This morning’s New York Times had a headline reading: “Girls Lead in Science Exam, but Not in the United States.” The article started with a rather fascinating graph showing country-by-country performance on the OECD test with a display of the percentage gap between male and female students. In the United States, the average scores were 509 for males and 495 for females; thus the males outperformed females by 14 points, or around 2.7%. Compare this with Japan’s data: Average scores of 534 for males and 545 for females gave the girls about a 2% lead.

Both the graph and accompany article interested me enough that a printed copy can now be found on my office door, along with my own editorial remark at the top. (See photographic proof.)

I found the Times article through my Twitter feed. Other interesting articles that hit my feed were a blog post by Hariett Hall (“Gender Differences and Why They Don’t Matter So Much“) and a 2005 article from Time magazine on “The Iceland Exception: A Land Where Girls Rule in Math.” [Michael Shermer linked to Hall’s article, and I shared the link about to the Iceland article.]

After I posted the Iceland article, John Wilson (@jwilson1812) asked for my opinions “about what this report from Iceland might suggest, what’s generalizable, what isn’t, and so on.” In this post I’m hoping to capture a longer response than what 140-characters would allow.

**1. The United States has a gender discrepancy problem in mathematics.**

To me, this point seems somewhat obvious. But given the headline from Hall’s article, and other comments, conversations, and feedback I’ve received over the last decade or two, it also seems clear that it isn’t obvious to *everyone*. I mean “problem” in the above statement as in, “**Something we ought to be concerned with pondering and understanding, and (if possible) fixing.**”

**2. A partial fix could be fixing the educational and employment climate.**

As the Times article points out,

Researchers say cultural forces keeping girls away from scientific careers are strong in the United States, Britain and Canada.

Hall’s article points out that **men and women are different,** and that their skills, interests, and aptitudes are shaped both by biology and by culture. Talking about how biological differences may (or may not) influence mathematical aptitude gets murky very quickly, and I am certainly not qualified to say anything one way or the other. On the other hand, talking about how cultural differences influence mathematical aptitude is a conversation we ought to have frequently.

**3. How can we fix the problem?**

The real answer to this question is, “I don’t know.” But I have a lot of hunches.

**Hunch #1: We need more collaborative classrooms.
**Somewhere a long time ago I read about a study done on middle school aged children playing soccer during recess or physical education classes. The students were separated by gender. In each group, researchers looked at what happened if a soccer player were injured during the game. With the boys’ game, an injury momentarily paused play; a spectator was swapped for the missing team member; the game quickly resumed. With the girls’ game, an injury stopped play. The girls (on both teams) decided they’d rather not play without their injured friend on the field, and so they took up to doing another activity altogether.

I think this parable fits with how I picture what happens in math classrooms. While I’ve taken lots and lots of math classes, I was never able to take a class that would fit any description other than “traditional, chalk-talk, lecture-style, definition-theorem-proof.” The math classes I saw as a student were like the boys’ soccer game: If one student fell behind, or got confused, or failed at mastering a concept, the class would pause, remove the “injured” participant, and continue moving forward. The aim of the class was the soccer game itself and not who was playing and who wasn’t. In my experiences, math classrooms are places where students practice an individual sport (like tennis) concurrently. They are not places of collaboration or conversation or team work. The coach is interested in keeping the game moving forward (even if dropping players is necessary).

I think this is bad for a few reasons. But the top reason is that I think it gives everyone (both women and men) the false impression that mathematics is an individual sport where the performance of the athlete is a solo endeavor. But real mathematics is nothing like this. **As mathematicians, collaboration is essential.** We publish papers together. We give weekly colloquium addresses to teach each other new ideas and to solicit help on tough problems. We travel to conferences to have conversations with others and work through problems as a team. Why do our classrooms give the opposite impression of how mathematics is done?

Showing the world (and girls especially) that mathematics is not done in isolation is crucial. **I believe that marketing mathematics as a collaborative, socially-based adventure would attract more girls to become mathematicians and scientists of all types.**

**Hunch #2: Attract, hire, and retain more female math professors.**

I did my undergraduate work at U.C. San Diego where I was a “Pure Mathematics” major. At the time I was there (early 2000s), the department had about 55 full-time tenured math faculty members. Of those, 5 were female. [See their department directory today for comparison.] One of the women professors mentioned that, at the time, among the “Top 25” math departments, U.C.S.D. had the *highest* percentage of tenured female math professors. What percent is 5/55? About 9%. This statistic was quoted with pride: “We are so great to have so many women! Among the math professors, only 90% of them are male here! Fantastic job!”

I think our cultural conception of what “Mathematics Professor” looks like needs to change. Yes, there are plenty of math professors I know who fit the stereotype exactly. But then there are those who look like me. The way we shift the stereotype is to disprove it. We need more minority math professors, we need more female math professors, we need more math professors who aren’t 60-year-old white males with chalk dust on their pants.

**On keeping women in science: **One thing obviously in need of repair in academics is promoting careers that allow for a work-life balance. Right now, I am expecting my second child. When I complained recently to colleagues about the “Leave Policy for New Faculty Parents,” one responded, “Well, when each of my five children were born, I was back at work the next week.”

I wish I could say this were not the norm. But it reminded me of a conversation I had 10+ years ago, when one of the women faculty at U.C.S.D. told me about giving birth on Thursday and being back teaching classes the following Monday.

**I love my job,** I love my co-workers, I love my students, I love being in the classroom. But my employer’s Leave Policy, combined with the remarkable and surprising lack of empathy from colleagues about said Leave Policy, has certainly made me consider jumping ship. Academia needs to wake up and offer a family-friendly, parent-friendly work environment where people are valued for being *people* first (and professors second).

**Hunch #3: We need to teach teachers differently.**

As an educator, it’s difficult to structure one’s classroom in a way dramatically different from the one you were in as a student. You think back, “How was I taught this idea?” and that’s the easiest answer to, “How will I teach this idea to my own students?” You can see this all over the math community as the traditional, blackboard-based, definition-theorem-proof machine chugs chugs chugs along. Thankfully, there’s been a giant movement in recent years toward changing the idea of what a classroom should look like. (See my earlier ideas about collaboration.)

Given that we are all inclined to teach the way we were taught, and given that for a very long time it was accepted dogma that boys always outperform girls in mathematics, it’s easy to see how this idea could still linger. Not that I think any particular person goes into their calculus classroom and says, “Sorry ladies, everyone knows you don’t have the skills to be really good at this.” But I do think (and I have seen ways) that this underlying stereotype has affected the way people teach.

**My Conclusions**

1. The gender imbalance in mathematics has some cultural factors.

2. We ought to be concerned with what those factors are, and how to change them.

3. Changing them is a process that will definitely take a lot of time and probably take a lot of money.

4. My best strategy at overcoming this problem is this: **Become a female math prof who posts blog articles about the gender imbalance in mathematics. **Unfortunately, this strategy is probably not widely implementable. It definitely takes a lot of time. An easier thing to do is to support and encourage those who are doing this or things similar to it.

5. My next best strategy for overcoming the problem is: **Seek out like-minded people and work together to figure out how we can change the math culture. **

As I said at the beginning of this, I know there is a problem and I don’t know it’s solution. But I’d be happy to hear what you think it might be.