In 2004, Harvard University announced generous new financial-aid policies under which families making less than $40,000 a year would not have to contribute to their child’s education. The university also said it would increase its efforts to recruit disadvantaged students. Lawrence H. Summers, Harvard’s president at the time, described the moves as a way to narrow the gap in opportunities available to students from different backgrounds.
So, did the backgrounds of students attending Harvard change after the aid policies did? The short answer: yes, especially in the first year.
A new paper by a Harvard senior, Nicholas Galat, takes up that question. The paper, “Addresses and the Aid Initiative: A Geospatial Analysis of the Harvard Student Body, Classes of 2003-2011,” examines whether and how the makeup of classes changed after the policies were introduced.
Mr. Galat, who expects to graduate this spring with a degree in statistics, has long been interested in the composition of the student body, he said. “From the moment I first stepped on Harvard’s campus four years ago, I was impressed by the diverse set of backgrounds represented here.”
For his research, he linked students’ home addresses listed in The Harvard Freshman Register with census block groups from the U.S. Census Bureau to determine any change in the mix of students coming from low-income neighborhoods, particularly rural ones, and neighborhoods with low levels of educational attainment.
Whether Harvard enrolled more low-income students from rural areas was of particular interest to Mr. Galat. Prior research has suggested that elite colleges focus their recruitment of low-income students in urban areas—with greater population density—and that students in rural areas have less access to good information about college.
In general, Mr. Galat found that Harvard’s shift made the biggest difference for the Class of 2009, the first to be recruited under the new policies. In that class, which matriculated in the fall of 2005, the neighborhoods students came from had lower median family income, educational attainment, and population density.
For instance, the class “contained approximately 36 more students from neighborhoods with median family incomes less than $60,000 than one would expect based on the average historical proportion of such students,” the paper says.
Most of the increase in the share of students from low-income neighborhoods was the result of more students’ coming from rural, low-income neighborhoods, the paper says. That’s probably because Harvard was already affordable for low-income students, Mr. Galat writes, and it already recruited in urban areas.
Over all, the class that entered in 2005 included more of the kinds of students the aid and outreach policies were meant to enroll, he found.
Other research has also noted the different composition of that class, attributing it to an increase in the number of qualified low-income applicants.
Because of privacy laws, Mr. Galat did not have access to student-specific admissions or financial-aid information.”That’s a fundamental limitation of my data set,” he said. At the same time, he said, “the small neighborhoods students come from are still very powerful” measures. Census block groups usually contain 600 to 3,000 people.
After 2005, Mr. Galat found, the picture changed. For the next two years, the entering classes looked more like those before the new policies were introduced. Mr. Galat offers two possible explanations for that. One is that similar aid policies introduced by other elite universities after Harvard did so attracted some of the talented, low-income students who may otherwise have ended up at Harvard. Another possibility is that the news-media attention surrounding Harvard’s announcement increased its effect that first year.
The paper does not consider students who entered Harvard after 2007 because the university changed its aid policies again that year. The analysis considers only students who reside in the United States and list their addresses in the register, a number that dropped over time.
For its part, the university maintains that its policy change in 2004 was effective. “Harvard’s decision to launch the Harvard Financial Aid Initiative (HFAI) in 2004 led to profound changes in our applicant pool and our undergraduate student body, and to similar efforts at many of our peers,” Jeff Neal, senior communications officer, wrote in an e-mail. “That in turn created strong competition for talented low-income applicants.”Return to Top