Watson Glaser A: Inference
Part of the 40-question Watson Glaser III · Within the 30-minute overall limit (~45 seconds per question)
What it tests. Inductive reasoning and probabilistic analysis from nuanced verbal clues, without overextending the data.
Worked example. If a passage discusses inflation in a fictional country, you cannot use real economics to upgrade a statement from Probably True to True unless the text guarantees it.
Common traps. The 'Probably True' versus 'True' distinction, and bringing outside commercial knowledge into the text.
How to handle it. Treat the text as the entire universe. True only if it is impossible to be false; if you need more than one logical leap, the answer is almost certainly Insufficient Data.
Watson Glaser B: Recognition of Assumptions
Part of the 40-question Watson Glaser III · Within the 30-minute overall limit
What it tests. Deconstructive logic - identifying unstated pillars an argument depends on, as when picking apart an opponent's statement of case.
Worked example. If the argument stands perfectly well without the proposed premise being true, the assumption has not been made.
Common traps. Confusing an assumption with a logical consequence or a sensible-sounding conclusion.
How to handle it. Use the negative-reversal technique: negate the assumption and reinsert it. If the argument collapses, the assumption is Made; if it still functions, it is Not Made.
Watson Glaser C: Deduction
Part of the 40-question Watson Glaser III · Within the 30-minute overall limit
What it tests. Deductive and syllogistic reasoning - whether a conclusion must inevitably follow from the premises.
Worked example. If the premises state all corporate lawyers are Olympic swimmers and no Olympic swimmers can read music, then 'no corporate lawyers can read music' Follows, however absurd in reality.
Common traps. Letting real-world truth override formal validity; conflating 'some' with 'all'.
How to handle it. Map relationships with Venn diagrams or symbolic logic. Watch absolute words (all, none, never, always) versus qualifiers (some, many, often). A premise about 'some' never supports a conclusion about 'all'.
Watson Glaser D: Interpretation
Part of the 40-question Watson Glaser III · Within the 30-minute overall limit
What it tests. Evidence synthesis and threshold evaluation - whether a point is proven on a closed set of evidence.
Worked example. Read the passage like a restrictive covenant: if you can imagine a realistic scenario where the text is accurate but the interpretation is still wrong, it does not follow.
Common traps. Accepting interpretations that are merely highly plausible while leaving equally reasonable alternatives unaddressed.
How to handle it. Do not fill narrative gaps with your own imagination; demand a necessary, uncontradicted explanation.
Watson Glaser E: Evaluation of Arguments
Part of the 40-question Watson Glaser III · Within the 30-minute overall limit
What it tests. Argument validation - an argument is Strong only if both highly relevant and genuinely important to the question.
Worked example. An emotionally compelling statement irrelevant to the precise question is Weak, however much you agree with it.
Common traps. Letting personal, political or moral beliefs colour the judgement.
How to handle it. Strip away rhetoric. Check direct relevance first, then importance. Speculation or unproven cause-and-effect is always Weak.
Immersive job simulation / SJT
Scenario-based · Generally untimed, though time taken may be recorded
What it tests. Professional judgement, prioritisation, ethical integrity and alignment with HL's values (ambition, collaboration, precision, client-centricity).
Worked example. Faced with conflicting deadlines from two partners, the optimal path is transparent communication and structured prioritisation - assess true commercial deadlines and speak to both supervisors.
Common traps. The 'hero' option (solving everything alone), the passive option (pushing the problem back without legwork), or hiding an error out of fear.
How to handle it. Never hide mistakes; escalate immediately, but arrive with one or two realistic solutions. Favour collaboration and risk awareness.
Proofreading and meticulousness
Detail-based · Generally untimed within the overall window
What it tests. Attention to detail and precision - the day-to-day reality of a first-seat trainee.
Worked example. A clause referencing a 'Section 14.2' that does not exist, or 'the Buyer' inconsistently written as 'the buyer'.
Common traps. Reading what you expect to see; missing subtle punctuation and inconsistent capitalisation of defined terms.
How to handle it. Slow your reading, anchor with the cursor line by line, and cross-check every number, date and percentage against the source.