While your anecdotal report in and of itself, doesn't hold considerable weight, I think it's likely that it's the result of a response bias. Parents tend to be especially defensive when it comes to how they raise their child, and it's likely they often - unconciously or conciously - missrepresent the truth because of the surrounding social factors. Respondants often provide the answer they believe is what they should do, rather than what they actually do.
I don't think there's anything on the ESA website that helps us understand any measures they might have taken to minimize this. Generally speaking if you just send out thousands of surveys, you're data quality isn't liable to be that high, and you're simply hoping that the sheer weight of the data compensates for that. With that said, elements like a considerable liability for response bias will still slant the data, and that's quite possibly what we're seeing on some of the questions related to parenting.
I'm not saying it's worthless, it has valuable bits and pieces, but what are our takeaways from this report relative to 2016s? That's not clearly presented, and what is presented, is quite limited in its scope, which is a shame.
