Hello everyone,
I've recently submitted a manuscript, which was sent back to me with reviewers' comments and suggestions. In one of them, they asked me to perform a sensitivity power analysis given my actual sample size. My experimental design was a three-group intervention (i.e., one control and two experimental conditions) with a binary outcome. I've been struggling to define which family and statistical test I should choose for that purpose (I'm using GPower), for example: an F test family, in which I can add the number of groups and select an ANOVA fixed effects test, or a z test, in which I cannot include the number of groups but I it is the type of model I used to test my hypothesis.
I'm not sure if my reasoning is correct, but I'd much appreciate any help you can provide!
Hello Isabel Tostes Ribeiro
If, as you say, your Outcome is binary, then I must say you are on the wrong path.
The secret of which way is right lies in your study Dependent (Binary).
I don't really understand your problem... You had analyzed the data, presumably using some test. This should be the test used in the power analysis. Given your outcome is binary, this should be (and have been) either a binomial test or some kind of Chi-squared test.
I don't really understand your problem... You had analyzed the data, presumably using some test. This should be the test used in the power analysis. Given your outcome is binary, this should be (and have been) either a binomial test or some kind of Chi-squared test on simulations.
A post-hoc power analysis is generally non-sense. But you may not want to discuss this with the reviewer...
Initially, I did not understand the question either. But a search on took me to a document by my colleague Karl Wuensch. It includes the screen capture image I attached to this post. In Karl's example, d=0.093136 is the minimum value of Cohen's d that would have been significant with the observed sample size (and all other things being equal).
Karl's document: https://core.ecu.edu/wuenschk/stathelp/poweranalysis_overview.pdf
Isabel Tostes Ribeiro, what kind of statistical test did you use? That will determine which Test Family and Statistical Test options you must choose (see the 2nd attached image).
HTH.
- Badges
- Science topic
- Similar topics
- Filing