On Fri, 6 Jun 2014, Mike Palij went:
if one does retrospective power analysis and finds that for the sample sizes used, there is low levels of power, then should request that a new study with a larger sample size be used
I used to use, and request, post hoc power analyses to accompany seemingly null findings. But now I use and request confidence intervals instead, per papers such as these: "Confidence intervals are a more useful complement to nonsignificant tests than are power calculations" <http://beheco.oxfordjournals.org/content/14/3/446.full> "The Use of Predicted Confidence Intervals When Planning Experiments and the Misuse of Power When Interpreting Results" <http://annals.org/article.aspx?articleid=707593> --David Epstein da...@neverdave.com --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=37065 or send a blank email to leave-37065-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
