Skip to Main Content
 

Global Search Box

 
 
 
 

Files

File List


25730.pdf (3.62 MB)

ETD Abstract Container

Abstract Header

Bayesian Approximate Measurement Invariance Approach

Wang, Shanshan
http://rave.ohiolink.edu/etdc/view?acc_num=ucin150512542310924

Abstract Details

2017, PhD, University of Cincinnati, Education, Criminal Justice, and Human Services: Educational Studies.
Measurement invariance is an important prerequire for comparing factor means across many groups and over time. This methodological case study sought to apply the Bayesian Approximate Measurement Invariance Approach (i.e., the random items effect model) to study 15-year-old students’ performance in PISA 2009 and PISA 2012 mathematics tests. This methodological case study served as the first step to demonstrate the flexibility and capacity of the random items effect model to compare factor means and variances not only across many groups but also over time. The subsequent Monte Carlo simulation study was carried out to check the stability of the random items effect model and showed good model fit results. The results of this dissertation showed that the Bayesian Approximate Measurement Invariance Approach offers a valuable alternative to the traditional multigroup confirmatory factor analysis approach to testing invariance for cross-cultural studies where a confirmatory factor analysis model with exactly zero cross-loadings is often unrealistic. More importantly, this dissertation addressed some practical issues (e.g., sample size of Level 2 unit) when using the random items effect model and the results of the simulation study can be used as general guidelines that ensure consistent and unbiased estimation of parameters and their standard errors and minimize potential estimation problems.
Benjamin Kelcey, Ph.D. (Committee Chair)
Ying Guo, Ph.D. (Committee Member)
George Richardson, Ph.D. (Committee Member)
Christopher Swoboda, Ph.D. (Committee Member)
128 p.
Educational Evaluation
Measurement invariance; IRT; Random items effect

Recommended Citations

Citations

  • Wang, S. (2017). Bayesian Approximate Measurement Invariance Approach [Doctoral dissertation, University of Cincinnati]. OhioLINK Electronic Theses and Dissertations Center. http://rave.ohiolink.edu/etdc/view?acc_num=ucin150512542310924

    APA Style (7th edition)

  • Wang, Shanshan. Bayesian Approximate Measurement Invariance Approach. 2017. University of Cincinnati, Doctoral dissertation. OhioLINK Electronic Theses and Dissertations Center, http://rave.ohiolink.edu/etdc/view?acc_num=ucin150512542310924.

    MLA Style (8th edition)

  • Wang, Shanshan. "Bayesian Approximate Measurement Invariance Approach." Doctoral dissertation, University of Cincinnati, 2017. http://rave.ohiolink.edu/etdc/view?acc_num=ucin150512542310924

    Chicago Manual of Style (17th edition)

ucin150512542310924
211
© 2017, all rights reserved.
This open access ETD is published by University of Cincinnati and OhioLINK.