Summer melt is defined as the attrition of students who complete high school with the intent to enroll in college but do not begin in the fall. The existing literature related to summer melt primarily focuses on interventions in an effort to minimize summer melt. These studies point to the benefit of summer assistance, either through advising or nudge text messaging, as strategies to decrease the rate of melt. Research to date primarily focuses on the macro-level of whether a student went to college or not.

This study uses a different approach in that it investigates the type and frequency of change between intended enrollment and actual enrollment. I conducted a quantitative study using records from four graduating high school classes (2016 – 2019) of a national charter public network. Type is analyzed by categorizing intended and actual enrollment as four-year college, two-year college, or no college. This allowed for a more nuanced view of shifts between plans. In addition, I conducted a multiple logistic regression analysis of four-year intending students to test candidate predictors of melt. The final regression model included student demographic characteristics, academic record and intended institution characteristics. I fit the same model with two-year intending students to identify similarities and differences in the variables associated with melt based on the type of institution a student plans to attend.

Of the 17,343 records used in this dataset, 22.8% of students had actual plans that did not match their intended plan. For four-year intending students, three variables were found to have statistically significant associations with melt: Expected Family Contribution (EFC), grade point average (GPA), and the six-year minority graduation rate of the intended institution. An implication from this study is that the lower the EFC, the higher expected probability of melt. This affirms the systemic challenges that exist for low-income students to gain access to higher education, perpetuating a system of social reproduction where those who are more affluent have higher odds of beginning college and greater chances to reap the benefits afforded by a college degree.