Past research has shown that student problem‐solving skills may be used to determine student final exam performance. This study reports on the relationship between student perceived problem‐solving skills and academic performance in introductory programming, in formative and summative programming assessment tasks. We found that the more effective problem solvers achieved better final exam scores. There was no significant difference in formative assessment performances between effective and poor problem solvers. It is also possible to categorize students on the basis of problem‐solving skills, in order to exploit opportunities to improve learning around constructivist learning theory. Finally, our study identified transferability skills and the study may be extended to identify the impact of problem solving transfer skills on student problem solving for programming.

In this article, we report the results of the impact of prior programming knowledge (PPK) on lecture attendance (LA) and on subsequent final programming exam performance in a university level introductory programming course. This study used Spearman's rank correlation coefficient, multiple regression, Kruskal-Wallis, and Bonferroni correction statistical techniques via SPSS software to analyze the student data for academic years 2012, 2013, and 2014 to test the hypotheses. Only LA, PPK, and final exam (FE) scores were considered for this analysis. Research suggests that PPK influences student LA and FE performance. Similar analysis was conducted on the impact of LA on FE results regardless of students' PPK levels. The results delivered mixed conclusions. Furthermore, the correlation coefficient results indicated that LA and FE were negatively correlated. However, the coefficient value was not sufficiently statistically significant to conclude that LA does not have an impact on FE results. On the other hand, the results of average LA on student FE results, with linear regression results, revealed that nonattendance of lectures had no effect on student performance in the FE. The multiple regression results of our study identified that, PPK in a regression model, is a good fit of the data, but LA in a regression model is not a good fit of the data.

Abstract-In this paper, the correlation between lecture attendance and assessment tasks on final exam performance of introductory programming students has been analyzed to identify if lecture attendance, and completion of in-class and take home formative assessment tasks affects student performance in the final examination. In this study, only lecture attendance, homework exercises and class demonstration scores, and final exam marks have been considered. This study used Spearman's Rank correlation coefficient and multiple regression techniques via SPSS software to analyze the student data of the academic years 2012, 2013 and 2014 of an introductory programming course to test the hypotheses. It is found that, there is a significant correlation between homework exercises and final exam scores. However, formal lecture attendance and final exam performance were negatively correlated. Moreover, multiple regression results of assessment tasks such as homework exercises, class activities and lecture attendance on final exam scores, did not provide any significant value to support the statement "Marks achieved in homework, class demonstrations, and lecture attendance, have a significant positive impact on final examination results".

No abstract

This article presents a study aimed at examining the novice student answers in an introductory programming final e-exam to identify misconceptions and types of errors. Our study used the Delphi concept inventory to identify student misconceptions and skill, rule, and knowledge-based errors approach to identify the types of errors made by novices in Python programming. The students' responses to each question were scrutinized by using the Delphi concept inventory, heuristic-analytic theory, and neo-Piagetian theory of cognitive development for qualitative data analysis. Moreover, the motivation for this exploratory study was to also address the misconceptions that students held in programming and help educators to redefine the teaching methods to correct those alternative conceptions. Student misconceptions were spotted in list referencing and inbuilt functions in Python. In a further quantitative analysis, the study found that students who had misconceptions made knowledge errors and failed to complete the coding tasks. Surprisingly, and coincidentally, it was identified that only a few students were able to write code related to mathematical problems.