Assessing Experiential Learning in Construction Education by Modeling Student Performance

The typical engineering curriculum has become less effective in training construction professionals because of the evolving construction industry needs. The latest National Science Foundation and the National Academies report indicate that industry-valued skills are changing. The Associated General Contractors of America recently stated that contractors expect growth in all sectors; however, companies are worried about the supply of skilled professionals. Workforce development has been of a growing interest in the construction industry, and this study approaches it by conducting an exploratory analysis applied to students that have completed a mandatory internship as part of their construction program at Arizona State University, in the School of Sustainable Engineering and the Built Environment. Data is collected from surveys, including grades by a direct evaluator from the company reflecting each student’s performance based on recent Student Learning Objectives. Preliminary correlations are computed between scores received on the 15 metrics in the survey and the final industry suggested grade. Based on the factors identified as highest predictors: ingenuity and creativity, punctuality and attendance, and initiative; a prognostic model of student performance in the construction industry is generated. With regard to graduate employability, student performance in the industry and human predispositions are also tested in order to evaluate their contribution to the generated model. The study finally identifies threats to validity and opportunities presented in a dynamic learning environment presented by internships. Results indicate that measuring student performance during internships in the construction industry creates challenges for the evaluator from the host company. Scoring definitions are introduced to standardize the evaluators’ grading based on observations of student behavior. 12 questions covering more Student Learning Objectives identified by the industry are added to the survey, potentially improving the reliability of the predictive model.