Recurrent Institutional Cycle Design

This is actually more of a concept than a specific design. This approach is appropriate in situations where a treatment is repeatedly being applied, on a cyclical basis, to a new group of respondents. What you basically do is fine-tune the design on subsequent cycles to investigate questions raised on previous cycles. Here are three examples:

In design A, the experimental treatment is applied to an incoming class (Class A). The posttest is administered at the same time a pretest is administered to the next incoming class (Class B). Class B then receives the experimental treatment and then a posttest. Comparing 01 and 02 is similar to comparing a treated population with an untreated population (though not as confidently). Comparing 02 with 03 helps to answer questions about gain or loss due to the experimental treatment, and comparing 03 to 01 provides information about the possible effect of the pretest on posttest scores.

Designs B and C may be useful if you have the luxury of being able to randomize a group into two groups, one pretested and one not. Note that the only difference in these two designs is that:

01 and 02 occur at the same time in design B but at different times in design C (and on a different group of participants).

03, 04, and 05 occur at the same time in design B, but in design C 05 occurs in a subsequent cycle (and on a different group of participants)

Let's try something a little different. This time, how about some of our readers telling me the different questions that these two designs can help to answer and do a little comparison and contrast. Try to think in terms of extraneous factors as much as you can.

Return to quasi-experimental designs.