# Which statistical test to use?

I have acquired data with multiple variables that I need to test. My dependent variable, Seed number, is tested between Ecotypes but also between Treatments (hot and cold). I need to see whether Seed Number between Ecotypes is the same when different Treatments are applied. In other words I might expect to see a difference in Seed Number between Ecotypes in the hot but not cold. The test is complicated by the fact that multiple repeats were taken that are likely to be significantly different and therefore grouping the repeats may confound the results.

I am using SPSS so any help there would be useful.

You may try an approach similar to the store-price example on this page: http://ww2.coastal.edu/kingw/statistics/R-tutorials/repeated.html

If your variable (column) names are SubjectNo, SeedNo, Ecotype and Treatment, the statement can be:

aov.out=aov(SeedNo ~ Treatment*Ecotype + Error(SubjectNo/Treatment), data=mydata)


Edit: Actually, following simpler form may be right for this situation:

aov.out=aov(SeedNo ~ Treatment*Ecotype, data=mydata)

• Unfortunately it's not repeated measures. I wonder if its two-way ANOVA focussing on the interaction between Ecotype*Treatment – statsconfusion Apr 16 '15 at 15:27
• Please see the edit above. – rnso Apr 16 '15 at 15:32

If you are interested in using SPSS, you can use ANCOVA for your objective.

DV-Seed number (continuous)

IDV- Ecotypes (categorical)

By controling for

Covariate- Treatment ( categorical)