Computer-Based Assessment of Student Performance in Programing Courses

In this study, the score equivalence of computer-based assessment (CBA) and paper-and-pencil testing (PPT) in an undergraduate course of introductory computer programing at the National Technical University of Athens' School of Chemical Engineering, was examined. The research was carried out with 211 students participating, and was realized by a specially constructed CBA platform as well as the classical PPT. Both types of assessment included the same questions of multiple-choice type, completing and correcting code in Fortran, as well as writing Fortran code from scratch. The results showed that, in general, students performed better and received higher scores under CBA compared to PPT. In particular, the very good students benefited from the advantages that the computer compiler offers in debugging code and completing code exercises and, thus, received relatively higher scores in these parts of the test. Moreover, switching answers and devoting more time-on-task was found to be beneficial for the students' performance. Overall, the findings of this study revealed that the actual programing abilities of the students are better evaluated using such CBA tests. (c) 2011 Wiley Periodicals, Inc. Comput Appl Eng Educ 21: 671-683, 2013