I second the suggestion to use a program other than SPSS (such as the free G*Power) to do the power analysis. SPSS may compute power using an unreasonable assumption -- that the actual effect in the population is of the same magnitude as the observed effect in the sample. Better to find power for the smallest effect that you would consider not to be trivial in magnitude. If that power is high, then you can make a strong statement regardless of whether your effect is "statistically significant" or not.
-----Original Message----- From: Noel Kinder [mailto:nkinder@;psyc.umd.edu] Sent: Tuesday, November 12, 2002 8:47 AM To: Teaching in the Psychological Sciences Subject: Re: SPSS & Power Dear Rob, In SPSS 11.0 it is possible to calculate the estimated power for ANOVAs. I have not tried this for a one-way ANOVA however. >From the Help guide for "GLM Univariate Options..." "Display...Select Observed power to obtain the power of the test when the alternative hypothesis is set based on the observed value. Select Parameter estimates to produce the parameter estimates, standard errors, t tests, confidence intervals, and the observed power for each test..." My favorite power resource is G*Power. G*Power is an excellent power estimation tool. While there are limitations to the tests it applies to, it does cover ANOVAs. In addition, I recommend G*Power to people BEFORE they start research for getting your needed sample size. It is easy to get your sample size when you enter your desired power and effect size. To get G*Power (free), you go to: http://www.psycho.uni-duesseldorf.de/aap/projects/gpower/index.html How to Use G*Power http://www.psycho.uni-duesseldorf.de/aap/projects/gpower/how_to_use_gpower.h tml Enjoy Noel Kinder, MA was University of Maryland PhD Student now Child Welfare League of America Research Specialist --- You are currently subscribed to tips as: [EMAIL PROTECTED] To unsubscribe send a blank email to [EMAIL PROTECTED]
