|
|
|
|
|
|
by maxerickson
4219 days ago
|
|
|
The RAND meta analysis at your link (reference 3) rates the quality of the studies it analyzed. Among the groups that were assigned randomly (to have educational intervention or not), the reduction in recidivism was about 40%. (I had wondered how self selection factored into the numbers...edit to add: I did not mean to suggest that the 40% I quote is directly comparable to the numbers in your post, I meant to share the factoid that the effect was strong when a quality control for self selection was present) |
|
|