What is impact evaluation?
An impact evaluation assesses changes in the well-being of individuals, households, communities or firms that can be attributed to a particular project, program or policy. The central impact evaluation question is what would have happened to those receiving the intervention if they had not in fact received the program. Since we cannot observe this group both with and without the intervention, the key challenge is to develop a counterfactual – that is, a group which is as similar as possible (in observable and unobservable dimensions) to those receiving the intervention. This comparison allows for the establishment of definitive causality – attributing observed changes in welfare to the program, while removing confounding factors.
Impact evaluation is aimed at providing feedback to help improve the design of programs and policies. In addition to providing for improved accountability, impact evaluations are a tool for dynamic learning, allowing policymakers to improve ongoing programs and ultimately better allocate funds across programs. There are other types of program assessments including organizational reviews and process monitoring, but these do not estimate the magnitude of effects with clear causation. Such a causal analysis is essential for understanding the relative role of alternative interventions in reducing poverty.
Why conduct an impact evaluation?
Information generated by impact evaluations informs decisions on whether to expand, modify, or eliminate a particular policy or program and can be used in prioritizing public actions. In addition, impact evaluations contribute to improve the effectiveness of policies and programs by addressing the following questions:
- Does the program achieve the intended goal?
- Should this pilot program be scaled up? Should this large scale program be continued?
- Can the changes in outcomes be explained by the program, or are they the result of some other factors occurring simultaneously?
- Do program impacts vary across different groups of intended beneficiaries (males, females, indigenous people), regions, and over time?
- Are there any unintended effects of the program, either positive or negative?
- How effective is the program in comparison with alternative interventions?
- Is the program worth the resources it costs?
When to conduct an impact evaluation?
Impact evaluations demand a substantial amount of information, time and resources. Therefore, it is important to select carefully the interventions that will be evaluated. One of the important considerations that could govern the selection of interventions (whether they be projects, programs or policies) for impact evaluation is the potential of evaluation results for learning.
Four questions can help guide the decision of when to conduct an impact evaluation:
Is the policy or program considered to be of strategic relevance for poverty reduction? The decision of what to evaluate depends on what are the most critical public actions to reduce poverty. Interventions that are expected to have the highest poverty impacts may be evaluated to ensure that poverty reduction efforts are on the right track and allow for any necessary corrections.
Is the intervention testing an innovative approach to poverty reduction? Impact evaluations can help to test pioneering approaches and decide whether they should be expanded and pursued at a larger scale. Hence, the innovative character of policies or programs also provides a strong reason to evaluate. This can be built into project design where, before committing large amounts of resources, multiple variations of the intervention are tested against each other.
Is there sufficient evidence that this type of intervention works well in a number of different contexts?
If the answer to this question is yes, then the scarce resources may best be devoted to helping adapt this intervention to local conditions and paying close attention to monitoring and supervision. If, however, there are significant differences in local conditions and/or the target population that cast doubt on the applicability of results from elsewhere, then an evaluation may be worth considering. (See our database of completed evaluations
to compare results.)
When do we expect outcomes to show an effect? Certain outcomes/impacts take time to materialize. In some cases this may mean that it is better to delay the final stage of the evaluation until these will show an effect. In other cases it may be better to choose a proximate set of indicators which are causally linked to the ultimate outcomes and are likely to show an effect earlier. Of course, the most comprehensive strategy is to combine both of these types of indicators.
How to evaluate the impact of interventions?
An impact evaluation must estimate the counterfactual, which attempts to define a hypothetical situation that would occur in the absence of the program, and to measure the welfare levels of individuals or other identifiable units that correspond with this hypothetical situation. In order to identify the proper comparison group, a careful understanding of how beneficiaries enter the program is critical. Ideally an evaluation should be built into a program as the program is designed which means that the evaluation should be planned as early as possible. Guidance for task managers on managing this process can be found in Impact Evaluation and the Project Cycle.
Note that it is important to distinguish between programs which cover some sub-set of the population (e.g. targeted social programs) and those that are national in scope (e.g. trade reforms). By nature, national programs do not have a group which is unaffected by the program. Hence, these programs require different techniques for evaluation such as CGE modeling, simulations, and before and after (reflexive) comparisons. These are not impact evaluations in the sense that it will be impossible to rule out confounding effects in order to establish causality.
In cases where there is variation in eligibility (e.g. by individual, household or geographical location), impact evaluation is likely to be feasible. This variation can be used to identify the appropriate comparison group through a range of methods. See Methods and Techniques for a more in depth discussion.
What is the role of impact evaluation in monitoring & evaluation?
As a component of the monitoring and evaluation process, impact evaluations are an essential instrument to test the validity of specific approaches to development and poverty alleviation. Impact evaluations help those involved on a project to establish whether or not there is a causal link between an intervention and those outcomes that are of importance to the policymaker. The counterfactual analysis used by impact evaluations is a critical tool for assessing the effectiveness of development interventions. By providing critical feedback with respect to what works and what does not, impact evaluations can help to solidify a results-based project structure.
In order to measure the impact of an intervention, a clear, well-designed evaluation strategy is necessary. Incorporating an impact evaluation into a development program requires a well-structured monitoring and evaluation plan. Every impact evaluation requires a specific methodological design, many of which are described on this site. Through conversations among project managers, government officials, and researchers, the appropriate methodology is chosen and incorporated into the monitoring and evaluation process.
Impact evaluations fit into the chain of monitoring and evaluation process in several ways. First, they help to assess the casual link between an intervention and an outcome of interest. Second, impact evaluations provide baseline evidence for the effectiveness of an intervention, which can be compared with other similar interventions. Through this process, impact evaluations assist in establishing credible cost-effectiveness comparisons. Third, impact evaluations can serve to build the knowledge base of what works in development. With and increasing demand for evidence of aid effectiveness, rigorous evaluations offer a method through which development successes can be highlighted.
What is the Development Impact Evaluation (DIME) Initiative?
The Development Impact Evaluation (DIME) initiative is a World Bank-led effort involving thematic networks and regional units under the guidance of the Bank’s Chief Economist. Its objectives are:
To increase the number of Bank projects with impact evaluation components;
To increase staff capacity to design and carry out such evaluations;
To build a process of systematic learning based on effective development interventions with lessons learned from completed evaluations.
Impact evaluations assess the specific outcomes attributable to a particular intervention or program. They do so by comparing outcomes where the intervention is applied against outcomes where the intervention does not exist. The appropriate comparison group is that which represents what would have happened in the absence of the intervention. By establishing a good comparison of outcomes for these two groups, an impact evaluation seeks to provide direct evidence of the extent to which the intervention changes outcomes. The goal is to verify the effect on outcomes that is caused by the intervention.
Impact evaluations generate knowledge on which sorts of programs create substantive results and which do not, and under what circumstances. Such information is critical not only for the policymakers directly in charge of the program evaluated, but also for others who may be considering adapting its approach for use in their own circumstances. Particularly when used strategically to test the effectiveness of specific approaches in addressing key development challenges, impact evaluations constitute the preferred approach to assessing results. Further, they can provide critical inputs (benchmarks) to other monitoring and evaluation activities.
Under DIME, groups of impact evaluations of strategic interventions are initiated in a coordinated fashion across countries in different regions of the world. This allows a comparative analysis of results in different settings and produces more robust estimates of program impact to inform policy and program design in the future. The initial themes include interventions to improve education service delivery, conditional cash transfers, urban upgrading programs, and early childhood development programs.