Anyone who thinks an adult telling about being molested as a child is a liar --

[Donald Ian Rankin]

213. Some back-of-an-envelope calculations:

If your figure of 4,000 children reported as abused in Kentucky in one year is typical for Kentucky, then a population of 4,000,000 means 1 person in 1000 is reported as abused per year, so a 70-year lifetime means about 1 person in 13 or 14 is reported as abused - which, I admit, is higher than I'd expected.

That obviously omits at least four things, though - reports of abuse that turn out not to be accurate, multiple reports of abuse of the same child, abuse that never gets reported, and the possibility that other states have more or less abuse per capita than Kentucky. But, as the report says, we don't have any data on at least two of those.

It's probably a safe assumption that there as the fraction of reports of child abuse that are not genuine is "very small" and the fraction of child abuse that gets reported is "not that large", but baldly presenting a number like "1 in 4" as fact, rather than "probably significantly worse than 1 in 13 in Kentucky, although we can't be certain" is not evidence-based discussion.

Willful ignorance is a bad thing, but in this case ignorance is the only option - the data simply isn't there. The choice is whether to admit that we are ignorant and leave it there, or to make guesses. But if you take the latter choice, they *must* be heavily fenced around with qualifiers and admissions that guesses are all they are, not simply presented as "common facts".