Gay men and straight women have similar brains

[g.johnson]

At least we've identified where the confusion is.

 

Of course I am not suggesting anyone is mis-reporting their brain symmetry. But assume an extreme case: that every single subject who claimed to be straight was in fact not straight. I can't for the life of me see how this would strengthen your confidence in the result.

 

ETA: This is the equivalent of every single marble turning out to be in fact black; little could then be said about the average weight difference between black and white marbles.

But in that case it would be highly unlikely to find a statistically significant difference in weight between the marbles which have, in essence, been selected entirely randomly. But in this case we did find a significant difference. That's the point. Any randomization of categorization can serve only to reduce the measured difference, not increase it, as I hoped my marble example would illustrate.

[g.johnson]

Subjects in another study might well not know whether they suffer from an asymptomatic health outcome, but if you were trying to measure the relative risk of that outcome being associated with exposure X, you'd be stymied by a systematic bias among the subjects in reporting whether they were exposed to X or not.

Again, this sort of misclassification would only serve to reduce the measured difference. If the subjects are indeed ignorant of their health status they are not likely to be biased to falsely reporting exposure. Someone who worked in the Glasgow shipyards and who doesn't know that he suffers from asymptotic asbestosis is no more likely to over-report exposure than his mate who worked the same job but happens to be disease free.

[g.johnson]

The question is really this: why would men with symmetric brains be more likely to describe themselves as gay than men with asymmetric brains?

[g.johnson]

Again, I'm raising the issue of whether systematic bias might be a problem in a fairly small study like this, not randomization. It is surely uncontroversial that a statistically significant positive result can be rendered non-significant (or even reversed) if systematic misclassification has taken place.

 

Maybe I am communicating very badly, but it is surely obvious that one's estimation of the risk of A causing B can be artificially inflated if your exposure group contains plenty of misclassifiied subjects and your unexposed group does not.

Right. But your problem is identifying a mechanism by which bias can be introduced when the subjects are entirely ignorant of the dependent variable. See my question above.

[Wilfrid1]

Okay, you are finally getting through to me.

[yvonne johnson]

But misclassification bias is different than systematic error, no? The researcher doesn't have a say in who is gay or straight (misclassification bias). I happen to think that the self-identifier (gay/straight) is OK (I realize you'd like another measure). There would be bias if the researcher did all the images then decided who would be gay or straight, or something similar.

 

And we have no evidence that there was anything wrong with the imaging machine (the main systematic error that could occur in this study).

 

Maybe I am communicating very badly, but it is surely obvious that one's estimation of the risk of A causing B can be artificially inflated if your exposure group contains plenty of misclassifiied subjects and your unexposed group does not.

That's the result of a poorly designed study.

[Daniel]

Wow.. For a liberal crowd its pretty funny that people are actually discussing this seriously.. Scientist don't know a damn thing about the brain but, somehow they know this?

[g.johnson]

The study doesn't say anything very much about brain function. It's concerned only with rather simple (and easily measured) anatomical observations about hemisphere size.

 

And we do know rather more about the way the brain works than you realize. We don't know much about qualia.

[Orik]

Wow.. For a liberal crowd its pretty funny that people are actually discussing this seriously.. Scientist don't know a damn thing about the brain but, somehow they know this?

 

Another study finds that many foodies express conservative opinion, but are later discovered to have been closet liberals.

[Kikujiro]

I realise I'm just showing off my superior verbal fluency here, but doesn't the Guardian report linked to up front precisely misreport the earlier "bad drivers" study as far as lesbians are concerned?

 

The study, led by the neurobiologist Ivanka Savic, builds on previous research that has identified differences in spatial and verbal abilities related to sex and sexual orientation. Tests have found gay men and straight women fare better at certain language tasks, while heterosexual men and lesbians tend to have better spatial awareness.

Grauniad

 

Were they any differences between gay and heterosexual women?

Homosexual and heterosexual women are alike in cognitive profiles in almost all studies, except in a study we conducted on verbal fluency, where lesbian women are male-like.

Interview with the author of the aforementioned "previous research" on spatial awareness

 

I'm waiting for the study on butches vs femmes, myself.

[omnivorette]

Be still my beating heart. It's Kiku.

[Orik]

I think kiku is expressing his usual concern - that actual scientific insight into the operation of the brain might turn whole fields of study into mere curiosities.

 

Kiku

[Kikujiro]

I think kiku is expressing his usual concern - that actual scientific insight into the operation of the brain might turn whole fields of study into mere curiosities.

 

You mean this? I'm not very well read on the Churchlands, but I assume y'all debated them thoroughly while I wasn't paying attention.

[omnivorette]

Oh my god. He posted again. I may have to go lay down.

[Orik]

I think kiku is expressing his usual concern - that actual scientific insight into the operation of the brain might turn whole fields of study into mere curiosities.

 

You mean this? I'm not very well read on the Churchlands, but I assume y'all debated them thoroughly while I wasn't paying attention.

 

We's all did that a while back. Of course, in keeping with the nature of food board discussion, the parties involved have all denied any possibility that something exists between this:

 

http://en.wikipedia.org/wiki/Daniel_Dennett

 

And that:

 

http://en.wikipedia.org/wiki/Tabula_rasa