Assessing the psychometric properties of University of Cape Coast teaching practice observation protocol using generalizability and item response theories
Loading...
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Cape Coast
DOI
Abstract
The study aimed to assess the psychometric properties of the teaching practice observation of the University of Cape Coast through the application of G-theory and the MFRM of the IRT. The study which was rooted in the positivists’ paradigm adopted the hybrid quantitative design in its investigation. First, the three-facet partially nested random balanced design, with occasion nested in rater nested in person, all crossed with items denoted by (o: r: p) × i, was employed for the G-theory, whereas the hybrid model of MFRM was also adopted with student-teacher, occasion, rater, and item as the facets. The population was made up of 4,379 cases of observations done on 1,170 student-teachers by 304 supervisors for the 2020/2021 academic year regular off-campus teaching practice exercise. The “aov” function of R programming language and EduG were used for the G-theory whereas FACET software was used for MFRM analysis. The residual, person-by-item interaction and persons contributed the most variations. The generalizability and dependability coefficients for the overall inventory as well as the dimensions of the inventory were at least .90, except for the professional commitment dimension which was .51 for both. The raters differed in the severity of their ratings. Additionally, the teaching practice observation protocol holistically had low-scale functioning quality. It was concluded that the regular off-campus teaching practice exercise measured student-teachers’ teaching competencies with high precision. The study recommends that CTPD consider modifying and calibrating the observation protocol to suit students’ competency.
Description
xv, 243p. :ill.
