Another score for the curvy burds.. - Page 3

endless psych · 15th November 2007, 3:12pm

Quote:

Originally Posted by hotblonde

Many general IQ tests are education based (altho, they shouldn't be) which does not measure IQ in any way. Also, stats tests used to qualify IQ vary to the degree that a post hoc test can be significant simply by employing Tukey rather than Bonferroni (or whatever).

What I am getting at is that IQ is not the important thing which is the problem - neither is IQ particularly difficult to analyse statistically. Saying (with any degree of confidence) that IQ relates to actual intelligence is the hard bit.

The problem is not IQ its intelligence itself.

(Also, given my knowledge of stats is a little rusty, would it not have to be one hell of a random population so that you could consider using both the Bonoferroni or Tukey test on it? Although this could just be because I'm thinking of a Bonoferroni T-test and the non-parametric Tukey wotsit.)

hotblonde · 15th November 2007, 3:16pm

Quote:

Originally Posted by endless psych

What I am getting at is that IQ is not the important thing which is the problem - neither is IQ particularly difficult to analyse statistically. Saying (with any degree of confidence) that IQ relates to actual intelligence is the hard bit.

The problem is not IQ its intelligence itself.

(Also, given my knowledge of stats is a little rusty, would it not have to be one hell of a random population so that you could consider using both the Bonoferroni or Tukey test on it?)

Post hoc tests are generally just for mulitple comparisons (being that if you look hard enough, you'll find a significance in any group comparison, so post hoc tests such as Tukey, etc add an extra layer of complexity). Given that most human populations are heterogeneous and IQ is a quantitative trait , these tests are subject to post hoc analysis as well as adjustment for various other factors. That was just an example. Many quantifiable traits are subject to stat manipulation. If you think IQ is easily statistically measured or quanlified, you must be in the minority.

endless psych · 15th November 2007, 3:19pm

Quote:

Originally Posted by hotblonde

If you think IQ is easily statistically manipulated, you must be in the minority.

Nope, I just think measurement and analysis are slightly different. Although thinking about it from a positivist point of view that statement probably sounds batshit crazy.

To try and put it more succintly I don't think theres any inalienable complexity in measuring anything remotely psychometric. The problem arises when you try to validate you are measuring and testing what you say you are.

hotblonde · 15th November 2007, 3:23pm

Quote:

Originally Posted by endless psych

Nope, I just think measurement and analysis are slightly different. Although thinking about it from a positivist point of view that statement probably sounds batshit crazy.

Most phenotypic measurements, such as IQ are very much at the mercy of those completing the survey or examination. As many of these surveys/studies are not generally streamlined to the point that they can be examined by multivariate analysis as the phenotying is not universally measured by a general consensus (all participants reading off the same hymn sheet so to speak), nobody and no one group can say that any trait is easily measured. That is the beauty of phenotypic analysis...it's all down to how good the actual measurements are and how well designed the study is. So, I still think that you're in the minority on IQ being an easily measurable trait.

*LesMTS* · 15th November 2007, 3:24pm

Their hypothesis (low waist to hip ratio will correlate with cognitive ability) is based on the idea that lower body fat has positive effects on the supply of fatty acids that are important for neurodevelopment. Their findings are that low waist to hip ratio women do indeed perform better on cognitive tests, as do the children of low waist to hip ratio women. Teenage mums (who are still developing themselves) and their kids may suffer from the competition for these neurodevelopmentary resources.
No idea about numbers of participants or what the tests were. Which would be good to know.

"Curvy" is just a word the BBC News have used to make it appealing.

Here's the abstract:

Quote:

Waist-hip ratio and cognitive ability: is gluteofemoral fat a privileged store of neurodevelopmental resources?

William D. Lassekab, Steven J.C. Gaulinb
Received 16 April 2007; accepted 26 July 2007. published online 4 October 2007.
Corrected Proof

Abstract
Upper-body fat has negative effects and lower-body fat has positive effects on the supply of long-chain polyunsaturated fatty acids that are essential for neurodevelopment. Thus, waist-hip ratio (WHR), a useful proxy for the ratio of upper-body fat to lower-body fat, should predict cognitive ability in women and their offspring. Moreover, because teenage mothers and their children compete for these resources, their cognitive development should be compromised, but less so for mothers with lower WHRs. These predictions are supported by data from the Third National Health and Nutrition Examination Survey. Controlling for other correlates of cognitive ability, women with lower WHRs and their children have significantly higher cognitive test scores, and teenage mothers with lower WHRs and their children are protected from cognitive decrements associated with teen births. These findings support the idea that WHR reflects the availability of neurodevelopmental resources and thus offer a new explanation for men's preference for low WHR.

hotblonde · 15th November 2007, 3:26pm

What journal is in Les?

*LesMTS* · 15th November 2007, 3:27pm

Quote:

Originally Posted by hotblonde

What journal is in Les?

Evolution and Human Behaviour.

endless psych · 15th November 2007, 3:27pm

Quote:

Originally Posted by hotblonde

Most phenotypic measurements, such as IQ are very much at the mercy of those completing the survey or examination.

Phenotypic makes the assumption that IQ's purely genetic no? Or am I misunderstanding the meaning of phenoytypic?

TSR · 15th November 2007, 3:29pm

Quote:

Originally Posted by Timmy

Fat burds are stupid. I have proof.

Fat burds want men, men want skinny women... fat women won't/don't lose weight in order to get men. Fat women are stupid.

your views intrigue me and i wish to subscribe to your newsletter

Quote:

Originally Posted by Largest of Als

We're not talking about fat burds, we're talking about curvy burds

Less healthy that a stick insect model type who only eats 10 calories a day?

There's nothing to say that girls cant be curvy and healthy at the same time...

Al

there's a line between curvy and fat, and it's often ignored/distorted

some people can eat anything they want and be skinny. doesn't mean they eat 10 calories a day. then you get those who eat a shitload and go to the gym 3 times a week. don't be so quick to call skinny girls "unhealthy stick insects," cos some of those bodies take a lot of effort to get/maintain

curvy and healthy, yes. but only if you observe the line. some fat people appear to have their own gravitational effect which bends the line the curse of family guy strikes again

*LesMTS* · 15th November 2007, 3:29pm

Quote:

Originally Posted by endless psych

Phenotypic makes the assumption that IQ's purely genetic no? Or am I misunderstanding the meaning of phenoytypic?

Yeah, I wouldn't say IQ was a phenotype.

hotblonde · 15th November 2007, 3:30pm

Quote:

Originally Posted by endless psych

Phenotypic makes the assumption that IQ's purely genetic no? Or am I misunderstanding the meaning of phenoytypic?

No, phenotype does not mean purely genetic. IQ is a quantifiable trait, in other words a QTL (such as height, weight, heart weight, blood pressure, etc). So, yes both yourself and Les are misunderstanding what phenotype means.

endless psych · 15th November 2007, 3:30pm

Quote:

Originally Posted by LesMTS

Yeah, I wouldn't say IQ was a phenotype.

Well I dunno there is a case for intelligence being as much about nature as nurture but probably less so for Intelligence quotient given the illiterate will nearly always have low IQ scores. (Although there are IQ tests specifically designed for folks without language comprehension now IIRC)

hotblonde · 15th November 2007, 3:33pm

The environmental influence can never be discounted (even in bad studies, it seems

)

endless psych · 15th November 2007, 3:33pm

Quote:

Originally Posted by hotblonde

No, phenotype does not mean purely genetic. IQ is a quantifiable trait, in other words a QTL (such as height, weight, heart weight, blood pressure, etc). So, yes both yourself and Les are misunderstanding what phenotype means.

Right oh.

IQ isn't a trait btw. Although it is generally (erroneosly in my opinion) treated as one I suppose.

wee scary burd · 15th November 2007, 3:33pm

Quote:

Originally Posted by endless psych

What I am getting at is that IQ is not the important thing which is the problem - neither is IQ particularly difficult to analyse statistically. Saying (with any degree of confidence) that IQ relates to actual intelligence is the hard bit.

The problem is not IQ its intelligence itself.

(Also, given my knowledge of stats is a little rusty, would it not have to be one hell of a random population so that you could consider using both the Bonoferroni or Tukey test on it? Although this could just be because I'm thinking of a Bonoferroni T-test and the non-parametric Tukey wotsit.)

I read that as Turkey Wotsits. Yuck

p.s. i have an IQ of 144