or an experimental effectiveness (or a randomized controlled trial) study in which three intervention groups and a control group undergoing an attentional network test (three networks to be measured) , what would be an appropriate statistical procedure for testing the hypotheses? I would appreciate answers, or at least some light shedding.
Hello Inge,
The answer depends a lot on what the outcome variable is, how it is quantified, and how data are collected.
At first blush, the 4 groups x 3 networks (conditions?) could be:
1. (ordinary) factorial anova, if the scores are of interval strength, cell variances are homogeneous, and residuals are normally distributed;
2. one-between subjects (groups) and one within-subjects (network), if all cases are exposed to all three network conditions. This involves a repeated measures factor (network), so the analysis is a bit different from #1. Again, though, there are assumptions involved.
There are "non-parametric" analogs to these options, if the scores are at least of ordinal strength.
Good luck with your work.
I see that you can choose either factorial arrangement with RCBD or a split plot. The split plot needs to qualify which variable is most important to you for detailed information, you put it as the second variable. Wish you luck.
- Badges
- Science method