[R] Running effect size tests on subsets of data

[Giovanna Badia]

Hello all,

I have seen instructions for conducting post-hoc pairwise comparisons in R after running an ANOVA, Kruskal-Wallis Test, or Yuen�s Test.

Is there a method to subset data automatically to run an effect size test across different pairs? For example, if I want to run Cohen�s d to examine the differences in data between different days of the week. My X variables would be the pairs below. (Note that no data was collected on Saturday.) I can subset the data manually and apply the effect size to each pair. Is there a way in R to create the pairs and run the effect size test all at once?

Monday-Tuesday
Monday-Wednesday
Monday-Thursday
Monday-Friday
Monday-Sunday
Tuesday-Wednesday
Tuesday-Thursday
Tuesday-Friday
Tuesday-Sunday
Wednesday-Thursday
Wednesday-Friday
Wednesday-Sunday
Thursday-Friday
Thursday-Sunday
Friday-Sunday

Any assistance you can provide would be much appreciated.

Thank you,
Giovanna

GIOVANNA BADIA | Assessment Librarian, McGill University | Office of the Dean of Libraries | giovanna.badia using mcgill.ca

	[[alternative HTML version deleted]]