Nearly 40% of software developers report a mental health condition or neurodivergence — and it's a workforce characteristic to understand, not a problem to solve.
The tech industry talks constantly about inclusion. The data tells a more specific story. The Stack Overflow Developer Survey — the largest annual survey of software developers with 73,268 respondents — provides the clearest picture. The 2022 edition is the most recent to include full demographic questions (gender, ethnicity, accessibility, mental health); subsequent years dropped these fields. The data is from 2022 but represents the most comprehensive developer demographic snapshot available at this scale, showing who actually builds software, what barriers persist, and where the workforce is changing.
This guide presents the demographic reality of the software development workforce, examines where representation gaps exist, and connects inclusion to the practical hiring and team-building decisions that engineering leaders make daily. The numbers are specific, sourced, and in several cases uncomfortable.
Most inclusion discussions start with aspirations. This one starts with data. The Stack Overflow 2022 Developer Survey collected demographic information from 73,268 developers across 180+ countries. Here's what the workforce actually looks like.
The gender split in software development remains stark:
| Gender | Percentage | Respondents |
|---|---|---|
| Man | 90.9% | 65,097 |
| Woman | 5.1% | 3,662 |
| Non-binary / genderqueer | 1.7% | 1,186 |
| Prefer not to say | 1.6% | 1,172 |
Software development remains overwhelmingly male at 91%. Women represent just 5.1% of survey respondents. The ratio has improved slowly over the past decade — Stack Overflow's earlier surveys showed women at 3-4% — but the pace of change is incremental, not transformational.
The gender pay gap exists but is smaller than many industries. Male developers earn a global median of $67,845 versus $63,985 for women — a 5.7% gap. Non-binary developers earn a $75,400 median, likely reflecting concentration in higher-paying Western markets and senior roles.
The global developer workforce reflects the industry's geographic spread:
| Ethnicity | Percentage |
|---|---|
| White | 28.8% |
| European | 27.2% |
| Indian | 7.1% |
| Asian | 6.9% |
| Hispanic or Latino/a | 4.2% |
| Middle Eastern | 3.0% |
| South American | 2.8% |
The global developer workforce is ethnically diverse by geography — developers come from everywhere, but within individual companies and countries, representation gaps persist. The 56% White/European share in global survey data reflects the concentration of English-speaking survey respondents in Western markets.
Software development is a young profession:
| Age Group | Percentage |
|---|---|
| 18–24 | 22.7% |
| 25–34 | 38.4% |
| 35–44 | 19.1% |
| 45–54 | 7.2% |
| 55–64 | 2.7% |
| 65+ | 0.8% |
Software development skews young. Over 61% of developers are under 35. Only 11% are over 45. The age distribution reflects the industry's rapid growth (younger cohorts entered a larger job market) but also suggests potential age-related barriers for older workers entering or remaining in the field.
The mental health and neurodivergence rates among developers are higher than general population baselines:
| Condition | Percentage |
|---|---|
| None reported | 61.6% |
| ADHD or concentration disorder | 9.2% |
| Anxiety disorder | 9.0% |
| Mood or emotional disorder | 8.5% |
| Autism spectrum | 3.7% |
Nearly 40% of developers report some form of mental health condition or neurodivergence. The 9.2% ADHD rate and 3.7% autism spectrum rate are significantly higher than general population estimates (typically 5% and 1-2% respectively), suggesting that neurodivergent individuals may be drawn to software development's structured, logic-driven work, or that the profession's demands create conditions that surface these diagnoses.
That's not a problem to solve. It's a workforce characteristic to understand and accommodate. Teams that treat neurodiversity as an asset — providing flexible work arrangements, clear communication protocols, sensory-appropriate environments, and structured feedback cadences — tap into a significant talent pool that rigid workplace norms exclude.
The overall gender distribution underscores how far the industry still has to go:
The data reveals a more nuanced picture than "tech isn't diverse." Some dimensions show persistent gaps. Others show more progress than headlines suggest.
Women make up 8.7% of developers with less than one year of professional experience. That entry rate, while still low, is nearly double the 5.1% representation in the overall workforce, suggesting the pipeline is slowly widening at the entry level, but retention and advancement remain challenges. Women's share decreases at senior experience levels.
The gender gap varies dramatically by role and geography. Our analysis of 4,145 software development companies across 83 countries shows that the workforce spans every major region, with South Asia (34.5%) and North America (35.1%) providing the largest developer populations. Different regions have different gender dynamics in tech — Scandinavian countries, for example, have higher female participation rates than South Asian markets.
Here's an inclusion story that rarely gets told: the global software development market is already one of the most geographically diverse industries in the world.
Our company data shows firms operating across 83 countries spanning 10 major regions:
| Region | Firms | Market Share |
|---|---|---|
| North America | 1,454 | 35.1% |
| South Asia | 1,429 | 34.5% |
| Eastern Europe | 395 | 9.5% |
| Western Europe | 282 | 6.8% |
| Middle East | 175 | 4.2% |
| Southeast Asia | 82 | 2.0% |
| Oceania | 78 | 1.9% |
| Africa | 30 | 0.7% |
| South America | 20 | 0.5% |
When organizations outsource or build distributed teams, they're creating cross-cultural, cross-national teams by default. The 93.4% of firms in our dataset that operate across multiple locations aren't just managing logistics. They're practicing inclusion at the geographic level whether they frame it that way or not.
Only 6.7% of developers report any form of disability or accessibility need, but this likely understates the reality given stigma around disclosure. The 1.7% reporting visual impairment and 0.8% reporting hearing impairment suggests a workforce where physical accessibility is partially addressed by the nature of the work (screen-based, increasingly voice-enabled), but where workplace accommodations and inclusive design practices still lag.
Inclusion in tech isn't a mission statement. It's a set of operational decisions that engineering leaders make about hiring, team structure, and workplace norms.
Credential requirements are the most common structural barrier to inclusion. The SO survey data makes the scale of this barrier concrete:
| Education Level | Percentage of Developers |
|---|---|
| Bachelor's degree | 42.3% |
| Master's degree | 21.6% |
| Some college, no degree | 13.0% |
| Secondary school only | 11.0% |
| Associate degree | 3.1% |
| Doctoral degree | 3.0% |
Nearly 24% of working developers don't have a bachelor's degree. They entered the profession through self-teaching, bootcamps, community college, or direct experience. When job descriptions require a bachelor's degree as a baseline, they exclude almost a quarter of the existing developer workforce — people who are already doing the work successfully.
The shift to skills-based hiring — evaluating what candidates can do rather than where they studied — is one of the most effective inclusion interventions available to engineering leaders. It costs nothing to implement and immediately widens the candidate pool to include the 24% already proving that degrees aren't prerequisites for developer competence.
Remote and hybrid work arrangements are inclusion infrastructure. The Stack Overflow data shows 80% of developers work at least partially remote. For developers with disabilities, caregiving responsibilities, or those located outside major tech hubs, remote options aren't a perk. Teams managing remote development across locations are inherently more accessible than office-only teams. They're the difference between participating in the industry or being excluded from it.
Organizations that insist on in-person-only work are making an implicit decision about who can work for them. That decision disproportionately excludes people with mobility challenges, developers in regions without tech hub access, and caregivers (predominantly women) who can't commit to rigid office schedules.
With 9.2% of developers reporting ADHD and 3.7% on the autism spectrum, the communication norms that teams adopt have inclusion implications whether leaders recognize it or not.
Written async communication (Slack threads, documented decisions, recorded meetings) accommodates developers who process information differently than neurotypical norms assume. Structured sprint ceremonies with clear agendas serve neurodivergent team members better than ad-hoc discussions. These aren't special accommodations. They're good engineering practices that happen to be inclusive.
McKinsey's "Diversity Matters Even More" report (2023) found that companies in the top quartile for ethnic diversity were 39% more likely to outperform financially. The finding has been challenged on methodological grounds by researchers who argue the correlation may reflect reverse causation (successful companies diversify their leadership, not the other way around).
The honest answer: the financial case for inclusion is directionally positive but not as clean as consultancy reports suggest. What is clear from the data is that the software development talent pool is global, multilingual, and neurodiverse. Organizations that can only hire from one demographic, one geography, or one educational background are fishing in a smaller pond. The custom software development market spans 83 countries precisely because talent is global. Inclusion, framed as access to the full talent market rather than as a moral obligation, is a competitive hiring advantage.
5.1% of Stack Overflow 2022 survey respondents identified as women (n=3,662 out of 73,268). At the entry level (less than 1 year experience), women represent 8.7%, suggesting the pipeline is slowly widening. The gender pay gap exists but is relatively narrow at 5.7% ($67,845 median for men vs $63,985 for women globally).
Considerably more than the general population. 9.2% of developers report ADHD (vs ~5% general population), 3.7% report autism spectrum conditions (vs ~1-2% general), and 9.0% report anxiety disorders. Nearly 40% report some form of mental health condition or neurodivergence.
Yes. The global software development market spans 83 countries across every major region. When companies build distributed teams through outsourcing or staff augmentation, they're creating cross-cultural, cross-national teams. This geographic diversity brings different perspectives, problem-solving approaches, and market understanding.
Skills-based hiring. Removing degree requirements and evaluating candidates on demonstrated capability rather than credentials immediately widens the talent pool to include career changers, bootcamp graduates, self-taught developers, and professionals transitioning from adjacent fields. It costs nothing to implement.
Directionally, yes. McKinsey's research shows correlation between diverse leadership and financial performance, though causation is debated. The practical case is clearer: the developer talent pool is global and diverse. Organizations that can only hire from narrow demographics are competing for a smaller share of available talent.
[1] Stack Overflow Developer Survey 2022 — 73,268 respondents. Gender, ethnicity, accessibility, mental health, salary, and experience data. Licensed ODbL v1.0.
[2] McKinsey — Diversity Matters Even More (2023) — Top-quartile ethnic diversity → 39% more likely to outperform.
[3] Econ Journal Watch — McKinsey's Results Revisited (2024) — Methodological critique of McKinsey's diversity-performance correlation.
[4] Internal analysis of 4,145 software development company profiles aggregated from Clutch, TechReviewer, and proprietary scoring datasets (January 2026 snapshot). Geographic distribution based on companies with disclosed country data.