I'm currently trying to figure out if there's anything wrong with how I am choosing to analyze my data. Specifically, I have three groups, one dependent variable (y), and one independent variable (x). I am trying to predict the same y from the same x for each group. I suspect that one group shows a relationship while others do not. In order to do this, I've filtered a large data set, creating three smaller data sets for each group. I then ran a linear regression on each of these groups using the same DV and IV. My goal is to determine if x predicts y, but only for one group. Is there anything wrong with running an analysis like this?