How to choose a personality assessment
Seven questions to ask any publisher before your organization buys an assessment. They apply to the ICA as much as to anything else.
1. What exactly will you use it for?
Development, coaching and team work ask much less of an instrument than selection does. Decide the use first, because it decides how much evidence you need. For the ICA, see appropriate use.
2. What evidence does the publisher show?
Ask for the technical evidence: reliability, the structure of the scores, and how results relate to other measures. The Standards for Educational and Psychological Testing describe what that evidence should cover. Our guide to assessment accuracy explains the terms. A publisher should say plainly what is still unproven, as our claims ledger does.
3. Is it normative or forced-choice?
Normative scales compare people with a reference group. Forced-choice (ipsative) scales show the order of one person's own preferences, so scores should not be compared between people. Each suits different uses. See ipsative and normative in the glossary.
4. How easy is it to fake?
In high-stakes settings people answer to look good. A meta-analysis found forced-choice formats less susceptible to faking than single-statement formats, though not immune (Cao and Drasgow, 2019).
5. How is it priced?
Most workplace assessments charge once to certify, then per assessment or per platform subscription. Some charge a yearly practitioner fee. Model the cost over three years, not the first purchase. Our cost comparison gives published figures.
6. Where does the data go?
Ask where answers are stored, who can see them, and how long they are kept. The ICA runs in each person's browser, and results are shared as files you control. See procurement.
7. Does it fit the law where you will use it?
If any use touches hiring or promotion, check the rules where you operate. Our hiring guide summarizes US federal and state rules as of 3 November 2026.
A one-page checklist
| Question | What a good answer looks like |
|---|---|
| Use | Written down, and matched to the evidence available |
| Evidence | Published, with limits stated |
| Format | Explained, with guidance on comparing scores |
| Faking | Addressed in the design or the guidance |
| Cost | Clear over three years |
| Data | You know where it lives and who sees it |
| Law | Checked for every place you will use it |
Questions
Should we choose the most popular test?
Popularity is not evidence. Choose the instrument whose evidence matches your intended use.
Sources
Keep reading
Take the next step.
Seven questions to ask before buying a personality or behavioral assessment: purpose, evidence, format, faking, cost model, data handling and legal fit.