Develop a psychological instrument with the Lab
From a clear construct definition to a scored, documented test. The Lab can guide the whole project, or only the stages you cannot cover in-house.
Test development is a project on request. Scope, planning and price are agreed after the intake.
How a development project runs
Each stage ends with a decision: carry on, revise or stop. That keeps weak items and vague scales out of the final test.
Define the construct
What exactly should the test measure, for whom and for which decision? We write a construct definition with facets, related and unrelated constructs, and the intended use.
Write and review items
Items are written against the definition, reviewed by content experts for relevance, clarity and cultural fit, and checked for wording that only some groups would understand.
Pilot and analyse
A pilot in the target group shows how items behave: difficulty, discrimination, item–total correlations, response patterns and a first look at the factor structure.
Set scoring and document
We fix scales, scoring rules and reverse-scored items, and document every choice, so the test can move on to validation and norming.
Choices to settle early
These decisions shape everything that follows. Changing them after the pilot usually means piloting again.
Intended use
Selection, development or research. A test for decisions about individuals needs stronger evidence than one for group-level insight.
Target group
Who will take the test: education level, career stage, sector. Items and norms have to fit that population.
Languages
Which language versions you need now and later. Each version is adapted and analysed in its own right, not just translated.
Item format
Multiple choice, Likert scales, forced choice or situational items, each with its own strengths and response biases.
Length and time
Enough items per scale for reliable scores, but no more than respondents can answer attentively.
Administration
Supervised or unsupervised, time limits, device. Unsupervised online use asks for extra attention to item exposure and response quality.
Built to run on Match.hr
A test developed with the Lab can be offered on Match.hr, privately for your organisation or in the library, with scoring and reports handled by the platform. Which of the points below apply to your test, and from when, we agree in the intake.
- Scales, subscales and reverse-scored items as defined in the scoring rules
- Norm scores such as percentiles, stanines and T-scores once a norm group exists
- Separate analysis for each language version
- Item statistics recalculated as administrations grow
- A psychometric label that shows the review level and the evidence so far
Our review criteria follow the structure of the COTAN and EFPA test review models. Match.hr is not affiliated with or endorsed by either organisation.
A new test starts without evidence
A pilot shows whether items behave as intended. It does not show that the test predicts anything. Evidence for a specific use, such as selection, needs a validation study with enough participants and a relevant criterion. Until then, present the test as in development and do not use it on its own for decisions about people.
Frequently asked questions
How many people does a pilot need?
It depends on the analyses. Item statistics become reasonably stable with a few hundred respondents per language version; a factor analysis usually needs more, especially with many scales. In the intake we work out a sample size that fits your questions and your access to the target group.
Can you develop a test in several languages at once?
Yes, but we treat each language as its own version. Items are adapted by native speakers with subject knowledge and reviewed for meaning, not only for wording. Whether the versions measure the same thing is a question for the data, so each version is piloted and analysed separately.
Who owns the instrument?
Rights and licences are agreed in writing before the work starts. If you later want to offer the test to other organisations, Publish your test explains how instruments are listed on Match.hr.
What does the documentation contain?
The construct definition, how items were written and reviewed, the pilot sample and results, the scoring rules, the intended use and its limits. It is the basis for a technical manual once validation and norming have been done.
Related pages
Test your first candidate today
Free account, no credit card, results in minutes.