Effects of problem-based learning on cognitive and affective variables in learning statistics among university students

• Main Author: Bayat, Sahar
• Format: Thesis
• Language: English
• Published: 2012
• Online Access: http://psasir.upm.edu.my/id/eprint/67270/1/IPM%202012%209%20IR.pdf; http://psasir.upm.edu.my/id/eprint/67270/

Two phases of experimental study with randomized post-test only control group design were conducted to investigate the effects of problem-based learning on the cognitive and affect variables in learning statistics among university students. Experiments in Phase I and Phase II were conducted over a duration of 10 weeks lessons, each week consisted of three hours lesson. In the first phase of the study, two experimental groups, namely the Traditional Problembased Learning (PBL-Tr) and the Web-basedProblem-based Learning (PBLWeb), and one control group were involved. The subjects of the study were graduate students who enrolled in an Educational Statistics course at a public university in Malaysia. The PBL-Tr group underwent learning using the traditional problem-based learning approach with guided questions and the PBL-Web group underwent learning using problem-based learning approach with guided questions by using a website; and the control group underwent learning using conventional instruction. In Phase I, the three groups were compared on cognitive variables (overall statistical performance, statistical conceptual knowledge and statistical procedural knowledge) and affect variables (metacognitive awareness and motivation). In Phase II another batch of graduate students who then enrolled in an Educational Statistics course at a public university in Malaysia participated in the study. In Phase II, three groups of students were administered the respective treatments and tested once again for the same attributes as that in Phase I and additional measures such as number of of errors made by the subjects during solving test problems, students’ mental load assessment for all three groups and hence instructional efficiency index were obtained. Four instruments were used in this study, namely Statistics Performance Tests, Paas Mental Load Rating Scale, Metacognitive Awareness Inventory and Motivation toward Learning Survey Questionnaire. The data for Phases I and II were analysed by using independent t-test and planned comparison test, ANOVA, ANCOVA and repeated measure ANOVA tests. The study shows that the problem-based learning with guided questions approach enhanced students’ overall performance and induced higher levels of metacognitive awareness and motivation toward learning with less mental load invested during the learning. These findings indicated that the PBL with guided questions with two different modes: traditional and by using a website instruction is superior in comparison to the conventional instruction, hence implying that integrating the use of this approach in teaching and learning of mathematics was more efficient than the conventional instruction strategy. Further, students in experimental groups showed an overall favourable view toward integrating the use of the PBL with guided questions in the teaching and learning of Statistics. Even though some students experienced difficulties during PBL with guided questions at the beginning, they confirmed that PBL improved their understanding of statistics. Hence, it may be concluded that both forms of PBL were effective for student learning.