kitchen table math, the sequel: SAT Gender Gap

Tuesday, June 14, 2011

SAT Gender Gap


The College Board Trend Report shows boys scoring higher in math than girls.

I have to believe this is cultural rather than wiring.






Cross Posted on Perfect Score Project



19 comments:

Luke Holzmann said...

Gladwell's Blink and Dr. Sax's Why Gender Matters seem to agree with you: Cultural (and who-knows-what-other) influences dramatically affect test scores. I'm surprised at how well males do in general, considering what I've read about how school tends to be geared more toward females ...but I forget which book that was in...

~Luke

Anonymous said...

Why do you "have" to believe it?

Couldn't you just believe it or not believe it?

Crimson Wife said...

Why are you so convinced that it's cultural? Dr. Lise Elliott wrote a fascinating book called Pink Brain, Blue Brain that discussed the scientific evidence showing boys have superior spatial reasoning abilities from infancy and girls have superior verbal reasoning. The evidence points to prenatal testosterone exposure as the cause. Girls who have a male co-twin perform more like boys than girls who have a female co-twin.

The important thing for parents of girls to remember is that score variations between individual girls dwarf the difference between female and male group averages. Group differences tell us very little about how any one individual will score.

Linda Seebach said...

Culture and wiring influence each other. Welsh people whose families and communities have been deeply involved with eisteddfods since before they were born are likely to be more skilled at choral performance; but Wales wouldn't have a culture of eisteddfods if the typical person in Wales couldn't carry a tune in a bucket.

Bonnie said...

The main point of Pink Brain Blue Brain was actually that culture matters. The book is about how small hardwired differences are dramatically amplified by the different ways we respond to boys and girls. Look at the subtitle: "How Small Differences Grow into Troublesome Gaps And What We Can Do about It"

Bonnie said...

Just to add - the author of Pink Brain Blue Brain (Lise Eliot) has sparred publicly with Michael Gurian, whose work she describes as pseudoscience. I rather agree with her.
http://ascd.typepad.com/blog/2010/11/myth-of-pink-blue-brains.html

cranberry said...

Boys outscore girls on the SAT verbal scores as well.

Why does it matter whether it's cultural or wiring?

As it is, the emphasis upon graded homework, arts and crafts, and group projects in K-12 education depresses the (typical) boy's academic performance.

Crimson Wife said...

When I read Dr. Elliott's book, the impression I took away from it is that it isn't so much differential treatment by others but rather that children seek out and spend more time doing activities related to the inborn strength. Hence my DS spending hours building with Legos and my DD spending hours reading. It's the whole "Matthew" effect again.

Bonnie said...

It matters whether a difference is hardwired or cultural because differences that are due to acculturation can be changed, if we decide that the skill is important enough. For example, we as a society decided that it is important for girls to do well in math, a couple of decades ago. One of the things that people realized was that we were giving girls overwhelming messages that they could expect to be bad at math. So everyone worked on that. And it helped. Girls have dramatically improved their performance in math in the last 20 years. This result was widely reported a couple of years ago.
http://berkeley.edu/news/media/releases/2008/07/24_math.shtml
Have we truly erased the gap between boys and girls? No! There is probably always going to be some difference, especially at the extremes (more boys tend to score at the highest reaches in math). But focusing on cultural aspects of the problem worked to improve things! There are now many more girls going into math, and into mathematical fields.

I believe very strongly that we are seeing a similar problem with boys. Teachers have such low expectations for boys. I have seen even MOMS stand in front of their 5 year old boys and joke about "He's all boy, he doesn't know his alphabet". What kind of message is being sent? It is so much like the message that girls used to get about math (and I grew up in that era, so I do remember teachers telling me that "girls can't do math").If we were able to improve girls scores in math by working on expectations, why can't we do the same for boys?

One of the problems I see is that all the teachers have latched onto this Gurian guy with his sweeping generalizations and stereotyping. My oldest son was a victim of this - when he entered kindergarten, he was reading simple chapter books. The teacher would not believe me, and didn't finally believe me until she did the reading assessment in December. That meant he was essentially ignored for several months. At the parent teacher conference, she told me she was so surprised, because "after all, boys don't read". As it turned out, the top 3 readers in that class were all boys.

As the mom to two boys who are heavy readers, and who don't like to jump around in class or disrupt things, and also to one very active, athletic, disruptive 5 year old girl who is still struggling with her letters, I would like educators to dump all the gender stereotyping and start seeing children as individuals. And to have high expectations for both boys and girls, and to stop with this "girls just can't do math" and "boys struggle to read" business.

cranberry said...

Bonnie, the study you listed based its results on the state-level NCLB tests. Our state's test is one of the hardest in the nation, and it isn't challenging to the top kids.

Again, even if some teachers expect boys to struggle with reading, on these SAT stats, boys outscored girls.

You may want to believe that boys and girls would post equal scores in a gender-blind culture, but we haven't attained a gender-blind culture, thus that's a belief, not a fact. It would be nice if teachers would assess each student individually. I continually had to point out to elementary school teachers that my daughter was a fast reader, and really did understand the books she read.

We know that intelligence has a strong genetic component. We also know that many, many genes are involved, and it isn't simple. On the other hand, males will express more recessive genes than females, as they are XY and girls are XX. Sex-linked diseases such as Hemophilia, color blindness, etc., will show up in males more frequently than females. Some genes might help academic performance, and some might hurt academic performance.

In terms of "wiring," one might expect what experience generally shows--the boys tend to have a different distribution of test scores, a flatter curve, with a greater percentage of very high scorers and very low scorers. The SAT isn't a survey of the entire field of high school juniors. Most of the students who would post very low scores on the SAT don't take it. If everyone did take the SAT, the male and female scores might seem to be even, as the very low scores might balance out the very high scores. If we looked at only the testers who scored 500 and below on such a theoretical test population, we might be discussing, "why aren't boys scoring as well as girls?" because the sample would include more very, very low-scoring boys.

Crimson Wife said...

Doesn't Maine require all high school juniors to take the SAT? It would be interesting to see if that state shows any differences in the gender gap compared to the overall national average. I know that Maine has a much lower SAT average because of all the extra test-takers (1389 vs. 1509 nationally) despite having higher-than-average NAEP scores.

cranberry said...

Good thought! Found it! http://professionals.collegeboard.com/profdownload/ME_10_03_03_01.pdf

At a glance, the Maine scores do show more high & low scoring males. On Critical Reading, 52.8% of the high scorers (700-800) are male. Males are also 63.5% of the lowest scorers (200-290). In Math, males are 67.6% of those who scored above 700. They're also 55.17% of the lowest scorers (200-290).

There are more males (7941 vs. 7442), but at a glance, the shapes of the curves differ.

Female students are stronger than male students at writing. On the writing subtest, females post more very high scores, and far fewer very low scores.

Bonnie said...

I am fully aware of the type of test that the study used. In fact, I said in my post : "Have we truly erased the gap between boys and girls? No! There is probably always going to be some difference, especially at the extremes (more boys tend to score at the highest reaches in math)."

However, there was a lot of improvement in the last 20 years. The improvement probably affected the kids in the middle more than those at the highest levels, but that is where most of our kids are anyway. The point is - changing our expectations of girls abilities in a field that we deem important *worked*. Girls did improve. So if boys are not reading well, maybe we should try a similar approach. Stop wringing our hands and saying "boys just can't...", and instead start expecting and encouraging them to do well.

Anonymous said...

This site:

http://cliftonchadwick.wordpress.com/2010/12/04/perfect-sat-math-scores-male-female-ratio-of-21/

has some interesting data and analysis.

It looks to me that what helped narrow the male/female gap was that the girls just worked harder than the boys. More AP and honors courses. Better grades. More years of math in high school.

This should be replicatable for boys and reading.

But if the girls have to work harder than the boys to narrow (not close) the "natural" gap, then this suggests that there is a non-trivial biological component that you can partially overcome via culture. This doesn't mean that the "natural" gap isn't real and it doesn't mean that the gap is cultural instead of biological.

-Mark Roulo

Bonnie said...

So you have to work hard? Is that a bad thing? Isn't that the view that Asian students typically have - that math ability has more to do with how hard you work than some innate talent? I have often heard that cited as a reason why so many more Asian students do well at math. Perhaps that is the secret - stop worrying about who does or does not have some magical innate talent, and start worrying about getting everyone to work harder.

Anonymous said...

"So you have to work hard? Is that a bad thing?"

Sigh.

No, it is not a bad thing.

And I didn't say it is a bad thing.

In fact, I suggested that this might well be used to narrow the reading gap between boys and girls.

I was responding to your belief that the boy-girl gap in math scores isn't (or can't be?) biological.

Unless you think that working hard is some innate advantage that girls have over boys, then once you normalize for working hard, the boys are still going to have higher test scores.

-Mark Roulo

Bostonian said...

Bonnie wrote, "Isn't that the view that Asian students typically have - that math ability has more to do with how hard you work than some innate talent?"

East Asians have an average IQ of about 105, compared to 100 for whites, and they have a relative strength in math. This likely explains their higher average SAT math scores and their over-representation in STEM fields.

Bonnie said...

And you are ignoring the fact that I already said that I think there is some biological difference. I just think there is a lot more that is due to culture and life experience. That was the central point of Lise Eliot's book, which is what led to this entire tangent.

If we believe reading and math are important skills, then we need to remove all barriers to excelling in those areas, including the barriers of telling half the population that they can't do something well because of their gender. I went through it as a girl in the 70's, and I see it happening with my sons now in the 00's. Sad.

Bostonian said...

Bonnie wrote, "If we believe reading and math are important skills, then we need to remove all barriers to excelling in those areas, including the barriers of telling half the population that they can't do something well because of their gender."

Your reading comprehension appears to be poor.

No one in this thread is telling girls they cannot do math well because of their sex. We understand that there is a distribution of math ability within each sex and that the distributions overlap considerably. There are lots of girls who are better than the average boy at math. The empirical (not moral) question is whether the two distributions of male and female math ability are the same. Some evidence suggest that they are not, with the male distribution having a higher mean. Regardless of whether this is true, each child should be evaluated and taught based on measures of his or her own achievement and aptitude.