A multitrait-multimethod validation of the Implicit Association Test: implicit and explicit attitudes are related but...

Area of Science:

• Social Psychology
• Psychometrics

Background:

• Recent research distinguishes between implicit and explicit evaluations.
• The construct validity of implicit attitudes requires rigorous testing.

Purpose of the Study:

• To apply Campbell and Fiske's multitrait-multimethod design to validate implicit attitudes measured by the Implicit Association Test (IAT).
• To differentiate between implicit and explicit attitude factors using latent-variable structural modeling.

Main Methods:

• Utilized a multitrait-multimethod design with 287 participants.
• Assessed up to seven attitude domains using both self-report and IAT measures.
• Employed latent-variable structural models to distinguish method variance from attitude variance.

Main Results:

• A correlated two-factors-per-attitude model (implicit and explicit factors) demonstrated superior fit over a single-factor model.
• Implicit and explicit attitude factors showed strong, yet distinct, relationships.
• Collapsing indicators into a single factor resulted in inferior model fit, supporting separate constructs.

Conclusions:

• Implicit and explicit measures assess related but distinct attitude constructs.
• Findings support theories positing dual-process or dual-representation mechanisms in attitudes.