 |
 |
Thanks to all who responded, apologies to any whose responses I missed; it's been a very busy time, and I may have inadvertently deleted some post in my urge to delete spam. I had asked about programs to do power analysis, particularly for repeated measures. Jeremy Miles wrote <<< I wrote a paper, published in BMC Medical Research Methodology, that describes how to use a structural equation model to estimate the power of designs, such as repeated measures or mixed ANOVA designs. The paper uses examples in Mx, which is freeware, although any SEM program could be used. Mx is a bit tricky to use, to start with, but you should be able to easily modify the examples that are given. You can find it here. http://www.biomedcentral.com/1471-2288/3/27 Another approach is to use the method described by D'Amico, Neilands and Zambarano, Power analysis for multivariate and repeated measures designs: A flexible approach using the SPSS MANOVA procedure, Behavior Research Methods, Instruments, & Computers, 33, 4, 479 -- 484 (although obviously you are going to have to use SPSS for that.) >>> Len Thomas wrote <<< Several years ago a colleague and I undertook a review of power analysis software. The citation is Thomas, L. and CJ Krebs. 1997. A review of statistical power analysis software. Bulletin of the Ecological Society of America 78 (2): 126-139. More conveniently, a copy is online at http://www.zoology.ubc.ca/~krebs/power.html Unfortunately for you, we concluded that PASS was at the time the easiest to use of the software available. We did review 28 other programs however, so perhaps you can find something suitable there. I think we looked at whether the programs explicitly accommodated repeated measures analysis. >>> Stephen Sen wrote <<< I use nQuery and find it useful and easy to use (but I am on the advisory board for Statistical Solutions!). It covers repeated measures ANOVA (although I have no personal experience of using this). See http://www.statsol.ie/nquery/features.htm >>> I also discovered that SAS v 9, which I recently upgraded to, does some power analysis, although not for the sort of repeated measures I needed. In addition, I have been in contact with Jerry Hintze, who wrote PASS, which I use now. While I still find PASS somewhat difficult to use, especially for repeated measures, I must say that he is extraordinarily available, and very helpful personally. Thanks again to all Peter Peter L. Flom, PhD Assistant Director, Statistics and Data Analysis Core Center for Drug Use and HIV Research National Development and Research Institutes 71 W. 23rd St www.peterflom.com New York, NY 10010 (212) 845-4485 (voice) (917) 438-0894 (fax)