There are 3 variables in my study. Cue type (valid, invalid), emotional type (negative, positive, neutral), and distractor (yes, no). all variables and levels are administered within the subject. How can I calculate how many participants I would need for such an experiment? I couldn't do this analysis in G*power. Could you help? What do you recommend?
@Ertan-Kaya The same query has been discussed in the below thread on Research Gate.
https://www.researchgate.net/post/What-is-the-best-way-to-determine-the-necessary-sample-size-for-a-two-way-ANOVA-in-a-psychological-study
Thank you very much for the reply. but 2way ANOVA is discussed there. that is, an experimental design made with independent variables. in mine, all variables and levels are repeated. so it cannot be done in g*power. I'm wondering if anyone has done/calculated in another way.
Hello Öslem,
G*Power wasn't really configured to offer precise estimates when you have multiple within-subject/repeated measures factors. You could:
1. Treat the two RM factors as a single dimension (e.g., a 3 x 4 RM design could be re-expressed as a 12-level, single RM dimension carrying all the relevant information of the two within-subjects factors). Doing so combines both of the individual main effects and their interaction, so choose the smallest ES that makes sense across those three terms.
2. Use a different software package to help guide your determination of a priori sample size. Here's one suggestion: https://aaroncaldwell.us/SuperpowerBook/
Good luck with your work.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size