Measurement and control of response styles using anchoring vignettes: a model-based approach

Area of Science:

• Psychological Measurement
• Psychometrics
• Survey Methodology

Background:

• Response styles significantly impact the reliability of psychological survey instruments using rating scales.
• Existing latent trait models offer control for response styles but are limited to a small number of response style types.
• Anchoring vignettes present a novel approach to address the limitations of current response style control methods.

Purpose of the Study:

• To introduce a new latent trait model designed to measure and control for any form of response style.
• To leverage vignette responses within a latent trait framework to enhance the accuracy of psychological measurements.
• To overcome the limitations of existing models in accommodating diverse response style variations.

Main Methods:

• Development of a novel latent trait model incorporating anchoring vignette data.
• Application of the model to analyze self-reported conscientiousness data from a cross-national study.
• Utilizing vignette responses to identify and adjust for various response styles.

Main Results:

• The proposed model successfully measures and controls for a wide range of response styles.
• Demonstrated effectiveness in enhancing the precision of self-reported data by accounting for response style variations.
• The model's utility is illustrated with real-world data from a cross-national conscientiousness study.

Conclusions:

• Anchoring vignettes, integrated into a latent trait model, provide a powerful tool for addressing response style effects.
• This new methodology offers a more comprehensive approach to controlling for response styles in psychological surveys.
• The findings contribute to more accurate and reliable cross-cultural comparisons of psychological constructs.