Accurate data about student persistence (retention) and academic progress, particularly for first-year and lower division students, is essential to understanding and addressing factors affecting student success in engineering. Unfortunately, nationally standardized methods of measuring freshman persistence and graduation rates provide an incomplete, and in some cases inaccurate, picture of engineering student retention issues. Standard assessment methodology is based on a simplistic model of higher education – the expectation that freshman enter a university, choose a major, attend full time, and emerge four to six years later from that same university with a degree. All other students – those who transfer into or out of a university, those who work off campus and attend college part-time, those who take a semester off for personal, military or financial reasons – are not included in the standard model or the reported statistics. In today’s higher education environment where nearly 60% of students attend more than one institution, this simplistic model is becoming outmoded. And in our metropolitan university that provides educational access to a wide range of students – where many of our engineering students arrive on campus needing preparatory coursework in math, where almost all students work off campus, and where at least 40% attend college part time – the standard model becomes ineffectual.

To better understand factors affecting persistence and success in our student population at Boise State University in Boise, Idaho, our team developed a model that measures retention and academic progress of all engineering students. Additionally, our team focused special emphasis on freshman level students. Unlike the rigidly linear model described above, our model recognizes multiple entry and exit points and differing rates of progress along the engineering education route. Over three academic years, our team developed and refined an algorithm to query the Boise State University PeopleSoft database and ask meaningful questions about our students and their progress. In our algorithm:

In this paper we will explain how this model and algorithm have provided greater clarity on retention issues and trends affecting pre-freshman and freshman level engineering students at Boise State University. The data prompted our engineering college to respond in several ways, such as creating a new introductory engineering class for pre-freshman level students, enhancing emphasis on advising, and supporting math education through a variety of programs. We will also explain how the sorting algorithm is a method easily adaptable and portable to database systems at other universities.