Assessing the data analysis training of engineering undergraduates

The need for acquiring data analysis skills has become ubiquitous across many professions. In engineering, this need has been recognized through elements such as the current ABET student outcome 3.b, which states that engineering graduates should have “an ability to design and conduct experiments, as well as to analyze and interpret data.” While this outcome is a requirement of engineering programs, the length and depth of the data analysis training of undergraduate students vary significantly across engineering majors, leading some scientists to criticize what they perceive as inadequate quantitative training for engineers. The evolution in the capacity to produce and store data requires an exploration of the current state of data analysis training provided by engineering programs. We begin this exploration through the following research question: What data analysis training has been offered to and acquired by engineering students through undergraduate courses? In this work, we aim to answer this question through a sequential exploratory mixed methods design. Using the Multiple-Institution Database for Investigating Engineering Longitudinal Development (MIDFIELD), we qualitatively coded records of courses offered to engineering students at one public institution between 1989 and 2011 to generate profiles reflecting different levels of data analysis preparation. These profiles were then quantitatively clustered into six distinct levels. The cluster analysis revealed variable patterns of data analysis preparation across different engineering majors. Results from this study also provide a baseline for employers to evaluate the data analysis training of engineers, especially as it relates to the expanding employment opportunities related to data analysis skills. Further, these results may help to inform potential programmatic evaluations and changes.