— Cognitive science

Test-Retest Reliability

Term

r-tt

Domain

Cognitive psychology

First introduced

20th century

IQ relevance

Direct

Test-Retest Reliability (r-tt) refers to the correlation between scores when the same test is taken twice; WAIS achieves ~0.96. The concept is central to modern cognitive psychology and psychometrics — particularly to how IQ is theorized, measured, and interpreted.

What test-retest reliability actually means

In cognitive psychology, test-retest reliability is operationalized as the correlation between scores when the same test is taken twice; WAIS achieves ~0.96. It is distinct from related constructs in that it specifically captures a psychometric property of the test instrument rather than a cognitive trait.

How it's measured

Measured through standardized testing batteries and laboratory paradigms specific to the construct.

Relationship to IQ

Test-Retest Reliability is integrated into the broader theory of IQ and cognitive function. It contributes to and is partially measured by composite IQ scores.

Test your test-retest reliability and 5 other cognitive axes

Core Brain's 20-minute test decomposes IQ into six cognitive components so you see your specific profile.

Why this matters

Understanding test-retest reliability is foundational for anyone trying to interpret an IQ score meaningfully. The single composite IQ number conceals significant variation across cognitive functions; understanding the components is how you go from "I scored 130" to "I scored 130, with particular strength in [X] and relative weakness in [Y]."

Frequently asked

What is test-retest reliability?

Test-Retest Reliability (r-tt) refers to the correlation between scores when the same test is taken twice; WAIS achieves ~0.96. It is a central concept in cognitive psychology and IQ measurement.

How is test-retest reliability different from general IQ?

Composite IQ is a weighted aggregate of multiple cognitive abilities. Test-Retest Reliability is one specific component or property that contributes to but is not identical to the IQ score.

Can test-retest reliability be improved?

Generally limited room for improvement in trait-level cognitive properties; significant variability in measured performance day-to-day.