Communications of the ACM

Follow us on Twitter at http://twitter.com/blogCACM The Communications Web site, http://cacm.acm.org, features more than a dozen bloggers in the BLOG@CACM community. In each issue of Communications, we’ll publish selected posts or excerpts.

those states. We don’t have data on who is taking CS across the U.S., due to our state-centric, decentralized model of primary and secondary school education.

California and Texas are among the leaders in implementing CS education. Both are members of the Expanding Computing Education Pathways (ECEP) Alliance ( https://ecepalliance.org/). Reports from ECEP states tell a similar story. For example, in Indiana, the most popular high school CS course enrolled only 2% of students in the state (http:// bit.ly/2oAq6Q6). Miranda Parker is finishing her dissertation on factors influencing CS adoption at the high school level, and she’s found only 1% of high school students in Georgia take a CS course. I suspect states not involved in ECEP have lower participation in CS, not higher. Given these numbers, it would be hard to believe more than 4% of U.S. high school students take a CS course.

One surprise for me in these reports:
many U.S. high schools offer CS. I wrote
a BLOG@CACM post in 2012 (http://bit.
ly/2ox5NDl) estimating about 10% of
high schools in the U.S. offered comput-
er science. Today, that number is much
larger. In Texas, it is 43%. In California, it
is 39%. In Georgia, it is around 40%, and
about 1/3 of high schools in Indiana of-
fer CS classes. The high schools that do
offer CS are most often in larger (wealth-
ier) high schools. Many students could
take those classes. But they do not.

A decade ago, I thought the reason so few students were taking CS was access. Today’s data paints a different story.

Even if CS is in the high school, few students sign up for it. Those who do are mostly male (71% in California, 74% in Texas). Some states are better at getting underrepresented minorities into CS classes. It’s 48% in Texas. In California, only 16% of AP CS A exam takers are members of underrepresented groups.

Some states might decide this is not a problem; 3% may be enough. U.S. universities still reeling from overwhelming numbers of students in CS classes ( http://bit.ly/2oCXoOs) might not want more students to become interested in CS at the high school level. I hear from schools that most introductory CS students they get have already had high school CS. These are compatible numbers. There are 15. 3 million high school students in the U.S.; 3% of that is around 450,000 students—which I expect is more than the number of students in university intro CS in the U.S. If CS participation numbers increased at the high school level, probably more students would want to take CS at the university level (even if not as a CS major). Universities would have to restructure to manage such an enormous load.

Both the U.K. Computing at School

Mark Guzdial
A Minuscule Percentage
of Students Take High
School Computer
Science in
the United States:
Access Isn’t Enough

September 2, 2019 http://bit.ly/2oupbkc

If I told you only 4% of all high school students in the U.S. were taking science or math classes, you’d be aghast. That’s such a tiny percentage! If 96% of students were not getting science or math classes, you could reasonably argue it does not exist in any practical sense.

Over the last few months, several reports provided new insights about U.S. high school computer science (CS):

˲ In California, only 3% of 1. 9 million high school students enrolled in CS in 2017 ( http://bit.ly/2pq5k TH).

˲ In Texas, 3.76% of its students completed a CS course in 2017–2018 ( http://bit.ly/2ox7cK6).

California and Texas are two largest states based on U.S. population, but we can’t generalize to everyone based on

Getting High School,
College Students
Interested in CS

Mark Guzdial considers how few U.S. high school students take computer science, while Robin K. Hill shares what she’s learned in teaching first-year college computing students.