Sign In

Communications of the ACM

[email protected]

Closing the Computing Labor Shortage with Computing in Schools


View as: Print Mobile App Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook
Mark Guzdial

Georgia Institute of Technology professor Mark Guzdial

The Code.org video  argues that there is a large computing labor shortage in the United States. Code.org suggests that there will be 1.4 million jobs for IT professionals over the next ten years, but only 400,000 graduates to fill those jobs. That’s why it’s important to get more computer science (explicitly "coding") into schools, to give students the opportunity to see real computing and discover an affinity for it.

There are experts that question whether we really have a STEM labor shortage in the United States. Even if we do, will improving STEM education actually have any effect? A 2007 report suggests that STEM education has actually improved in the last decades in the United States. Why should improving it more make a difference?  The STEM labor shortage may because of a low yield. Students may be interested in studying STEM, but might not be interested in a STEM career. Improving STEM education may further may have little impact on the economy.

The argument about computing is different than the argument about STEM generally. Most people in computing agree that we really do have a labor shortage — we need more people who know computing. The evidence is stronger that improving computing education in schools will have an impact on computing careers.

First, there is very little computer science in high schools today. There are 2,000 Advanced Placement Computer Science teachers in the United States for some 25,000 high schools. There is science and mathematics education already in schools, and marginal improvements may not have a big impact on careers. For the most of the US, we are talking about providing computing education where there’s nothing.

Second, the research evidence that we have suggests that adding computing education into a high school has a dramatic impact, especially on under-represented minorities. In the study Stuck in the Shallow End, Jane Margolis and her colleagues described minority students in the Los Angeles Unified School District who wanted to study computer science, but simply couldn’t get access to a course. Those are exactly the students that we need in order to fill the computing labor gap. The computing industry today mostly draws from white or Asian males, which are only 30% of the US population. Broadening access will be critical to getting enough computing professionals.

We have recently re-analyzed a large data set (described in last year’s ACM ICER conference) that we collected as part of Georgia Computes. We surveyed 1400+ students from across Georgia in their first year of coursework. We wanted to know if having computing education opportuniteis in high school had more impact on under-represented minority students than on the white or Asian students. Our dataset only included students who were in computing courses, so we don’t have a control group.

Shelly Engelman’s report does show that there are some significant differences in the impact of high school computer science between under-represented minority and majority students.  

  • Statistically significant (p<.01) more under-represented minority students participated in computing activities (e.g., clubs or summer camps) than majority students (44% vs. 27%).  
  • Under-represented minority students were more likely to say that a specific early computing experience (like a computer club, or a visit from an IT professional) that motivated them to study computing in college (21% vs. 16%, p<.05).
  • To be specific about outcomes, the early computing experiences of under-represented minority students significantly positively impacted their satisfaction with computing and the likelihood that they would pursue a computing career.

We don’t know if there were other under-represented minority students who had the same early computing experiences but didn’t go into computing. We do know that having early computing experiences is having a significant impact on the under-represented minorities students who pursue computing, and it plays a bigger role than for the majority students.

The evidence that I see gives us good reason to believe that there is a value in computing education in schools in terms of computing careers. Unlike the rest of STEM, there is little computer science out there. Going from zero to something is way more than a marginal improvement. Getting more access to quality computer science education is likely to turn on more kids to computing, especially among the under-represented minorities that we most need.


 

No entries found