Use of a comparison group is an important component of a well-designed study. In order to determine if the program is effective, it is important to make comparisons between the intervention group and another group. The choice of the comparison group is important. There are a couple options:

• Pre-post comparison. In this design, you compare a group that has had the intervention to cases in the same site prior to the intervention. For example, if you implement a mediation program, you compare cases that closed in 2017 (prior to mediation being implemented) to cases that closed in 2019 (after mediation was implemented in a site). This method has the advantage that site characteristics should be similar between pre and post because you are using the same site. The disadvantage is that you cannot account for history effects (that is, something else may have happened in the site during the time period of your study that explains your findings).
• Comparison between sites. In this design, you compare a site that implemented a program with a site that did not. The most robust way to do this is to compare both sites using pre and post samples (before and after the intervention). The disadvantage of this is that the sites may look different for a number of reasons, including geographic makeup, local court practice, etc. and this may influence findings.
• Matched sample comparison between sites. A matched sample comparison group is a more rigorous research design. It allows for cases to be matched between an intervention and a comparison group on key factors. In child welfare these would be case factors that we know from the research are related to outcomes (e.g., age, substance abuse). Cases are matched so that they are similar and then compared to see if the program is related to an outcome.
• Random assignment to groups. The only comparison group that allows you to say that your program caused a change in outcomes is an experimental group where cases are randomly assigned to be either part of the intervention or part of a control group. This is the most rigorous comparison group and allows for the best possible chance of learning whether your program is effective.