Colleges and Upward Mobility in the US Over the Last Century

Access to college education is widely regarded as an important path to economic mobility. However, to the extent that children from richer families attend more highly ranked colleges than children from poorer families, higher education might actually amplify the persistence of economic status across generations rather than “leveling the playing field”. This project will bring new data and a long-term perspective to the study of higher education and inequality and intergenerational mobility in the US. To do so, the investigators will construct a large dataset with information on the socioeconomic backgrounds and post-college outcomes of US college students over the last 100 years. To collect these data, the investigators will first digitize the rosters of students attending different US colleges from the early 1900s and until the 1950s. In a second step, they will link these students to US population censuses, enabling them to observe students’ childhood circumstances and their own adult outcomes (such as income). The researchers will use this dataset to investigate important questions related to higher education in the US, for instance: (1) To what extent has access to “elite” colleges increased over the last century for students from poorer socioeconomic backgrounds?, (2) Which colleges have been the most effective at improving the labor market prospects of their students, particularly those from poorer families? (3) What have been the effects of changes in colleges’ admission practices (for instance, the introduction of standardized testing) on the socioeconomic composition of the students that they enroll? This new dataset, which the investigators will make publicly available, will also offer many exciting research opportunities for other scholars. Moreover, by identifying colleges in which students have better outcomes today than in the past, this dataset will also help identify the institutional changes associated with these improvements.

This project will bring new data and a long-term perspective on the role of higher education in inequality and intergenerational mobility in the US. To do so, the investigators will construct a large individual-level dataset of students who attended different US colleges over the course of the 20th century. This data collection will proceed in two steps. In a first step, they will digitize historical college registers for students attending different US colleges from the early 1900s and until the 1950s. In a second step, they will link these students to US population censuses, enabling them to observe both their childhood circumstances and their own longer-term adult outcomes (such as income). Linking these two sources of data is made possible by the fact that the both the college registers and the census include identifying information such as students’ names and hometowns. By combining these data with similarly constructed data for students attending college in more recent years, the investigators will be able to characterize the socioeconomic backgrounds and post-college outcomes of US college students over the last 100 years. The investigators will use these data to address a number of questions related to access to higher education in the US: (1) To what extent has access to “elite” colleges increased over the last century for students from poorer socioeconomic backgrounds? (2) What has been the influence of changes in colleges’ admission practices (for instance, the introduction of standardized testing) on the socioeconomic backgrounds of the students that they enroll? (3) Which colleges have been the most effective at improving the labor market prospects of their students, particularly for those at the bottom of the income distribution? This new dataset, which the investigators will make publicly available, will offer many exciting research opportunities for scholars in fields such as economics, history, and education.