Improving College Students' Learning Performance In Computer Programming: The Case of Using The Polya Model

https://doi.org/10.37178/ca-c.21.5.081

Abstract

Many studies have shown that students with poor problem-solving skills are prone to failing programming courses. To help mitigate this problem, several problem-solving methods have been recommended, including the Polya Model. Therefore, this study aims to determine the impact of this model on matriculation college students’ performances in learning programming algorithms. A quasi-experimental study with pre-test post-test research design was conducted to assess the performances in programming algorithm among a group of 30 college students, who were selected using the random sampling method. They were assigned into an experimental group and a control group who underwent an intervention treatment that lasted for seven weeks on two groups of students: control and treatment. The experimental and control groups learned programming algorithms using the Polya Model and conventional method, respectively.  The research instrument used for the pre-test and post-test consisted of six subjective questions with three levels of difficulty based on Bloom’s taxonomy. Descriptive statistics and a series of t-tests were used to analyze the data. The results of the t-test showed that there was a significant difference in the mean scores of the post-test measurements between the two groups, which favored the experimental group, especially for questions with a higher level of difficulty. This finding suggests that the Polya Model is an effective problem-solving technique that programming students need to learn to help improve their performances in learning programming algorithms at the collegiate level.