Name: Eunbee Kim

Ph.D. Dissertation Proposal Meeting

Date: Monday, April 15th, 2024

Time: 2:00PM

Location: Link

Dissertation Chair/Advisor:

Dr. Susan Embretson—School of Psychology, Georgia Institute of Technology

Dissertation Committee Members:

Dr. James Roberts—School of Psychology, Georgia Institute of Technology

Dr. Rick Thomas—School of Psychology, Georgia Institute of Technology

Dr. Mark Himmelstein—School of Psychology, Georgia Institute of Technology

Dr. Michael Hunter—Human Development and Family Studies, Pennsylvania State University

Title: Applications of the Partial Credit Model (PCM) Accounting for Extreme Response Styles (ERS) Considering the Dependence between ERS and the Measured Trait

Abstract: Among diverse response styles inherent in measurement data, often collected through Likert-type scales, extreme response styles (ERS) have gained considerable attention. This study proposes the application of the partial credit model (PCM) accounting for extreme response styles (ERS-PCM), wherein the ERS tendency is treated as dependent of the measured trait. It is worth noting that a dependence between ERS and the measured trait is intrinsic to measurement using Likert-type scale, as respondents with extremely high or low trait levels are expected to endorse more extreme options compared to those with average trait levels, compared to respondents with middle trait levels. Moreover, previous literature has established associations between ERS tendency and various individual differences, supporting the practicality of assuming dependency between the measured trait and ERS tendency. In Study 1, the ERS-PCM accounting for dependence between ERS and the measured trait (ERS-PCM-D), with different specifications, will be applied to big-five personality measurements. It is hypothesized that incorporating the quadratic relationship between ERS and the trait enhances interpretability for parameter estimates and yields the better model fit. Study 2 examines parameter recovery for the ERS-PCM-D under various conditions, including different correlations between ERS tendency and the trait. Study 2 aims to examine the effectiveness of the ERS-PCM-D. The findings have the potential to demonstrate that the ERS-PCM-D improves the applicability and generalizability of the ERS-PCM by incorporating dependence between ERS and the trait.