>sue gerber wrote: >> >> Silly question? What is the non-parametric equivalent to a 2x3 repeated >> measures ANOVA design? >> >> Thanks > >Unfortunaetly, > There are no simple non-parametric equivalents to a 2X3 Anova. >The Friedman test will allow you to do a 1-way Repeated measures ANOVA >with non-parametric data. > > M. Kruger

Sue, a quick fix would be to first transform the data to ranks and THEN use the general factorial analysis in SPSS, etc. Course the p-values will be approximations, etc. Essentially this is how the Freidman Test is done. That is, it's a simple one-way ANOVA on the ranks.

Now my fellow statisticians might get a heart burn with this solution. But I don't.

Take care - Ed

************* http://maclab.tamucc.edu/math/ejones.html ************** Edward R. Jones | Office: (512) 994-5850 Seabreeze #10 | Fax: (512) 994-5848 Texas A&M Univ. - Corpus Christi | Internet: ejones@tamucc.edu 6300 Ocean Drive | AMPRnet: ab5eh@ab5eh.ampr.org Corpus Christi, TX 78412 | Packet: ab5eh@kf5px.#stx.tx.usa.na ****************************************************************************