Another name for this quasi-experimental design is the "Multiple Time-Series Design". It can be visualized as follows:
This design is simply the Time Series Design with the addition of a comparison group. (Ideally, the experimental treatment 'X' would be randomly assigned to one group or the other.)
In our diabetes example, this would be equivalent to selecting two (or more) similar clinic sites and implementing the intervention in one but not in the other. The addition of the comparison group allows you to rule out the extraneous factor of history as a potential explanation for any results observed and contributes to generalizability.
This design produces outcome patterns similar to those exemplified under the Time Series quasi-experimental design, except that both experimental and control group data each produce a line. Comparison of the two lines can be informative.
Return to quasi-experimental designs.