Using data to improve student outcomes at Michigan State

Michigan State University is an American original: a land-grant, research university established more than 150 years ago to serve the public good. Using scientific knowledge and skills to solve problems is what land-grant universities do. And one of the problems that we have turned our attention to is improving our own record of student graduation rates. Eight in 10 admitted students goes on to earn a degree with us, but we know that we can do better, which is why we are using data analytics to identify and solve problems related to student success.

Analytics play an important role in our proactive outreach to students, in changing the classroom, in uncovering institutional shortcomings, in changing the institution to better support students, in helping us better understand our students’ perspectives and needs when they arrive on campus, and in breaking down information silos about students.

Three recent initiatives in which we have used institutional data to transform the university in the direction of student success are our Go Green, Go 15 credit momentum campaign, reforming our gateway mathematics courses, and the formation of a campus-wide learning analytics group.

Our Go Green, Go 15 credit momentum campaign plays on our university colours (green and white) and has yielded positive early outcomes. A typical MSU degree requires 120 credit hours – or four years if a student completes 30 credits a year, 15 per semester. From 2006 to 2016, the number of new students taking 15 or more credits had declined to 28 per cent. After implementing our campaign to influence academic advisors, students, and families, that increased to 43 per cent in autumn 2017 and 50 per cent in 2018. 

Enrolling in at least 30 credits in the first calendar year at MSU sets the pace to graduate in four years, reducing time to degree and potentially saving students up to $50,000 in their cost of attendance. Without analytics, we wouldn’t have known to direct our attention at credit momentum.

Maths reform has also benefitted from thoughtful use of data analytics. A decade of data clearly showed how our old maths requirement created obstacles to student success. Multiple units from across campus collaborated to revise our gateway mathematics curriculum to better meet the needs of our students. We created two requirement pathways: one for students who need college algebra and a quantitative literacy sequence for students who do not. We eliminated developmental maths and created a scaffolded curriculum for students to succeed in either pathway. We also started a maths advising professional learning community to improve learning. This group continues to rely on data analytics to inform discussions and shape new directions for the campus.

Finally, to deepen campus-wide inquiry, we have created a learning analytics group, a cross-institutional group of faculty, staff, and administrators who work together to identify and analyse institutional data that can help drive initiatives for student success. 

Their work contributed to the success of the Go Green, Go 15 and maths reform and is also looking at signals in data about housing, fraternity-sorority participation, undergraduate research, use of services such as tutoring and help rooms, and instructor early alerts about student performance. 

While we don’t think that these are all causal – changing roommates doesn’t cause someone to leave MSU – they can serve as signals. We have found that students who change rooms in their first year are less likely to return to MSU. We can use that data as a signal to direct resources towards students who change rooms and identify any other transition or adjustment challenges that they may be experiencing.

If data analytics is capable of genuinely making a difference in terms of improving student success, it will be because of its ability to uncover institutional shortcomings and guide us in resolving those shortcomings, and because of its capacity to convince people that they can and should act. Its potential to improve student success rests in a combination of helpfulness and hype – helpfulness in making institutional roadblocks to student success more evident, and hype in how it can be capable of compelling action.

To not use our research skills to collect and analyse data that could transform our student success at our institution would be not only unethical, but antithetical to our land-grant roots. What we do at MSU is apply scientific knowledge to solve real world problems. What better place to start than at home. 

June Pierce Youatt is provost and executive vice-president for academic affairs at Michigan State University.