Cohort pass rates, decoded

Every time the SRA publishes a new statistical report, the same headline number gets shared everywhere: one pass rate, for one sitting, as if it describes your odds. It doesn't. It describes a cohort that isn't you, sitting under conditions that aren't yours, with a mix of first-timers and resitters you can't see from the outside. Here's how to actually read the numbers.

Start with the headline, then immediately break it apart

The SRA's report for the January 2026 SQE1 sitting shows an overall pass rate of 53% across all candidates. Split that number and it moves a lot: first-time candidates passed at 58%, while the overall figure is dragged down by a much lower resit pass rate. By component, FLK1 came in at 62% overall (65% first attempt) and FLK2 at 57% overall (61% first attempt).

That gap between "all candidates" and "first-time candidates" is not a rounding difference. It's the single most important adjustment to make before you compare yourself to any published rate. A cohort that includes a large block of resitters will always report a lower blended pass rate than a cohort that doesn't, regardless of how well-prepared any individual candidate is.

SQE2 tells a similar story from a different angle. The SRA's report for the January 2026 sitting put the overall SQE2 pass rate at 80%, with first-time candidates at 79%, barely different from the cohort average. That's not because SQE2 resitters are stronger than SQE1 resitters. It's because the SQE2 population has already been filtered once: you can only sit SQE2 after passing SQE1, so the pool is smaller, further along, and has already cleared one bar. Comparing a raw SQE1 rate to a raw SQE2 rate and concluding SQE2 is "easier" skips over the selection effect that produced the SQE2 cohort in the first place.

Why headline rates mislead

Three things move a published pass rate around that have nothing to do with how hard the assessment was that sitting:

• Cohort composition. The proportion of first-sitters versus resitters shifts from one sitting to the next, and it moves the blended rate mechanically. The SRA's data shows the July 2025 SQE1 sitting had an unusually high proportion of resitters, and that sitting also recorded the lowest overall pass rate in the series, at 41% against 46% for first-timers. Some of that drop is cohort mix, not difficulty.
• Selection into the sitting. Candidates choose when to sit, and that choice isn't random. People who feel underprepared sometimes delay; people on a training contract clock sometimes sit regardless. Each sitting's cohort is self-selected in ways the headline number can't show you.
• Attempt number. A first attempt and a third attempt are not the same event statistically, even for the same candidate. Someone resitting has already had one full diagnostic pass at the assessment and, in principle, a chance to close specific gaps, but resit pass rates are still consistently lower than first-attempt rates across the SRA's published series. That tells you resitting is hard, not that resitters are careless.

None of this means the numbers are meaningless. It means a single topline figure answers a narrower question than it looks like it's answering. It tells you what proportion of this specific, mixed group passed this specific sitting, not what your own chances are.

The attempt-number pattern, across sittings

Look at the SRA's four-years-on analysis rather than a single sitting and the pattern holds consistently: across the full SQE1 candidate population to date, first-attempt pass rates run well ahead of resit pass rates, and the great majority of candidates who eventually pass do so on their first try rather than grinding it out over multiple sittings. SQE2 shows the same shape in sharper relief: the large majority of eventual passers pass first time, and resit performance lags noticeably behind.

The practical reading isn't "you must pass first time or you're in trouble." It's that first-attempt preparation quality carries a lot of weight, and that resitting works best when it's targeted at genuinely diagnosed gaps rather than a general "do more revision" reattempt. A resit that repeats the same preparation that produced the original result tends to produce a similar result.

The attainment-gap data, and why it isn't a verdict on individuals

The SRA has also published research into differential outcomes by candidate characteristics, and it's worth reading rather than skipping past. Its analysis found measurable pass-rate gaps by ethnicity, and by whether English is a candidate's first language, alongside a smaller gap by gender. It also ran multivariate analysis to ask how much of the variation in scores those characteristics actually explain.

The answer is: less than the raw gap suggests, and less than most other factors. The SRA's research found that ethnicity alone explains a single-digit share of the variance in SQE1 scores, with a similarly small share for socio-economic background, and with the SRA's own commentary noting that most of the variance in outcomes isn't explained by the demographic factors it measured at all. That's an uncomfortable pair of facts to hold at once: real, documented attainment gaps exist in the published data, and no demographic category comes close to determining an individual's result. Both are true. The SRA has commissioned further independent research (including with the University of Exeter) specifically because a gap that size, unexplained by the obvious variables, deserves a proper answer rather than a shrug.

What this means for you reading it as an individual candidate: population-level statistics describe populations. They're evidence that the system needs to keep examining itself. They are not a prediction about what you, specifically, will score.

The number that predicts your result isn't in any SRA report. It's whatever you can honestly say about your own topic coverage, practice volume and mock performance right now.

Reading a report like an analyst, not a headline-scanner

A short checklist for the next time a new statistical report lands:

1. Find the first-attempt figure, not just the blended figure, and compare like with like.
2. Check the resit proportion for that sitting before comparing it to a previous one, because a shift in cohort mix alone can move the headline several points.
3. Note the sample size. Smaller sittings produce noisier percentages; a two-point move on a few hundred candidates is a different thing to a two-point move on several thousand.
4. Treat SQE1-to-SQE2 comparisons with the selection effect in mind, because SQE2's cohort has already cleared SQE1, so its higher pass rate isn't purely "SQE2 is easier."
5. Read demographic breakdowns as evidence the system is examining itself, not as odds assigned to you personally.

None of this is a reason to ignore the data. The SRA publishes it precisely so candidates, providers and the SRA itself can hold the system to account. It's a reason to read it at the right altitude, as a description of how a whole system is performing, not as a fortune-telling device pointed at you.

The takeaway

Published pass rates are real, useful, and about something other than you. They tell you how a mixed cohort performed on a given day, shaped by resit proportions, selection effects and attempt number, not how a well-prepared individual candidate will do. The gap between first-attempt and blended rates, and between SQE1 and SQE2 cohorts, is mostly composition, not difficulty swinging wildly between sittings.

The pass rate that actually matters is the one you build for yourself: honest coverage of the syllabus, enough timed practice to know where you're weak, and a realistic read on whether you're ready to sit or better served waiting a sitting. Preparation quality is the variable in your control, and it's also the variable the published statistics can't measure about you. If you want a live, honest version of that number, tracking your own topic-by-topic mastery and sitting full timed mocks tells you far more than any cohort average. Start with mastery tracking and a full mock rather than a headline.

Sources: SRA SQE1 January 2026 Statistical Report; SRA SQE2 January 2026 Statistical Report; SRA, "The SQE four years on, facts and figures"; SRA research on differential outcomes by ethnicity in legal professional assessments, all published at sra.org.uk and sqe.sra.org.uk.