Joe - thanks for taking the time to read the research, and for your thoughtful question.

As you know, the reason researchers assess interventions like "Scared Straight" in isolation (i.e., vs. "doing nothing" in this case) is that it makes it clearer whether the program is working or not. If, as in this situation, you have multiple studies of the intervention in multiple sites that all conclude that "Scared Straight" either has no effect or is harmful, there's no basis to suppose that combining it with interventions that have been shown to have a positive impact will make it successful. In fact, as far as I'm aware, there are no independent research studies that show even the slightest positive effect for "Scared Straight," so at this point, policymakers and practitioners have very strong evidence that suggests they'd be doing harm by using it at all.

Then, too, it's important to remember that "Scared Straight's" premise is that the shock of being exposed to hardened inmates in a prison setting will cause young people to modify their behavior over the long term. The research as it stands -- which has been performed more than once by different researchers in different places -- shows that doing nothing with the kids either has the same effect as "Scared Straight" (in which case, there's no point in doing it) or (as this meta-analysis concludes) it makes kids *more* likely to commit crimes. Either way, the premise that kids can be scared straight doesn't hold water.