Law & Civil Service — 2026 Guide

Watson Glaser Critical Thinking Test 2026: Complete Guide for Law & Civil Service

All 5 Watson Glaser question types with worked examples, Magic Circle law firm and Civil Service Fast Stream context, scoring explained, and expert preparation strategies.

5Question types with examples

40 minTypical test duration

Magic Circle+ Civil Service employers

2026Fully updated

Section 01

What is the Watson Glaser Critical Thinking Test?

The Watson Glaser Critical Thinking Appraisal (WGCTA) is a standardised psychometric test that measures five specific dimensions of critical thinking: the ability to draw inferences, recognise assumptions, perform deductions, interpret evidence, and evaluate arguments. It was developed by Goodwin Watson and Edward Glaser and is published by Pearson TalentLens.

Watson Glaser is the dominant verbal reasoning assessment in two specific sectors: legal recruitment (used by virtually all Magic Circle, Silver Circle, and US Big Law firms hiring in the UK) and civil service recruitment (UK Civil Service Fast Stream, GCHQ, FCA, Bank of England). It is also used by some investment banks, financial regulators, and consultancies.

The test measures how well you reason with text — not just whether you can extract facts from a passage (like SHL Verbal), but whether you can evaluate the quality of reasoning, identify logical entailments, distinguish assumptions from inferences, and assess argument strength. These are the core skills of legal and policy analysis.

⚖️

Why law firms use Watson Glaser specifically

Solicitors and barristers spend their careers evaluating arguments, identifying logical weaknesses, distinguishing what is proven from what is assumed, and assessing the strength of opposing positions. Watson Glaser tests exactly these skills — which is why it has been the standard screening tool for top law firm recruitment for decades. Performing well requires the same analytical rigour that effective legal reasoning demands.

Section 02

Watson Glaser vs SHL Verbal Reasoning

Candidates who have prepared for SHL Verbal Reasoning often underestimate how different Watson Glaser is. The two tests both use written passages, but they measure fundamentally different cognitive processes.

Feature	SHL Verbal Reasoning	Watson Glaser
Core skill	Comprehension and fact extraction	Critical thinking and argument evaluation
Answer format	True / False / Cannot Say (3 options)	5 different formats depending on question type
Question types	1 type (T/F/CS) applied to all questions	5 distinct types requiring different cognitive modes
Difficulty source	Time pressure + qualifier word traps	Logical precision + distinguishing near-synonymous categories
Preparation approach	Framework application + reading speed	5 separate type-specific frameworks + logical reasoning practice
Primary users	Finance, consulting, engineering, all industries	Law firms, civil service, financial regulators
Duration	~25 minutes / 30 questions	~35–45 minutes / 40 questions

⚠️

SHL Verbal preparation does not adequately prepare you for Watson Glaser

The True/False/Cannot Say framework from SHL does not transfer directly to Watson Glaser — the Inference section has 5 options (True, Probably True, Insufficient Data, Probably False, False), and the other four sections use completely different evaluation frameworks. Candidates who treat Watson Glaser as "SHL Verbal but harder" consistently underperform. Each section requires separate, specific preparation.

Section 03

Test Format & Scoring

Feature	Detail
Total questions	40 questions across 5 sections (8 questions per section)
Duration	35–45 minutes (some versions have no time limit; check your specific format)
Format	Multiple choice — varies by section (2, 3, or 5 options depending on question type)
Scoring	Raw score converted to percentile against a norm group; no penalty for wrong answers
Versions	Watson Glaser II (standard) and Watson Glaser Short Form (24 questions, 30 minutes)
Platform	Pearson TalentLens — online delivery; some firms use paper-based versions in supervised settings

💡

Law firm cut scores are typically higher than corporate employer SHL scores

Magic Circle and US Big Law firms use Watson Glaser to screen very large applicant pools — typically several thousand applicants for under 100 training contract places. Estimated cut scores at the most competitive firms are around the 80th percentile. At Silver Circle and national firms, 70th–75th percentile is more typical. Always aim to maximise.

Section 04

All 5 Question Types with Worked Examples

Section 1: Inference 5-option scale

Inference: A conclusion drawn from observed or assumed facts. Your job is to evaluate how likely the inference is, given only the stated facts. Options: True / Probably True / Insufficient Data / Probably False / False.

"A local council has recorded a 40% increase in cycling journeys over the past three years. The council invested in 15 km of new protected cycle lanes during this period."

Inference: The cycle lane investment caused the increase in cycling journeys.

True

Probably True

Insufficient Data

Probably False

False

✓ Insufficient Data

The passage states two facts: cycling increased 40%, and cycle lanes were built. It does not establish causation — both could be explained by other factors (fuel prices, COVID-era habits, weather, cycling culture shift). Causation requires explicit evidence of a causal link, not just temporal co-occurrence. The correct answer is Insufficient Data — we cannot determine from the facts alone whether the lanes caused the increase.

Section 2: Recognition of Assumptions 2-option: Assumption Made / Not Made

An assumption is something taken for granted or presupposed in a statement, without explicit evidence. Your job is to identify whether a given assumption is necessarily being made for the statement to be true.

Statement: "We should invest in renewable energy now, before fossil fuel prices rise further."

Proposed assumption: Fossil fuel prices will continue to rise.

Assumption Made

Assumption Not Made

✓ Assumption Made

The argument uses "before fossil fuel prices rise further" — the phrase "further" presupposes they will continue rising. The recommendation only makes logical sense if further price rises are expected. This assumption is embedded in the statement and must be accepted for the argument to hold. Compare: "Fossil fuel prices may fluctuate" — this would NOT be an assumption made, as the statement doesn't require fluctuation, only further rises.

Section 3: Deduction 2-option: Follows / Does Not Follow

Deduction: Does the conclusion follow necessarily and with logical certainty from the given premises? Accept the premises as true even if they conflict with real-world knowledge. Only two options: Conclusion Follows / Conclusion Does Not Follow.

Premises: All registered solicitors are members of the Law Society. Some members of the Law Society are also barristers.

Conclusion: Some barristers are registered solicitors.

Conclusion Follows

Conclusion Does Not Follow

✓ Does Not Follow

Premise 1: All solicitors → Law Society members. Premise 2: Some Law Society members are barristers. The "some" in Premise 2 could refer to members who are purely barristers (not solicitors). We cannot conclude that any of the "some barristers" are also solicitors — the overlap could exist entirely among non-solicitor Law Society members. A common error is to assume overlap implies all instances overlap. Strict logical deduction does not permit this.

Section 4: Interpretation 2-option: Conclusion Follows / Does Not Follow

Interpretation: Does the conclusion follow beyond reasonable doubt from the evidence? This is probabilistic (not certain) — more lenient than Deduction but stricter than mere possibility. The evidence must make the conclusion highly probable, not just possible.

Evidence: "In a survey of 800 employees, 74% said they felt more productive when working from home at least two days per week. Only 18% said they preferred working in the office full-time."

Conclusion: The majority of surveyed employees would prefer a hybrid working arrangement to full-time office work.

Conclusion Follows

Does Not Follow

✓ Conclusion Follows

74% reported feeling more productive working from home ≥2 days, and only 18% preferred full-time office. While the survey doesn't directly ask about "hybrid preference," the data makes it highly probable beyond reasonable doubt that the majority prefer hybrid to full-time office. Interpretation allows probabilistic conclusions — this one is strongly supported by both data points pointing in the same direction.

Section 5: Evaluation of Arguments 2-option: Strong / Weak

A Strong argument is directly relevant to the question and important enough to significantly influence the decision. A Weak argument is trivial, irrelevant, emotional, or based on minor considerations — even if factually true.

Question: Should the UK government make voting mandatory for all citizens?

Argument: "No, because compulsory voting would force people to choose candidates they know nothing about, potentially resulting in uniformed votes that distort democratic outcomes."

Strong Argument

Weak Argument

✓ Strong Argument

This argument is directly relevant to the question (it addresses a fundamental consequence of the proposed policy), based on a substantive concern (informed democratic participation), and important enough to significantly influence the decision. Compare with a weak argument like "No, because some people might have to travel far to vote" — this is trivially solved by postal voting and doesn't address a fundamental issue. Strong arguments address the core of the question with material weight.

Section 05

Watson Glaser in Law Firm Recruitment

Watson Glaser is the near-universal aptitude test for Magic Circle, Silver Circle, and leading US-headquartered law firms hiring in the UK. It is typically administered online after the initial application stage, before any interview. Most firms allow candidates to take it remotely within a set window.

The test is used as a primary filter in very high-volume recruitment — top Magic Circle firms receive 5,000–15,000 training contract applications for fewer than 100 places annually. Watson Glaser efficiently screens for the analytical precision and logical rigour that effective legal work demands.

Firm Type	When Administered	Est. Cut Score	Format
Magic Circle (A&O Shearman, Clifford Chance, Freshfields, Linklaters, Slaughter & May)	After application; before any interview	~80th percentile	Online, untimed or 35–45 min
Silver Circle & US firms (Herbert Smith Freehills, Norton Rose, Latham, Kirkland)	After application or at first interview stage	~70th–75th percentile	Online, typically timed
National and regional firms	Varies — often at assessment centre	~65th percentile	Online or paper-based

💡

Law firms care most about the Deduction and Inference sections

While all 5 sections are scored, Deduction (does this conclusion necessarily follow?) and Inference (how probable is this conclusion from the evidence?) most directly mirror legal analytical skills. Candidates who can distinguish necessary logical entailments from probable inferences — and identify assumptions baked into arguments — demonstrate exactly the precision that makes an effective solicitor.

Section 06

Watson Glaser in Civil Service Recruitment

The UK Civil Service Fast Stream uses Watson Glaser as part of its online assessment stage, alongside a numerical reasoning test, a situational judgement test, and a written exercise. It is used because civil servants are required to analyse complex policy evidence, evaluate competing arguments, and draft well-reasoned advice — skills that map directly to Watson Glaser's five dimensions.

Other civil service and regulatory employers that use Watson Glaser or similar critical thinking tests include the Financial Conduct Authority (FCA), the Bank of England, GCHQ, and the Government Legal Department (GLD). The GLD's process is particularly demanding, as it combines Watson Glaser with a written legal problem exercise.

Organisation	Role Types	Watson Glaser Stage	Other Assessments Alongside
Civil Service Fast Stream	Policy, operational delivery, commercial, digital	Online sift stage	Numerical, SJT, e-tray exercise
Government Legal Department	Government lawyers	Online sift stage	Written legal problem
Financial Conduct Authority	Regulatory, policy, supervision	Online testing stage	SHL numerical, verbal
Bank of England	Economics, research, supervision	Online testing stage	Numerical, situational

Section 07

Which Employers Use Watson Glaser?

Magic Circle Law

⚖️ Magic Circle Firms

A&O Shearman
Clifford Chance
Freshfields Bruckhaus Deringer
Linklaters
Slaughter and May

US Big Law & Silver Circle

🏛️ Top International Firms

Herbert Smith Freehills
Norton Rose Fulbright
Latham & Watkins
Kirkland & Ellis
DLA Piper
Allen & Overy (legacy)

UK Civil Service

🏛️ Government Employers

Civil Service Fast Stream
Government Legal Department
GCHQ
Cabinet Office
HM Treasury

Financial Regulation

🏦 Regulators & Central Banks

Financial Conduct Authority (FCA)
Bank of England
Prudential Regulation Authority
Competition & Markets Authority

Section 08

Preparation Strategies

Master each section separately before mixing them. The 5 question types require different cognitive modes. Practising all 5 together before mastering each individually is inefficient — spend a dedicated session per type until you fully understand the framework, then move to mixed practice.
Learn the Inference scale precisely. The 5-point scale (True / Probably True / Insufficient Data / Probably False / False) is the most distinctive and most difficult aspect of Watson Glaser. The key distinction is between "Probably True" and "Insufficient Data" — if the facts make something probable but not certain, choose Probably True; if there genuinely isn't enough information to form a probabilistic view, choose Insufficient Data.
For Assumptions: test whether the assumption is necessary, not just plausible. An assumption must be logically required for the statement to make sense. Assumptions that are plausible but not required should be marked "Not Made."
For Deduction: be ruthlessly strict. "Follows" means necessarily follows with logical certainty, not just "seems likely." When in doubt, mark "Does Not Follow."
For Evaluation of Arguments: focus on relevance + importance. A strong argument directly addresses the core question with substantial weight. Arguments that are factually true but tangential, or emotionally appealing but logically minor, are Weak.
Practise analytical reading daily. Legal judgments, policy papers, and analytical journalism all exercise exactly the skills Watson Glaser tests. 15 minutes of daily analytical reading builds the critical thinking habits that the test measures.
Use official practice materials. Pearson TalentLens (the Watson Glaser publisher) provides official practice questions. Law firms often link to these from their graduate recruitment pages. Use these alongside third-party providers for varied practice.

Section 09

Frequently Asked Questions

Is the Watson Glaser test timed?+

It depends on the format and employer. Watson Glaser II (the standard version used by most law firms) is typically administered either untimed (candidates take as long as they need) or with a 35–45 minute limit. The Watson Glaser Short Form (24 questions) typically has a 30-minute limit. Some employers use it as a pure ability measure (untimed); others use time pressure as part of the assessment. Check your specific employer's instructions.

How is Watson Glaser scored?+

Raw scores are converted to percentiles compared against a norm group relevant to the employer (professional services, legal, or general graduate). No penalty is applied for wrong answers — always guess if unsure. Law firm cut scores are estimated at 70th–80th percentile depending on firm tier.

What is the difference between Deduction and Interpretation?+

Deduction is a strict logical test — the conclusion must follow with certainty from the premises. Interpretation is a probabilistic test — the conclusion must follow "beyond reasonable doubt" from the evidence. Deduction: "Does this necessarily follow?" (Yes/No). Interpretation: "Is this conclusion highly probable from this evidence?" (Yes/No). Deduction is stricter — many conclusions that follow Interpretively do not follow Deductively.

Can I retake the Watson Glaser test if I score poorly?+

Most law firms allow one attempt per application cycle (typically one academic year). A low score cannot be retaken within the same cycle. This is why preparation matters — there's no second chance in the same cycle, and strong preparation typically produces meaningful score improvements.

Is Watson Glaser harder than SHL Verbal Reasoning?+

Watson Glaser is harder than SHL Verbal Reasoning in terms of logical precision required — but not because the passages are harder to read. The challenge comes from the 5 distinct question types (especially the 5-point Inference scale and the Deduction section), each requiring a different analytical framework. Many candidates find Watson Glaser more demanding than SHL Verbal because it cannot be solved with a simple True/False/Cannot Say heuristic alone.

Ready to Prepare for Watson Glaser?

Build the critical thinking skills that law firms and the civil service demand — practice the 5 question types, starting with Inference and Deduction.

Start Free Practice Tests →SHL Verbal Reasoning Guide