Psychometrics Glossary

A quick glossary of key terms you’ll often see in psychological tests and result pages. Understanding these terms helps you interpret scores responsibly.

How to use this glossary

Use these definitions as practical guidance. In online testing, terms are often simplified. If you want a deeper explanation, see Reliability and validity explained and How to interpret results.

Key terms

Construct

A psychological concept being measured (e.g., anxiety, stress, impulsivity). Constructs are not directly observable; tests approximate them through items.

Item

A single question or statement in a questionnaire.

Scale

A set of items scored together to measure a construct (e.g., an anxiety scale).

Subscale

A subset of items within a scale that measures a specific dimension (e.g., cognitive vs physical anxiety symptoms).

Cutoff

A threshold used to categorize scores (for example low/moderate/high). Helpful for screening; not absolute proof of diagnosis.

Norms

Reference values from a group that help interpret a score by comparison (e.g., average scores for a population).

Percentile

A ranking score. For example, the 70th percentile means your score is higher than 70% of the reference group.

Reliability

Consistency: whether the test gives similar results under similar conditions.

Validity

Meaningful accuracy: whether the test actually measures what it claims to measure.

Internal consistency

A form of reliability showing whether items within a scale tend to correlate (they “move together”).

Test-retest reliability

Whether results are similar when you repeat the test after a short period (assuming the underlying trait/state is stable).

Sensitivity

How well a screening test detects potential cases (fewer false negatives).

Specificity

How well a screening test avoids false alarms (fewer false positives).

False positive

The test flags risk even though a diagnosis is not present. Can happen due to stress, overlapping symptoms, or cutoffs prioritizing sensitivity.

False negative

The test does not flag risk even though the difficulty may be present. Can happen due to under-reporting or atypical symptom presentation.

Measurement error

The natural uncertainty in any test score. No score is perfectly precise; small differences can occur due to context or random variation.

Screening

A quick method to estimate risk level. Screening is not diagnosis.

Diagnosis

A clinical judgment made by a qualified professional after a broader assessment including context, history, impairment and differential considerations.

Explore tests

Browse screening and self-reflection tools here: All Tests.

FAQ

What is a cutoff score?

A cutoff is a threshold used to categorize a score (for example low, moderate, high). Cutoffs are helpful for screening but are not absolute proof of a diagnosis.

What are norms in psychological testing?

Norms are reference values from a group that help interpret a score by comparison (for example average scores for a certain population).

What is the difference between reliability and validity?

Reliability is consistency (similar results under similar conditions). Validity is whether the test measures what it claims to measure.