Interview Evidence and Answer Quality Assessment for Job Seekers and Career ChangersEverybody can tell a good story about their own work. The question is what would still be true if somebody checked.

Methodology: Thirty-six original exercises across six formats: fourteen claim-and-receipt items, six select-every-that-applies exercises, six keyed claims, four most-and-least forced choices, three matching exercises and three ordering exercises. The declared response instruction is knowledge and applied judgement throughout - which single thing would make this claim checkable - and one instruction covers the whole form. Construct statement: it measures whether somebody can tell a claim that could be tested from a claim that can only be agreed with, whether they name their own part in shared work, whether they can distinguish what they did from what changed, and whether an answer answers the question that was asked. It does not measure interview delivery, accent, fluency, confidence or likeability, it does not predict whether any particular interview will be passed, and it is not a personality measure. Scoring is signal detection. Seven of the fourteen claim items can be substantiated by something on offer and seven cannot, and the correct receipt has to be NAMED rather than merely accepted, so the discrimination figure and the threshold figure are computed with the log-linear correction and reported separately: accepting every claim and rejecting every claim both land at zero discrimination as arithmetic rather than as a threshold. The threshold is reported in the instrument's own words as taking claims on trust or doubting all of them, with neither direction treated as the good one, because a person who doubts everything answers a different question badly. Raw accuracy is printed as a contributor with the note that it is not the headline, precisely because a fixed habit can reach half of it. Every other strand is chance-corrected against the authored answer priors carried on each option. Constructs and sources: the structured behavioural interview and the evidence standards behind competency-based questioning (Campion, Palmer and Campion on structure; Janz's patterned behaviour description interviewing); the finding that verifiability rather than elaborateness distinguishes accounts, drawn from the deception-detection literature on checkable detail (Nahari and colleagues on verifiability); impression management and faking in selection interviews (Levashina and Campion's taxonomy of slight image creation, extensive image creation and deceptive ingratiation); attribution of shared outcomes and the self-serving bias in accounts of team work; signal-detection theory and the separation of discrimination from response bias (Green and Swets; Macmillan and Creelman on the log-linear correction); the distinction between activity and outcome in performance description (the programme-logic separation of outputs from outcomes); and item-writing guidance from Haladyna, Downing and Rodriguez. All items are original works written for this instrument. No test, vendor, employer, framework or coaching programme is reproduced, named or implied, and no endorsement or affiliation exists or is suggested.