My students and I recently completed a pilot intervention study for anxiety in Division 1 volleyball players. We studied the players from one volleyball team over the course of a season, measuring self-reported anxiety before the season, 8 weeks into the season, and one-month after the end of the season. We had three groups, two intervention groups and one control group. The team has 21 players, so we started with 7 in each condition. Due to drop out across the intervention, the 8-week time point had 6, 5, and 4 players in each group, and the post-season time point had 6, 4, and 3 players in each group. My question is whether we could still use Hedge's g as an initial measure of "effect" difference in the various measures or whether our very small groups preclude us from using this statistic. Any references either way would be appreciated!
Brydges, C. R. (2019). Effect size guidelines, sample size calculations, and statistical power in gerontology. Innovation in aging, 3(4), igz036.
this may prove a tiny bit useful! :)
As far as I know, nothing about the small sample size makes the effect size estimate invalid, but you might calculate confidence intervals for it. The interval might be quite wide.
- Badges
- Science topic
- Similar topics
- Social Psychology
- Attitude