jamovi

Hi!
Thanks for this neat software!

My apologies if this suggestion has already been made, but might it be possible to include an option to split the dataset along one variable for analyses. It is already possible to split the file in the Descriptives tab, but it would be great to have this option elsewhere too, for example when running one-sample t-tests (to give a specific example, I'm trying to test performance against chance level for each one of several experimental groups).

Thanks very much in advance for your comments!
H

hi,

splitting data is a little way off yet, but there are some ways you can achieve the same thing.

you can compute a new variable, with the FILTER(). something like:

= FILTER(group, group == 'group1')

in this way, you can create a new variable for each group, and analyse them separately.

it's not quite as slick as being able to 'split' the dataset, but it will get you there at a pinch.

cheers

jonathon

Thanks for the suggestion! I will try that.

Excellent work overall on that software

jonathon wrote: in this way, you can create a new variable for each group, and analyse them separately.

My dataset already contains a 'group' variable, but I dont understand how to analyze the groups separately. For example how do you subset the data so that analysis is only performed on group==1?

hi,

in the following example, we run an ANOVA on *only* the `dose == 'VC'` rows. we do this by creating a computed variable called `len (only VC)`:
makes sense?

cheers

ah gotcha. So theres no direct way to subset the data without creating new variables? That would be pretty useful if you're testing out models using different variables. It seems that currently you'd have to recreate a new variable(s) whenever you 1) want to work with a different subset or 2) want to test out different variables in the model

we're working on filtering at the moment, so hang tight.