It's probably fine from a sample size perspective. The problem is that sample size potentially has inherent bias because it only includes people who game online.
Without any data, we can't say for sure, but I think it is a reasonable assumption to make that if you play an earlier edition of D&D, it is less likely that you play over a VTT, compared to more recent editions of D&D.
My assumption is on the basis that older editions of D&D are more likely to be played by older people (i.e. playing the edition they grew up with). I would further assume that the older you are, the less likely you are to play D&D over a VTT (for various reasons).
That said, they are all assumptions I have made, with absolutely no data to back up whether those assumptions are correct or not.