The impact of education histories on the decision to become self-employed: a study of young, aspiring, minority business owners

For example, Brock et al. (1986) find that college graduates are more likely to become self-employed while Fairlie (1999) concludes that the link between education and self-employment is weak for both whites and blacks. Fairlie (2006) finds that low levels of education are partly responsible for the relatively low self-employment entry rates for blacks and Hispanics. Bates and Lofstrom (2007) find that education positively predicts self-employment entry for industries requiring high-level skills. In a meta-analysis of research on industrial countries, van der Sluis et al. (2003) find that the relationship between education and selection into entrepreneurship is insignificant. See Fairlie and Robb (2008) for a review of literature on education and self-employment.

Following Tyler et al. (2003), I weight observations equally in fitting the models.

It is important to note that extensive variation in self-employment rates exists within racial groups (Fairlie and Meyer 1996), and the broad racial and ethnic categories I control for in my aspiration model cannot account for that variation.

I also include dummy variables for missing observations on income.

See Aldrich and Kim (2007) for a review of adolescent occupational aspirations and the link to parental self-employment.

With most locus of control testing instruments, low scores indicate an internal locus of control.

See Becker (1962), Arrow (1973), and Spence (1973) for formal development of the economic theories of human capital and signaling.