Mercury News passes the buck: Can early CS education boost number of women in tech?

February 26, 2014 at 1:07 pm 12 comments

Check out the headline “Can early computer science education boost number of women in tech?”  Then read the part (quoted below) where they show what works at Harvey Mudd.  I don’t read anything there about early CS education.  I do believe that we need CS in high schools to improve diversity in computing, but I’m not sure that much earlier than high school helps much.  I worry about higher education giving up on issues of diversity, by changing the discussion to K12.

I wish that Mercury News would have really said what they found: University Computing Programs, you have the power to improve your diversity! You can change your classes and your culture! Don’t just pass the buck to K12 schools!

“The difference is, females in general are much more interested in what you can do with the technology, than with just the technology itself,” says Harvey Mudd President Maria Klawe, a computer scientist herself.

So administrators created an introductory course specifically for students without programming experience. They emphasized coding’s connection to other disciplines. They paid for freshman women to attend the annual Grace Hopper Celebration of Women in Computing, a chance to meet programming role models in diverse fields. And they provided early research opportunities for women students to inspire them to stick with the field.

The result? The percentage of female computer science majors at Harvey Mudd increased from about 10 percent before the initiatives to 43 percent today.

via Can early computer science education boost number of women in tech? – San Jose Mercury News.

Entry filed under: Uncategorized. Tags: , , , , .

Are MOOCs the New Textbooks? Misunderstanding the role of the teacher There are no computer science teachers in NY

12 Comments Add your own

  • […] Read more from the original source: Mercury News passes the buck: Can early CS education boost … […]

  • 2. lizaloop  |  February 27, 2014 at 12:03 am

    3 reasons for early (K-12) computer literacy (including coding) courses:
    1. Every citizen needs to understand computing in order to make informed policy decisions in a democracy. In this sense CS is part of social studies and as fundamental as history, geography and student government.
    2. Coding is an excellent introduction to algebra, often more motivating and less painful because the result can be a fun computer game.
    3. Girls and “underprivileged” kids need do be told early and often that CS is an available career for them. Since they are less likely to get this message at home it is important for schools to intervene.

    I have taught BASIC, Pilot and LOGO to hundreds of kids and have one warning to those of you who love to code. Understand that only about 1/3 of your students are likely to share your enthusiasm for coding itself. Don’t spoil the experience for the other 2/3s: They can have a good time with simulations and applications. The important concepts are: 1. that people should control computers and not the other way around, and 2. garbage in – garbage out.

  • 3. tomas  |  February 27, 2014 at 12:35 am

    Would people please stop praising Harvey Mudd in general for this ridiculousness? They pulled a horrible self-serving PR move that is not a good model for education or the welfare of anyone, students, faculty, or the public overall.

    They simply changed their admission policies, in short, behind the black box that the media doesn’t cover. It’s not clear what exactly went into that except any numbers ever released show that the overall changes of their CS department weren’t particularly positive.

    They didn’t meaningfully increase the number of female computer science majors. The actual number was something like a dozen or two total people. They correspondingly decreased the number of men majoring in computer science (through not being admitted in the first place), which of course changes the ratio.

    Overall, admitting people to your institution is the end-all, be-all answer, ahead of all pathetic stunts. Shocking, but there is not a lack of potential students out there. Every school that admits more wealthy foreign students than various minorities, that brings in generic business majors, they are choosing not to have CS majors, simple as that. Enough with cheap stunts to make certain individuals’ careers and administrations look good while purposefully avoiding programs and meaningful action that helps underprivileged groups.

    • 4. Mark Guzdial  |  February 27, 2014 at 11:29 am

      Tomas, what evidence do you have for your claim? I’ve seen evidence that CMU increased the percentage of women in their computing program by simply changing admissions policies. But the evidence that I’ve seen suggests that Harvey Mudd successfully convinced students to become computer scientists by activities on campus. There’s even a paper in this month’s CACM going over that data, where the lead author (Christine Alvarado) is now at UCSD not Harvey Mudd. She has no incentive to feed a “PR move.” She’s analyzing data and finding which of the interventions at Harvey Mudd had the greatest effect (and has been studying the data for years).

      • 5. tomas  |  February 28, 2014 at 1:25 am

        The total number of male CS graduates went down. The number of degrees in total also didn’t really change, which of course it has to to mathematically match. The statistics are wholly inconsistent with higher education trends in general, except of course for HM being a selective institution that just artificially caps and chooses the make up of any of its classes.

        You have to look at the numbers to see that Harvey Mudd is a private institution with a tiny number of students. They don’t seem to share detailed info with all third party organizations that track university stats so I don’t know where you’d have access to look. But you can still look up numbers on degrees, race/ethnic ratios and so on (of course they are almost entirely white and asian). Still without any intrusion on anyone’s privacy someone like our blog author could perhaps even ask a colleague.

        They only confer like 30 total degrees a year in the CS field over the past decade, so what happened is they went from something like 4 female students to 13. Check the numbers yourself, but their total contribution to benefitting women, minorities, anyone is tiny and irrelevant.

        Journalists are just terrible. It’s not clear what other administrative, public relations, donor/money people on any side of things are dishonest as opposed to just inattentive and sloppy, of course.

        “Weathy private university now has like 10 more female students” isn’t the headline anyone wants. “200 percent increase in high achieving women in formerly privileged male STEM discipline” is the sort of thing that sounds a lot better.

        If a school that serves the broad public like UCLA made strides to graduate another hundred Computer Science majors each year, of any underrepresented group, that would be great. (It would be great in any corresponding Sci/Math/Eng field.) Why, if Harvard did the same it would still be something, their Computer Science department probably isn’t that big in the first place either.

        The bottom line is Harvey Mudd is a rich private school that simply poaches students from other universities, wealthy students who might have been Ivy League bound anyway. There’s no reason to thinkHarvey Mudd has really increased outreach to anyone. Their total student enrollment and number of degrees hasn’t meaningfully changed, definitely not in CS, highly qualified students who don’t go to Mudd simply end up going somewhere else, and they get publicity out of this sort of thing.

        • 6. Mark Guzdial  |  February 28, 2014 at 9:25 am

          I totally agree that Harvey Mudd is highly selective, and the numbers involved are small. That’s who they are.

          But they did get change (large percentage, small in number) in female participation in CS by working at it. They got funding to send women to Grace Hopper. The point for me is other schools can do this! They don’t have to be rich, or have a charismatic leader like Maria Klawe (who is probably really good at raising funds). The amount of money spent on spare football uniforms in a year is plenty to send bunches of women from most state universities to Grace Hopper. It’s possible, if they made increasing the number of women in computing a priority.

          The point of the Harvey Mudd story isn’t the number of women. It’s that the College decided to take action and were successful. That’s why I find the Mercury News headline faulty. It can’t just be about getting CS into schools earlier. Universities can change these numbers if they make it a priority.

          • 7. Michael S. Kirkpatrick  |  February 28, 2014 at 10:55 am

            While I share some of Tomas’s skepticism about the Harvey Mudd story, I don’t interpret it quite as cynically. Yes, Mudd is doing some great things that we should all emulate to increase female representation. At the same time, yes, some of Mudd’s increased numbers may be the result of a zero-sum change in admissions policies. (I hate to label it as affirmative action-like quotas, because I do not mean to be dismissive of it in any way. Given Mudd’s stature and exclusivity, I have no doubt that every incoming student–and many turned away–are perfectly qualified. But I can’t shake the feeling that changes in secretive admissions policies played a part.)

            My skepticism is about extrapolating too much from the Mudd story. Mudd’s success derives from several sources that other institutions would struggle to replicate. Changes in curriculum, a high profile, inspiring female president–who happens to be a computer scientist to boot!–of the college, high-level institutional support, spectacularly exclusive admissions, national media attention…it is the confluence of all of these factors that drive Mudd’s success. What if your institution’s president is a stodgy old male with stellar academic credentials in, say, biology or history? Does the CS department there have any chance of replicating the Mudd story? Doubtful.

            You bring up the idea of institutional funding to take female students to Grace Hopper. That’s easy to do at Mudd, because the president and the upper administration have bought into the mission. Without such support, the case is harder to make. Why should CS be different? Why shouldn’t the college fund ALL students to go to professional conferences? Sure, we are an area of national need and we have a gender problem. But so do some other fields, like nursing.

            As for admissions demographics, Mudd pulls from a pool made up of the best and brightest across the country. Most schools can’t compete. In fact, most schools can’t even pull from the best and brightest in their state. (It’s the old Lake Wobegon problem…not all children are above average…) When all the top female students who are interested in CS go to Mudd, Berkeley, and Stanford, other departments are left with the remaining pool. And considering the high school experience plays a huge role in students’ major interests, it means we must work to overcome factors over which we have no control.

            Again, I am not trying to say there are things we can do better. We certainly can, and Mudd is a great model for a lot of that. But we must take the Mudd story with a rather large heaping of salt, because their model just doesn’t scale currently. Without sweeping cultural changes, drastically altered administration priorities, fixing the K-12 computing education problems, and more visible female leaders, anything that CS departments can do will have a small effect. Unfortunately, while I see plenty of departments trying to do their part, I just do not see adequate progress in the former areas.

            • 8. Mark Guzdial  |  February 28, 2014 at 8:47 pm

              Nothing scales to all higher-ed institutions immediately. Instead, we want a near-peer of Harvey Mudd (maybe a Rose Hulman, or a University of Michigan) to say, “We could do that!” Or even, “I think we can be the first state university to do that!” More press for Harvey Mudd makes the potential reward greater for the next institution to try it.

              Fixing K-12 is far harder than finding an ambitious higher-education administrator.

  • 9. Peter Donaldson  |  March 1, 2014 at 3:29 am

    As a perspective from a complete outsider, wouldn’t studying the differences in approach, content and student attitudes between UC Berkeley’s Beauty and Joy course and other CS courses be more helpful? This year the course is just over 50% female and it’s a much larger sample size.

    • 10. Mark Guzdial  |  March 1, 2014 at 9:29 am

      I have a blog post coming out about the Berkeley course post-SIGCSE because of the great article that came out about it recently. There’s no evidence that I can see that it’s a causal relationship. Do students coming into the course know that it’s a great course, and that’s why they’re taking it? Or are they interested in more CS, and then discover it’s a great course? All the evidence I know about student course choice and the way that information flows from undergraduate-to-high-school (in brief: it doesn’t) suggests the latter, not the former.

      • 11. Peter Donaldson  |  March 2, 2014 at 4:48 pm

        Hi Mark,

        if there was no causal relationship because of a lack of information flow to students before they started the course then I’d expect cs0 courses in other universities to roughly have a 50/50 split in them. Is this the case?

        If it isn’t then it’s likely that there is some kind of interaction going on that’s worth investigating further. It would also be interesting to see the retention rate after BjC.

        With such a small number of students I’m not sure how much Harvey Mudd can tell us about cs for all compared to programmes with a much larger intake.

        • 12. Mark Guzdial  |  March 2, 2014 at 5:18 pm

          Some do. MediaComp at Georgia Tech has been over 50% female for most of the last 10 years. The folks at UCSD have shown a dramatic increase in retention into the Sophomore year after adopting peer instruction, pair programming, and MediaComp in 2007. There’s been an increase in women being retained in the class. I wouldn’t be surprised if they are now seeing more women in the class. There are lots of variables that influence going into a class, and one of them is time and factors in the general society.

          Folks at Berkeley could go a long way towards supporting the claim of causality with some interviews (even surveys) at the start of the course about why students are taking the course. Sometimes the answers aren’t what you want. When we first adopted Threads, we had a boom in enrollment, and the administration claimed that it was because of Threads. Then Maureen Biggers surveyed the new students — only 25% of the students even knew what Threads were, and few students claimed that it was why they chose CS.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Trackback this post  |  Subscribe to the comments via RSS Feed

Recent Posts

February 2014
« Jan   Mar »


Blog Stats

  • 1,528,060 hits

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 5,291 other followers

CS Teaching Tips

%d bloggers like this: