The baking of preferences throughout high school

Abstract

The purpose of this study is to see if teenagers’ preferences for time and risk change during their time in high school. We are examining whether older teenagers exhibit different preferences and how these differences evolve between girls and boys. To do that, We use a large and powered sample (n=4830) of non-self-selected teenagers from 22 Spanish schools with very different backgrounds. Besides the task to measure preferences for time and risk, we also collected information on the 207 classes (size, students’ background, and the network of friends and enemies) and an extensive array of students’ characteristics, including cognitive abilities (reflection and probability abilities, GPA and expectations to go to the university). We also computed popularity in friends and enemies networks and a measure of centrality (betweenness) for each participant. All these measures allow us to control for potential omitted variables that were not considered in other papers. By taking all these factors as exogenous, we find that girls and boys are not different in patience and risk preferences, but age (measured by grade) only affects the last one. We also find that cognitive abilities and homophily (measured as the average measure of each economic preference on the subject’s friends) strongly predict both time and risk preferences. Additionally, we explore the existence of different mechanisms of gender differences and find that cognitive abilities explain gender differences in both economic preferences.

JEL codes: C91, D81.

Status: Working Paper