But you could draw the same conclusion about people named Cuthbert or Aloysius or Nehemiah. Or, for that matter, about people named John or Peter or Michael. Take it to the extreme, and you’d conclude that marriage is good for anyone whose name you don’t know, but that once you learn that person’s name, whatever it is, you suddenly can’t conclude anything any more.
That’s not the hypothesis being tested. What’s being tested is married parents vs. unmarried parents. The genital configurations of those pairings (like their names) are irrelevant variables which are not being tested.
No.
sigh
Do you know what’s meant by the terms “dependent variable” and “independent variable?”
They are irrelevant until you start claiming that the study shows they are irrelevant. Once you make that claim, you need to show that the study controlled them (or for them).
No. They are irrelevant regardless (unless you prove otherwise, which you can’t). Irrelevance is the null assumption here. The burden is on you.
The study tested the following hypothesis: Do married parents produce better outcomes than other family arrangements? And it answered it in the affirmative. From that study one can conclude that as a general rule that married parents produce better outcomes for children than do unmarried parents.
Some may be exceptions from that general rule, but without evidence that a specific couple is that exception the study suggests that the general rule applies as the default.
You were asked
And responded
And sorry but that is as absurd as saying that despite the fact that I know that in general Lipitor is a wonderful cholestrol lowering agent, I must state that I have no idea if Lipitor will lower the cholesterol level of a Spanish man with a pet poodle because Spanish men with pet poodles have not been specifically studied… Sure it is possible that there is some correlate of being Spanish and owning a poodle that blocks the effect of Lipitor, some genetic condition that both leads for a poodle preference and that becomes manifest only when people nap mid-day, but without evidence of such I can suggest Lipitor for this particular man as likely to be effective. I can state that it is, based on that study, likely that married Asians also have better child outcomes on average than do unmarried Asians.
Now as for the hypothesis that marriage of gays is a specific exception to the rule suggested by the study, that married couples lead to better child outcomes - the onus is on those who would propose that to support the contention.
Here is where we differ- you believe the hypothesis being tested is “Gay and straight marriage produce similar outcomes.” I believe that the hypothesis is that marriage between parents is good for kids, which this study strongly supports (at least as a correlation). The burden of proof in my mind would be to provide evidence that gays are the exception to that general rule. There is no study that does that and certainly not this one. You are completely correct that the study does not inform on the hypothesis you believe is being tested.
Again, by your standard we could never recommend any dietary, activity, or medication intervention unless the exact subgroup under consideration has been studied in statistically significant sample sizes. That is absurd. Unless we learn from independent evidence that Lipitor is specifically ineffective for a hypercholesterolemic male Spanish poodle owner we can assume it is as effective for him as for the general population studied. If someone says that I should not prescribe it for that specific person despite general population studies that suggest it works well the burden of proof is not on me to prove that Spanish poodle owners respond the same, but on the other person to prove they do not. Lipitor should be advised for him and marriage for gay parents by the same exact logic.
OK, I’m done for the moment. Anyone else who believes in fighting ignorance is welcome to step up.
Sometimes you really are just being sophist, Bricker. This is one of those times.
What should I tell the hypercholesterolemic Spanish poodle owner about Lipitor and why? In what ways is it, in your mind, a different circumstance than the question of whether marriage should be advised for gay parents?
(Honestly, just calling those who disagree with you “ignorant” is poor style.)
Who is testing that hypothesis?
Minnery tried to use this study to argue that SSM is bad for kids. You agree that the study did NOT show that, right?
So the position we find ourselves in is that we have the only data we have seems to suggest that married partners are better for kids, and the data is not sliced in such a way that it distinguishes whether the partners are same-sex or opposite sex (or by what their first names are). It’s not a slam dunk, but I haven’t heard any reasons to suspect that it wouldn’t apply to same-sex partners.
In short, there’s not a hypothesis that we’re testing. Someone tried to use some existing data to support his point, when the existing data reasonably seems to work against it.
I’ll bump this thread with this bit of info:
Psychological association calls for legalization of same-sex marriage
Thanks for the share. It’s a start.
Although I often disagree with Bricker in this case he is right on the money.
Here is the test by fire. Suppose that FoF was right. Suppose that the instant they are married gays become child eating satanists. In such a universe how would the results of this study look. The answer is it would look exactly the same way as it did in the current universe. It would find that married couples are better for children then single parents, because it didn’t happen to include any of the child eating satanists.
Given that the results are equally likely in both the current universe and the alternate child devouring universe, this study makes no claim as to which universe is reality.
In the case of a drug trial done it New York and extending its results to Mississippi, that may be a good thing to do. But if the entire point you are trying to make is that there is no difference between Mississippi and New York. You are going to need a new study.
That’s incorrect. The study did not exclude same-sex couples.
And applying this rationale to our hypothetical, if parents named Egbert suddenly, upon marriage, became child-eating satanists, the study would not have found that either.
Now someone comes to you proposing to ban marriage to parents named Egbert because of concern for kids. I tell him that the data we have tells us that in general, kids are better off with married parents, and unless he comes up with some reason to question whether this study wouldn’t apply to parents named Egbert, the data we have seems to argue against his position.
He then takes your gambit and tells me that the study didn’t include any Egberts, so the study has no bearing on the issue.
He’d be wrong. Unless he presents any kind of a reason why the data we have would not apply to Egberts, the only reasonable approach is to accept that it does.
Nobody ever said that science can’t get the wrong answer sometimes. But it’s usually right, and if we throw out science for the sake of those occasional wrong answers, there’s not much of anything left to replace it with.
But the person in the anti-Egbert league says that they believe that Egberts are fundamentally different. At which point without more data you get into a “No they aren’t”, “Yes, they are” type argument, convincing no one. What is needed are other studies to show that Egberts are like everyone else and the anti-Egberts are full of it.
As far as coming up with reasons why the data wouldn’t apply to homosexuals, I’m sure that the Fof can come up with lots. They are all wrong, but we don’t know that a-priori without more data.
Taking this further, if this study says marriage is good, doesn’t that extend to people who are already married, so that children will clearly do better in polygamous households, than ones with only one marriage.
But it likely didn’t include a sample large enough to reach any conclusions about it. One gay child eating satanist among a couple of hundred normal married couples isn’t going to change the results.
If it did include a large number of gay couples than these would need to be analyzed separately to show that they as a group also performed better than the single parents.
No they wouldn’t, no more than couples with names starting with ‘M’ would. it’s an irrelevant variable.
But Focus on Family’s claim is that it is a relevant variable. They’re wrong, of course, but you can’t prove that they’re wrong about it being a relevant variable by referring to a study that starts with the assumption that it’s an irrelevant variable.