An impact evaluation can be said to have the following elements:
Input: An impact evaluation is always conducted with respect to an input (a project, program, policy or campaign). The program itself may be a vehicle to carry a product (say, a drug, a nutritious diet, a communication message, etc) which is actually supposed to cause the change or improvement.
Population: The population of interest to the evaluators refers to the group of people who are immediate or indirect/ secondary stakeholders in the program. The evaluation draws its sample from the population.
Sample: A sample is a set of population units (individuals, groups, households, etc) selected either at random or purposively (on the basis of certain pre-defined criteria). The random sample may be selected either by drawing randomly from a list (simple random) or by drawing from a list after providing for a sampling interval (systematic random). The purposive sample may make use of expert opinion (judgmental sampling), secondary sources of data, historical trends, or practical considerations such as cost, time or convenience to determine the criteria for sampling.
Change: The change refers to the benefit/ improvement that the input is supposed to lead to. It may, in some cases, be tangible (e.g. weight gain due to a diet or dosage) and in other cases, less so (e.g. improvement in awareness level, efficiency, or application of new knowledge)
Control Group: The control group refers to the population that did not – intentionally or unintentionally - receive the benefits of the program. It is this group from where the evaluators draw the “counterfactual” (i.e. what the situation in the population of interest would have been in the absence of the program).
Null Hypothesis: A hypothesis represents the assumption that is to be tested through an evaluation. The null hypothesis is one that assumes zero causation (i.e. zero relationship between the input and the change). The alternative hypothesis is opposite of a null hypothesis.
Measurement Tool: The impact evaluation makes use of various tools to measure the impact on various parameters. The most commonly used tools include sample surveys.
Evaluator: The evaluator may be an internal one (i.e. one of those responsible for the management or execution of the program) or a third party (i.e. a neutral or private research agency or auditor). In certain cases, the evaluation may be completely planned and managed by the representatives of the primary stakeholders (the program beneficiaries). In the latter case, where the element of public participation is extremely strong, the evaluation acquires the shape of a social audit.
Design: Each impact evaluation is based on a specific evaluation design. Although evaluation designs may produce useful information about a program's effectiveness, some may produce more useful information than others. For example, designs that track effects over extended time periods (time series designs) are generally superior to those that simply compare periods before and after intervention (pre-post designs); comparison group designs are superior to those that lack any basis for comparison; and designs that use true control groups (experimental designs) have the greatest potential for producing authoritative results.
