AIM GlossaryBrowse the glossary:
measurement invariance
— The extent to which a measure operates equivalently across groups.
Full explanation:
Measurement invariance analyses indicate whether or not the construct being measured is understood similarly and has the same meaning for all groups. If a measure does not have measurement invariance any observed group differences could be due to some form of bias rather than true differences on the variable of interest.
Measurement Analysis: Multigroup Confirmatory Factor Analysis (MGCFA)
Interpretation:
• Configural Invariance: Similar pattern across groups (no constraints)
• Metric Invariance: Similar item-construct relationships (factor loadings)
• Scalar Invariance: Similar intercepts/thresholds across groups
• Indicators of measurement invariance = CFI change ≤ 0.01, RMSEA increase ≤ 0.015
Measurement Analysis: Multigroup Confirmatory Factor Analysis (MGCFA)
Interpretation:
• Configural Invariance: Similar pattern across groups (no constraints)
• Metric Invariance: Similar item-construct relationships (factor loadings)
• Scalar Invariance: Similar intercepts/thresholds across groups
• Indicators of measurement invariance = CFI change ≤ 0.01, RMSEA increase ≤ 0.015
Still don't get it?
If you feel that this definition hasn't helped you to understand the term, click on our monkey to let us know.
← measurement bias mechanism →