The Missing Pipeline: Why the Technology Skills Gap Is an Education Problem First

There are roughly 4.8 million unfilled cybersecurity positions globally right now. That figure, from ISC2's 2024 Cybersecurity Workforce Study, tends to get discussed as a hiring problem — not enough qualified people entering the field, not enough salaries to retain the ones who do, not enough organizations with the budget to address it. Those are real constraints. But they describe the middle of the pipeline, not the beginning.

The beginning is a classroom. Ideally in middle school, ideally taught by someone who has actually worked in technology, ideally in a school where the subject is treated as fundamental rather than elective. And that classroom — for a substantial portion of students in underserved communities across the country — either doesn't exist or isn't functioning as anything more than an introduction to office software.

The skills gap isn't primarily a recruitment problem. It's the downstream result of an education system that has treated technology as a subject for some students rather than a foundation for all of them, and that concentrated even the limited instruction it had in the places least likely to produce a diverse professional workforce.

According to the 2024 State of Computer Science Education report — published by Code.org, the Computer Science Teachers Association, and the Expanding Computing Education Pathways Alliance — 60 percent of U.S. public high schools now offer foundational computer science. That's real growth from where the number was eight years ago, and state-level policy investment has driven it.

But 40 percent of U.S. public high schools still offer nothing. And that 40 percent is not distributed evenly. Rural schools, smaller schools, and urban schools serving low-income communities are consistently underrepresented in the schools that do offer CS. Black, Hispanic/Latino, and Native American students are, as a group, less likely than their peers to attend a school with foundational CS access — which means the gap tracks race and income as reliably as it tracks geography. Women make up only 33 percent of students enrolled in CS courses nationally, a figure that has held essentially flat for years despite considerable attention to it.

This is before accounting for quality. A course on the roster is not the same as a course that works.

A University of Maryland study cited in the Code.org report found that a single computer science course in high school increases earnings by at least 8 percent by age 24. The benefit was most pronounced for low-income students, Black students, and female students. The course that would most change a student's economic trajectory is the course they're least likely to be sitting in.

The teacher problem underneath the access problem is structural.

Computer science graduates who can teach the subject meaningfully — who can connect algorithms to real decisions, who can show students what a career in security or systems administration actually looks like for someone from their neighborhood — can earn two to four times a teaching salary in industry. Most of them don't go into classrooms. Those who do tend to go to schools with the resources to compete for experienced instructors, which is not, generally, where the access gaps are most severe.

Research published in the Journal of Computer Science Integration found that working in underserved communities was negatively associated with prospective teachers' interest in teaching CS — a finding shaped by salary gaps, institutional support deficits, and the compounding difficulty of the work itself. The communities where CS education would have the largest economic impact are the communities where finding someone to teach it is hardest.

Some systems have tried to address this by retraining teachers from adjacent subjects. A history or math teacher who completes summer CS training can cover introductory content, but teaching CS well also requires being able to make it personally meaningful — to help a student see how networking or security or data analysis connects to something real in their life and in their professional future. That's not a criticism of good teachers who cross-train. It's an observation about what the subject actually demands, and about how far a short-term workaround gets you when the structural problem is this deep.

Chicago offers a useful counterpoint. Out-of-school CS programs supported by the CME Group Foundation found that 90 percent of participating students — drawn from underserved areas of the city — increased proficiency in computer science or STEM. Seventy percent reported interest in STEM careers. These programs reached students who had limited or no access to CS in their regular schools. The gains were real and they came from structured engagement with material that had been designed to connect to students' actual context, delivered by people with relevant background.

The results weren't magic. They reflected what happens when you build instruction intentionally rather than hoping it emerges from a device distribution.

The pandemic made this concrete in a way that was hard to misread. Chromebooks and hotspots reached millions of students who hadn't previously had them. The infrastructure gap narrowed meaningfully in a short period. And then the research arrived.

Digital Promise's 2024 findings showed that access to digital tools alone does not bridge the opportunity gap. Students from historically excluded communities were more likely to use assistive and passive features — text-to-speech, basic navigation — while higher-achieving peers used more active and productive digital tools. The tools were in both students' hands. What they knew to do with them was not equivalent. Pierce and Cleary's 2024 study in PLOS ONE reached a related conclusion at broader scale: without skilled instruction in using technology purposefully, the societal benefits of digital integration cannot be realized. Hardware is necessary and not sufficient.

This distinction matters practically for any organization trying to address the issue. A laptop in a student's home does not teach them how networks function, what distinguishes a phishing attempt from a legitimate request, how to reason about data security, or what the entry points into a technology career actually look like. That requires a teacher with relevant knowledge, curriculum with genuine depth, and an institutional context that treats the subject seriously.

The ISC2 data adds another layer worth holding alongside the access numbers: 90 percent of organizations report skills shortages, and 64 percent say skills gaps present a greater challenge than headcount shortages alone. The problem isn't simply that there aren't enough bodies in seats. It's that the people coming into the field don't have the preparation to do the work that needs doing. Skills gaps are a training and pipeline problem, not just a hiring problem.

The Stanford Center for Racial Justice found in 2024 that underserved populations face heightened risk of falling behind on emerging technologies, including AI. That risk isn't abstract. It describes the practical position of students who enter a technology-saturated labor market without the foundational preparation to understand what the systems around them are doing, let alone shape or secure them. They become consumers of infrastructure they can't interrogate. They fill roles below the level where the economic return on technology skills is most significant.

The NIST NICE Workforce Framework for Cybersecurity identifies dozens of distinct work roles across the technology and security fields — many of which do not require a four-year computer science degree and are genuinely accessible to students who built foundational literacy in high school or community college. That accessibility is real. But only if the foundation was built. For students whose schools spent a decade treating technology education as an add-on rather than a core subject, the pathways that look open on paper are practically much narrower.

The conversation about fixing this tends to arrive at funding, policy, and awareness — and those needs are genuine. More than $88 million was allocated to CS education in state budgets in 2024, a record. Eleven states now require CS credit for graduation. The trend is in the right direction.

But funding and policy exist within systems that have historically made consistent choices about whose education constitutes an investment. The U.S. Department of Education's 2024 National Educational Technology Plan named three divides: access, design, and use. Access to devices and connectivity has received most of the attention and resources. The design and use divides — meaning whether the technology is deployed in educationally meaningful ways, and whether students develop genuine competency rather than surface familiarity — have been much harder to close, because closing them requires skilled instruction, which requires teachers, which loops back to the structural problem at the center.

What operationally effective technology education looks like in underserved communities is not fundamentally mysterious. It means curriculum that connects to reachable careers rather than generic computing concepts. It means assessments that measure actual competency. It means instructors who understand both the technical content and the communities they're teaching in. It means institutions — school systems, workforce programs, community organizations — that treat technology education as core to student outcomes rather than responsive to grant cycles.

Most of that is more expensive and more complicated than another device purchase or another awareness campaign. That discomfort about the cost and complexity is part of why we keep having the same conversation about the skills gap without getting further upstream.

RELATED ARTICLES