Definition
Concurrent validity is a form of criterion-related validity that assesses the degree to which the results of a particular test or measurement correspond with those of an established, concurrent benchmark or criterion that is measured at the same time. It indicates how well a new instrument predicts outcomes measured by an existing, validated instrument administered concurrently.
Overview
In the field of psychometrics and measurement theory, validity refers to the extent to which an instrument measures what it purports to measure. Concurrent validity is one of several subtypes of criterion-related validity, alongside predictive validity. While predictive validity evaluates how well a test forecasts future performance on a criterion, concurrent validity examines the correlation between test scores and criterion measures collected simultaneously. It is frequently employed during the development of new psychological assessments, educational tests, health‑related questionnaires, and other measurement tools where an established standard already exists.
Typical applications include:
- Comparing a new depression inventory with the Beck Depression Inventory administered to the same participants.
- Validating a new computer‑based skill test against a traditional paper‑pencil version taken in the same session.
- Assessing a novel physiological sensor against a gold‑standard clinical device measured concurrently.
Etymology/Origin
The term combines “concurrent,” from the Latin concurrere meaning “to run together,” and “validity,” from the Latin validus meaning “strong” or “effective.” The concept emerged in the early to mid‑20th century alongside the formalization of validity theory in psychological testing, particularly within the work of Lee J. Cronbach and others who distinguished between predictive and concurrent forms of criterion validity.
Characteristics
- Simultaneous measurement: Both the test under evaluation and the criterion are administered within the same time frame, often to the same sample.
- Correlation coefficient: Validity is quantified using statistical correlation (e.g., Pearson’s r) or other concordance measures such as intraclass correlation coefficients. Higher correlations indicate stronger concurrent validity.
- Reference standard: Requires an established, widely accepted instrument or criterion that is considered a gold standard for the construct being measured.
- Sample considerations: The sampled population should be representative of the intended test users; otherwise, validity estimates may not generalize.
- Potential confounds: Shared method variance, testing fatigue, or situational factors can inflate correlations, so methodological controls (e.g., counterbalancing order of administration) are important.
- Application limits: High concurrent validity does not guarantee that the test will predict future outcomes (predictive validity) or that it fully captures the construct’s theoretical breadth.
Related Topics
- Predictive validity – assessment of how well a test forecasts future performance on a criterion.
- Construct validity – overall evidence that a test measures the theoretical construct it is intended to assess.
- Criterion-related validity – umbrella term for validity evidence derived from the relationship between test scores and external criteria (includes both concurrent and predictive validity).
- Face validity – the degree to which a test appears effective in terms of its stated aims, based on subjective judgment.
- Reliability – the consistency of a measurement instrument, which interacts with validity but is a distinct property.
- Standardization – procedures for administering and scoring a test uniformly, important for generating comparable concurrent validity data.