Pre-Academic Learning Self-Efficacy scale (PAL-SE) – Are the psychometric properties comparable across three higher education sectors in Denmark?

Research output: Contribution to conference without a publisher/journalAbstractResearchpeer-review

Abstract

The Motivated Strategies for Learning Questionnaire self-efficacy scale (MSLQ-SE) with an improved response scale was shown in 2017 (DOI: 10.1016/j.stueduc.2017.04.003) to consist of two separate scales providing measures of Specific Academic Learning Self-efficacy and Specific Academic Exam Self Efficacy (SAL-SE & SAE-SE), each fitting a Rasch model. Due to the course-specific nature of the scales, they do not lend themselves to studies of pre-academic self-efficacy or large-scale studies including many degree programs. In 2019 (DOI: 10.14786/flr.v7i3.503), the scales were adapted to target the first courses in the students’ future studies, thus measuring Pre-Academic Learning Self-Efficacy and Pre-Academic Exam Self-Efficacy (PAL-SE & PAE-SE, and their psychometric properties were investigated using graphical loglinear Rasch models (GLLRM) with a large sample of university students (N = 2686) admitted to nine degree program represented in two universities. For university students, the PAL-SE fitted a GLLRM with substantial local dependence between items 1 and 3 and differential item functioning (DIF) of item 2 relative to degree program, targeting varied from good to very good for subgroups defined by degree program, and average reliability was acceptable for statistical purposes (0.81). The two present studies aim to investigate the degree to which the psychometric properties of the PAL-SE scale are the same or differ for students in two other Danish higher education sectors, while employing the same methods as the study on university students to maximize comparability of results. Study (A) is conducted with a sample of students just admitted, but not yet commenced, to 22 different academy profession degree programs (N = 3757). For academy profession degree students, the PAL-SE scale fitted a GLLRM with weak to moderate local dependence between items 1 and 3 and items 2 and 3 as well as DIF for item 3 relative to degree program and DIF for item 1 relative to age, while targeting varied from poor to very good for subgroups defined by the DIF variables, and average reliability was 0.77. Study (B) is conducted with a sample of students likewise admitted but not commenced to the teacher degree program across all university colleges in Denmark (N = 1249). For teacher students the PAL-SE scale fitted a GLLRM with weak to moderate local dependence between items 1 and 3 and items 2 and 3, no DIF, and targeting was poor and reliability was 0.76.
Original languageEnglish
Publication date3 Nov 2021
Publication statusPublished - 3 Nov 2021
EventNordic Educational Research Association (NERA) - 2021 - Odense, Odense, Denmark
Duration: 3 Nov 20215 Nov 2021

Conference

ConferenceNordic Educational Research Association (NERA) - 2021
LocationOdense
Country/TerritoryDenmark
CityOdense
Period03/11/2105/11/21

Cite this