Counting every student: Still the best way to measure success
Counting every student: Still the best way to measure success
With the federal government’s ongoing decisions to tear down years of carefully built data systems, it’s bleak out there for higher education policy folks. We can no longer count on having access to accurate and timely data about the areas we study or represent. That’s why the annual release of success rate data from Michigan’s Center for Educational Performance and Information (CEPI) is an even bigger deal than normal this year.
As usual, the CEPI success rates show a more complete picture of success at Michigan’s public universities than the federal data typically do. CEPI’s methodology adjusts the calculation to account for those students who transfer in from or out to another institution or are part-time. The federal data system (IPEDS) doesn’t do that, sticking primarily to first-time and full-time students. This doesn’t have a huge effect on some of the largest research universities where enrollments are typically drawn from first-time, full-time students, but for our regional universities where many students have transferred to or are attending part-time while working, this can mean a big swing in the data. On top of that, four of our regional universities in northern regions of the state also fulfill a mission similar to that of a community college and grant associate degrees and certificates.
| IPEDS 6-Year Graduation Rates vs. CEPI 6-Year Success Rates | |||
| IPEDS | CEPI | Diff. | |
| Central Michigan University | 61.8% | 77.2% | 15.4% |
| Eastern Michigan University | 45.5% | 64.4% | 18.9% |
| Ferris State University | 47.9% | 67.6% | 19.7% |
| Grand Valley State University | 68.6% | 80.6% | 12.0% |
| Lake Superior State University | 56.9% | 70.6% | 13.7% |
| Michigan State University | 82.2% | 90.8% | 8.6% |
| Michigan Technological University | 68.7% | 83.7% | 15.0% |
| Northern Michigan University | 50.8% | 68.4% | 17.6% |
| Oakland University | 56.9% | 73.4% | 16.5% |
| Saginaw Valley State University | 53.8% | 62.4% | 8.6% |
| University of Michigan-Ann Arbor | 93.2% | 95.0% | 1.8% |
| University of Michigan-Dearborn | 56.6% | 71.9% | 15.3% |
| University of Michigan-Flint | 44.6% | 60.6% | 16.0% |
| Wayne State University | 56.5% | 67.4% | 10.9% |
| Western Michigan University | 57.8% | 74.8% | 17.0% |
| Aggregate: | 68.3% | 78.9% | 10.6% |
| Average: | 60.1% | 73.9% | 13.8% |
First, my annual disclaimer: these two datasets are based on different student cohort years because IPEDS data is so delayed. But what does the above mean? Well, there’s functionally a 10.6 percentage point difference between what the federal data show as the graduation outcomes of students in our sector versus what the state data show. (I prefer analyzing aggregate totals rather than averaging averages). A difference that significant is equal to literally thousands of students (4,345 to be precise) being uncounted. 4,345 graduates going back to their hometowns to start businesses. 4,345 graduates becoming part of the turnaround story in downtown Detroit. 4,345 graduates having families, becoming graduate students, becoming officers in the armed forces, all uncounted by federal data.
My annual touting of the CEPI data cannot be mistaken for joy at potentially losing IPEDS. IPEDS is an invaluable resource that compares higher education institutions and outcomes in standard ways across states and time. Losing IPEDS – and most associated National Center for Education Statistics staff have been terminated by the administration – would set back higher education policy by decades. CEPI’s methodology reveals the holes in IPEDS, but that doesn’t mean IPEDS should be discarded. In the interim, state datasets like CEPI’s are only going to become more important, and we have to tout their utility publicly.
Mia Murphy is the Chief Policy Officer at the Michigan Association of State Universities.