Insurance Underwriting and Risk Selection Assessment for Underwriters and Risk TeamsEvery underwriter has one instrument they reach for first. Yours is not the problem until it is the only one.

Methodology: Thirty-six original exercises across five formats: sixteen most-and-least forced-choice blocks, six select-every-that-applies exercises, six binary claims, four keyed questions and four quantity estimates. The declared response instruction is behavioural tendency throughout - what would you most likely do - and one instruction covers the whole form. Construct statement: it measures which underwriting instrument somebody reaches for first, how well their choices match what each risk actually warrants, what they take a piece of information to establish, and whether the file they leave is answerable to a later reader. It does not test any country's insurance regulation, any insurer's own guidelines, any rating software or any class of business in technical depth, and it is not a personality measure. Scoring is a quasi-ipsative forced choice. Each block offers one option from each of four instruments - price it, shape it with terms, find out more first, walk away - re-worded for the risk in front of the reader so no instrument can be recognised by its phrasing. Each instrument is keyed as the best move on exactly four of the sixteen blocks and as the worst move on four others, which is the design's whole defence: a respondent who always prices, always adds terms, always asks for more or always declines scores identically to the other three habits and behind anybody reading the cases. The report demonstrates that on the reader's own blocks rather than asserting it. The keyed positions are spread four to a slot on each pole and the offset between the two keyed positions varies across blocks, because a fixed distance between most and least is a third key. Each pole carries its OWN declared prior, because the best move and the worst move are two different questions that people answer differently, and a single set of marginals would make one of the two corrections wrong. The tendency profile is centred within person - each instrument's chosen-most count minus its chosen-least count, with the respondent's own mean removed - so elevation and decisiveness drop out and a high score on everything is arithmetically impossible, which the report states rather than leaving the reader to infer. Six blocks per instrument would be needed for a figure and there are four, so each lean is reported as a placement with the cut printed, never as a number and never as a percentile. Every lean is printed with its cost as well as its use, because a profile of four legitimate instruments that names only strengths is a horoscope. Constructs and sources: the forced-choice and quasi-ipsative measurement literature, in which blocks matched for desirability across different dimensions resist faking better than single-stimulus rating scales, and normative information survives when the blocks are quasi-ipsative rather than purely ipsative (Brown and Maydeu-Olivares on Thurstonian forced choice; Bartram on the applicant-setting effect sizes, which are far smaller than laboratory ones); the underwriting-cycle and adverse-selection literature on why a rate cannot reach every exposure and why capacity, competence and absent information are structural rather than priced constraints; the value-of-information principle that a question is worth asking only when its answer could change the decision; the distinction between exposure-changing instruments (exclusions, warranties, deductibles, sub-limits) and consideration-changing ones (rate), which is the most common confusion in junior underwriting files; the outcome-bias literature on portfolio review, where decisions are judged by results rather than by what was knowable at the time; the documentation research on defensibility, in which a file that records what was set aside is distinguishable from one that records only what was done; McDaniel and colleagues on behavioural-tendency response instructions; and Haladyna, Downing and Rodriguez on plausible distractors and cue control. All exercises are original works; no commercial instrument's items or name is reproduced, no insurer, broker or rating product is named, and no affiliation is implied.