How to select subgroups for a t-test procedure in a variable with more than two groups?
Hey friends! I was trying to use the filter function to make a grouping variable for a t-test with 4 levels use just 2 for a t-test. However I found that I got an error message regardless of filter status. Any thoughts? Screenshot attached.
Comments
Hmmm interesting... This looks like a bug to me
That said, there's probably a couple of ways to get the t test:
1) create a new variable with two values and missing if not in either of the two conditions of interest
2) do an ANOVA with post hoc pairwise comparisons, though you might be forced to use the pooled variance from the 4 groups
This is 0.19? If so, please report this on our GitHub page. I have seen this before and hoped we fixed it...
EJ