An individual student's probability of enrollment is calculated by combining data in his or her ACT record with weights. The weights are calculated from variables in students' ACT records and from enrollment data that institutions submit to the ACT Class Profile Service.
The table below lists the primary variables that predict enrollment at every institution.
| Type | Variables |
|---|---|
| Academic | ACT Composite High School GPA |
| Demographic |
Distance from home to campus High school size Family income Size of home community |
| Enrollment Preferences |
Level of college choice (1–6) Preferred college size Preferred tuition Preferred major (7 groups) Planned work hours |
Predictive modeling data for ACT score senders are appended to ACT electronic records of score senders to your institution. Predictive modeling data for EOS names selected are appended to EOS records and provided to you when you purchase EOS names.
The Predictive Power of the Student Choice Set
When students take the ACT, they can choose to send their scores to up to six institutions in preference order. ACT calls this mix of campuses the choice set. Through extensive research, ACT has determined that this choice set carries a predictive power in the models that far outweighs other variables.
- The preference order, or level of college choice, is the single most important factor in predicting that a student will enroll at a specific institution.
- Likewise, the mix of campuses (in-state or out-of-state, public or private) is the single most important factor in the Mobility Index, used to predict enrollment at an out-of-state institution, and the Institution Type Index, used to predict enrollment at a private institution.