Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


InhanKang_Dissertation.pdf (17.91 MB)

ETD Abstract Container

Abstract Header

Psychometric Process Modeling: A Modeling Framework to Study Intra-individual Processes Underlying Responses and Response Times in Psychological Measurement

Kang, Inhan
http://rave.ohiolink.edu/etdc/view?acc_num=osu1658353404080671

Abstract Details

2022, Doctor of Philosophy, Ohio State University, Psychology.
Despite their successful accounts for latent structures of observations and individual differences as a function of variations in latent variables, psychometric models are oblivious of the intra-individual processes of a respondent during the measurement procedure. Without theoretical accounts for what cognitive components drive the response processes and how this causality arises, some important questions on validity and measurement remain unanswered. To address this issue, we propose psychometric process modeling that integrates psychometric models with decision-making theories in perceptual and cognitive psychology to study the internal processes of an individual in psychological and educational measurement. This approach redefines latent variables as cognitive components in psychological processes of measurement and establishes their causal relationships with manifest variables. We provide examples of psychometric process models and discuss their theoretical implications and practical applicability. The first three studies in the dissertation show that psychometric process models provide a theoretical explanation of the existence and cognitive sources of conditional dependence in and between responses and response times (RTs) and a method to derive practically useful information and diagnosis for respondents and items. Another study demonstrates how we can develop psychometric process models for ordinal responses and RTs from personality and attitude measurement, based on various confidence judgment theories and psychometric models. The resulting models have different representations of intra-individual processes of measurement and we can find a better theoretical account by comparing these models. We also present several ongoing projects on psychometric process modeling, including 1) a proposal for a racing accumulator framework to reinterpret traditional psychometric models as different information processing rules and extend them to develop process models for personality and attitude measurement and 2) a discussion on process models with discrete latent variables.
Roger Ratcliff (Advisor)
Paul De Boeck (Committee Member)
Brandon Turner (Committee Member)
371 p.
Psychology
Psychometrics, Quantitative Psychology, Psychological Process, Decision-making, Sequential Sampling Models, Evidence Accumulation, Diffusion Model, Latent Variable Modeling

Recommended Citations

Citations

  • Kang, I. (2022). Psychometric Process Modeling: A Modeling Framework to Study Intra-individual Processes Underlying Responses and Response Times in Psychological Measurement [Doctoral dissertation, Ohio State University]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=osu1658353404080671

    APA Style (7th edition)

  • Kang, Inhan. Psychometric Process Modeling: A Modeling Framework to Study Intra-individual Processes Underlying Responses and Response Times in Psychological Measurement. 2022. Ohio State University, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=osu1658353404080671.

    MLA Style (8th edition)

  • Kang, Inhan. "Psychometric Process Modeling: A Modeling Framework to Study Intra-individual Processes Underlying Responses and Response Times in Psychological Measurement." Doctoral dissertation, Ohio State University, 2022. http://rave.ohiolink.edu/etdc/view?acc_num=osu1658353404080671

    Chicago Manual of Style (17th edition)

osu1658353404080671
293
© 2022, all rights reserved.
This open access ETD is published by The Ohio State University and OhioLINK.