How Many College Applications Is Enough?

A new study demonstrates how a relatively arbitrary option within college entrance exams has a substantial impact on the number of colleges to which students apply, and could partially improve issues of undermatching prevalent among low-income college applicants. The study, produced by Harvard and the National Bureau of Economic Research, demonstrates that when the ACT increased the number of free score reports that students could send to schools, test-takers applied to more colleges—and more selective colleges—regardless of income level.

In 1997, ACT changed its policy to allow students to send four free score reports instead of just three. By 2000, the percentage of ACT test-takers who sent exactly four score reports jumped from less than 5 percent to nearly 75 percent. Meanwhile, by 2000 only 10 percent of ACT-takers sent three score reports, when just a few years earlier, more than 80 percent did. Most test-takers responded to the policy change by taking advantage of the extra free report, while very few students opted to shell out the $6 for an additional score report.

While it seems strange that only an extra $6 would dissuade so many students from sending an additional report, the findings are consistent with evidence that peoples’ decisions are often based on the underlying default option. Authors Richard Thaler and Cass Sunstein famously pointed out a few examples of the same occurrence in their book, Nudge: When people are faced with a decision, they often choose the default choice—even if better options are available. This phenomenon is shown to have powerful effects on 401k contributions, Medicare Part D enrollments, and even organ donor rates(not to mentionstudent loan repayment options).

The researchers in this study found that not only were ACT-takers sending more score reports, but they were also applying to more colleges and universities. (Just because the student sends a report to a college does not necessarily mean he will follow through and submit an application to that fourth school.) Middle- and high-income students who took the ACT after the policy change sent out 0.19 more applications on average than they did before the policy change. For low-income students, the average increase in applications was 0.14.

And the researchers showed that ACT-takers applied to more selective schools (as measured by the 25^th percentile of ACT scores for the incoming freshman class) after the policy change. The effect was especially strong for low-income students, who applied to schools with 25^th-percentile ACT scores that were 0.93 points higher on average than the schools they applied to before the policy change.

Low-income students who took the ACT saw higher rates of admissions and attendance at more selective schools following the policy change. This finding supports existing evidence that high-achieving, low-income students may not be attending more selective schools simply because they never apply. This is particularly important because we know that low-income students benefit even more (as measured by lifetime earnings) from an education at a more-selective school than their middle- and high-income counterparts.

Should these results encourage ACT and SAT to increase the number of free score reports a student can send? Maybe. Should these organizations extend the registration fee waiver to cover the cost of sending more than four score reports? Probably—and the SAT especially may want to consider it as a way to start recouping market share lost to the ACT. But the real point of this research is to show that even small tweaks in the college application process can have big consequences.

The College Board’s website asks and answers the question, “How many applications are enough?” for students beginning to apply to college. It admits there is no concrete number for students to follow, but recommends sending anywhere between five and eight applications to ensure admission into a college of choice. But hardly every college-going student uses the College Board website. Instead, these students are more likely to use cues like the number of free score reports to help them form an estimate of the number of college applications to send.

Sending a corresponding number of applications to the four score reports probably seems reasonable, especially for a high school student who only gets about 38 minutes of college advising each year from his guidance counselor. The problem is that the arbitrary number can create an anchoring effect, causing students to apply to fewer colleges than they would if the number of score reports sent were higher. And applying to fewer schools can decrease a student’s chances of attending a more selective school, potentially depriving them of the opportunity for higher earnings. It seems that with stakes that high, we’d want to ensure prospective college students aren’t using rules of thumb and “magic numbers” from a testing company to guide them during the college application process.