public sociology at HIRED
I spoke about work and crime today at HIRED, an organization providing employment and training services to a client base that includes former felons. I initially expected an informal break-room discussion with a few staff members, but I was given two hours and an audience of 40-50 smart professionals. In principle I relish this sort of public sociology opportunity. In practice, however, it can be daunting. I once worked in a small employment and training nonprofit in Madison (the Dane County Private Industry Council). In preparing, I thought how I would have reacted to some university pointy-head telling me how to do my job at a mandatory two-hour training session (even if said pointy-head does bring cool slides). I tried to hit some key ideas and empirical findings from the good research literature, but offered as many caveats (e.g., would experimental results from the 1970s and 1980s hold today?) as generalizations. It is always a challenge to summarize oceans of research without glossing over important details, or putting an overly optimistic or pessimistic spin on the findings. Fortunately, I had copies of a short review piece that Jer Staff and I had written for corrections administrators so I could at least provide citations when needed.
In the end, I got a lot of perceptive questions and comments. I felt somewhat at home answering the big-picture questions about stigma, punishment, and reintegration. I stumbled, however, when pushed for existing models of exemplary programs that have been rigorously assessed. I know something about what people are doing today and I know a lot about what "worked" in the past, but I cannot say whether what people are doing now is working now. More distressingly, I cannot say for certain why things worked or failed to work. That is, even where a causal effect of work on crime can be identified through an experiment or other means, social scientists can often only speculate about the mechanism connecting them (e.g., money, stability, age-graded role transition). I'm certainly guilty of either emphasizing the "true effect" but remaining agnostic as to the mechanism, or focusing on potential mechanisms but hedging on causality. I came back with three (pointy-headed yet also simple-minded!) reactions:
- Run more experiments and vary the mechanisms. If money is the hypothesized mechanism, for example, we could pay group A $10/hour and group B $20/hour. But experiments are expensive and random assignment is a tough sell to practitioners even under the best conditions (unlimited resources, no harm to participants, no guilt in denying treatment to non-participants).
- Let theory guide the mechanisms (economic vs. learning vs. informal control vs. all of the above).
- Push harder for evaluation of the myriad "interventions" that employment and training and corrections professionals make every day.