This is the second in a series of articles describing ongoing research that involves studying engineering college student graduation using Cox proportional hazards models. The first article, called "Proportional hazards models of graduation," was based on main effects models of graduation controlling for descriptors such as in-state residence, hometown population, and student major. This second article attends to first-order interaction terms between pairs of previously considered main effects. Survival analysis of graduation data here suggests significance of standardized math scores, and English and Science ACT scores, under certain circumstances that might not have been discovered without the examination of interaction.

 

Key words: Graduation, proportional hazards, retention, survival analysis.