>That seems to be highly unscientific/prone to skewing the sample.
I'd argue the emails they did send are MORE prone to skewing the sample.
If I had no CCPA plan and got an email from someone introducing themselves as researchers, I'd tell them I haven't gotten around to it, and that I intend to comply but haven't had anything prompt me to put in effort regarding CCPA.
If I got the email they did send, that would be exactly the request that makes me go do the legal research, and my response would as narrow as possible.
Yes, perhaps they should have hired real people to do real requests. That would have been a bulletproof way of studying this ethically so your conclusion is moot.
That seems to be highly unscientific/prone to skewing the sample. It probably would have rendered the results useless.