The inference section of the Watson Glaser Critical Thinking Appraisal is widely considered the most challenging part of the test. Unlike other sections where answers feel more clear-cut, inference questions require you to rate conclusions on a precise 5-point scale — and the difference between adjacent ratings can be razor-thin. This guide breaks down exactly how the scale works, why candidates consistently struggle with it, and what you can do to sharpen your judgment before test day.
Whether you are preparing for a role at a law firm, a management consultancy, or a graduate assessment centre, understanding the inference section is essential. Start with our Watson Glaser practice test and return to this guide to refine your approach.
Every inference question presents you with a short passage of facts followed by a proposed inference — a conclusion someone might draw from those facts. Your job is to rate that inference using one of five labels: True, Probably True, Insufficient Data, Probably False, or False.
True means the inference follows directly and necessarily from the stated facts — it cannot be wrong given what is written. Probably True means the inference is more likely correct than not, given the available information, but is not certain. Insufficient Data (sometimes called "More Data Needed") means you simply cannot determine whether the inference is true or false — the passage neither supports nor contradicts it. Probably False means the inference is more likely wrong than right based on what is stated. False means the inference directly contradicts the stated facts.
The scale is symmetrical: True and False sit at the poles, while Probably True and Probably False mirror each other on either side of the neutral Insufficient Data midpoint. This symmetry is intentional — Watson Glaser tests whether you can calibrate your confidence, not just pick a direction. For a broader overview of the test, see our complete Watson Glaser guide.
The inference section trips up even well-prepared candidates for a predictable set of reasons. First, our brains are wired to fill gaps — when we read a passage, we naturally import background knowledge and common sense to build a richer picture than what is literally written. The Watson Glaser test exploits this tendency relentlessly. An inference may feel obviously true because it aligns with what you know about the world, but if the passage itself does not provide the supporting evidence, "Probably True" or even "Insufficient Data" may be the correct rating.
Second, the five categories create decision paralysis. Most people intuitively think in binary: true or false. Adding three intermediate options — and requiring you to choose between them — demands a level of calibration that takes practice to develop. Many candidates default to Insufficient Data too often, treating it as a safe middle ground rather than a specific claim that relevant evidence is absent from the passage.
Third, time pressure distorts judgment. Under exam conditions, candidates rush past the critical step of re-reading the passage after seeing the inference. A second reading, with the specific inference in mind, almost always reveals evidence that changes the rating. Practise this two-pass discipline now so it becomes automatic. Explore how this section compares with the deduction section, which follows different logical rules entirely.
The single most important skill in the inference section is learning to distinguish what the passage states from what it merely makes plausible. Consider a passage that says: "Company X's revenue grew 15% last year, driven entirely by international sales." An inference that says "Company X's domestic sales declined last year" is Probably True — the passage implies domestic growth was flat or negative, but it does not state this explicitly. An inference that says "Company X has no domestic customers" is Probably False — too strong a conclusion from the available data. An inference that says "Company X will grow 15% again next year" is Insufficient Data — the passage says nothing about future performance.
Practising this distinction — stated vs. plausible — is the fastest way to improve your score. For every inference, ask yourself: "Is this directly supported by words in the passage, or am I adding my own logic?" If you are adding logic, the question becomes how much logic you are adding and whether that logic is sound given the passage content. See our tips in the 7 tips to pass your Watson Glaser assessment for additional frameworks.
Judge every inference only on what the passage states — not on what you know from outside the passage, not on what seems likely from common sense, and not on what the passage implies but does not say. Your job is to act as a strict logician, not a reasonable person making everyday assumptions. Every time you catch yourself thinking "well, obviously…" — stop. That instinct is exactly what the test is designed to trigger and exploit.
A helpful mental reset: imagine you are a visitor from another planet who knows nothing about Earth except what is written in the passage. Would this inference hold up? If not, downgrade your rating.