Exploring latent factors in teaching statistical literacy using non-Bayesian factor analysis and Bayesian factor analysis

• Main Authors: Diokno, Felipe Jose O., Jumangit, Ryan U.
• Format: text
• Language: English
• Published: Animo Repository 2010
• Subjects: Bayesian statistical decision theory; Transformative learning; Factor analysis; Mathematics
• Online Access: https://animorepository.dlsu.edu.ph/etd_bachelors/5342
• Institution: De La Salle University

De La Salle University-Manila has been a fixture in the Philippines for over a hundred years. In the year 2006, the University changed its pedagogy for several general education courses to what is known as transformative learning. Transformative learning was previously studied using non-Bayesian factor analysis by two professors in the University. In this paper, the researchers were able to conduct a study regarding the topic using the Bayesian approach. In order to name the factors with traditional Bayesian Factor Analysis and with Bayesian Factor Analysis with jointly distributed means and loadings, data were taken from the paper conducted using the non-Bayesian approach and this data were run through the program Matlab. Factor scores were generated by the researchers in doing so. It was found through the comparison of the factor scores of the Bayesian approach and the non-Bayesian approach that the factors produced were highly similar with the exception of a few variables, and that the factor scores of the non-Bayesian approach were higher than the ones achieved through the Bayesian approach.