DELAYED BY TRANSFER. Inside Higher Ed presents A ‘Penalty’ for Starting at a Community College.

A working paper circulated in September by the National Bureau of Economic Research takes a look at the entire pathway, from community college to bachelor’s degree, and compares students’ success over a nine-year period with their peers who began at four-year colleges. Like previous studies, it found a significant “penalty,” or decreased likelihood of completing a degree, for students who started out in community colleges compared to those who started at four-year institutions.

The commentary that follows suggests the research considers sample selection biases.

The difficulty with comparing degree completion rates between two-year and four-year colleges is that the profile of students attending each isn’t necessarily comparable. On average, community college students are older, more likely to be minorities and to come from more disadvantaged backgrounds. In addition, not all students who enroll in a community college intend to earn a bachelor’s degree, and even those who do can change their minds.

That mind-changing is self-selection. There are methods for capturing the effects of self-selection in statistical inference. (I'm uncomfortable with the language of "correcting" or "controlling for" that practitioners often use, and I sometimes wonder if the extra econometric make-work doesn't produce rounding errors that swamp any of the theoretical biases in the estimators.)

But separating the students with a “demonstrated intent” of graduating with a four-year degree solves the apples-and-oranges problem of comparing students from different types of institutions. While the results still show a penalty for community college students, it’s smaller than it otherwise would be and suggests possible solutions.

The article doesn't mention what those solutions (more language that makes me uncomfortable: better to speak of tradeoffs) are.

At Northern Illinois, close to half of the undergraduate students are transfers, and some transfer credit on the transcript is common. The article interviews price-sensitive transfer students and notes some university efforts to ease the transition. Left unanswered: whether some of those transfer credits are the consequence of years of attrition of the faculty and staff, such that students pick up a course here or there to complete a degree in four or five years rather than wait for that course to be offered here.