VerifiedEvidence: highv1.0.0

Convergent Validity

A form of construct validity shown when an instrument's scores correlate strongly with scores from other established instruments measuring a related concept.

Last reviewedDarrin Baines IP Ltd

Concept Architecture

Concept

Theoretically, Convergent Validity is a component of construct validity that assesses whether measures intended to evaluate the same or closely related theoretical constructs are strongly associated with one another. It is founded on psychometric theory and the multitrait-multimethod framework, which predicts that instruments measuring the same construct should demonstrate substantial agreement. Convergent validity exists to provide empirical evidence that a measurement instrument captures the intended underlying construct.

Mathematically, convergent validity is evaluated by examining the magnitude of associations between theoretically related measures. Correlation coefficients, factor loadings and average variance extracted are commonly used to quantify the degree of convergence. Strong positive relationships consistent with theoretical expectations provide evidence supporting convergent validity.

In practice, convergent validity is assessed during the validation of clinical outcome assessments, patient-reported outcome measures and health-related quality-of-life instruments by comparing scores with established reference instruments measuring similar constructs. Health economists evaluate convergent validity to ensure that utility measures and economic outcome instruments appropriately reflect the health concepts they are intended to measure.


Purpose

Used to determine whether instruments designed to measure the same construct demonstrate the expected degree of agreement, supporting the validity of health outcome measures used in research and economic evaluation.


Mathematical Formulae

Primary Formula

There is no universally recognised canonical mathematical formula.

Supporting Formulae

Pearson correlation coefficient:

r = Cov(X,Y) / (�?�?)

Average Variance Extracted:

AVE = ???� / (???� + ???)

where:

?? = standardised factor loading

?? = measurement error variance

Related Mathematical Methods

Construct Validity

Pearson Correlation

Spearman Rank Correlation

Confirmatory Factor Analysis

Average Variance Extracted

Multitrait-Multimethod Analysis


Example

A newly developed fatigue questionnaire is compared with an established fatigue scale used in oncology research. The two instruments demonstrate a Pearson correlation coefficient of r = 0.84, consistent with theoretical expectations that both measure the same underlying construct. This strong positive association provides evidence of good convergent validity.


Excel Implementation

FunctionExample FormulaHealth Economics Application
CORREL=CORREL(B2:B101,C2:C101)Calculate correlations between related health outcome measures
RSQ=RSQ(B2:B101,C2:C101)Assess the proportion of shared variance between instruments
LINEST=LINEST(Y_range,X_range,TRUE,TRUE)Examine linear relationships between related constructs
AVERAGE=AVERAGE(B2:B101)Summarise questionnaire scores
STDEV.S=STDEV.S(B2:B101)Assess score variability during validation

VBA (Optional)

VBA can automate convergent validity analyses by generating correlation matrices, validation summaries and psychometric reports for multiple outcome instruments.


Sources

  • Campbell DT, Fiske DW. Convergent and Discriminant Validation by the Multitrait-Multimethod Matrix.
  • Cronbach LJ, Meehl PE. Construct Validity in Psychological Tests.
  • Streiner DL, Norman GR, Cairney J. Health Measurement Scales: A Practical Guide to Their Development and Use.
  • COSMIN Initiative. COSMIN Methodology for Evaluating Measurement Properties.
  • DeVellis RF. Scale Development: Theory and Applications.

Library

Publications

1
  • Journal article

    GRADE Guidelines: 1. Introduction — GRADE Evidence Profiles and Summary of Findings Tables — Guyatt, Oxman, Akl, Kunz, Vist, Brozek, et al., GRADE Series ed., 2011 (Journal of Clinical Epidemiology)

    The introductory paper of the GRADE (Grading of Recommendations Assessment, Development and Evaluation) series, setting out how to rate certainty of evidence (high/moderate/low/very low) and build evidence profiles and summary-of-findings tables.

Frequently Asked Questions (6)

  • What is convergent validity?

    A form of construct validity shown when an instrument's scores correlate strongly with scores from other established instruments measuring a related concept.

    Source: Campbell & Fiske 1959

  • What does convergent validity expect of related measures?

    Convergent validity expects that an instrument's scores will line up with those from other measures of the same or a closely related concept, since tools aimed at the same target should broadly agree. A new quality-of-life questionnaire, for example, should correlate with an established one and with related indicators such as physical function. Strong correlation supports the claim that the instrument taps the intended construct rather than something unrelated. Agreement with kindred measures is the test. Streiner and Norman (2008) describe this form of validity.

    Source: Streiner & Norman 2008

  • How is convergent validity assessed?

    Convergent validity is assessed by correlating an instrument's scores with those of other established measures of the same or related concepts and examining whether the correlations are strong, as theory predicts. A high correlation with a measure expected to be related supports convergent validity. The comparison measures should be well established and genuinely related to the construct. So convergent validity is demonstrated by showing that the instrument's scores agree with other measures of related concepts, providing evidence that it captures the intended construct, as part of building the case for construct validity.

    Source: Campbell & Fiske 1959

  • Why is convergent validity important?

    Convergent validity is important because it provides evidence that an instrument measures the intended construct: if scores correlate with other measures of related concepts, as they should, this supports the claim that the instrument captures the construct rather than something unrelated. Without such evidence, an instrument's validity is uncertain. Convergent validity, together with discriminant validity, forms a key part of construct validation. So demonstrating convergent validity helps establish that an instrument's scores meaningfully reflect the concept of interest, which is necessary for confidence in research and decisions using the instrument.

    Source: Campbell & Fiske 1959

  • How does convergent validity relate to discriminant validity?

    Convergent validity and discriminant validity are complementary components of construct validity: convergent validity is shown when scores correlate strongly with measures of related concepts, as expected, while discriminant validity is shown when scores do not correlate strongly with measures of unrelated, distinct concepts. Together they demonstrate that the instrument captures the intended construct and not others, agreeing with what it should and differing from what it should. Campbell and Fiske proposed assessing both through the multitrait-multimethod matrix. So the two provide a fuller test of construct validity than either alone.

    Source: Campbell & Fiske 1959

  • What are the limitations of convergent validity?

    The limitations of convergent validity include that a strong correlation with another measure shows agreement but not that either measure is truly valid, since both could share the same bias; that the comparison measure must itself be well validated and genuinely related; and that convergent evidence alone is insufficient, requiring discriminant validity to show the instrument is distinct from unrelated concepts. Correlations also depend on the sample and measures used. So convergent validity is one necessary but not sufficient component of construct validation, interpreted alongside discriminant and other evidence rather than as proof of validity on its own.

    Source: Campbell & Fiske 1959

Trust Record

Verified by Dr Darrin Baines

British health economist

Professional identity: darrinbaines.org

Verification date: 25 Nov 2025

Content version: 1.0.0

Canonical Identity

Term code
HE-ES-EA-009

Stable URI · Machine-readable · Resolvable · CC BY 4.0