Yet it is indeed sometimes appropriate to extrapolate from several specifics to a general rule.
A drug is tested. The clinical trial includes Whites, Blacks, Hispanics, males and females, but not a statistically significant number of Asians (even though they were not excluded from the study). The drug is highly effective for treating the disease state in all groups that it has been tested in. Those without that treatment have an adverse outcome in comparison. A Japanese patient has the disease.
Is it reasonable to treat the patient with the drug? Should a drug only be used on a member of each specific sub-group … oh, Hawaiian or people who are of both Jewish and of African descent … if it has been tested on that particular sub-population? Or is it reasonable to take evidence of a beneficial effect on one broad set of humans as presumptive evidence that it is likely effective for all groups? We do make assumptions of generality from a randomized group to populations that are not adequately represented in the group.
This is not Statistics at the college sophomore level; you may need to go a bit higher. It is a serious issue in the medical literature, not just one I am making up.
No, it was not massive voter fraud. :smack: Lord knows, Coleman made them recount enough times. Good riddance to bad rubbish. Senator Franken will be re-elected.
I can’t believe this particular discussion has gone on so long. You’ve more than adequately explained things, Bricker, and more than enough times, IMO.
Please let me echo Miller’s +1 with another.
That should put you up to at least +2, if my own remembrance of simple mathematics isn’t faulty.
Perhaps Paul Stanley should actually look at the study (warning: PDF). It’s on page 172, in the section labelled “Definitions of Selected Terms”, oddly enough. The fact that it is labelled so clearly, and that the document even has bookmarks with said labels, must have thrown him off. It’s also pretty clear that married gays fall under nuclear family if you read the entire definition section on “family structure.”
Bricker, you’re adamantly taking a strict interpretation of the outcome of the study. Using that approach, we could also say that the study does not give any useful information about the relative outcomes for married and unmarried families where the father is named “Egbert.”
But it’s reasonable to infer that this study would apply to the situation where the father is named Egbert. Is it reasonable to infer that it applies to same-sex couples as well? Seems reasonable to me, but I’d like to see a study with more specifics to be sure.
What is certain is that the study does NOT tell us what the Minnery said that it did - that children of same-sex couples are worse off because of it.
After five pages of this ridiculousness, I’m still in the dark as to why the hell congress is even entertaining this report? Personally I don’t care what they come up with as to which type of married couple will raise the best kids. I’m sure if we really dug deep, we might find that children of rich people fare better in this world. What does that mean? We should only allow rich people to get married?
Dissallowing gay marriage doesn’t address a single frickin’ thing in the world as it relates to children. The Right isn’t concerned with the quality of child-rearing, they’re concerned with frequency of adult-‘rearing’ which grosses them out.
Well, I can’t (obviously) join in the universal condemnation of the Right that Jack urges, but on this point, he’s absolutely correct. This was not a case of neutral devotion to the truth, come what may; the FoF folks advanced this report because they thought they it supported their ultimate goal of forbidding same-sex marriage. This was not motivated by a genuine desire for the best outcomes for children – if it were, then clearly permitting same-sex marriage woul dbe the best thing, as numerous other studies no doubt show.
It’s irritating to see the Left cherry-pick data and refuse to acknowledge clear evidence. (I saw what I believe were some choice examples in this thread!)
But it’s downright disenheartening when I see “my own side” do it, because it forces me to confront the fact that the dishonesty – on this issue – runs far more rampant from folks that identify as “Right” than from the other side.
So… yes. This study is meaningless as relates to marriage policy anyway, and the only reason it was tossed into the ring was out of a forlorn hope that the wrong conclusion would be drawn and the wrong goal supported… and much as it pains me, that’s all from the Right.
And you’d be correct. There seems no particular reason that this would make a difference, but suppose I advance the theory that parents with odd given names suffer teasing and ostracizatin as children, and thus grow up to be poor parents. Therefore, say I, you would expect more negative outcomes from Egbert parents than Bob, Steve, or Phil parents.
Can you use this study to refute me?
And you’d need them. All you’re saying is that to you, the same-sex factor should be a trivial change. That’s fine, but there’s no evidence to support it in the study. It just seems reasonable to you.
Well, it seems reasonable to me. But since it doesn’t seem reasonable to everyone, you cannot say that the study supports it. What you can say is that nothing in the study contradicts it.
Bricker, I think you’re cogently and correctly arguing your side of the question you’re debating. The problem isn’t in that; it’s that you’re arguing a different question than others in this thread. The two questions are “How do gays compare to straights as parents?” and “Should gay marriage be prohibited?”. You are correct that this study offers no insight at all on the first question. It does, however, offer insight on the second question.
No. But to most the burden of proof would fall upon you to prove the assertion (actually beyond that assertion, that it makes them bad parents in a manner that makes the outcomes for their children not improved by marriage), not for me to falsify it. In the absence of such proof the study suggests that the children of Egbert and Agnes are more likely to have better outcomes (as defined by the study) if they are a married couple than if they are not. There is no need to have studied specifically children of Egbert and Agnes pairings to make that conclusion.
Likewise in the absence of any a priori evidence to believe that the children of gay parents are uniquely resistent to the beneficial effects of marriage, this study, on its own, is highly suggestive that gay marriage benefits the children of gay parents. Yes, I still agree that it does not prove such, but it does highly suggest it.
By your standard there is no reason to believe that any medicine will be effective for any condition I, as unique individual, have as there have been no studies that have included me, let alone a statistically significant sample of me’s, with all my unique combination of genetics and history, in it. Just because Lipitor worked for a random group of humans does not prove it works for me’s. My doctor would be foolish for suggesting it as likely effective. He doesn’t understand statistics; there is no evidence that Lipitor will work for me.
We generalize to me, to Egbert, to gay parents, from the broad sample, even if I, Egbert, or gay parents, are not a statistically significant portion of the broad sample … unless we have evidence that there is something different about me, or Egbert, or gay parents, that makes it non-applicable.
Then please quote the study’s conclusions that support the proposition that gay marriage should not be prohibited.
But wait. Do NOT proceed with the assumption that gay marriage and straight marriage should produce similar outcomes, because that assumption is central to the proposition we’re testing by asking if gay marriage should be prohibited. In other words, you can’t say, “We of course assume that same-sex marriage should provide similar benefits for children, because there’s no reason that it shouldn’t.”
That assumption is what we’re testing – or more accurately, that assumption is as challenged as the “should gay marriage be prohibited” message. You can’t assume the truth of your claim and then say the study supports you.
So – does the study affirmatively support the claim that gay marriage should be permitted? Where?
Again, it’s the context of the question. There’s no “in the absence of proof” going on here; the need for proof comes because you’re claiming support from the study. The burden is on you to show the study support a specific claim.
No. That’s the hypothesis we’re testing. You cannot begin by assuming the hypothesis false.
Actually, in the ordinary course of events, you can. That’s exactly what you should do. But you cannot claim this study tests that hypothesis.
You’re confusing two states: the general one, in which I must prove any claim I make; and the specific one here, in which we have begun with your claim that the study suggests a particular conclusion.
In the first instance, of course I would have to advance evidence for my claim, and of course that evidence does not exist. But we’re not there. We’re in a special world where you’ve claimed teh study suggests a conclusion. To weigh that claim, we can’t start by assuming the truth of that claim.
I think I’m done with this. I appreciate the couple of posts that have “+1” me, but it would be great if someone else could explain it differently so as to make sense to the folks that don’t get it. I’m tapped out.
Parallel Minnery of Focus on the Family presents a report to the Senate, stating that people named “Egbert” should not be allowed to marry, citing a study that nuclear families with married parents have better outcomes with kids if they’re married vs. if they’re not married. Minnery also states that the study defined a nuclear family to be one that did not include a parent named Egbert.
Senator Parallel Franken admonishes Minnery, pointing out that the study did not define nuclear family that way, and if anything, the study would support allowing Egberts to marry.
A poster named stpauler on the PSDMB creates a thread that Franken pwned Minnery.
However, Parallel Bricker disagrees, saying that the study does not suggest that Egberts would have better outcomes with raising kids, because the study probably didn’t even include anyone named Egbert.
Parallel DSeid points out that although the study didn’t look specifically at the names, the reasonable conclusion would be to accept that it would apply to Egberts, given the lack of any reason to think that it wouldn’t apply.
Parallel Bricker then issues a challenge: does the study affirmatively support the claim that Egbert marriage should be permitted? Where?
Now, back on our planet, would you agree that Parallel Bricker has a good point? He has not offered one bit of information that would cast doubt on whether the study would apply to Egberts, so is he correct in saying that without a study of Egberts specifically, we can make no inferences from the data that we do have?
Yes, Parallel Bricker has a point; in fact, he’s right on the money. The only way we can make any inferences from the study is to add in some “common sense” conclusions from our own experience – to basically declare that we can’t imagine any reason Egberts would affect the outcome, so we’re going to start off by assuming that it doesn’t. But as Parallel Bricker would undoubtedly point out, that’s an impermissible assumption when it’s the very characteristic you’re questioning.
The multitudes of Parallel Posters decalre Parallel Bricker a genius, and there is much parallel rejoicing.
Really? You’re going to stick to your guns and maintain that, without any reason to think otherwise, that this study doesn’t reasonably support the idea that Egberts would have better families if they’re married? Really?
If we start with the assumption (the quite reasonable assumption, I might add) that being named Egbert has no effect on outcome, the yes, the study supports it.
The problem is that if we’re testing that hypothesis, we’re not allowed to start with that assumption. We must be neutral on the point, unless the study has data to move us.
So, yes: without any reason to think otherwise, this study doesn’t reasonably support the idea that Egberts would have better families if they’re married. Because there is no data to support it, and we’re not allowed to assume it. When Parallel Dseid said, “The reasonable conclusion would be to accept that it would apply to Egberts, given the lack of any reason to think that it wouldn’t apply,” he was cheating – he was adding an assumption that he’s not allowed to make.