I agree with you and what really takes the cake away from taking their research as reliable and valid is not only that they skimmed over any kind of educational background and programming concepts (for the sake of the argument I'll assume that they consider them as granted) but calling out loud that engineers that work for Microsoft are worse than the rest is to my eyes a sad generalization that is used just to pinch nerves and win smiles from people who consciously and for reasons of their own don't like Microsoft products (myself being one of them).

That would be like hiring designers based on whether they know photoshop or coreldraw.

I don't think companies are doing that. The causality just isn't being addressed properly.

I can come up with a lot of plausible reasons why you're more likely to find "good" engineers focusing on Python or Java, and why Ruby developers might be more cross-disciplined, and none of them are because the engineers are defined by the language.

This "study" is laughably poor in presentation, and their interpretation of the results they got is almost certainly unforgivably flawed.