For decades, applicants have rushed their applications to Harvard University, hoping to be identified as the best of their generation. More than 42,000 applied to be admitted to the entry-level class last year. But how do those applicants, and all those who apply to other colleges, decide where to seek admission?
In recent years, college admissions officials -- and high school counselors -- have wondered if they could influence that behavior. Not necessarily to get any more students to go to college in Cambridge, but how to get more students to apply to college -- any college.
Educators have an ally in their campaign. Christine Mulhern, a Ph.D. candidate at Harvard, recently spent time looking at how students use Naviance, a Hobson's online tool that currently reaches 40 percent of the U.S. population of high school students. What Mulhern did was to examine how students actually use Naviance at an unnamed public school district, described only as being in the mid-Atlantic, having about 4,000 graduates a year with 10 to 15 high schools.
She describes Naviance as a popular tool that students in the district -- regardless of their background -- value. Naviance gives students a sense of how they do compared to others from their local area. Given that students generally attend high schools from where they live, Naviance gives a sense of how they are doing academically.
Diversity is an important issue in the district. Eight percent of its students identify as Latino, 20 percent as black, 17 percent as Asian and 49 percent as white. More than one in five (21 percent) received free or reduced lunch at least once while in the district.
While Naviance has been criticized for helping students with many advantages in the college application process, Mulhern finds that black, Latino and economically disadvantaged students are among that biggest beneficiaries. To illustrate a point about how students seem to value the information contained in scattergrams, she writes, "Scattergram visibility increases applications by 40 percent among students who received [free lunch] and 36 percent for black and Hispanic students. These students are the most likely to lack information about college."
She adds, "Scattergram visibility also has larger impacts for in-state public colleges, increasing applications by 62 percent. Students may view scattergrams for in-state public colleges more than other colleges, because they are nearby and inexpensive, or because they are more likely to have heard about these colleges."
In many ways, Mulhern agrees that students benefit from knowing more about the socioeconomic status of more institutions. She finds that these students use the tool to identify institutions where they will be comfortable.
"Dividing the attendance estimates by the application estimates indicates that 22 percent of black and Hispanic [students] induced to apply to a college by a scattergram go on to attend that college. This jumps to 38 percent for students near the typical acceptee lines, probably because they are more likely to be admitted to the college (and not many better ones) than students far from the lines. In addition, 29 percent of students induced to apply to an in-state public college by a scattergram go on to attend it," she adds.
The Big Names
Of course some colleges are especially interested in what the data suggest about the most competitive of students.
Mulhern notes that "in terms of percent change, moving closer to the typical acceptee has the largest impact on application probabilities at highly selective colleges." Further, admission rates are lower at these institutions, so the relationship between an applicant's admit rate and the average is smaller.
"Admissibility signals may be most relevant to decisions about applying to highly selective colleges because admissions probabilities are much lower at these colleges," she write. "This heterogeneity could also be driven by the types of scattergrams available to students. Students have access to more scattergrams for highly selective colleges than less selective ones. If students only see a few less selective schools, the decision of which to apply to may be relatively simple. In contrast, choosing among 15-20 highly selective colleges may seem a daunting task, which could lead students to rely on a heuristic, such as the lines, to narrow down their choice set."
Then there is the question of who sees the selective colleges.
"This story is also consistent with the larger impact of admissibility signals on the application choices of students who never received [a free lunch] (and white and Asian students) compared to students who received [free lunch], black and Hispanic students. The latter group had access to about half as many relevant scattergrams as the first group, making the choice of which of these colleges to apply to simpler," she writes. "The higher-income and white and Asian students had many relevant colleges to choose among, and appear to use the admissibility signals to narrow down their choice set."
The desire of students to attend "good" colleges also is reinforced, she said, by colleges (competitive ones) receiving a disproportionate number of applications from those students.
"Information on the typical acceptee lines also deters some students from attending highly selective colleges. Students just below the GPA line are less likely to apply to a college than students just above it, especially among highly selective colleges. This reduces the selectivity of students’ application portfolios and college attended, which is concerning because less selective colleges tend to have lower completion rates and are associated with lower earnings," she writes.
She said she found the link largely positive except for the reluctance of students to apply to colleges where they were only marginally (if that) unprepared.
In an interview, Mulhern said that her goal in the project was to see the process, "to see how students apply to college." She said she was pleased with the experience and was considering further studies for her dissertation.
Monica Morrell, general manager at Hobson's, said she thought it was significant that Naviance had influence in the college-decision process. "Scattergrams are just one component of the college application process."
"Our goal is to show people options," she said. Some people may find that information in the scattergrams, and others may find it in the in-depth information about colleges.
"It's not just the academics," Morrell said, noting the range of services Naviance provides.