The Atheist Experience blog has an interesting post about how Chuck Colson and Prison Fellowship faked the data to make it appear as though taking part in their Christian rehabilitation program had a much higher rate of success in avoiding recidivism than it actually did.
What Colson claimed was that they studied the recidivism rates of prisoners who completed his ministry program and compared them with those who did not. Recidivism means that within a certain amount of time after they were released from prison, they were reincarcerated for committing new crimes. Colson always argued the study demonstrated that those who completed the program experienced a significantly decreased recidivism rate.
What he didn’t tell you is that the standards for “completing” the program dramatically skew the numbers in his favor. A person is only defined as a graduate if they stick with the program for a period of time, then are released from jail, and get a job after their release. In other words, a person who sat in on the ministry classes for the required amount of time, left the program, and then couldn’t find a job, wouldn’t be considered to have completed a program. Therefore, if they were arrested later, that would be counted as a win for Colson, because they didn’t do what they what they were supposed to, therefore this proves that failing to “complete” the program was correlated with their arrest.But this is a total cheat. If you simply removed the ministry from the equation, and only compared prisoners who got a job to those who didn’t get a job, obviously the employed prisoners would be far less likely to go back to jail. They don’t need to steal stuff to get money! So here we have Chuck Colson deliberately excluding the group most likely to go back to jail, and then giving his ministry credit for something that happens after they leave. The study doesn’t even attempt to demonstrate that people who take the program are more likely to get jobs.
In fact, what the study showed when you looked at the raw numbers was that among prisoners who simply entered the program — including both graduates and “dropouts”, the recidivism rates were slightly higher than the control group that wasn’t involved at all. Or to put it simply, if the program had not existed at all, it’s possible that fewer of them would have returned to jail.
I’m certainly not surprised by that. Prison Fellowship brought in a lot of money, including a lot of government grants, based on its claim to be effective in turning prison inmates into good, law-abiding citizens.