Predicting student success in an introductory programming course

Abstract

Abstrac tThis paper examines to what extent a student's aptitude in computer programming may b e predicted through measuring certain cognitiv e skills, personality traits and past academi c achievement . The primary purpose of this study was to build a practical and reliable model fo r predicting success in programming, with hope s of better counseling students . Results fro m correlating predictor variables with a student' s final numerical score confirmed past studie s which showed the diagramming and reasoning test s of the Computer Programmer Aptitude Battery an d a student's GPA to be the predictors mos t closely associated with success . A multipl e regression equation developed from 5 predictor s correctly classified 61 of 79 students (77 .2% ) into low and high aptitude groups .

Mathematics Ability and Anxiety, Computer and Programming Anxieties, Age and Gender as Determinants of Achievement in Basic Programming

“…Previous researches indicate that students with programming background particularly in structured programming [7,8], investigative personality [9] mathematical background [10,11,12,13] have better programming achievement. Some other researchers found that scores in aptitude tests, prior records of academic achievement (such as school GPA) and effort or self motivation explain significantly the variations in programming achievement of students [14,15,16]. Specifically studies on cognitive factors' prediction of computer programming achievement revealed that demographic, cognitive and academic variables and computer exposure strongly predicted class performance [15,16].…”

“…Some other researchers found that scores in aptitude tests, prior records of academic achievement (such as school GPA) and effort or self motivation explain significantly the variations in programming achievement of students [14,15,16]. Specifically studies on cognitive factors' prediction of computer programming achievement revealed that demographic, cognitive and academic variables and computer exposure strongly predicted class performance [15,16]. Other studies considered variables such as learning styles [2], self efficacy [18], comfort level [19,20], personality type [21,22], mental model [7] as positive predictors of programming achievement.…”

Predicting introductory programming performance: A multi-institutional multivariate study

“…Austin (1987), Barker and Unger (1983), Gibbs (2000), Hostetler (1983), Kurtz (1980), and Mayer et al (1986) have investigated certain cognitive factors, including cognitive style and abstract reasoning ability and provide useful insights into the role of cognition in learning to program. In recent studies researchers have examined various psychological factors including students' perceived comfort-level when learning to program.…”

Will Johnny/Joanie Make a Good Software Engineer? Are Course Grades Showing the Whole Picture?

“…Several studies have been performed on predicting success in a Computer Science course or major, particularly in early CS courses [18,23,16,20]. Alexander et al performed a case study on predicting future success in a college computer science curriculum based on high school experiences and grades.…”

“…Here, each error of commission is counted once, and each error of omission is counted k times, where k is the ratio of the effort needed to fix errors of each type. In the case study, we have looked at four such ratios, covering a reasonably large span of possibilities: k = 1, 4,8,16. When k=1, we assume that one error of omission "costs" exactly one error of commission, when k=4, we assume that one error of omission costs four (4) errors of commission, etc.…”