The "LDS young adults" don't commit suicide study

[Indy Coug]

Originally Posted by Archaea

Your objections to the definitions and conclusions are noted, but I'm not certain a different standard could be measured.

What other standard is measurable?

Well, that's the fundamental problem here. There are no clear, unmistakable delineations in the population.

[jay santos]

Originally Posted by Archaea

Your objections to the definitions and conclusions are noted, but I'm not certain a different standard could be measured.

What other standard is measurable?

Oh I can think of a lot of ways to do it, all cost time and money. I don't think what the guy did was necessarily a bad way to go about it. I judge studies like this based on the conclusions they draw. If they go too far then it makes the whole study feel manipulative and unprofessional, like Indy says is common (I'll say "sometimes happens") in the social sciences. Also the guy that did the study is a statistician not a behavioral scientist which really makes me wonder in the first place.

[pelagius]

Originally Posted by Archaea

Your objections to the definitions and conclusions are noted, but I'm not certain a different standard could be measured.

What other standard is measurable?

In my view Jay is being to kind to the study. His objection is valid but the empirical work is even more flawed than that. Jay said,

You can say, living the standards of the church makes you less prone to suicide

I don't think you can even make this inference. Some factor that makes activity difficult could also be correlated with suicide. I don't have a problem with the definition of "active" but the authors attempt to suggest that it is uncorrelated with mental health (broadly defined) is almost laughable in my view. The study needs an instrument really badly and it doesn't have one. I don't share Indy's disdain for this kind of empirical work in general but this study can only support a very limited set of inferences.

[Archaea]

Originally Posted by pelagius

In my view Jay is being to kind to the study. His objection is valid but the empirical work is even more flawed than that. Jay said,



I don't think you can even make this inference. Some factor that makes activity difficult could also be correlated with suicide. I don't have a problem with the definition of "active" but the authors attempt to suggest that it is uncorrelated with mental health (broadly defined) is almost laughable in my view. The study needs an instrument really badly and it doesn't have one. I don't share Indy's disdain for this kind of empirical work in general but this study can only support a very limited set of inferences.

That makes sense. The inferences were unreasonable.

[MikeWaters]

Originally Posted by jay santos

Oh I can think of a lot of ways to do it, all cost time and money. I don't think what the guy did was necessarily a bad way to go about it. I judge studies like this based on the conclusions they draw. If they go too far then it makes the whole study feel manipulative and unprofessional, like Indy says is common (I'll say "sometimes happens") in the social sciences. Also the guy that did the study is a statistician not a behavioral scientist which really makes me wonder in the first place.

Not only that, his research focus is not in this area.

https://facultyprofile.byu.edu/Publi.../vita_sch2.doc

That's his CV.

[MikeWaters]

Originally Posted by pelagius

I don't think you can even make this inference. Some factor that makes activity difficult could also be correlated with suicide. I don't have a problem with the definition of "active" but the authors attempt to suggest that it is uncorrelated with mental health (broadly defined) is almost laughable in my view. The study needs an instrument really badly and it doesn't have one. I don't share Indy's disdain for this kind of empirical work in general but this study can only support a very limited set of inferences.

This is basically what I have said so far, but spread over a number of posts and not as eloquently.

My personal experience is that the more troubled the kid and the family, the less likely that kid is to advance in the priesthood as a young man, i.e. less likely he is to stay active.

[Jeff Lebowski]

Originally Posted by Indy Coug

Sociology/psychology studies have by default questionable value. The "value" they have is for those who want their own personal biases/gut instincts validated by any study that happens to agree with their position.

And just like that, Indy dismisses an entire field of academic research. Again.

Validating personal biases indeed.

[pelagius]

Originally Posted by MikeWaters

This is basically what I have said so far, but spread over a number of posts and not as eloquently.

My personal experience is that the more troubled the kid and the family, the less likely that kid is to advance in the priesthood as a young man, i.e. less likely he is to stay active.

Agreed, this a really tricky question to get at empirically. Your best bet is probably to get Mormon twins between 15-19 and let one twin go to church and not let the other twin go, read her scriptures, pray, etc . Triplets would work too. Any volunteers?

[MikeWaters]

Say they did the study in Texas. Active LDS vs. everyone else.

Using the same methods. And they said that Non LDS was 20 times more likely to commit suicide.

What would that tell us? Nothing. Because you haven't taken into account all the other factors in such a diverse state.

By the way, this is real stuff. A family in my hometown had a daughter commit suicide.

[Indy Coug]

Originally Posted by Jeff Lebowski

And just like that, Indy dismisses an entire field of academic research. Again.

Validating personal biases indeed.

Dismisses an entire field? No.

Claim that sociological research is more prone to error and fraud than other scientific fields? Yes.

Sociology carries a heavier burden of proof than others? Yes.