Material Type  

 학위논문

 

0017165153

Date and Time of Latest Transaction  

20250211153130

ISBN  

9798346875994

DDC  

370

Author  

Zou, Tongtong.

Title/Author  

Person-Level Measurement Validity Issues in Noncognitive Assessment.

Publish Info  

[S.l.] : The University of Wisconsin - Madison., 2024

Publish Info  

Ann Arbor : ProQuest Dissertations & Theses, 2024

Material Info  

126 p.

General Note  

Source: Dissertations Abstracts International, Volume: 86-06, Section: B.

General Note  

Advisor: Bolt, Daniel.

학위논문주기  

Thesis (Ph.D.)--The University of Wisconsin - Madison, 2024.

Abstracts/Etc  

요약Self-report rating scales are widely used for measuring noncognitive psychological constructs such as personality and attitudes, where response style behavior has been documented to be widely present. Defined as the stylistic tendencies in respondents' use of rating scales, response styles are theoretically independent of the intended-to-be-measured content trait(s) and thus frequently pose a threat to the validity of the instrument. From the item response theory (IRT) perspective, such a detrimental effect is primarily due to violation of the local independence assumption among category selection, which is illustrated with a real data example in the introductory chapter of the dissertation. This thesis investigates the influence of response style on various psychometric analyses using both empirical datasets and simulation studies. These analyses range from traditional person-appropriateness indices to the increasingly popular topic of response time modeling.The first chapter outlines the three empirical datasets used in the dissertation, along with their specific features that made them suitable for the study of response style. In the second chapter, different person-fit indices are compared in terms of their performance when response style is present, based on the datasets in the first chapter. In the third chapter, response time is introduced as an additional outcome for the modeling of noncognitive item responses, with an extensive methodological review of existing approaches. The fourth chapter entails a simulation study investigating the theory that response style is the potential cause of asymmetry effects that Molenaar et al. (2021) observed in the "Inverted-U" response time model. A new response time model that incorporates the response style factor is subsequently proposed in the fifth chapter, followed by a comparison of model fit between the new model and the original models when applied to empirical datasets.The last chapter offers a reflective summary of the research questions covered by the current thesis while highlighting some recommendations for future directions in studying response style. Overall, the findings throughout the thesis demonstrate how response style factors can interfere with measurement in multiple ways and suggest possible solutions to minimize confounding effects on IRT parameters. Collectively, this work demonstrates the significant role of response style in noncognitive assessments and provides methodological guidelines for handling rating scale data in practice.

Subject Added Entry-Topical Term  

Educational psychology.

Subject Added Entry-Topical Term  

Cognitive psychology.

Subject Added Entry-Topical Term  

Counseling psychology.

Index Term-Uncontrolled  

Item response theory

Index Term-Uncontrolled  

Noncognitive psychological constructs

Index Term-Uncontrolled  

Self-report rating scales

Index Term-Uncontrolled  

Psychometric analyses

Added Entry-Corporate Name  

The University of Wisconsin - Madison Educational Psychology

Host Item Entry  

Dissertations Abstracts International. 86-06B.

Electronic Location and Access  

로그인을 한후 보실 수 있는 자료입니다.

Control Number  

joongbu:657547