Please help settle a disagreement, One-Way vs. Repeated Measures ANOVA

Hello all, please help with a disagreement in statistical analysis for our research project. My colleague and I disagree on the correct procedure to use.

Research Design
a. There are 2 groups, an experimental group (n = 21) and a control group (n = 22).

b. There are 7 dependent variables, each with its own measure. Some of the DV's are related to one another: a) there are 2 measure of mental shift/attention, b) 2 measures of working memory, c) 2 measures of fluid processing, and d) 1 measure of attitude.

c. The 2 groups were given the measures as a pre-test and post-test.

Here is where the disagree is...We disagree on the type of analysis to run. Here are the options:

1. One-way anova. One of us feels that since the DV's and measures are pretty independent of one another, the best way to compare the 2 groups is using a one-way anova for each of the different sub-groups of measures. This would be done for the pre-test and post-test measures.

2. Repeated measures anova. One of use feels that the one-way anova is running too many tests and therefore increasing type 1 error. This person feels that a repeated measures anova is the correct way to run this.

Or...are we both wrong. Please help us decide the best way to run this analysis given our research design.

The research questions were pretty simple. We just predicted that the experimental group would outperform the control group on each of the measures. We did not predict any relationships among the DV's.


TS Contributor
Since you want to treat the measurements independently from each other,
7 "mixed" analyses of variance (with a repetead-measures factor and a
between-subjects factor) could be used.

With kind regards