Freethought & Rationalism Archive

HelenM · 07-08-2003, 07:17 AM

Quote:

Jesus Saves:
How President Bush found himself hugging a murderer in the White House.

That's not quite the way a University of Pennsylvania report puts it. But that's the underlying message of a just-released study confirming sharply reduced recidivism rates for Texas inmates who've completed an innovative joint venture between Chuck Colson's Prison Fellowship Ministries and the Texas Department of Criminal Justice. [...]

[...]

[...] the positive findings about the InnerChange Freedom Initiative parallel more than 500 other studies showing that the "faith factor" often makes faith-based programs more effective than their purely secular counterparts.

To put it another way, critics of the faith-based approach may claim that their only issue is with religion. But if these results are any clue, increasingly the argument against such programs requires turning a blind eye to science.

Comments, anyone?

Roland98 · 07-08-2003, 07:21 AM

Helen,

Can you look up the original study on that? They say "matched" but I'd like to know how they did the matching. There was a similar study out a few months ago, in which similar results were published--but the kicker was, the prisoners had a choice to enroll or not enroll in the program, rather than having a group of volunteers being randomly enrolled and matched against a true control group. So in the previous study, the results were completely invalidated due to selection bias. I'd be interested to know if the same thing happened in this study.

Jamie_L · 07-08-2003, 07:42 AM

Well, Roland's got a good point about the potential for self-selection bias.

Assuming that the study is good, I can think of a number of reasons why a faith-based program might gain advantages over a secular program - none of which have to do with the truth of the faith-based belief system.

Ignoring both of those, however, the question might then be: What do we do with this information? If faith-based programs work better, are they a good idea? What about faith-based program makes it work, and can the lessons-learned be applied to secular programs.

I think there is a danger in the government turning prisons into recruiting grounds for religions. Certainly you can't have 20 different faiths running programs in prisons, so how do you avoid favoritism? Suppose a Baptist program works better than the others? Is there a problem with the government endorsing the Baptist program over others? What if a Muslim program works better? What if a Wiccan program wants in on the action, but doesn't have proven results yet? Can you deny them access to the prison system?

Furthermore, there needs to be a secular option. If it's not working, why isn't it? A faith-based system with good numbers might further discourage investment in improving the secular system, leaving those who need help but who don't want to join that particular faith with no where to turn.

Which isn't necessarily to say that a faith-based program can't be part of the equation, but it seems like this is a thorny area. Government should avoid the pitfall of getting heavily entangled with/dependent on religious services.

Jamie

HelenM · 07-08-2003, 07:50 AM

Quote:

Originally posted by Roland98
Helen,

Can you look up the original study on that?

It looks like this is all there is online:

Study: Faith-Based Rehabilitation Program Shows Promise

I guess you need to call the number in there if you want more information.

Practically speaking, enrollment would have to be voluntary, really, wouldn't it? I mean, an atheist who doesn't want to be in a faith-based program is hardly going to respond to it in the same way as a theist, and vice-versa.

I think it's reasonable to compare the best results of voluntary enrollment secular programs with voluntary enrollment faith-based ones. It's also reasonable, if a difference is found, to investigate why.

I wonder if I should have put this topic in the Political Discussions Forum.

Helen

Roland98 · 07-08-2003, 08:04 AM

Quote:

Originally posted by HelenM
[B]It looks like this is all there is online:

Study: Faith-Based Rehabilitation Program Shows Promise

I guess you need to call the number in there if you want more information.

I emailed Dr. Johnson, because that still doesn't go into the methodology. I'd assume the study was published in a journal somewhere--if not, that right there it is a bit sketchy to me.

Quote:

Practically speaking, enrollment would have to be voluntary, really, wouldn't it? I mean, an atheist who doesn't want to be in a faith-based program is hardly going to respond to it in the same way as a theist, and vice-versa.

Right--but what you need to do is take this group who would be willing to enroll, and then divide them up into at least two groups. One would get the "treatment" described in the article, another would be treated the same way ordinary prisoners are, and you could even use a third group to try another method (e.g., secular based). Then you compare these groups to each other. By randomizing the self-selected prisoners, you more effectively eliminate the self-selection bias. That wasn't done in the previous study (I think it was in Iowa). What they did there was compare the results of the religious group to the prison population at large--which does nothing to eliminate the selection bias inherent in the study, and therefore makes the data worthless.

Quote:

I think it's reasonable to compare the best results of voluntary enrollment secular programs with voluntary enrollment faith-based ones. It's also reasonable, if a difference is found, to investigate why.

Yep--that's the way it should be done, if it's going to be carried out.

Roland98 · 07-10-2003, 06:48 AM

Just to update in case anyone is interested--

Received an email from Dr. Johnson.

1) No, this study wasn't published in a peer-reviewed journal. The link to the pdf (published on the web only) is here. Strike one.

2) Enrollment was optional (applicants applied), and did not need to be Christian beforehand, but were told it was a Christian-based program. So there was obvious selection bias. This was not controlled for, as "cases" were matched to "controls" in the general population--not to controls who'd applied to the program but were either denied admission or who were admitted but then went through a secular program of a similar time course. (Admitttedly, they tried to do this, but the group that volunteered for the study was too small. They try some hand-waving to get around this, but if this had been sent for peer-review, they would have been turned down flat due to this limitation.) Strike two.

3) Ugh--personal testimony. Couldn't get more subjective--particularly since they don't include any that speaks against the program. Strike 3--game over.

It does sound like the main benefit is that these guys get treated with respect by the volunteers--they are treated well, they read the bible, they are expected to live up to certain standards, and they are taught they are loved and they are part of a family. But all that could be done with a secular program, as well (sure, even throw in some Bible readings--but add other literature too).

It's a real shame that all this time and money was spent without a secular control group that taught similar things...I do think there's a better way to do this without invoking the big Sky Daddy.

lunachick · 08-06-2003, 05:53 AM

Thought you might be interested in this:

http://slate.msn.com/id/2086617/

Quote:

But when you look carefully at the Penn study, it's clear that the program didn't work. The InnerChange participants did somewhat worse than the controls: They were slightly more likely to be rearrested and noticeably more likely (24 percent versus 20 percent) to be reimprisoned. If faith is, as Paul told the Hebrews, the evidence of things not seen, then InnerChange is an opportunity to cultivate faith; we certainly haven't seen any results.

So, how did the Penn study get perverted into evidence that InnerChange worked? Through one of the oldest tricks in the book, one almost guaranteed to make a success of any program: counting the winners and ignoring the losers. The technical term for this in statistics is "selection bias"; program managers know it as "creaming." Harvard public policy professor Anne Piehl, who reviewed the study before it was published, calls this instance of it "cooking the books."

Queen of Swords · 08-06-2003, 06:18 AM

There was a recent article in the Gulf News about how three men in prison in the United Arab Emirates had embraced Islam. One guy was British and one Nigerian; I forgot what the third was. Anyway, the lieutenant in charge of the Human Rights Section said that lectures and teaching sessions on Islam were provided to all prisoners, and the prison library was well stocked with literature, presumably of the "There is no god but Allah" variety.

Which only goes to show that all religions benefit from a captive audience.

Roland98 · 08-06-2003, 11:47 AM

Quote:

Originally posted by lunachick
Thought you might be interested in this:

http://slate.msn.com/id/2086617/

Heh heh, went looking for this thread in the older pages to post that exact article...but you beat me to it!

I think it pretty succintly shows that, despite all the hoopla and claims to the contrary, there has been not a single study that supports the idea that faith-based programs help. How much more money are we going to throw at this before something else is tried?

Thread Tools	Search this Thread
Show Printable Version	Search this Thread: Advanced Search

07-08-2003, 07:21 AM	#2
Roland98 Contributor Join Date: Apr 2003 Location: Gilead Posts: 11,186	Helen, Can you look up the original study on that? They say "matched" but I'd like to know how they did the matching. There was a similar study out a few months ago, in which similar results were published--but the kicker was, the prisoners had a choice to enroll or not enroll in the program, rather than having a group of volunteers being randomly enrolled and matched against a true control group. So in the previous study, the results were completely invalidated due to selection bias. I'd be interested to know if the same thing happened in this study.

07-08-2003, 07:42 AM	#3
Jamie_L Veteran Member Join Date: Oct 2001 Location: U.S. Posts: 2,565	Well, Roland's got a good point about the potential for self-selection bias. Assuming that the study is good, I can think of a number of reasons why a faith-based program might gain advantages over a secular program - none of which have to do with the truth of the faith-based belief system. Ignoring both of those, however, the question might then be: What do we do with this information? If faith-based programs work better, are they a good idea? What about faith-based program makes it work, and can the lessons-learned be applied to secular programs. I think there is a danger in the government turning prisons into recruiting grounds for religions. Certainly you can't have 20 different faiths running programs in prisons, so how do you avoid favoritism? Suppose a Baptist program works better than the others? Is there a problem with the government endorsing the Baptist program over others? What if a Muslim program works better? What if a Wiccan program wants in on the action, but doesn't have proven results yet? Can you deny them access to the prison system? Furthermore, there needs to be a secular option. If it's not working, why isn't it? A faith-based system with good numbers might further discourage investment in improving the secular system, leaving those who need help but who don't want to join that particular faith with no where to turn. Which isn't necessarily to say that a faith-based program can't be part of the equation, but it seems like this is a thorny area. Government should avoid the pitfall of getting heavily entangled with/dependent on religious services. Jamie

07-10-2003, 06:48 AM	#6
Roland98 Contributor Join Date: Apr 2003 Location: Gilead Posts: 11,186	Just to update in case anyone is interested-- Received an email from Dr. Johnson. 1) No, this study wasn't published in a peer-reviewed journal. The link to the pdf (published on the web only) is here. Strike one. 2) Enrollment was optional (applicants applied), and did not need to be Christian beforehand, but were told it was a Christian-based program. So there was obvious selection bias. This was not controlled for, as "cases" were matched to "controls" in the general population--not to controls who'd applied to the program but were either denied admission or who were admitted but then went through a secular program of a similar time course. (Admitttedly, they tried to do this, but the group that volunteered for the study was too small. They try some hand-waving to get around this, but if this had been sent for peer-review, they would have been turned down flat due to this limitation.) Strike two. 3) Ugh--personal testimony. Couldn't get more subjective--particularly since they don't include any that speaks against the program. Strike 3--game over. It does sound like the main benefit is that these guys get treated with respect by the volunteers--they are treated well, they read the bible, they are expected to live up to certain standards, and they are taught they are loved and they are part of a family. But all that could be done with a secular program, as well (sure, even throw in some Bible readings--but add other literature too). It's a real shame that all this time and money was spent without a secular control group that taught similar things...I do think there's a better way to do this without invoking the big Sky Daddy.

08-06-2003, 06:18 AM	#8
Queen of Swords Contributor Join Date: Nov 2000 Location: Canada. Finally. Posts: 10,155	There was a recent article in the Gulf News about how three men in prison in the United Arab Emirates had embraced Islam. One guy was British and one Nigerian; I forgot what the third was. Anyway, the lieutenant in charge of the Human Rights Section said that lectures and teaching sessions on Islam were provided to all prisoners, and the prison library was well stocked with literature, presumably of the "There is no god but Allah" variety. Which only goes to show that all religions benefit from a captive audience.

Freethought & Rationalism Archive

The archives are read only.