Impact of Work-Integrated Learning on Career Readiness Through the Mediating Role of Self-Efficacy and Learning Satisfaction: A Longitudinal Growth Modeling Approach
Description
The participants were undergraduate students undertaking a credit bearing and compulsory work-integrated learning (WIL) placement as an integrated part of their degree programme at Anonymized University. A multistage sampling design was used. In the first stage, eight academic departments that include WIL as a mandatory, credit bearing component of the undergraduate curriculum were purposively selected; departments that do not have students participate in WIL, or have voluntary, non-credit placements, were excluded because they did not have a placement exposure, assessment, or experience that was structured and comparable across programmes. The second stage involved students enrolled in the 7th and 8th semester of these departments were selected by proportionate stratified random sampling with departments as strata. Stratified selection was used to make the sample representative of the departmental composition of the target population, and the participants in each stratum were selected using the proportional-allocation equation nh = (Nh / N) × n, where Nh is the number of individuals in stratum h, N is the number of individuals in the target population, and n is the total number of individuals in the sample. Proportional allocation was used in such a way that every department provided participants based on the population of that department as a percentage of the total population (Cochran, 1977; Lohr, 2021). Based on a medium effect size of f² = 0.15, which is typical in educational research (Cohen, 1988); conventional benchmarks for latent growth modelling (Bollen & Curran, 2006); and an alpha of .05 and statistical power of .95, the minimum sample required was 395, which was well surpassed by the achieved sample, and also surpassed the conventional benchmarks for the latent growth modelling, providing sufficient power to detect the cross-process slope couplings, and to support bias-corrected bootstrapping of indirect effects. Of the students with four-wave data, 900 (90.0% of the sample) had complete data at placement commencement (T1, n = 944), 868 (9.6% of the sample) at the midpoint of the placement (T2), 868 (9.2% of the sample) in the late, intensive placement phase (T3), and 850 (8.9% of the sample) at placement completion (T4); analyses were thus limited to students who had complete data at placement commencement, placement midpoint, late placement, and placement completion, and a final analytic sample of 850 students remained; all students participated in the final analytic sample and had an 89.0% complete-case retention rate. Attrition was slow and even rather than focused at any particular wave, and the high retention is most likely a result of the fact that the survey was built into the compulsory placement curriculum and that completion of the survey was an integral part of a prescribed programme point rather than a choice of study.
Files
Institutions
- University of SargodhaPunjab, Sargodha