| This is a compilation of three applied economic analyses. In the first analysis, college choice for a sample of students from the National Educational Longitudinal Study of 1988 who have expressed a desire to attend college is analyzed using a multinomial logit framework. For this select sample of students, family income is found to have a negative impact on the probability of attending a four-year college. Exploration of this unexpected result suggests it stems from signaling by low income/high aid and high income/low aid families. Family aid is found to have a positive impact on the probability of attending a four-year college, though this is likely also a result of signaling. In the second analysis, persistence through till their sophomore year of college is analyzed for students at four-year colleges. A multinomial logit of college choice is used to construct estimated inverse Mills' ratios, which are used to control for self-selection into college type in the binomial probit of persistence. Family income is found to be insignificant for students in most tuition categories and only a very high level of family aid is associated with the probability of persisting. The relationships are most likely not causal. The third analysis is the result of a joint project with Elizabeth T. Powers. We use Census data from 1980, 1990, and 2000 to look for evidence that likely Supplemental Security Income (SSI) participants make interstate moves to take advantage of higher SSI benefits available in some states. Using five different approaches, we are unable to find any strong evidence to support the welfare magnets hypothesis. |
