Estimation of Response Styles Using the Multidimensional Nominal Response Model: A Tutorial and Comparison With Sum...

Area of Science:

• Psychometrics
• Statistical Modeling

Background:

• Item response theory (IRT) models are increasingly used to measure response styles on Likert-type items.
• Traditional sum-score methods may confound response styles with substantive content.

Purpose of the Study:

• To provide a tutorial on the multidimensional nominal response model (MNRM).
• To offer guidance on estimating the MNRM using flexMIRT®, mirt, and Mplus.
• To compare different scoring approaches and their impact on substantive conclusions.

Main Methods:

• The paper details the specification and interpretation of response functions within the MNRM.
• Analytical methods for converting sum scores to scale scores using the MNRM are presented.
• A real data example is used to compare the MNRM with traditional sum-score approaches.

Main Results:

• The multidimensional nominal response model (MNRM) provides a flexible, model-based alternative to sum-score methods.
• Sum-score approaches can lead to scores that are confounded with substantive content.
• The study highlights the importance of choosing appropriate methods for measuring response styles.

Conclusions:

• The MNRM offers a more nuanced approach to measuring response styles compared to traditional sum scores.
• Researchers should consider the potential for confounding in sum-score-based analyses.
• This work facilitates further research into the impact of different response style measurement approaches on study findings.