Same Variable, Opposite Direction
We grouped institutions by their share of distance education enrollment and split each group by non-profit versus for-profit control. This is a useful way to inspect a possible interaction, but the cell counts and closure counts must stay visible.
Online delivery can plausibly change costs, reach, and regulatory exposure, but those mechanisms are not identified by this comparison. They remain hypotheses for a larger longitudinal study.
The Case for Interaction Effects
An interaction term is worth testing; it is not yet a finding. A flexible model can represent different effects by sector, but it cannot manufacture information that is absent from the outcome sample.
The fully-online group contains 7 non-profits and 3 for-profits, with no observed closures in either subgroup. The sample is too small to draw a comparative conclusion.
In the hybrid group, 2 of 683 non-profits were observed closed, compared with 0 of 11 for-profits. Those counts do not support a stable sector contrast.
Online share from the IPEDS distance-education field retained in the public College Scorecard/IPEDS output. “Online share” is the fraction of enrollment in exclusively online programs. Results are descriptive; no causal or deployment claim is made.