Improving automatic code assessment

Authors:
Brad Vander Zanden;Michael W. Berry
Affiliations:
University of Tennessee, Knoxville, TN;University of Tennessee, Knoxville, TN
Venue:
Journal of Computing Sciences in Colleges
Year:
2013

Citing 5
Cited 0

Generation of problems, answers, grade, and feedback---case study of a fully automated tutor

Journal on Educational Resources in Computing (JERIC)
Two experiments using learning rate to evaluate an experimenter developed tool for splay trees

Proceedings of the 42nd ACM technical symposium on Computer science education
MyProgrammingLab with Pearson eText -- Access Card -- for Problem Solving with C++

MyProgrammingLab with Pearson eText -- Access Card -- for Problem Solving with C++
Experiments with algorithm visualization tool development

Proceedings of the 43rd ACM technical symposium on Computer Science Education
Codeassessor: an interactive, web-based tool for introductory programming

Journal of Computing Sciences in Colleges

Quantified Score

Hi-index	0.00

Visualization

Abstract

In Fall Semester 2012 we started using an automatic code assessment system named CodeAssessor to grade students' coding problems on exams. This paper describes some of the lessons we have learned in using automatic code assessment and the improvements to CodeAssessor we have made as a result of them. The students' results on the exams have improved from a median of 8/25 on the first exam given in Fall Semester 2012 to a median of 19/25 on the first exam given in Spring Semester 2013.