Exploring the Influence of Response Styles on Continuous Scale Assessments: Insights From a Novel Modeling Approach.

Area of Science:

• Psychometrics
• Psychological Measurement
• Statistical Modeling

Background:

• Traditional discrete formats for psychological traits are embedded in item response theory.
• Continuous response formats offer advantages in computer-based testing.
• Response styles (e.g., extreme, acquiescence) can bias trait inferences and skew scores.

Purpose of the Study:

• To investigate the impact of response styles on continuous scale responses.
• To propose extensions to the Continuous Response Model (CRM) to capture extreme response style (ERS) and acquiescence response style (ARS).
• To quantitatively model individual variations in response styles within a continuous context.

Main Methods:

• Development of the CRM-ERS and CRM-ARS models.
• Evaluation through simulation studies.
• Parameter calibration using Bayesian methods.

Main Results:

• Proposed models demonstrated satisfactory parameter recovery.
• Neglecting response styles resulted in biased estimation.
• Estimation accuracy improved with increased test length and sample size.

Conclusions:

• Accounting for response styles is crucial for accurate psychological trait assessment.
• The proposed CRM extensions effectively capture individual response style variations.
• The models have practical applications for improving the validity of continuous scale data.