I am a big fan of Stat iQ, but have run into some issues when relating a variable with check-all-that-apply (e.g., types of discrimination) and a category (e.g., gender). The kicker is that the responses to the types of discrimination is based on display logic of those who responded "yes" to the previous question asking whether or not they have experienced discrimination in the workplace. So, not everyone who responded to the gender question is responding to the types of discrimination question.
When I export the results after relating them in Stats iQ, I get a lot of wonderful graphs, some of which are based on a sample size for all who responded to gender and another sample size based on all the boxes checked for each response option. But there doesn't appear to be an option to relate to the response count of the types of discrimination question for each gender category. I ran the numbers manually (using filters in the results section) and came up with very different results, which is disconcerting. See attached screenshot of the checked counts and sample sizes. The highlighted column is the one I compiled manually and then used to create my own graphs.
I'm wondering if this is limitation with Stat iQ, or is this due to me not understanding how to set up this type of analyse. I am hoping there is a way to set this up for Stat iQ to run rather than me having to run these manually. Any insights would be greatly appreciated!