Name: Na Liu

Masters Thesis Proposal Meeting

Date: Monday, May 6th, 2024

Time: 10:30 am - 12:00pm

Location: Virtual, Meeting link click here.  

Thesis Chair/Advisor:

James Roberts, Ph.D. (Georgia Tech)

Thesis Committee Members:

Susan Embretson, Ph.D. (Georgia Tech)

Rick Thomas, Ph.D. (Georgia Tech)

Title: Enhancing Precision in the Generalized Graded Unfolding Model (GGUM) Using Multiple Responses Indicative of Item Location 

Abstract: This proposal describes an extension of the Generalized Graded Unfolding Model (GGUM) to enhance precision in estimating item locations on a latent continuum for non-cognitive psychological constructs, such as attitudes, emotions, and personality. The traditional GGUM requires relatively large sample sizes than the typical samples found in psychological research; N>750 has been previously recommended. The new hybrid model integrates the Method of Successive Intervals (MSI) with GGUM, aiming to reduce sample size requirements without compromising precision. This research involves simulating data and examining parameter recovery using a Markov Chain Monte Carlo (MCMC) estimation procedure.  Additionally, real data on attitudes on abortion collected from undergraduate students at the Georgia Institute of Technology will also be examined with the new model.  Anticipated outcomes include more precise item location estimates and the ability to apply GGUM in studies with smaller sample sizes, thus broadening its applicability in psychological research.