jamovi

A suggestion for a new feature: For t tests, allow the user to specify the two levels of the independent variable (IV) to be used in the test, rather than require that the data file include only two levels of that IV.

yes, good idea. i've been toying with this idea for a while.

this may also get solved once we allow the user to make subsets, etc.

thanks for the suggestion

Yes, PLEASE add this feature soon - also including for the TOSTER module. It is so annoying to go back to Excel all the time in order to create a new data file...

I've always been irritated that SPSS requires me to enter the two values. It seems that if they are 1,2 I'll guess 0,1 or vice versa (I'm frequently analyzing other people's data), generating the error message, "at least one of the groups is empty." So having this as an option would be very useful, but if I had to pick just one, I'd stick with requiring that the factor have two levels & select a subset of the data set when there are more. The 2-only perspective also helps beginners understand that if they have more than 2 levels, perhaps there's a more appropriate approach like a 1-way ANOVA (this is regard to beginners, not others in this forum who I'm sure are well aware of that.)

ah yes. i should have followed this up.

this is how you can do it with the 'computed variables' (this is using the tooth growth example).