Measurement Validity and Reliability

Flashcards covering key concepts related to measurement validity and reliability.

Face Validity

The degree to which a measurement appears, on the face of it, to measure what it is supposed to.

Content Validity

The extent to which variables cover the entire content, or all the major dimensions, of the concept being measured.

Criterion Validity

The measure is valid if scores correlate with other measures of the same concept (usually a well-accepted or ‘gold standard’ method of measurement).

Construct Validity

How well a measure conforms to theoretical expectations, or how well it measures a theoretical or abstract construct.

Measurement Reliability

The extent to which a measurement instrument or performance is dependable, stable, and consistent when assessed under identical conditions.

Correlation Coefficient

A correlation statistic that represents the strength of the association between two measurements. Commonly reported correlation coefficients include Pearson’s r and the intraclass correlation coefficient (ICC).

Test-retest reliability

Evaluates the stability of a measurement obtained on two different occasions when we would expect no change in the construct being measured.

Intra-rater reliability

Evaluates the ability of a single rater to obtain the same result when presented repeatedly with the same observations.

Inter-rater reliability

Evaluates the ability of different raters to obtain the same measurement relative to each other.

Random Error

Unpredictable and is scattered about the true value.

Systematic Error

Usually predictable and therefore directional.