To make a long story short, my research showed that measurement errors in regression don't really matter as long as you sample enough, which is kind of what I expected anyway.

I'd elaborate further and illustrate with pictures and tables of data, but I feel like not looking at this stuff ever again.