Study on people who leave religion

[BriansThoughtMirror]

I have only read half of this, but it was too interesting not to share:
http://www.inquiriesjournal.com/article ... us-exiting

This is a peer reviewed study (undergrad level, so take that for what it's worth) of the reasons and experiences of people leaving various faiths. The author's observations reflect my experience quite strongly.

[Jeffret]

The study is extremely well-done and very interesting. It's a long read, admittedly, but there are lots of interesting bits. Nothing terribly new or revelatory, but mostly the sort of stuff that makes sense. It's interesting to see the different responses from different people in different religions and how similar and how different they are.


Having said that, I'm going to focus in this post more on criticizing the study. I'm not saying it's not worth reading or invalid, but I want to talk about its weaknesses.

It's peer-reviewed, but in reading the journal's description of their peer-review the review is more about the general quality of the writing, the citations, the methodologies, and the idea than it is about peers' knowledgeable in the field reviewing the content. The peers are other students, though they may be graduate students. They state that they review for proper use of grammar and style, but there are several problems in that area in the paper. But, regular peer review certainly has its problems and failures also. I'd say the peer-review in this case is pretty good.

The data set in the study is pretty weak. It's a decent number of responses (610 survey responses), but not great. The number of interviews is kind of small (17), but that's a decent amount for a single undergraduate student to undertake. As LeCount mentions a number of times, the biggest problem with the sample set is its self-selection. Each quote or response is valid, but the aggregate numbers may not be reliable. In particular, the ranking of which factors were most significant may not be reliable. In spite of these weaknesses, I'm quite impressed at how much LeCount was able to do with what she had. She presents lots of great breakdowns, comparisons, and results. As a description of the characteristics of her sample set, it's very good. It's just not clear how well it describes a more generic sample set.

She notes, "While not surprising, another item of note is the inconsistent responses of participants when comparing their surveys and interviews. Because my interviewees were selected from survey participants, I could compare their responses from both formats. Survey responses were shorter, less complete responses and were not always truly reflective of their experience. At times survey responses showed very different responses than in person responses". With only 17 interviews, it is hard to know how reflective the survey responses were of the individuals' actual experiences. There is reason for concern.

LeCount states, "The single largest factor that led participants to and through this first doubt stage is their reliance on logic and individual rationale. This was the case no matter which religion was exited or demographic factors, with 50% of all participants mentioning a form of logic as a factor for their exit. Whether the participant was an apostate or currently religious, this self-identification as logical was consistent." LeCount mentions one of the biggest issues, but doesn't explain further. This expression of the importance of logic is a self-identification. Humans generally want to consider themselves and to be considered as being logical, but do an awful job of actually being logical. Lots and lots of examples and studies show that humans aren't really very logical. Rather, logic is generally a post-justification we impose upon our choices and decisions. That's a difficulty in many surveys such as this.

Instead, many other studies have shown that far and away the biggest reasons people join or leave are social. People join a religion because it meets their social needs. People leave, and join another one or none at all, because it doesn't. Logic can be a factor. It can be something that helps us to start reevaluating our social needs. We can realize that our social needs aren't what we thought we were. However, for most people social needs are the bigger issue, while logic may be how we self-justify it. LeCount touches on social and community needs in a number of different ways, but I feel she kind of ignores their importance. She doesn't cite the studies on the impact and importance of social needs. This may well be a weakness in her data collection, as she collected respondents from Reddit, from younger people who have moved to more online construction of their social interactions. It's possible also that she may be over-relying on what her respondents tell her where it could be valuable to dig deeper.

[Reuben]

Cool. It'll be interesting to compare it with John Dehlin's study on Mormons who stop believing.

http://www.whymormonsquestion.org/wp-co ... r20121.pdf

[Jeffret]

Again, I want to stress that my cautions or elaborations regarding LeCount's excellent paper are not meant to dismiss it or to suggest it isn't worthwhile. I'm not sure the numbers and rankings are completely reliable or replicable to other groups, but the analysis of this group is very interesting, well-done, and explanatory.

[BriansThoughtMirror]

Again, I want to stress that my cautions or elaborations regarding LeCount's excellent paper are not meant to dismiss it or to suggest it isn't worthwhile. I'm not sure the numbers and rankings are completely reliable or replicable to other groups, but the analysis of this group is very interesting, well-done, and explanatory.

Thanks for reading it, and for the great response! I agree with pretty much all of your criticisms. It is just an undergrad study, but given that, it's very interesting! She does have problems with selection bias, which she freely admits. She also states that "logical" was a self chosen label, and it only means that the respondents considers themselves to be logical, not that the respondents are actually any more logical than other humans. It's also hard to quantify any of it since she doesn't present any of her raw data. It may also be true that she interpreted the data to fit the ideas about role-exits she had already been studying, but we can't say for sure.

For me, it was less valuable as a piece of real science than it was as an emotional catharsis. To me, it was powerful to read an analysis of apostasy from a non-Mormon specific perspective that hit so close to home. It was neat to see so many people having similar experiences to mine, even outside of Mormonism, even if it is just a large collection of anecdotes that can't be used to make broad generalizations.