Reliability in Psychological Testing

These flashcards cover key concepts and definitions related to the topic of reliability in psychological testing.

Reliability

The consistency of measurement in psychological testing.

Measurement Error

Discrepancies between true ability and measured ability.

Classical Test Score Theory (CTT)

A reliability theory that assumes each person has a true score free of measurement errors.

True Score

The actual ability or characteristic of an individual that a test aims to measure.

Observed Score

The score obtained from a test, which includes both the true score and error score.

Systematic Error

Predictable error that consistently affects the results in the same direction.

Random Error

Unpredictable error caused by random factors that affect scores inconsistently.

Domain Sampling Model

A model that addresses issues of using a limited number of items to represent a larger construct.

Item Response Theory (IRT)

A theory that focuses on measuring an individual's ability through item difficulty and discrimination.

Generalizability Theory

Theory that evaluates the impact of various facets in the testing situation on test scores.

Test-retest Reliability

A method of estimating reliability by correlating scores from two administrations of the same test.

Parallel Forms Reliability

Estimate of reliability by correlating scores from different forms of the same test.

Internal Consistency

A measure of how closely related the items on a single test are to each other.

Kuder-Richardson Formula (KR20)

A formula used to assess reliability in tests with dichotomous items.

Coefficient Alpha

A general measure of internal consistency that can be used for dichotomous and nondichotomous items.

Inter-scorer Reliability

The degree of agreement between different scorers evaluating the same performance.

Standard Error of Measurement (SEM)

A measure of how much error is present in an observed score.

Confidence Interval

A range around a test score that indicates where the true score is likely to fall.

Spearman-Brown Formula

A formula used to estimate the reliability of a test based on its length.