ACT develops two types of predictions: enrollment probabilities and indexes.
Enrollment probabilities for ACT score senders, EOS names selected, and inquiry pools are campus specific. That is, they provide the probability that a given student will attend a specific institution.
The indexes provide a probability that a student will show a specific enrollment behavior.
- The Mobility Index indicates the likelihood that a student will enroll at an out-of-state institution.
- The Institution Type Index predicts the likelihood that a student will enroll at a private institution.
| Campus-Specific Predictions |
Student-Specific Predictions Based Primarily on Student Choice Set |
|---|---|
|
|
Predicting Out-of-State Enrollment
The Mobility Index predicts the likelihood that a student will enroll at an out-of-state institution, using (1) the student choice set or "campus mix" (campuses to which students send scores), (2) ACT Composite score, (3) preferred distance from home to campus, and (4) selected other variables.
The pie chart below shows the relative importance of specific factors in the Mobility Index model.
The campus mix comprises 84% or more than 4/5 of the Mobility Index model. This large percentage indicates that student enrollment behavior is very intentional and predictable. If students send scores to out-of-state institutions, they are much more likely to enroll out-of-state. If students do not send scores to out-of-state institutions, they are very unlikely to enroll out-of-state.
Predicting Enrollment at a Private Institution
The Institution Type Index predicts the likelihood that a student will enroll at a private institution, using (1) the student choice set (campuses to which students send scores), (2) ACT Composite score, (3) preferred institutional type, and (4) selected other variables.
The pie chart below shows the relative importance of specific factors in the Institution Type Index model.
The campus mix comprises 65% or 2/3 of the Institution Type Index model. Preferred institution type accounts for another quarter of the model. Together, the campus mix and preferred institution type account for almost 90% of the model.
This large percentage indicates that student enrollment behavior is very intentional and predictable.