Applicants who held both offers picked one. Choose two CAA programs and the comparison finds the applicants accepted to both, then reads which one they went to.

The question it answers

Not which program is better, but which one applicants chose when they had a place at each. That is a decision made with both offers in hand, so it carries information no ranking does.

Accepted by both

How many applicants in the dataset reported acceptances from both programs.

Chose one of these two

How many of them went on to attend one of the pair. This is the number the split is calculated from.

Chose another or unknown

The rest, who attended somewhere else or never reported where they landed. They are counted and shown rather than quietly dropped.

The split

Each program's count and its share of the applicants who picked one of the two, with the more-selected side marked, or both marked Tied.

How small samples are handled

Programs are only offered as a pair when applicants have reported deciding between them, so the second dropdown greys out the programs there is nothing to say about rather than returning an empty comparison. Below three reported decisions no side is called ahead, and the result carries a warning to read the split as a rough early signal rather than a settled pattern.

The accepted applicant profile

Under the split, both programs are set on identical rows so the difference is read across rather than remembered between two screens. Each figure is the median reported by that program's accepted applicants, with the middle half of reports beneath it and the number of reports behind it, so a five-person figure is never mistaken for a fifty-person one. A measure no one has reported enough of says so instead of printing a number.

Academics

Cumulative GPA, science GPA, and GRE. MCAT is not among the rows here; Program Analytics is where to read it.

Experience

Shadowing hours and work hours.

What the counts include

Only applicants who listed acceptances from both selected programs, matched against the program they reported attending. Schools running more than one campus are combined into a single program, so a choice between two campuses of the same school is not counted as a head to head.