Proceedings of the 2nd International Conference of Teaching and Learning (ICTL 2009) INTI University College, Malaysia

PREDICTING ACADEMIC PERFORMANCE OF ENGINEERING DIPLOMA STUDENTS Adele H.T. Kam1 and P.C. Ch’ng2 INTI International College Penang, Malaysia ([email protected]; [email protected])

ABSTRACT The main purpose of this study is to explore whether the Malaysian Certificate of Education (SPM) subject grades are reliable predictors for future performance in an engineering diploma program. The need for such a study arises because the admission requirements for the program, based primarily on the SPM grades, are not very high. There is a high variability in the academic achievement of students entering the program, leading to significant attrition rates. Secondary objectives of this study include exploring relationships between first semester performance and the overall performance in the program. Students may gauge their probability of success in the program from their first semester results. The study group consisted of 453 students, sampled using the cluster sampling method, at INTI International College Penang, Malaysia. These students were enrolled for the Diploma in Electrical and Electronic Engineering Program. The findings revealed that SPM grades for science subjects, namely, Physics, Chemistry and Biology and the grade for Additional Mathematics are strong predictors for achievement in the program. The performance in the first semester courses, namely Basic Electrical Technology and Introduction to Computer Systems and Programming, could also be used to predict the overall performance in the program. Prediction models using SPM subject grades, SPM cumulative aggregate and first semester subject marks were generated to predict overall performance, indicated by cumulative average (CAVG) scores. These models would be useful for counselling prospective students as well as for remediation considerations for weak students.

KEYWORDS Academic performance, Engineering diploma, Predictor, Prediction model, SPM grades, CAVG

INTRODUCTION There are reportedly more than 500 private higher learning institutions in Malaysia offering degrees and diplomas in various fields of study. The demand for private higher education in Malaysia is spurred largely by the country’s rapid economic growth. The passing of the Private Higher Educational Institutions Act (PHEIA) in 1996 also opened opportunities for the establishment of colleges offering tertiary programmes that could be completed in a shorter time after the Malaysian Certificate of Education (SPM) (Ayob and Yaakub, 1999). The SPM is a national examination equivalent to the British General Certificate of Secondary Education (GCSE). The number of students enrolled for the engineering undergraduate and diploma programs in private higher education institutions has been increasing, in tandem with the greater accessibility of higher education. The Malaysian Qualifications Agency (MQA), a statutory body in Malaysia which accredits academic programs provided by private higher education institutions, has set a minimum entry requirement of 3 SPM level credits (which must include

1

Mathematics and one science subject) and a pass in English for admission into the engineering diploma program. The lower admission requirement (compared to public universities), coupled with the increase of college enrolment, produces a greater variability in the academic achievement of students attempting engineering studies in private colleges. This variability has led to a significant number of students dropping out of the program or requiring extended time to complete the studies. Consequently, the ability to predict achievement in engineering studies would be invaluable for two good reasons. Firstly, it would serve as an indicator to potential applicants of their probability in succeeding in the program. Secondly, it allows the identification of students who may not be able to meet the demands of engineering studies, thus enabling remedial interventions that could help improve their chances of success. The SPM level achievement is the main admission criteria for most of the locally offered diploma and pre-university programs. If a student did well in the SPM examinations, would he also do well in tertiary education? Are SPM results a reliable predictor for academic performance in engineering studies? To date, very few studies have been conducted to answer the latter question. Elsewhere, studies have been conducted to seek correlation between higher education achievement and high school performance or Grade Point Average (GPA), A-Level achievement, Scholastic Achievement Test (SAT) scores and such. However, differing conclusions were found through some of the studies. Levin and Wyckoff (1990), French et al. (2005), Klomegah and Yao (2007) and Hall et al. (2008) found that High School GPA significantly contributed to college cumulative GPA (CGPA). However, researchers like Todd (2001) and Adamson and Clifford (2002) concluded that A-level grades and school examinations were poor predictors of higher education performance. Can Malaysia’s high school examination, therefore, predict college academic performance? With this in mind, the main objective of this study is to verify whether SPM results can be used as a predictor for achievement in an engineering diploma program. Specific research questions include: 1. Are SPM grades related to performance in the engineering diploma program? 2. Are first semester course achievements related to performance for the overall program? 3. Can a reliable model be developed to predict performance for the engineering diploma program?

METHODOLOGY Research Method A correlation research approach was used to investigate the relationship between student engineering course achievements and their SPM subjects’ grades. Participants The sample consisted of 453 college students enrolled in the Diploma in Electrical and Electronic Engineering program at INTI International College Penang from 2001 - 2009. 209 of these students have completed the program. The remainder consisted of students who are still in the program as well as those who have withdrawn. SPM grades, first semester subject

2

marks and CAVG (cumulative average) scores were obtained from the college admission records. 20% of the sample was female while 80% of them was male. A bivariate Pearson’s correlation was applied between the students’ CAVG scores and their SPM subjects’ grades. Multiple regression analysis was used to obtain the most suitable regression equivalent in explaining the students’ CAVG scores. Variables The students’ CAVG scores for the first semester and final semester were specified as the dependent variables. It ranged from 0 to 100. The SPM subjects such as English Language (BI), Malay Language (BM), Biology (BIO), Physics (PHY), Chemistry (CHEM), Mathematics (MATH) and Additional Mathematics (AM) were included as the independent variables. SPM subjects’ grades ranges from the highest grade (grade 1) to the lowest grade (grade 9). Traditionally, Grades 1 and 2 represent distinction, Grades 3 to 6 are for credit, Grades 7 and 8 for pass and Grade 9 represents fail. The first semester engineering subjects such as Basic Electrical Technology (EGE 161), Engineering Mathematics 1 (MAT 163), Introduction to Computer Systems and Programming (CSC 167) and Technical English (ENL 162) marks were included as the independent variables. All the four engineering subject marks ranged from 0 to a 100 marks. The significance level for all the statistical results in the study was accepted to be 0.05 and all the results were tested two-ways. For statistical analysis, the software used was SPSS 10.1.

FINDINGS Insightful relationships emerged from the data analysis. Descriptive statistics was presented first. This was followed by the correlations between the students’ CAVG scores and the other variables. Results from the multiple regression analysis were be stated and lastly, was a summary of the research findings. Data Analysis Table 1. Descriptive Statistics of CAVG Scores and SPM Subject Grades VARIABLES CAVG - 1st Semester CAVG - Final Semester BI BM MATH AM PHY BIO CHEM

Mean 57.3415 63.8316 3.0177 4.2870 1.5872 3.4406 3.8326 4.8182 4.2810

Std. Deviation 16.71208 10.34076 2.06416 2.04847 1.26893 2.35614 2.16916 2.21497 2.43906

N 453 209 453 453 453 438 436 363 427

3

Table 1 shows the descriptive statistics of the measurements. The mean values for the SPM subjects refer to the mean grades (which range from 1 to 9 as explained earlier). It can be observed from Table 1 that the engineering programme attracts students with good results in Mathematics, English Language, Additional Mathematics and Physics (above Grade 4). Table 2. Correlation of 1st Semester CAVG Scores and Final semester CAVG Scores with Each Other and with Individual SPM Subjects CAVG semester 1 r p CAVG.800 .000 Final AM -.674 .000 BI -.365 .000 BIO -.672 .000 BM -.421 .000 CHEM -.687 .000 MATH -.534 .000 PHY -.694 .000

CAVG-Final r

p

-.548 -.368 -.596 -.388 -.632 -.268 -.660

.000 .000 .000 .000 .000 .000 .000

The bivariate Pearson’s correlation coefficient was employed to determine the degree of relationship between the students’ first semester CAVG and the final semester CAVG scores. The degree of relationship between each of these CAVG scores with the individual SPM subject grades was also examined. All the SPM subject grades (BI, BM, BIO, PHY, CHEM, MATH, AM) were significantly correlated with each other and with the first and final semester CAVG scores at the p =.000 level (refer to Table 2). Highest correlations were found between first semester CAVG and PHY (r = -0.694), CHEM (r = -0.687), AM (r = 0.674) and BIO (r = -0.672). Weaker correlation was found between first semester CAVG and BM (r = -0.421), BI (r = -0.365), MATH (r = -0.534). A similar pattern of relationship is observed between the final semester CAVG scores and the SPM subjects. Very strong correlation was detected between the students’ first semester CAVG and their final semester CAVG scores (r = 0.8; p