Font Size: a A A

Two essays on college educational loan debt

Posted on:2003-10-01Degree:Ed.DType:Dissertation
University:Harvard UniversityCandidate:Fuller, Hester LouiseFull Text:PDF
GTID:1467390011483406Subject:Education
Abstract/Summary:
Following a short Introduction, the dissertation presents two self-standing essays on undergraduate educational loan debt.;The first essay combines a review of the literature on the distribution of undergraduate educational loan debt with an analysis of data from the Baccalaureate and Beyond [B&B] survey of 1993 college graduates. Using a strategy developed by Cragg, data from the Baccalaureate and Beyond survey, are employed to model debt at graduation on selected student characteristics (gender, age, ethnicity, background family wealth, and financial independence from, the birth family) and selected characteristics of the institutions from which they graduated (sector, sticker price and total costs of attendance). The models show that institutional factors affect both the propensity to owe anything at graduation, and the magnitude of the outstanding loan balance. The models also show that most of the personal characteristics studied influence the probability of having any debt---but not the size of that debt---at graduation. The exception to this rule is the characteristic of being Hispanic, which does not influence the odds of having debt, but is associated with a markedly smaller loan balance, on average, than amounts owed by debtor students of other ethnic groups, particularly White students.;The second essay uses logistic regression to investigate two questions with data from the Baccalaureate and Beyond survey of 1993 college graduates. First, does debt influence a new graduate's choice to teach in primary, middle or secondary school? Second, is debt's effect on this choice the same for both genders, all ethnic groups and all college majors? Analysis reveals evidence that propensity to teach in kindergarten through grade 12 (K12) soon after graduation is lower among those with higher levels of undergraduate educational loan debt, all else being equal, and that this effect is constant across gender, ethnic group and field of major undergraduate study. In practical terms, the model suggests that educational loan debt diminishes the annual pool of prospective teachers by numbers sufficient to staff all the K12 schools in a moderately sized American city, like Buffalo or Cincinnati.
Keywords/Search Tags:Educational loan debt, Data from the baccalaureate, College
Related items