1.) Skepchick Magazine will be late this month. Again. I suggest you consult with your own trusted psychic or spirit guide in order to console yourself until it comes out.
2.) I have a big report to file, but I have to hold off for a few reasons. Look forward to something either over the weekend or on Monday that I know you guys are going to dig.
3.) Men are smarter than women. Uh, wha? There are so many problems with this that I don’t even know where to start: how does one choose what factors to include when describing intelligence? How do you objectively test and quantify those attributes? Once you come up with your numbers, what the hell do they mean in the real world? That’s the biggest problem — the researcher pulling this crap is using a difference of less than 4 IQ points to explain why women don’t get equal rights, pay, and treatment in general. I’m in awe that this is even considered news. Hey news outlets, can we stop giving these assholes a venue on which to air their stupidity? Thanks!
And my thanks to anyone who reads more about that study and explains to me how it has any scientific merit.
At school I got IQ tested and did fairly well, but since school I have a) been so drunk I've made a complete arse of myself, b) betted money frivolously, c) thought that jumping over a bench in a park was a good idea, d) posted a Rush song on my blog. There is an addage from the Panglossian Forest Gump…
It's also interesting that John Philippe Rushton makes the assumption that women hit a glass ceiling because their brains are somehow inferior. I can only think he must of tested the IQs of *all* male senior executives, because his statement is predicated on the basis that a higher IQ equals more success, which I don't think is backed up statistically. HAIL TO THE CHIEF!
“Hey news outlets, can we stop giving these assholes a venue on which to air their stupidity? Thanks!”
Evidently not. Ever. I’m increasingly getting the impression that news outlets exist solely for the propagation of stupidity.
The journalist’s obligation to the public interest is a myth, just like dry land.
As a side note, I’m starting to be slightly uncomfortable with the number of first posts I’m getting here in the comments. Is it an evil conspiracy or just that Rebecca and I are in the same time zone and both participate in blog-related activities (blogtivities) while at work?
Isn’t it possible you just didn’t understand the study?
I’ve tried reading up on the subject but it goes over my IQ endowed head, so I’ve forwarded this on to Ben Goldacre of the Guardian’s Bad Science column. Hopefully he’ll have something to say on the subject. What strikes me most in my readings is how IQ seems to be an American obsession.
That’s “fewer than 4 IQ points”.
Fewer if you count it and less if you measure it..
Aside from that, IQ is just a measure of an abstract concept. I apparently have a high IQ but experience leads me to believe it’s not an indicator of any kind of success or wisdom. Any happiness is just a by-product.
The “research” is pure swill. No better are the two commentators at the bottom of the article. Let’s paraphrase:
1. Men are smater than women because I’m the dumb one in my family. [Yes, she’s the dumb one if she thinks a sample size of 4 and an unscientific survey are worth a thimble-full of spit.]
2. Women are smarter than men because they are so nuturing and competent and perfect…or at least I get laid way more when I talk this way.
I'm willing to accept that men might have a genetic predisposition in spacial understanding and women in linguistic. After all it might be easy to explain since men did the hunting and women the child-raising. But, let's be honest. A 3.6 degree difference in a scale that ranges from 95 to 170 (or more) is too marginal to be of significance. We also see no margin of error for this study, which is suspicious. I would bet that the margin of error was enough to nullify that 3.6 difference.
And let's not forget that psychologists are still divided over the accuracy of IQ tests or their actual meaning. Meh… we're still in the summer TV zone, when news is scarce. This looks nothing more than an attempt to fill in news-space.
While I certainly am not giving this particular test any merit, I do feel it necessary to dispel a few misconceptions about IQ (namely that a difference of 3.6 points is in fact quite large)
First of all, the scale does not range from 95 to 170 – in fact, the AVERAGE IQ is 100, so to say that the range starts at 95 would be to claim that nobody is less than a third of a standard deviation below average. It goes as far below 100 as it does above, and numbers below 0 are possible, just as numbers above 200 are.
How the scale works is that the relative intelligence of each of the people within a population is measured and plotted on a histogram. The result is a bell-curve with a standard distribution. The peak this curve is assigned to be 100 points, and 1 standard deviation in either direction is a change of 15. So, if a person were to have an IQ of 103.6, that would mean they are 0.24 standard deviations above normal, which corresponds to being more intelligent than 59.5% of the population (as opposed to an IQ of 100, which would mean one is more intelligent than 50% of the population).
IF this study were correct (which, if it has a sample size of 4, most certainly is not). it would mean that the average man (one who is more intelligent than 50% of men) would be more intelligent than 59.5% of women. Conversely, the average woman would be more intelligent than only 40.5% of men. So, despite only being 3.6 points on a scale of hundreds, a change of the average of 3.6 points is in fact quite significant.
The one thing that you are correct about though (EvanT) is that in a study with a sample size of 4, yes, the margin of error would be by far large enough to make it impossible to tell whether or not this difference is in fact real, as the error bars will be much much more than 3.6 points.
The sample size of 4 came from the columnist commenting on her own family, not the study. 3.6 average difference may indeed be statistically significant in the study. (BB is quite correct to warn against the fallacy of judging based on whether or not a number seems small instead of doing proper statistics.) They may have even done one of the “good” IQ tests.
Even if we stipulate all of this, the question still remains whether IQ is a measure of any practical value. I’ve never seen any evidence that it is.
Is that an error margin per type of test, or per person tested?
For what it's worth, it might mean women are actually "smarter" than men by up to 6.4 IQ points.
Standardized IQ measures have a margin of error of 5 points in either direction. The Scholastic tests are not an IQ test, they are an achievement test. Big difference.
For what it’s worth, it might mean women are actually “smarter†than men by up to 6.4 IQ points.
Error margins exist within a percentage confidence, frequently 95% (α value of 0.05). An error margin does not mean that the real value is definitely within that margin, it just means that it has a (1 – α) probability of being within that margin.
So, once you get into what "might" be true, you are opening the door wide open. Anything might be true. It's just a matter of how likely it is. I may be telling you something you already know, but I'm just being rigorous.