3 ms·
Harvey Mudd College: Increasing women in CS from 10 to 40% in 5 years
- tracker1 11y agoMy only concern when I see things like this, is I really don't want to see better gender equality at the expense of reverse-discrimination, or for programs that are only for one role|race|gender. It's one thing to try to balance out, it's another to create a situation of intolerance. My younger brother had a teacher (5th grade, about 20 years ago now), that openly said that boys in her class would average one grade score below girls in the class. Thing like that just aren't right, and don't serve to improve a situation. As it stands there are significantly more women than men involved in post high school education. Right, fair and equal are not always the same thing... but just should be.
- nkurz 11y agoI wonder how the faculty male/female ratio impacts the student ratio. Here's Harvey Mudd's CS Department: https://www.cs.hmc.edu/people/ https://www.cs.hmc.edu/people/ It lists 15 Faculty: 9 Male, 6 Female. Interestingly, 6/15 also equals 40%. How does Harvey Mudd's ratio compare to other schools? I'm guessing most CS departments are more male dominated, but don't know for sure.
- H0n3sty 11y agoTL;DR 1. split intro course into one class for students with no prior programming experience and one for students with prior experience. 2. send about 65, including 25 first year, female students to a conference geared towards women in technology. 3. encourage female students to apply for early research opportunities. 4. match female computer scientist mentors with female students. Interesting comment in original article: So to summarize you use special treatment to remedy a gender imbalance because you feel better if you have an equal number of male and female students in any given major. What majors have more women than men and what is being done to remedy those?
- dang 11y agohttps://news.ycombinator.com/item?id=11182080 https://news.ycombinator.com/item?id=11182080
- harry8 11y agoThe lack of women in any CS program is a symptom and indicates a problem that is wider than mere gender (or race or whatever obvious demogoraphic category you have). Harvey Mudd appear to have a desire to fix the actual problem, which is people who would do very well to go with CS being discouraged. As long as what anyone does to address demographic imbalance goes to the root cause and is not just a whole load of symptom masking I'm all for it. If $minority aren't present you can bet that there are also a bunch of white guys who would be really good at CS, better than the incumbents perhaps, who have also been discouraged because they're shy, are intimidated by those who loudly know 1% more at the start of the course or whatever it is. Just as there would be any other demographic group. In fixing that problem of discouraging the best talent the quality of their CS grads in terms of what they are able to get done, what novel insights they have should improve. The students should push the academics harder and make the CS program stronger. Will that happen? I hope so but we'll see. I don't know more than this blog post. The disaster would be if symptoms of the disease were masked and the manner in which they are masked means the grads have learned less. (Eg finding the subject matter is too intimidating so dumbing it down as a response.) Anytime you have a narrow demographic dominating a field it's symptom telling there's a problem but it tells you exactly nothing about what the true nature of the problem is. You have to do a bunch of additional work to find out if it's because of hostility to "outsiders", poor preparation earlier in education, societal cultural discouragement or if the field has no intellectual merit and is largely a con that has taken hold of a particular demographic or any one of a million other causes. Treating a demographic imbalance as the problem itself is crazy - the disease, whatever it is, will persist and be harder to track.