Impact evaluation assesses the changes that can be attributed
to a particular intervention, such as a project, program or policy. Before Jumping onto to methods there are few basic concepts that are needed to be defined. Casual Effect is the impact of a program that on to a specific situation or scenario. The Policy which is being analyzed is known as Treatment. Counterfactual is the imaginary situation which depicts the situation when the Treatment is not present. Control Group is a group of to be treated people that helps in estimation of Counterfactual. They are basically the people who do not take part in the program Treatment Group is group of people who have participated in the program Methods:
The simple difference method.
It is one of the most frequently used methods to measure impacts. Analysis based on simple differences measure the impact by comparing the post situation of those people who participated in a program with group of people that did not. The main assumption of this method is that those in the comparison group are totally same to those that participated in the program. A key advantage, and reason for its frequent use, is that this method does not require data on the situation prior to the treatment. However, there is a big drawback that if the treated and comparison groups are different in any way prior to the program, the method may be biased and may under- or overestimate the real impact of a policy; that is, selection bias.
Pre- vs. Post-Treatment Period
It is the simplest type of evaluation method. Instead of using a comparison group it only focuses on the treated group. In this methods the treated group is analyzed before the program and after the program. The data gathered after the analysis is then compared to come up to a solution. Situation of treated group before the implementation of program is considered as the counterfactual. The only benefit of this method is it does not require information on people that did not participate in the program. But this method also has a drawback that many factors that vary over time can affect an outcome. Difference-in-Differences Estimations A difference-in-differences evaluation is a combination of simple difference and pre-post evaluation method. The effect is calculated by measuring the change over time for the treated group and for the comparison group and then taking the difference between these two differences. The key assumption of this method is the assumption of common trends. The benefit of this method is it controls for all the characteristics that do not change over time (both observable and unobservable) and for all the changes over time that affect the treated and untreated group in the same manner. The drawback is that it is typically impossible to assess whether the two groups would have developed in the same way in the absence of the program. If this is not the case, the analysis will be biased.
Matching methods
Matching Method compare outcomes of treated individuals with those of similar
individuals that were not treated. participants are matched with individuals that are identical along selected characteristics but that did not participate in the treatment. The key assumption of this method is that those who participate in the program are, on average, identical to their matched peers, except for having participated in the program. The benefit of this method is it controls for observed characteristics that do not change over time. The drawback is that it is typically impossible to rule out that there are not also other, unobserved characteristics that differ between the groups, which would bias the impact estimation.