Difficulty in predicting performance in a project-based learning program

Abstract

Project-based engineering programs, where technical learning is contextualized with design projects, provide an alternative to traditional engineering programs. Currently, one such upperdivision program relies heavily on students' lower-division pre-engineering coursework as part of the application process, which is usually taught in lectures and formal, pre-defined lab experiences. However, it is not clear that a student's success in lecture-based courses would predict success in project-based courses. Thus, it is important to study the relationship between potential performance predictors and actual performance. Not only will this allow for refinement of the selection criteria for such programs, but it will also serve to inform those who are educating and advising potential applicants for these types of programs. Broadly, as engineering design projects are core to early career success, this work could have implications for all engineering curricula. This paper examines the relationship between students' performance in their pre-engineering coursework and their performance in the upper-division project-based program. The correlations between student pre-engineering academic records and their upper division performance are reported as potential performance indicators for success in a projectbased engineering program.