Dear all,
As part of my bachelor thesis, I am using a simple mediation with bootstrapping procedure. I would now like to calculate the power of my analysis. For guidance, I have read the 2007 paper "Required Sample Size to Detect the Mediated Effect" by Matthew S. Fritz and David P. MacKinnon (DOI: 10.1111/j.1467-9280.2007.01882.x). I would like to use the "empirical estimates of sample sizes needed for .8 power" (Table 3) to assess the power of my mediation analysis (using bootstrapping). As far as I know, the power of a test always depends on the sample size, the effect size and the type 1 error. Unfortunately, however, I could find no information on the probability of type I error on which Fritz and MacKinnon's percentile bootstrap and bias-corrected bootstrap power analyses are based. I wold now like to ask, if anybody knows which type 1 error ist used in the paper. It would be very helpful if somebody could help me with this question.
Kind regards
Leonie Aderhold
You are right. It is odd that Fritz and MacKinnon (2007) do not seem to provide the specified alpha (Type I) error rate explicitly in their Methods or Results sections. In the Discussion section of that paper, they mention "a Type I error of .05", so I would assume that they used .05 in their simulations (it is the most commonly used Type-I error rate), but it is not entirely clear. Perhaps send an email to Matthew Fritz (https://cehs.unl.edu/edpsych/faculty/matthew-fritz/) to confirm.
Dear Mr Geiser,
thank you for your response. I have contacted Mr MacKinnon and he confirmed that you were indeed correct about a type 1 error of .05.
Sincerely,
Leonie Aderhold
- Badges
- Science topic
- Similar topics
- Names