Skip to content

Open-source early-warning model

Compass: support, not a label

Research and build

Compass is an open-source early warning model built by MANGO students. It suggests check-ins, explains every flag, and is audited for fairness in public.

Prototype: tested on synthetic and public data

How it works

Signals in. Reasons out

School records go in. A plain-language reason to check in comes out. Here is how that compares with Wisconsin's system.

Wisconsin DEWS compared with Compass
IssueWisconsin DEWSCompass
InputsOver 40 features, including race and free or reduced-price lunch statusAttendance, behavior, and course performance only
OutputLow, moderate, or high risk label"Check-in suggested" with the specific reasons listed
ExplainabilityEnsemble model, hard to inspectInterpretable model you can check by hand
Fairness checksInternal analysis, made public through a records requestBuilt-in audit of error rates by group, published with every release
Data locationState systemRuns inside the division. MANGO never sees student records.
CodeNot publicOpen source
Interactive demo

Meet a synthetic student

Move the sliders or pick a preset. Compass runs right here in your browser, with the same published weights as the open-source code.

Synthetic student

18%
0
0
3.0
0
0.0

All students here are synthetic. Compass never sees real student data on this site.

Compass output

Check-in suggested

  • Missed 18% of school days this grading periodAttendance outreach

What moved the result

Compared with a typical synthetic student. Bars to the right push toward a check-in. The biggest push here: days absent this grading period.

  • Days absent
  • Suspensions or serious incidents
  • Core course failures
  • GPA
  • Credits behind on-track pace
  • Trend: GPA change since last period

Core course failures has a weight of zero in this version: GPA already carries that signal in the synthetic data. The model card explains why.

Fairness audit

Same students, two models

A false alarm is a student flagged for a check-in who would have graduated on time anyway. We compare Compass with a baseline built like Wisconsin's, which does use group and income.

False alarm rate by group. Largest gap between groups: 8.0 points for the DEWS-style baseline, 6.1 points for Compass.

Ranking quality (AUC)
Baseline0.775Compass0.785
Precision at capacity
Baseline43.0%Compass46.2%
Largest false alarm gap
Baseline8.0 pointsCompass6.1 points

AUDIT FAILED: Compass's false alarm rate ranges from 7.7% (Group A) to 13.7% (Group C), a gap of 6.1 percentage points (limit 5). Compass does not meet our own limit here yet, so it stays a prototype.

Synthetic data with built-in structural inequality. See the model card for methods.

Deeper analysis

Removing race isn't enough

Compass never sees a student's group, but other data can carry the same inequality: some groups get more recorded suspensions for the same behavior, and check-ins pile up where school resources are thin. Researchers found the same pattern in Wisconsin (Perdomo et al., 2025). That is a question about counselors and funding, not about any one student.

Methods and full results in the model card

With and without suspensions

Recorded suspensions vs. true incidents

Per student, on average. In our synthetic schools, Group C gets 1.8x as many recorded suspensions per real incident as Group A.

Average recorded suspensions and true incidents per student, by group
GroupTrue incidentsRecorded suspensions
Group A0.230.22
Group B0.260.34
Group C0.300.51

Compass false alarm rate by group, with the suspensions input. Largest gap: 6.1 points.

Here, dropping suspensions barely moved the gap (6.1 points with it, 6.1 points without). The rest of the gap comes from unequal school resources, which shape attendance and grades too.

Check-ins by school

School information alone ranked students less well (AUC 0.66 vs. 0.78 for Compass).

Each dot is a synthetic school. In the lowest-funded third, Compass suggests a check-in for 24.2% of students on average; in the best-funded third, 5.9%. Correlation between funding and check-in share: -0.95.

Show data table
Synthetic schools: funding, counselor ratio, and share of suggested check-ins
SchoolFunding indexStudents per counselorCheck-in share
S240.6159138.4%
S080.7450227.8%
S010.7448423.4%
S270.7650023.0%
S200.7857630.5%
S190.8254526.2%
S100.8449621.1%
S300.8446019.5%
S180.8744118.0%
S050.8843414.0%
S260.8945316.9%
S040.9048016.9%
S220.9445314.1%
S290.9546316.3%
S250.9641615.0%
S170.9648117.9%
S030.9744116.8%
S151.0142013.0%
S211.0244213.3%
S281.083878.2%
S131.113468.1%
S021.113999.2%
S091.123748.4%
S231.133418.5%
S141.203656.2%
S061.233346.6%
S161.263244.1%
S071.322673.4%
S121.342602.4%
S111.372662.0%
Six commitments

What Compass will never do

  1. 01

    No demographic inputs.

    Compass never scores students using race, ethnicity, sex, income, disability, English learner status, or home address.

  2. 02

    Every flag explains itself.

    Each result shows, in plain language, which attendance, behavior, or course indicators triggered it.

  3. 03

    Fairness is measured, not assumed.

    Compass reports error rates for every student group and publishes the results.

  4. 04

    Support, not labels.

    Compass suggests check-ins. It never labels a student and is never used for discipline or placement.

  5. 05

    Student data stays in the school.

    Compass runs inside a school division. MANGO never receives student records.

  6. 06

    Open source.

    The code, documentation, and evaluation results are public.

Roadmap

Prototype first. Proof later

  1. Phase 1In progress

    Synthetic data

    Built: data generator, model, audit, and this demo. Not yet passing our fairness limit.

  2. Phase 2Started

    Public research data

    First check on the UCI Student Performance data (649 students, two schools in Portugal). Small and not from Virginia, so a sanity check, not proof.

  3. Phase 3Planned

    Validation with a Virginia school division

    Under a formal data agreement consistent with FERPA, a division runs Compass on its own de-identified records and shares only aggregate results. We publish them, including any failures.

Work at a Virginia school division?

We're looking for a validation partner.