Interacting factors that predict success and failure in a CS1 course (original) (raw)
Related papers
Predicting and Improving Performance on Introductory Programming Courses (CS1)
2019
This thesis describes a longitudinal study on factors which predict academic success in introductory programming at undergraduate level, including the development of these factors into a fully automated web based system (which predicts students who are at risk of not succeeding early in the introductory programming module) and interventions to address attrition rates on introductory programming courses (CS1). Numerous studies have developed models for predicting success in CS1, however there is little evidence on their ability to generalise or on their use beyond early investigations. In addition, they are seldom followed up with interventions, after struggling students have been identified. The approach overcomes this by providing a web-based real time system, with a prediction model at its core that has been longitudinally developed and revalidated, with recommendations for interventions which educators could implement to support struggling students that have been identified. This t...
Programming: predicting student success early in CS1. a re-validation and replication study
Proceedings of the 23rd Annual ACM Conference on Innovation and Technology in Computer Science Education, 2018
This paper describes a large, multi-institutional revalidation study conducted in the academic year 2015-16. Six hundred and ninetytwo students participated in this study, from 11 institutions (ten institutions in Ireland and one in Denmark). The primary goal was to validate and further develop an existing computational prediction model called Predict Student Success (PreSS). In doing so, this study addressed a call from the 2015 ITiCSE working group (the second "Grand Challenge"), to "systematically analyse and verify previous studies using data from multiple contexts to tease out tacit factors that contribute to previously observed outcomes". PreSS was developed and validated in a longitudinal study conducted over a three year period (twelve years previous from 2004-06). PreSS could predict with near 80% accuracy, how a student would likely perform on an introductory programming module. Notably this could be achieved at a very early stage in the module. This paper describes a revalidation of the original PreSS model on a significantly larger multi-institutional data set twelve years after its initial development and looks at recent research on additional factors that may improve the model. The work involved the development of a fully automated end-to-end tool, which can predict student success early in CS1, with an accuracy of 71%. This paper describes, in detail the PreSS model, recent research, pilot studies and the re-validation and replication study of the PreSS model. CCS CONCEPTS • Social and professional topics → Computer science education; CS1;
A study to identify predictors of achievement in an introductory computer science course
Proceedings of the 2003 SIGMIS conference on Computer personnel research Freedom in Philadelphia--leveraging differences and diversity in the IT workforce - SIGMIS CPR '03, 2003
In the study reported on here, 65 prospective computer or information science majors (47 male, 18 female) worked through a tutorial on the basics of Perl. All actions were recorded and time-stamped, allowing us to investigate the relationship between six factors that we believed would predict performance in an introductory computer science (CS) course (as measured by course grade) and how much students would learn from the tutorial (as measured by gain score from pre-test to post-test). These factors are: preparation (SAT score, number of previous CS courses taken, and pre-test score), time spent on the tutorial as a whole and on individual sections, amount and type of experimentation, programming accuracy and/or proficiency, approach to materials that involve mathematical formalisms, and approach to learning highly unfamiliar material (string manipulation procedures). Gender differences with respect to these factors were also investigated.
Predicting outcomes in Introductory Programming using J48 classification
World Transactions on Engineering and Technology Education , 2017
In a computer science (CS) major, Introductory Programming becomes a substantial course, which determines whether students can complete that major or not. This study evaluates the correlation between student data with the students’ capacity to pass that course. Such correlation is exploited according to a data mining technique called J48. For each student, the work incorporates personal, prior education, admission and assessment data. Based on an evaluation of 41 pieces of student data, the national test score for mathematics in Indonesia presents the most promising attributes, followed by the admission test score. The results of this study are expected to provide a brief insight for CS lecturer and the university, so that they can handle emerging issues in CS education, especially, the low retention rate.
Predicting Student Success in CS2
Proceedings of the 53rd ACM Technical Symposium on Computer Science Education
A number of published studies indicate that many students who receive passing grades in CS1 may struggle in CS2. This can lead to higher attrition and failure rates in CS2, and perhaps also in subsequent courses in the curriculum. Many researchers have studied factors that lead to student success and common misconceptions that may arise in introductory courses. However, relatively little attention has been focused on the transition between CS1 and CS2. In this paper, we report on a study of a variety of types of CS1 exam questions, with the goal of finding questions that can predict a student's success (or lack of success) in CS2. Results indicate that code explanation and code completion questions can be especially good predictors, along with some code tracing questions. We discuss some of the factors that may make certain questions better predictors than others, in the process confirming some observations that have been reported by other researchers. The results from this experiment seem promising, and point to several possibilities for further research.
Educational data are increasing tremendously with little exploration by the education managers. Hidden knowledge can be discovered from the huge data sets available in educational databases through the use of data mining techniques. Educational data mining (EDM) is an emerging discipline, concerned with developing methods for exploring the unique types of data that reside in educational databases. One of the significant areas of the application of EDM is the development of student models for predicting students’ performances in their educational institutions. The focus of this work is to identify the optimal decision tree algorithms for predicting students’ performance in a computer programming course taken in 200 level based on their ordinary level results in Mathematics and Physics and their 100 level results in Mathematics and Physics courses. One hundred and thirty one (131) students’ records from computer science programme at Kwara State University (KWASU) between 2009 and 2013 were used. The attributes used are students’ ordinary level scores in Mathematics and Physics, 100 level results in Mathematics and Physics courses and the score in a 200 level computer programming course (CSC 203). C4.5 (known as J48 in WEKA) Classification and Regression Tree (CART), and Best-First Tree (BF Tree) decision tree algorithms were used in Waikato Environment for Knowledge Analysis data mining software to generate three classification models employed in predicting students’ performance in CSC 203. The results of these algorithms were compared using 10-fold cross validation method in terms of prediction accuracy and computational time. Our results showed that J48 tree has the highest prediction accuracy of 70.37% and least execution time of 0.02 seconds while CART and BFTree has prediction accuracy of 60.44% and 60.30% respectively and both having execution time of 0.22 seconds based on the data set used in this study. This study also revealed that previous knowledge of Mathematics and Physics both at Ordinary level and 100 level are essential determinants of students’ performance in a computer programming course.
Informatica
Data Mining is a growing field, a strand of which is Educational data mining (EDM). EDM is currently used to help institutions and students through creating accurate predictions that are considered in decision making. One of EDM's concerns is that of predicting students' academic performance and fundamental learning difficulties in a particular course. In fact, EDM can help computer science (CS)enrolled students to predict whether they can pass their courses without taking further action. An introductory programming course is usually the first challenging course faced by students in CS departments since a student's performance in such a course is highly based on their intellectual skills. This paper presents a real case study from one of Saudi Arabia's leading universities. This study used well-known prediction models-specifically, decision tree (DT), k-nearest neighbor (kNN), Naïve Bayes (NB), and support vector machine (SVM) models-to create a reliable prediction model for at-risk students in an introductory programming course using preliminary performance information showing their self-efficacy. The results of this study showed that the DT and SVM models yielded the best performance with the highest accuracy rate (99.18%). Furthermore, comparisons between the applied models were conducted with different evaluation metrics. Povzetek:Analizirana je uspešnost metod strojnega učenja pri predvidevanju uspešnosti izpita za prvi programerski predmet.
Can Early Programming Performance Predict Computer Science Students’ Success?
CogITo Smart Journal
Investigating the possibility of lower-level computer programming courses predicting future performance of computer science students has received a lot of attention from scholars. This study mainly aimed to predict the success of computer science students based on their performance in the first two computer programming courses, namely Computer Programming I and Computer Programming II. The study employed a quantitative correlational design. Six years of data from graduating students were analyzed. The results demonstrate that the better the grade on Computer Programming I and II, the shorter the study duration. When further analysis was conducted to find out whether gender diversity exists, the results demonstrated that in Computer Programming I and II, female students outperformed males. Statistically, this difference was only significant in Computer Programming I. A greater proportion of female students graduated on time, yet it is not statistically significant.
Predicting introductory programming performance: A multi-institutional multivariate study
Computer Science Education, 2006
A model for predicting student performance on introductory programming modules is presented. The model uses attributes identified in a study carried out at four third-level institutions in the Republic of Ireland. Four instruments were used to collect the data and over 25 attributes were examined. A data reduction technique was applied and a logistic regression model using 10-fold stratified cross validation was developed. The model used three attributes: Leaving Certificate Mathematics result (final mathematics examination at second level), number of hours playing computer games while taking the module and programming self-esteem. Prediction success was significant with 80% of students correctly classified. The model also works well on a per-institution level. A discussion on the implications of the model is provided and future work is outlined.