학술논문

Students' Satisfaction of Blended Problem-Based Learning: An Academic Experience at Kwame Nkrumah University of Science and Technology.
Document Type
Article
Source
Education Research International; 10/26/2023, p1-8, 8p
Subject
Problem-based learning
Blended learning
Innovations in higher education
Self-managed learning (Personnel management)
Students
Satisfaction
Language
ISSN
20904002
Abstract
Implementation of problem-based learning has been a challenge in most African institutions of higher learning including Kwame Nkrumah University of Science and Technology (KNUST) due to deficiencies in the secondary education system. This article seeks to investigate the effect of blended problem-based learning (BPBL) practices on students' satisfaction at KNUST during the Enhancing Entrepreneurship, Innovation and Sustainability in Higher Education in Africa (EEISHEA) 2019–2022 project. In line with this objective, a survey instrument was prepared and validated and a multistage sampling technique with an effective sample size of 1,304 students (621 female and 883 males), 71 lecturers, and 16 technicians across the six colleges of KNUST were obtained. The generalized linear models, simple and multiple logistic regression, were estimated from the data to compare effect of teacher-centered and BPBL paradigms on students' satisfaction. To prevent overfitting, a cross-validation was performed on the fitted models. Receiver operating curve and area under curve were employed to ascertain the predictive power of the models. Results indicate about 67.35% of students posited that lecturers have incorporated some problem-based learning (PBL) elements into their course delivery. About 72.13% of respondents submitted that the integration of the PBL ignited their self-directed and collaborative learning and improved learner's problem-solving skills. Furthermore, the fitted models suggest that BPBL, gender, level of program, and college influence students' satisfaction. The odds of satisfaction of an individual in the BPBL group is 7.4 times higher than respondents in the traditional group with a confidence interval for the odds ratio (OR), (OR 7.4, 95% CI: 4.31–13.16). [ABSTRACT FROM AUTHOR]