Aggregate Matching

In conditions where participants cannot be randomly assigned to program and control groups, the use of proper nonrandomized control groups is recommended to more accurately assess the effects of the independent variable under study. Aggregate matching is a procedure for devising matched controls in quasiexperimental evaluation research. Individuals are not matched, but the overall distributions in the experimental and control groups on each matching variable are made to correspond. For example, as a result of this procedure, similar proportions of characteristics such as gender and race would be found in both the program and comparison groups.