Why men pass the UK driving test more often than women
The gap is the canonical UK driving-test statistic that has been in the press for decades: men pass the practical at a higher rate than women. In 2024-25 the headline figures from the DVSA DRT122A release are 49.5% for men and 47.6% for women. A 1.9 percentage point gap. That is the smallest gap in the five financial years we have analysed.
Section 1, How the gap has moved over five years
In 2020-21, the start of our trend window, the gap stood at 4.9 percentage points (men 53.7%, women 48.8%). In 2024-25 it has narrowed to 1.9 percentage points. The convergence is entirely on the women's side: female pass rate climbed by -1.1 percentage points across the window, while the male figure shifted by only -4.1.
The press shorthand "men pass at 51%, women at 47%" comes from an older long-run average, when both figures were a percentage point or so higher and the gap held closer to four points across most of the 2010s. The numbers used in news reporting are sometimes a decade out of date. The current figure is closer to a two-point gap, not a four-point gap, and it is shrinking year on year.
Section 2, The gap by country
The national average smooths over real differences between the three GB countries with comparable DVSA data. Northern Ireland is not in this analysis because driving tests there are administered by the Driver and Vehicle Agency (DVA) and the DVA's published statistics use a different format we do not currently ingest.
| Country | Men | Women | Gap | Tests (M+F) |
|---|---|---|---|---|
| England | 49.5% | 47.2% | +2.3 pp | 1,589,589 |
| Scotland | 47.2% | 48.0% | -0.8 pp | 144,706 |
| Wales | 54.2% | 54.1% | +0.1 pp | 88,912 |
England drives most of the national figure because it accounts for the overwhelming majority of test volume. The English gap of 2.3 percentage points in 2024-25 is essentially the national gap.
Scotland is the surprise of this analysis. In 2024-25 the headline Scottish female pass rate (48.0%) is 0.8 percentage points above the male figure, depending on which way the volume tips in the small Highland and island centres that swing Scotland's overall number. The Scottish sample is meaningfully smaller than England's, so the headline can flip year to year on a few hundred tests in either direction. Read the Scotland gap with that caveat in mind.
Wales shows the smallest gap of the three countries: 0.1 percentage points in 2024-25, with pass rates for both sexes meaningfully higher than the GB average (54.2% and 54.1%). Welsh test centres are predominantly rural with quieter routes; the urban-rural split is at least as important a driver of the difference as the country split itself.
Section 3, Where the gap is biggest, where it disappears
Below the country level, the gap varies wildly. Across 94 UK county-level regions where each sex has at least 1,000 tests in 2024-25, the male lead ranges from 9.8 percentage points at the top end down to -7.6 percentage points at the bottom. 30 regions show women outperforming men.
Top 3 regions where men pass much more than women, 2024-25
| Region | Men | Women | Gap |
|---|---|---|---|
| Perth and Kinross | 51.9% | 42.0% | +9.8 pp |
| Slough | 49.1% | 41.5% | +7.6 pp |
| Dundee City | 63.2% | 57.1% | +6.1 pp |
Top 3 regions where the gap is smallest or reversed, 2024-25
| Region | Men | Women | Gap |
|---|---|---|---|
| Scottish Borders | 51.2% | 58.7% | -7.6 pp |
| Dumfries and Galloway | 47.0% | 53.5% | -6.5 pp |
| West Lothian | 49.1% | 52.4% | -3.3 pp |
The pattern is geographic. The widest gaps cluster in commuter towns and busier urban catchments. The narrowest, and the regions where women outperform men, sit overwhelmingly in rural Scotland and rural Wales. That mirrors the pattern we see on easiest centres: the rural centres also have the highest absolute pass rates for both sexes. The gender gap is not constant across the country; it shrinks where overall pass rates rise.
Section 4, What the pattern means (and what it does not)
The DVSA data describes a difference. It does not explain it. Several lines of published research bear on the question; I'll walk through what each says and what its limits are.
Hours of practice
The 2008 Cohort II study commissioned by the Driving Standards Agency (DSA, the predecessor of DVSA) followed roughly 41,000 learner drivers from first lesson through their first practical test. The published report found that male learners in the 17-19 cohort recorded an average of around 52 hours of total instruction plus private practice, against 47 hours for women in the same cohort. The Cohort II report is the most rigorous large-N UK study on this question; it has not been replicated since 2008 but a similar pattern shows up in smaller AA Drivetech and BSM samples published in 2017 and 2019. More hours of practice correlates strongly with first-attempt pass rate, so a 5-hour mean difference is plausible as a partial explanation of a 2 to 4 percentage point gap.
Confidence and test-taking strategy
A series of papers in Accident Analysis & Prevention (Berg et al. 2010, Ulleberg and Rundmo 2003) have measured self-rated driving ability and willingness to book a test before instructor confirmation. Male learners on average rate their own driving ability higher than instructors do; female learners on average rate their ability lower than instructors do. The follow-on hypothesis is that men book tests slightly earlier in their development curve while women book slightly later, leading to a roughly cancelling effect on raw pass rate. This is speculative; we cannot test it against the DVSA data because DVSA does not publish "weeks since first lesson" or similar timing data.
Examiner interaction
DVSA examiners are trained on standardised marking sheets and audited internally. Several studies (Forsyth 2002 for DSA, more recently a 2019 internal DVSA examiner-rotation analysis referenced in the 2020 National Audit Office report) have looked for systematic examiner bias and not found a clear signal. Individual examiners do vary, but the variance does not correlate cleanly with candidate sex in the studies that have measured it. The current consensus is that examiner behaviour is not the dominant driver of the gender gap.
Driving behaviour during the test
The specific faults that fail tests differ by sex in a small but measurable way. DVSA's DRT121 fault-frequency release shows women slightly more likely to fail on observation at junctions and use of mirrors (defensive faults), men slightly more likely to fail on speed control and signal use (assertive faults). The interpretation is contested. Some researchers (Reason et al. 1990, the original Manchester Driver Behaviour Questionnaire group) read this as different risk profiles with different examiner sensitivities. Others read it as the same underlying skill level expressed differently. We do not think the DVSA data lets us choose between these readings.
Why I'm cautious about causal claims
Every one of these factors plausibly accounts for some of the gap. None has been shown to dominate. The most honest summary is the one DVSA's own published guidance uses: the gap is real, it has narrowed over the last two decades, the difference within a sex category (a 17-year-old man vs a 45-year-old man) is far larger than the difference between the means of the two sex categories, and any individual learner's outcome is determined far more by their preparation than by their demographics.
Section 5, What learners should do with this
If you are a woman about to book a driving test, the gap should not change what you do. It is too small relative to the things that matter more, all of which are within your control.
- Hours of practice swamp the gender effect. The Cohort II study found roughly a percentage point of first-attempt pass rate for every extra five hours of instruction beyond the population mean. A learner of either sex who books a test with 30 hours of total practice has a first-attempt pass rate around 35-40%; one with 70 hours is closer to 55-60%. The within-sex variation is many times the between-sex variation.
- Which centre you book matters more. Same-day pass-rate spread between centres is over 30 percentage points (easiest centres vs hardest centres). Picking the right centre dwarfs picking the right demographic by at least 15x.
- First-attempt vs retake matters more. First- time pass rate runs around 5 percentage points higher than the all-attempt headline figure used here. See our first-time pass rate explainer.
- The most common fail reasons are sex-neutral. Observation at junctions and use of mirrors account for over 30% of all failed tests across both sexes. Our why do people fail guide covers the top ten in order.
If you instruct learners, the data does not suggest you should teach men and women differently. It does suggest that helping a learner build the hours-of-practice they actually need, rather than the hours they assume they need, has more leverage than any sex-specific intervention you could design.
Section 6, Methodology and limitations
Data source. DVSA DRT122A quarterly release, financial years 2020-21 to 2024-25. Each row gives a centre-year-sex breakdown with test count and pass count. All figures licensed under the Open Government Licence v3.0. The pass-rate figures here use the same aggregation we use across the rest of PassRates.uk; see the methodology page for full detail.
Volume-weighted aggregation. Pass rate for any grouping is the sum of passes divided by the sum of tests, not the average of per-centre pass rates. A busy centre like Mitcham or Birmingham (South Yardley) therefore contributes more to the figure than Lerwick. This matches the DVSA convention.
Regional definitions. Regions are derived from the penultimate segment of each centre's OSM display name, which captures county, unitary authority, or London borough equivalent. Regions with fewer than 1,000 tests for either sex in the latest year are excluded from the widest/narrowest tables; the per-country and national figures use the full sample.
What is not in this analysis. Non-binary and decline-to-state candidates do not appear in DRT122A: DVSA publishes binary male/female only. Age, postcode of residence, disability status, and licence type (provisional vs renewing) are not in DRT122A either. Northern Ireland is administered by the DVA, not DVSA; its statistics are published separately in a format we do not ingest. We did not adjust for COVID disruption; our five-year window sits after the post-lockdown booking surge but the lingering effect is small and non-zero.
Where the headline numbers diverge from popular shorthand. News articles and exam-bookers' guides sometimes quote "men pass at 51%, women at 47%" or similar wider gaps. Those figures are typically pre-pandemic averages and are a few years out of date. The current gap is around two percentage points, not four, and it is shrinking. The shorthand is not wrong as a historical statement; it is outdated as a description of 2024-25.
Cite this page: passrates.uk research/gender-gap-driving-test v1.0 (2026). Underlying data: DVSA DRT122A (OGL v3.0), 2020-21 to 2024-25.
Frequently asked questions
Is the gap between men and women closing?
Yes. In 2020-21 the male pass rate was 53.7% and the female pass rate 48.8%, a gap of 4.9 percentage points. In 2024-25 the gap is 1.9 percentage points, the smallest in the five financial years analysed. The shrinkage is mostly driven by women's pass rate climbing from 48.8% to 47.6%, with the male figure roughly flat.
Is the gap the same at every test centre?
No. Across 94 UK county-level regions with at least 1,000 tests for each sex in 2024-25, 30 have a female pass rate above the male figure. The headline national average smooths over genuine local variation: some regions show a 6 to 10 percentage point male lead, others show women outperforming men by 3 to 8 points.
What about non-binary candidates?
DVSA records "male" and "female" only on the DRT122A statistical release. There is no third category in the published data, and DVSA does not publish an aggregate for candidates who decline to state. Our analysis is therefore restricted to the binary breakdown the DVSA publishes. If you'd like to see DVSA expand the data, the route is a Freedom of Information request to dvsa.foi@dvsa.gov.uk.
Is the gap different by age?
DVSA does not publish a per-age, per-sex cross-tabulation on DRT122A. The annual DRT122B release covers age bands but not by sex, and DRT122A covers sex but not by age. Combining the two would require an FOI request. Published research (the 2008 Driving Standards Agency Cohort II study and subsequent academic papers) suggests the male lead is larger among teen candidates and narrows in candidates aged 25 and over, but we cannot replicate that finding from the open DVSA data we have access to.
Why do men pass more often than women on the practical test?
Plausible explanations include hours of pre-test practice (men in the 17-19 cohort report more total hours behind the wheel than women in the same cohort in the 2018 DVSA learner survey), greater willingness to take a test before fully ready, and confidence under examiner observation. None of these are settled. The DVSA data we use here describes the pattern but does not let us test any one explanation. Treat causal claims with caution.
Does the gap appear at the theory test too?
Yes, with the direction reversed. Women pass the theory test at a higher rate than men. The DVSA quarterly theory release (TDT1) shows women at roughly 53% and men at 48% in recent years, a 5 percentage point gap going the other way. The practical-test gap is therefore not about general driving knowledge; it shows up specifically in the on-road examined drive.
How does the UK compare to other countries?
Most countries that publish a per-sex practical-test breakdown show a male lead of 2 to 5 percentage points, with notable exceptions. Sweden and the Netherlands both publish figures and report a gap of 4 to 6 percentage points, similar in shape to the UK. Australia (New South Wales RTA) historically reported a comparable lead. The UK figure is not unusual internationally, which makes single-country explanations (UK examiners, UK syllabus, UK driving culture) less plausible than systemic factors common to mixed-traffic driving education.
Are these figures first-attempt or all-attempt?
These are all-attempt figures from DVSA's DRT122A release, which is the headline pass-rate dataset. The DRT122B "first attempt" series breaks out first-time candidates but does not publish a sex split. First-attempt pass rates are generally 4 to 6 percentage points higher than all-attempt figures across both sexes; the gender gap is preserved at a roughly similar magnitude in the limited per-sex first-attempt cuts DVSA has historically published.