Back

Assessing the Value of Education-Focused Laws in Reducing Child Sexual Abuse

Srivastava, K.; Thompson, B. M.

2022-02-01 health policy
10.1101/2022.01.31.22270166 medRxiv
Show abstract

I.We investigated the impact of new state laws requiring or encouraging education on Child Sexual Abuse ("CSA") in schools. From 2005 through June, 2019, 31 states enacted such laws. By comparing the states with and without those laws using a difference-in-differences calculation, and controlling for changes in population, wages, employment and other relevant CSA laws, we conclude that most likely such laws: O_LIIncreased the number of CSA reports made to state departments of child services each year, and C_LIO_LIDecreased the number of CSA cases that those departments were able to substantiate from the reported cases. C_LI We had hypothesized, however, that effective educational content would cause more reported cases to be substantiated as people learn what to report, but implementation of the new laws apparently instead increased the proportion of unsubstantiated reports of CSA. Our research does not tell us whether that is good or bad. In the best scenario, children are helped, but at the expense of potentially hurting those who are wrongly accused. In another scenario, children and the wrongly accused are hurt. In arguably the worst scenario, child are hurt and perpetrators are helped. We recommend that policymakers support research to learn why the laws are shifting more cases to unsubstantiated. Depending on the answer, the laws may be worthwhile to continue but with a modification to how the reporting is done, creating a new path for reporting troublesome but lawful activity. Such a modification would preserve the benefit to children while avoiding the harm to adults.

Matching journals

The top 1 journal accounts for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.