Good day Gurus!
I am new to research and which test to run as well as sample size.
I have three pre-tests (two scales and a writing prompt- gathering data from the writing), participants than randomized into three groups, separate intervention for each group, then three post-tests(writing prompt to test if there is a change from the intervention and same two scales from pre-test) with a survey at the end. I am evaluating the changes after the intervention, to hopefully see a positive change.
I have tried GPower, yet not sure how to pick the test which would give me sample size. I am thinking a multivariate comparison, yet that is why I am asking.
Independent Variables - Intervention
Dependent Variables - positive change, post-test scales, second writing prompt that should change due to intervention
Controls - Pre-test scales
Suggestions?
Thank you in advance. JDM
Hello Jason,
Assuming you're actually going for multivariate comparisons, you could use:
1. Multivariate regression (the grouping variable would be recast into two dummy variates to capture the information about three independent groups);
2. Manova
G*Power offers manova computations, which will mirror those of the multivariate regression, for a given effect (model term).
In either choice, how you treat the pre/post scores will make a difference. Your options are: (a) include both (this makes for a repeated measures dimension; in the regression approach, you'll have to include group X occasion interaction terms; you'll get these automatically in the manova; or (b) use the pre-treatment scores as covariates (just force them in the regression model as IVs along with the treatment variables), or declare them as covariates in the manova.
If you opt for the repeated measures framework, then the treatment X occasions interaction is usually the effect of interest. In the covariate framework, you're essentially comparing treatments on post-treatment scores, adjusted for pre-treatment differences.
Finally, though a lot of people say they want multivariate comparisons, it is quite common to see all of the attention focused on univariate outcomes in such studies.
Good luck with your work.
GPower will be pretty good to give you required sample estimates.
You can also utilize manual calculation using sample size formulae by published by many authors e.g. Yamene
Goodluck
Thank you Chukwuemeka. I appreciate your help and great suggestion.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size