Quad

DataPull request #41

Where students come from, what happens to them, and what colleges spend

Seven new kinds of federal data on every college. Where first-years come from, transfer students in and out, 8-year results for every student who starts and how many finish in 4, 6, or 8 years, graduation rates for Pell recipients and by race, spending and endowment per student, and full-time faculty share and pay.

What's new

  • Where first-years come from (Profile → Students): the share from the college's own state, from other states, and from abroad, the top five home states, and a state map. "Known for" adds "Draws students nationally" and, where most students come from one other state, "Most students are from {state}". Explore can sort by first-years from other states and filter to colleges that draw nationally (half or more from other states).
  • Transfers in and out (Profile → Students):
    • In: how many new transfer students enrolled this fall, and their share of all new undergraduates. Over time charts this every fall since 2008.
    • Out: of students who entered 8 years earlier, how many left and were enrolled at another college. Federal data doesn't count students leaving in a given year, so the page follows one entering class and says so.
    • Explore can sort by transfer share and has a column for it.
  • Everyone who starts here, 8 years later (Profile → Cost & outcomes): what happened to every student who started, including transfer and part-time students the usual graduation rate leaves out. It shows who earned a credential, who is still enrolled, and who went on to another college, with a switch between students who started in college here and students who transferred in. Explore has an 8-year completion column and sort.
  • How long it takes (in the same card): the share who finished within 4, 6, and 8 years, for all students, for students who started in college here, and for transfer students. Over time charts all three by entering class, and Explore can sort by finishing within 4 years.
  • Do lower-income students finish? (Profile → Cost & outcomes): the 6-year graduation rate for Pell Grant recipients next to students with no need-based federal aid, with the gap in points, and the rate by race and ethnicity. Groups under 30 students aren't shown. Explore can sort by the Pell graduation gap and how it changed over 10 years, and filter to gaps under 5 points. "Known for" adds "Pell students graduate at the same rate".
  • Spending and endowment (Profile → Academics): instruction spending and endowment per student, compared only with colleges that report on the same accounting form, so public and private colleges are never ranked together. "Known for" adds "Big endowment per student" (top 5% of its sector).
  • Faculty (Profile → Academics): the share of faculty who are full-time and the average salary. Explore can sort and filter by full-time share.
  • Compare has rows for all of these, and Over time charts how each one changed, most for 10 to 20 years.
  • New glossary terms explain each measure, including adjusted cohort, transfer-in and transfer-out, time to degree, GASB vs. FASB accounting, and 9-month equated salary.

Behind the scenes

  • Sources: all federal.
    • Fall Enrollment part C (fall 2024)
    • Fall Enrollment part A (fall 2024), for transfer students
    • Outcome Measures (students who entered in fall 2016)
    • Graduation Rates by Pell and loan status (entered fall 2018)
    • College Scorecard graduation by race
    • Finance (fiscal 2023–24)
    • Salaries (fall 2024)
    • College Scorecard full-time faculty share
  • No existing value changed.
  • Detail file: home-state counts are too big for the main dataset, so each college now has a detail file with its own checks. Majors will use it next.
  • Built to refresh safely: each new federal file updates these numbers and their years on its own, and the data sync stops rather than show something wrong if a file changes shape:
    • Transfer students: the counts must add up (full-time plus part-time) and match NCES's own derived totals. If NCES renumbers a level, the sync stops.
    • 4, 6, and 8 years: the shares must only go up from 4 to 8 years. If more than 1% of colleges break that, the sync stops.
    • Each spec's "Refresh and maintenance" section lists every guard, and tests prove each one catches a broken file.
  • Edge cases, kept as reported or treated as unreported:
    • Colleges report residence every other fall, so its history has a point every two years.
    • About 20 colleges report Pell gaps over 40 points, which usually means students were sorted into groups inconsistently. They're shown with a caution but not ranked.
    • Three colleges reported $0 of instruction spending; that's treated as not reported.
    • The 8-year results show where students enrolled, not whether they finished there, so "Known for" says "go on to another college".
    • For transfer students, "within 4 years" counts from when they arrived at this college.
  • Wave 2 is complete: residence, transfers, outcome measures and time to degree, graduation by group, finances, and faculty join student-to-faculty ratio, campus profile, and campus services. Metro area stays deferred. Next on the roadmap: majors and earnings by major.