Three conditions + one control, measured for IQ pre intervention and post intervention. Is this a two or one way repeated measures ANOVA?

Also... does anybody know if SPSS repeated measures option is a one way or two way ANOVA?

  • $\begingroup$ The organization of your intervention isn't quite clear. Did you measure each subject before intervention and again after intervention? You have three interventions and a control. Are these three, A, B, and C,related to each other in any way? For example, is C a combination of A and B, or are they increasing amounts of a treatment? $\endgroup$ Commented Jun 8, 2021 at 1:25
  • $\begingroup$ Hi! Thank you for your reply. Group A - music lessons, Group B - language lessons, Group C - language and music lessons. None of the participants participated in more than one group. IQ-scores were measured before the intervention and again after the intervention. $\endgroup$ Commented Jun 8, 2021 at 1:30

1 Answer 1


You have two treatment factors: one is language lessons-or L, the other is music lessons-or M. Some people get one or the other, some get both, and some get to go do something else.

There could be an interaction between language and music, called LM, which is the combined effect of L and M. If the combination is quite strong, then LM could be large. If language lessons and music lessons interfere with each other, then this could actually be negative. If there is no joint effect of L and M then the effect of both should be about the same as the sum of the two separate effects. The effect of L plus the effect of M should about add up to the effect of the combination. (For effect, also read the change from before to after.)

The way to describe this is as a two factor or two treatment experiment, with each factor at two levels, nothing or intervention.

There is a third factor, usually termed Pre-Post or something of the sort that isn't very interesting. This is a repeated measures factor if you do an analysis of the original individual observations. This analysis might not be useful or informative. I take it you are interested primarily in change and this could guide your analysis.

You could calculate the differences between the second measurement and the first measurement of you dependent variable. Then do an analysis of variance of the differences using your two factor design. This should tell you whether groups changed and if the combination (LM) added anything to the components L and M.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.