In an ipsative questionnaire the sum of scale scores from each respondent adds to a constant value, which is exactly what makes between-person comparison impossible.
Why it matters when the plan changes
The distinction decides what a score can be used for. In an ipsative questionnaire the sum of scale scores from each respondent adds to a constant value, so a high score on one attribute comes at the expense of another within the same person. That is informative for a conversation about someone's own pattern, but it cannot answer which of two candidates is stronger on a dimension, because each person's scores are relative to their own total rather than to a defined population. Using ipsative scores for selection is a common and quiet error.
The tension is that forced choice resists faking by making respondents rank statements against each other instead of rating themselves, and it pays for that resistance in this coin. The format that stops people endorsing everything also produces scores that are within-person by default. Recovering comparability requires a measurement model, such as Thurstonian IRT, which makes forced-choice instruments usable for high-stakes decisions.
In practice
Two candidates both rank decisiveness above consultation. Read as raw rankings, they look identical. Modelled into normative scores, one is marginally ahead on a narrow range and the other decisively ahead on a wide one. The comparison the decision needs exists only after the modelling step, and a report that skipped it would have shown a tie.
Evidence
In ipsative questionnaires the sum of scale scores from each respondent adds to a constant value, and the format contrasts with rating scales.
Ipsative, Wikipedia (2026)A norm-referenced test estimates the position of the tested individual within a defined population on the trait measured.
Norm-referenced test, Wikipedia (2026)
What it cannot tell you
The distinction tells you what a score can be compared to, not whether the underlying trait model is sound. A well-modelled normative score can still measure a weak construct, and an ipsative profile can still be a poor description of the person's pattern. Neither format guarantees validity of what is being measured, only the comparability of the result.
Questions
Because ipsative questionnaires are defined, per Wikipedia (2026), by scale scores from each respondent summing to a constant value, so a high score means high relative to that person's other scores, not high in absolute terms. Two people with identical profiles may differ substantially in absolute strength, and the profile cannot show it.
Describing a person's own pattern: which of their attributes is relatively strongest, where their energy goes, what they will reach for first. That is genuinely useful in a development conversation with the individual, and it is the use the format naturally supports.
Through a measurement model that treats each forced choice as a comparison between two statements, working backwards to the underlying dimensions that would explain the pattern. The output resembles a norm-referenced test, as described in Norm-referenced test, Wikipedia (2026), which estimates a person's position within a defined population rather than against their own total.
For selection and comparison, yes, because the question is about relative standing between people. For a development conversation with one person, the ipsative view is often more directly useful. The error is using one where the other is required, which usually runs in the selection direction.
Ask whether scores can be compared between candidates and what model produces them. A supplier offering a forced-choice instrument who cannot name the scoring model is probably reporting ipsative scores, and those should not be driving a comparison between people.