Research Design
What is Research Design?
A research design provides the structure of any scientific work. It is a plan outlining how information is to be gathered for an assessment or evaluation. Typical tasks include identifying the data-gathering method (or methods), the tools to be used for gathering information, the group or population from whom the data will be collected, and how the information will be organized and analyzed. A critical decision is the choice of a counterfactual, which measures what the possible outcomes or impact would have been if there hadn’t been an intervention. This can be done using a pre-test, a comparison or control group, or both.
What are the types of Research Design?
The choice of design should be influenced by the resources (e.g., money, skill level) available as well as the degree of scientific rigor desired. Typically, an outcome evaluation uses one of three designs: randomized control trial, quasi-experimental designs with comparison groups, or pre- and post-intervention comparison.
-
Randomized Control Trial (RCT): This design option uses two or more groups of participants who are randomly assigned to either the treatment in question or to a control group that is not exposed to the treatment. Members of both groups receive the same pre-treatment and post-treatment assessments. The RCT is considered the gold standard of research design, but is not required for most program evaluations.
-
Quasi-Experiment with Comparison Group: This design option is similar to the RCT design, except that the comparison group is not randomly assigned. Comparison groups are chosen so that participants are as similar as possible to those in the treatment service or system being evaluated.
- Pre- and post-intervention comparison: This design assesses participants on the same variables and over the same time period, before and after they complete treatment. This design option is not as rigorous as the two previous designs, but it is often more realistic for organizations with limited evaluation experience and/or resources. In this design, the only participants who are measured or assessed are those who received the treatment. Although pre- and post-intervention designs are less scientifically rigorous, they can produce useful results for purposes of accountability and program improvement.
What are the key features of Research Design?
-
Measures or observations can refer to a single measure (e.g., a measure of body weight), a single tool with multiple items (e.g., a 10-item self-esteem scale), a complex multipart tool (e.g., a survey), or a whole battery of tests or measures given out on one occasion.
-
Treatments or programs can refer to a simple intervention (e.g., a one-time treatment) or to a complex program (e.g., an employment training program).
- Sampling is the process of selecting units (e.g., people) from a population of interest so that by studying the sample one can fairly generalize the results back to the population from which they were chosen.