4 ms·

You are saying "leads to" but the data just say "is correlated with". As the article points out, assuming causality can't really be justified, because of select

by not2b 3y ago

You are saying "leads to" but the data just say "is correlated with". As the article points out, assuming causality can't really be justified, because of selection effects. For example, it may be that having lots of different sex partners before marriage doesn't "lead to" a worse marriage, rather perhaps those who tend to have lots of different partners may just be less happy with an exclusive arrangement.

atomicnumber3 3y ago
Correlation doesn't imply causation but all causal relationships will yield correlated variables. So ... correlation is a good place to dig in more to try to identify and then demonstrate a causal relationship.
- itsoktocry 3y ago
  There's a certain type of "correlation isn't causation" person who believes the phrase means correlation is useless,as opposed to interesting, but incomplete.
- wholinator2 3y ago
  It's also research done by an institution with the stated principles of faith and patriotism so take everything with a grain of salt. I bet if their data pointed to lawless polygamy they wouldn't be advertising this. Anyone can lie with statistics, even to themselves.
  - hackeraccount 3y ago
    I understand the logic of what you're saying but doesn't it tend towards thinking that are no trustworthy sources of information other then those that postulate things you already believe in or at least those that agree with conventional wisdom? Not that I have a better idea - generally chasing down methodological errors if not ought right falsehood is more trouble then it's worth (liars can lie faster then the truth can move) but it makes me edgy when I say it out loud. This could all be because in this particular case I don't particularly disagree with the premise.
phkahler 3y ago
>> For example, it may be that having lots of different sex partners before marriage doesn't "lead to" a worse marriage, rather perhaps those who tend to have lots of different partners may just be less happy with an exclusive arrangement. Either way, if you're looking for a long term relationship you should avoid those with high body counts. In some cases causality doesn't matter but you get an indicator either way.

by not2b 3y ago

atomicnumber3 3y ago

Correlation doesn't imply causation but all causal relationships will yield correlated variables. So ... correlation is a good place to dig in more to try to identify and then demonstrate a causal relationship.

itsoktocry 3y ago
There's a certain type of "correlation isn't causation" person who believes the phrase means correlation is useless,as opposed to interesting, but incomplete.
wholinator2 3y ago
It's also research done by an institution with the stated principles of faith and patriotism so take everything with a grain of salt. I bet if their data pointed to lawless polygamy they wouldn't be advertising this. Anyone can lie with statistics, even to themselves.
- hackeraccount 3y ago
  I understand the logic of what you're saying but doesn't it tend towards thinking that are no trustworthy sources of information other then those that postulate things you already believe in or at least those that agree with conventional wisdom? Not that I have a better idea - generally chasing down methodological errors if not ought right falsehood is more trouble then it's worth (liars can lie faster then the truth can move) but it makes me edgy when I say it out loud. This could all be because in this particular case I don't particularly disagree with the premise.

phkahler 3y ago

>> For example, it may be that having lots of different sex partners before marriage doesn't "lead to" a worse marriage, rather perhaps those who tend to have lots of different partners may just be less happy with an exclusive arrangement. Either way, if you're looking for a long term relationship you should avoid those with high body counts. In some cases causality doesn't matter but you get an indicator either way.