global challenge dialogue

Poverty has a woman’s face. ILRI is facilitating a global e-consultation to fight poverty through women and livestock.

Over the coming months ILRI will be facilitating a Global Challenge Dialogue on Women and Livestock. This e-consultation will involve knowledgeable and influential thinkers and doers from around the world. They will be invited to take up the challenge of fighting poverty through women and livestock; to create new ways to empower women livestock keepers to further develop themselves, their families, their communities and their nations.

The Challenge Dialogue is about deepening understanding of the challenge, seeking ideas, and devising a strategy and action plan that will realize tangible impacts.

At the end of six months participants will have developed:
A first assessment of the global value of livestock-keeping by women.
A list of major opportunities in enhancing women’s contribution to livestock development.
A roadmap showing how joint actions will lead to real improvements to poor women’s lives, communities and environments.
Proposals for activities that help poor women get even more benefits from livestock.

Why are livestock so important to women?

Poverty has a woman’s face. Women do two thirds of the world’s work, and produce half the world’s food, yet earn only a tenth of the world’s income and own less than a hundredth of the world’s property. Of the 600 million poor livestock keepers in the world, around two thirds are women and most live in rural areas.

There is a special relationship between women and livestock. Poor women can own livestock when they are denied land. Looking after livestock fits well with their work of running households and raising families. Hundreds of millions of women livestock farmers daily tend sheep, goats and chickens, milk cows, buy and prepare food, plant and harvest crops, weed their plots, look after children, clean their home, fetch and carry water and firewood, prepare every meal for the family, care for the sick and elderly, while often simultaneously running small informal businesses – selling milk, eggs, fruits and vegetables – in market centres and along roadsides.

Women are the great unsung heroes of agricultural development. They are the farm and market managers who make agriculture viable, the glue that holds families and communities together, the stewards who safeguard their environments for the generations to come.

What can be done to better lives through ‘livestock women’?

Broad change will enable women to get more out of livestock. Change is needed in the ways governments, NGOs, and researchers support women livestock keepers. Change is needed in the ways societies value women’s work. Change is needed in service delivery to women farmers.

Patti Kristjanson, leader of this Challenge Dialogue, says:

‘To alleviate severe poverty, we need to change institutions and to engage women. Knowledge and technology is important, but it’s not enough. We have to change our ways of working and give poor people the lead in building their own futures.

‘We already have the skills and tools to bring about meaningful changes. We need the will to make them happen. By working together, we can start to solve a problem too big for any one person, organisation or institution to address alone’ said Kristjanson.

Challenge Dialogue: a new kind of consultation

A ‘Challenge Dialogue’ is a disciplined process of defining a specific challenge, engaging diverse stakeholders in a productive conversation focused on co-creating solutions, and taking action towards the solutions.

It is a proven vehicle for taking groups of more than 100 people through a structured conversation over several months focused on developing alignment and agreement around a plan for solving complex tasks.

‘Challenge Dialogue’ is particularly useful when faced with a significant opportunity or problem to be solved, when you need to bring people together that don’t normally work as a team and get them collaborating quickly and effectively, and you want to move to action within a defined timeframe.

Patti Kristjanson, ILRI’s Innovation Works leader says ‘The idea behind the Challenge Dialogue is that we involve as many diverse participants as possible and engage them in a bigger conversation. Everyone’s opinions are encouraged – thus we get diversity of views and a free flow of innovative ideas and solutions.

The Challenge Dialogue System (CDS) has been developed by Innovation Expedition. See

The global Challenge Dialogue on Women and Livestock will involve people who are passionate about reducing poverty and improving poor women’s lives. Participants will generate excitement, interest and evidence about how ‘livestock women’ can improve lives and reduce world poverty. It will bring in partners, investments and actions that will better support women livestock keepers and, through them, their vulnerable children, communities and environments.

For more information about ILRI Challenge Dialogues visit the Innovation Works initiative at

If you are interested in participating in the Global Challenge Dialogue on Women and Livestock, please contact Patti Kristjanson. Please provide a brief summary of your background and interests.

Further Information contact:

Patti Kristjanson

Innovation Works Leader
International Livestock Research Institute (ILRI)

Evolution of poverty and inequality in Uganda

According to a new report, poverty in Uganda has reduced in more than 80 percent of the rural sub-counties though this reduction has been least in the Northern region.

Nature, Distribution and Evolution of Poverty and Inequality in Uganda, describes and summarizes the trends in poverty and inequality in Uganda over the period 1992-2002. The report is the culmination of a two-year research project conducted by the Uganda Bureau of Statistics (UBOS), World Resources Institute in collaboration with the International Livestock Research Institute (ILRI), with technical assistance from the World Bank and with financial support from the Rockefeller Foundation. These new poverty measures can be used to help target, design and implement pro-poor development strategies that are both effective and inclusive.

This report presents information using the most recent data from the National Population and Housing Census of 2002 and the National Household Survey of 2002/3 and examines the changes in poverty over the period 1992-2002 as well as providing estimates of Ugandan poverty and inequality at the district, county and sub-county levels. The new estimates of well-being presented in this report are based on statistical techniques that combine existing survey and census datasets. Within sub-counties, poverty and inequality measures are computed for rural and urban communities. The report also demonstrates how poverty maps can be combined with other indicators of well-being to better understand the phenomenon of poverty. The results from the analysis of changes in poverty levels for 1992–2002 show that rural and urban areas with low initial poverty rates also experienced significant decreases in the absolute number of poor people by 2002.

The analysis of this publication makes use of the information from the 2002 population and Housing Census and the 2002/3 Uganda National Household Survey (UNHS). It builds on the previous work presented in the publication ‘Where are the poor? Mapping patterns of well-being in Uganda 1992-1999‘ which presented poverty estimates for each administrative district and county in Uganda. Building on and improving upon previous work, this analysis utilizes the most recent available data and investigates the potential for improving service and development targeting by complementing the new poverty information with geographical infrastructural and service information for key sectors of the economy, an important concern in a growing country like Uganda.

‘The Government of Uganda had for many days been allocating resources to districts and communities with limited experiential basis for the decisions to target for example income inequality. Although this disbursement of funds was meant to reduce poverty and improve project implementation, there was a risk of achieving limited success, partly due to the lack of information. To that end, the poverty maps could go a long way in helping to make informed decisions. There is also need for government and policy makers as well as development partners, to rely on empirical indicators for targeting resources,’ says the Executive Director of the Uganda Bureau of Statistics, John B. Male –Mukasa.

The aim of this research is to support implementation of the Poverty Eradication Action Plan by providing information that can improve design and targeting of specific poverty interventions and assist in crafting better targeted programmes that help the poor in Uganda meet their basic consumptions needs.

How have poverty levels changed over the decade, if at all?

The highest drops in poverty in rural areas between 1992 and 1999 were in Central and parts of western regions, but the new analysis show that poverty has reduced in more than 80 percent of the rural sub-counties of Uganda, the report provides critical indicators for evidence-based pro-poor policy making and key benchmarks for measuring the progress made by the government of Uganda. A comparison of national poverty levels for 1992 and 2002 points towards an improvement in welfare over the decade, with the national poverty rate falling from 56% in 1992 to 39% in 2002. Between 1992 and 2002, estimated poverty incidence shows a marked decline in both urban and rural areas. In urban areas the incidence of poverty decreases by 16 percentage points compared to 18 percentage points in rural areas. Urban poverty was more concentrated in the Northern and Western regions in 2002 relative to the Central and Eastern regions. The absolute number of poor people increased in the Northern and Western regions by 130% and 112.5% respectively. In contrast, in the Central and Eastern regions, the absolute number of poor people declined by 62% and 45% respectively. The Northern Region had the highest urban poverty incidence in 1992 (50%) and 2002 (38%). It also had the highest poverty increment between 1992 and 2002. Central Region, with relatively low urban poverty incidence in 1992 (19%) and 2002 (17%), demonstrated the highest reduction in poverty incidence over the same period.

In the rural areas, the situation is slightly different with more poor people concentrated in the Northern and Eastern regions than in the Central and Western regions. The absolute number of poor people increased in the Northern and Eastern regions by 48% and 14% respectively. Conversely, in Central and Western regions, the absolute number of rural poor declined by 32% and 14% respectively. As is the case in the urban areas, Northern Region, with the highest poverty incidence in 1992 (75%) and 2002 (66%) also had the highest increase in absolute number of poor between 1992 and 2002. Similarly, in Central Region with relatively low poverty incidence in 1992 (54%) and in 2002 (27%), we see the highest poverty reduction over the same period.

The results from the analysis of changes in poverty levels for 1992–2002 show that rural and urban areas with low initial poverty rates also experienced significant decreases in the absolute number of poor people by 2002. Areas with high initial poverty rates instead witnessed increased absolute numbers of poor people by 2002.

Download the book:

Data for the changes in poverty can be obtained on request from the Uganda Bureau of Statistics (UBOS )and the International Livestock Research Institute (ILRI).

Snapshot: Uganda

Uganda, located in eastern Africa, has an estimated population of 25.3 million and an annual population growth rate of 2.7%. The country has been plagued by an on-going 20-year-old war in the north between the government and rebel fighters. The rebels have become notorious for their crimes against civilians and for the abduction and murder of children. This has resulted in about 20,000 displaced persons, many of whom live in government ‘internally displaced’ camps. Over a million women and children have been affected. Children too afraid to sleep at night for fear of being abducted by rebels during their dawn raids on villages leave their homes each evening. They have become known as the ‘night commuters’ of northern Uganda – traveling from their villages to the safety of towns to avoid capture. Of recent, some progress has been made in terms of return of peace to the region. Currently, the LRA rebels and government are engaged in peace talks aimed at ending the 20 year civil war.

Although this paints a bleak picture, Uganda as a whole has made great strides, taking advantage of significant growth in the 1990s and is considered a model of development in Africa. Crippling inflation rates have been brought under control and GDP growth has been impressive, at 7.9% in 1999 and with a projected growth rate of 6.2% in 2005. Agriculture is the most important sector of the economy, contributing over 32% of GDP and employing over 80% of the work force. Uganda’s principal export is coffee, along with fish, fish products, cotton and tea.

Despite Uganda’s progress and concerted poverty reduction efforts, poverty is still widespread, with an estimated 31% (in 2005/06) of the population living below the national poverty line. The latest figures show the average life expectancy of a Ugandan is 43 years (47 years in 1990), infant mortality is 83 per 1000 live births, and under-5 mortality is 141 per 1000 children. The annual number of births is 1.3 million, but an estimated 184,000 children under 5 die each year.

(Data sources: World Bank; UNICEF.)

Electronic version of important poverty mapping book for Uganda available here

An electronic version of an important book, Where are the Poor? Mapping Patterns of Well-Being in Uganda, is now available.

Uganda has some of the poorest people in the world. For the first time, the question Where are the poor in Uganda? can be answered, as a result of sophisticated poverty maps developed by the Uganda Bureau of Statistics and the International Livestock Research Institute (ILRI). These maps provide facts and figures on poverty and inequality by region, district and county, highlighting where the poorest are located and estimating the numbers of poor and levels of poverty. These maps are important because they can be used to ensure that resources are targeted at those most in need.

If you are interested in viewing the entire electronic version of the book, click open:

If you are interested in viewing this book by chapter, go to:

To view the maps from the Atlas of Estimated Measures of Poverty Below the Regional Level: 1992 Poverty Maps, go to:

To view the maps from the Atlas of Estimated Measures of Poverty Below the Regional Level: 1999 Poverty Maps and the Change in Poverty from 1992 to 1999, go to:

Poverty pathways to be mapped across Kenya

The Kenya Government on 23 June 2005 announced that it has enlisted the Nairobi-headquartered International Livestock Research Institute (ILRI) to undertake an ambitious study investigating how, when and why Kenyan households move into and out of poverty. A deeper understanding of poverty dynamics can help developing countries better target and tailor pro-poor poverty interventions. ILRI has previously undertaken two similar studies on ‘Pathways out of Poverty and the Role of Livestock’, one in western Kenya and the other in Peru. These were undertaken in collaboration with the Pro-poor Livestock Policy Initiative of the Food and Agricultural Organization of the United Nations and Dr. Anirudh Krishna of Duke University, in the USA, who developed the participatory methods used in the study for similar research he first conducted in India. Remarkably, members of poor communities in India, Kenya and Peru all site the same factors that force households into poverty or help people climb out of poverty: loss or acquisition of livestock is, respectively, key to both. The new study in Kenya will be conducted across the whole country and will include all three of the country’s major livestock systems: pastoral, agro-pastoral and mixed crop-and-livestock production. The information on poverty will be collected in participatory ways and will be coupled with results of Kenya’s formal Welfare Monitoring Survey undertaken by the Ministry of Planning, ILRI’s partner in this new initiative, along with the Ministry of Agriculture. The breadth of the information obtained will allow scientists to answer a wide range of questions about poverty. The better understanding of poverty dynamics gained will help government policymakers and donor agencies better target and tailor pro-poor poverty interventions in this and other developing countries. The Kenya Government has awarded ILRI US$250,000 to undertake this study. 2004 Western Kenya Study: This study revealed that poor families move through six stages of progress out of poverty – from being able to secure food (stage one) to purchasing a sheep or goat (stage six). Fourteen stages were identified and these stages highlight the relative importance of livestock to the poor. The main findings are summarised in an ILRI Top Story. Click here to link to From Poor to Well-Off: Livestock can make a difference. Deep-seated customs can play a significant role in a family's descent into poverty and were identified as such by individuals surveyed. Raising awareness of the crippling effects of these customs, through a media campaign, could help get communities talking about the problems, and this could lead groups to actively seek solutions. The main findings are summarised in an ILRI Top Story. Click here to link to Funerals, Thefts and Bride Price: Livestock Loss Leads to Poverty. Click here to link to the full report Pathways out of Poverty in Western Kenya and the Role of Livestock. 2005 Peru Study: This study found that, overall, the number of households in poverty declined by 19% over 25 years in the 40 Andean communities studied. However, it also found that while some households escaped poverty, other households in the same communities fell into poverty and became poor. In addition to helping households escape poverty, stopping or at least controlling descents is essential to reducing poverty. The hole at the bottom must be plugged before there is any chance of filling the bucket. Else, households will continue slipping into poverty even as other households escape. Diversification of income sources – from livestock, crops and non-agricultural sources – are positively and strongly related to escapes from poverty. Market access, gains from small businesses, and community organizations are also positively and significantly associated with escaping poverty. On the other hand, health, land division, and social expenses (on marriages and funerals) tend to perpetuate poverty. Source: Excerpted from the draft working paper: The Hole at the Bottom of the Bucket: Household Poverty Dynamics in Forty Communities of the Peruvian Andes, Anirudh Krishna, Patti Kristjanson, Judith Kuan, Gustavo Quilca, Maren Radeny, and Alicia Sanchez-Urrelo