4. Discussion
Students face complex decisions about their education and are unlikely to have full information about the consequences of their actions. While empirical results concerning both the beliefs of students and barriers to information have been accumulating, there has been little useful theory. In addition to the impressive Proyecto 3E data gathering effort, generating one of the largest data sets of student beliefs, HNRZ offer a theory of belief formation that is likely to be applicable in other contexts. This model provides a framing for the literature on student beliefs and allows the causal determinants of belief formation to be better studied and understood. Theoretical development here is important given the general growing interest in information-based behavioral intervention. Informational interventions targeting lowSES students, who are emphasized in the theory since they face different information costs, include work by Hoxby and Turner (2013), who inform high-performing low-income students about their potential to attend and afford selective colleges, garnering a large behavioral response. Bettinger, Long, Oreopoulos, and Sanbonmatsu (2012) inform students and families about the availability of financial aid for college, although the information alone was ineffective unless paired with aid in filling out the fi- nancial aid application. Hastings et al. (2015), as previously mentioned, inform students about the costs and earnings associated with their educational options, finding that low-income students have the largest marginal response to the intervention. Resting these sorts of interventions (and the study of student choice more broadly) on a coherent theory of belief formation tells us why these interventions seem to work, lets us understand when they would not be expected to, and points towards how they can be improved.
