Then, on the gender issue: This is a real issue. I have read Eileen Pollack’s book, The Only Woman in the Room, and been troubled by the circumstances that turned her away from physics. She completed a really wonderful senior thesis in physics at Yale, but was not encouraged by any of the professors to continue to grad school. In general, I think that undergrads (perhaps women especially) may greatly over-estimate the knowledge and experience distance between themselves and the faculty. This may be a bigger issue at top schools.
STEM undergrads do not need to go to grad school. But I think Prof. Pollack might have wanted to do that, and a little encouragement could have sent her in that direction. I doubt that her story is unique.
Without succumbing to gender stereotyping, I think it is true that women may react somewhat differently to setbacks than men do (on average). Data from a largish group of universities in my field tend to show lower odds of PhD completion by women than by men. If men who enter the university as grad students complete their PhDs with a probability of x (x < 1 virtually everywhere), then the women at the same university tend to complete their PhDs with the lower probability of x squared. We also lose women at the transition from undergrad work to grad school. Again, there is no need for people to go on to grad school–but we would like to offer the opportunity to anyone who is interested in it, and has a reasonably good undergraduate record.
In practical terms, there are a few steps that can be taken:
The use of set percentage guarantees for grades eliminates feelings of head-to-head competition, which women may enjoy less than men, or not enjoy at all.
If the course does not have set percentages, then it is useful if the distribution of scores is provided by the professor. Most of mine did that. There was a memorable midterm exam sophomore year where the mode was 4 (out of 100) and the median was 28. The great majority of the students who scored below the median and stuck with the class still passed. In a case like this, it may actually be useful to encourage daughters to compare their performance with the rest of the class. If the median is 28, then any score above 38 is great!
Encouraging daughters to talk with grad students (if they are at a university) is helpful, and it will give them a different perspective.
The New York Times Sunday magazine carried a story about the difference between “Worriers” and “Warriors” about a year ago, with reference to exam performance. The “Warriors” were those who loved competition, and tended to do better on standardized tests and other high-stakes exams than one might have predicted. “Worriers” worried about a large number of contingencies, became anxious, and tended to do worse on the tests than one might have predicted. The effects of this could be counteracted by showing the “Worriers” that a certain amount of nervousness would actually enhance their performance.
There is another gender issue that is less easily counteracted: As I understand it, women react differently than men to the molecules the body produces in stress situations (cortisol, I think? plus others). The only solution that I know of to this is to get beyond the stage where there are timed, written exams, and grow comfortable with conference presentations, by making them.
There is still bias, implicit and explicit, but as more women enter STEM fields, I think bias will be less common (though sadly, probably not gone).
