Tufts gets creative on admissions
Application process will go beyond SAT scores and seek original thinking
MEDFORD -- A majority of the 15,000 young people who apply to Tufts University each year look impeccable on paper: good grades and test scores, plenty of extracurricular activities.
But there's only room to admit a quarter of them, so who deserves to make the cut?
In choosing among its most promising applicants, Tufts and other selective colleges have leaned heavily on subjective judgments -- based on interviews, essays, and recommendations -- to pick the applicants they believe will do the best in the classroom and later in life.
This fall, Tufts will try a more scientific approach to its toughest decisions, using its application to measure aspects of intelligence that cannot be approximated by SAT scores. A top Tufts dean believes that creativity and practical skills -- the ability to implement ideas and win other people's backing -- are just as important as the analytical skills typically measured by standardized tests.
The university will ask applicants to show original thinking and imagine, for example, an alternative version of history: What if civil rights pioneer Rosa Parks had given up her seat on the bus? Or, they could be asked to write an off-the-wall ministory with the title, ``My roommate is a space alien."
The first question might not sound so different than those on a typical application essay, but this year's questions will be designed and evaluated based on psychological research. Tufts officials hope to better identify future leaders and predict college grades. Such methods could also boost diversity among those accepted , because research indicates that the assessment erases much of the gap between racial and socioeconomic groups seen on traditional standardized tests.
``If you want to admit people who are going to be leaders in tomorrow's world, which every university says it does, focusing on [grade point average] and SATs does not get you very far," said Robert J. Sternberg, Tufts' new dean of arts and sciences and a psychologist who is directing the pilot project, based on research he did as a Yale professor.
Word of Tufts' experiment generated buzz at an admissions training seminar at Harvard last month, and deans at MIT and Stanford said they would be watching closely in the next few years to see if Tufts is successful. Many admissions directors yearn for more effective ways to sort through growing volumes of applications that look polished, but similar. And they want tools to identify the spark of talent in disadvantaged students whose applications may be less refined.
Sternberg and Lee A. Coffin, dean of undergraduate admissions, are still hashing out the questions they will include on applications for the freshman class entering in fall 2007. The new questions will be optional, and each student will choose only one or two. Each question will be designed to measure creativity, practical abilities, or the potential for leadership.
The new questions might include one geared to evaluate a student's ability to make good judgments: Your roommate has the answer key to a test and has offered to share it with you. What do you do?
There will be no right answer, Sternberg and Coffin say. Rather, Tufts will evaluate the responses on the basis of their originality or feasibility, depending on the case.
Sternberg and Coffin have yet to decide how they will factor the scores into a student's overall rating. But they hope to end up with at least a small group of students whose answers essentially won them admission. Then they'll study how those students do at Tufts, compared with those admitted based on traditional measures.
Tufts began experimenting last year with optional questions. Coffin recalls one young man who answered a question about the legacy he hopes to leave. Bobby didn't have the best grades, but charmed the admissions office with his essay on how he wanted to build an independent Scottish republic that would remain friendly with England. He will be a freshman in the fall .
If Sternberg's assessment does a better job finding the kind of students who will become leaders, Tufts may expand the set of questions. But first, the school wants to see the results.
``A year from now, I could say this made no difference, or we could have four dozen Bobbys we wouldn't have seen before," Coffin said. ``I don't have any way of knowing yet."
Sternberg and Coffin said they don't expect the SAT scores of such students to be lower, because they will continue to draw from candidates with impressive credentials. Already, Tufts gives less weight to mediocre scores of some promising applicants.
Choate Rosemary Hall, a prep school in Connecticut, is experimenting with the same ideas. Choate officials are deciding whether to give applicants this fall the option of taking an assessment licensed from Sternberg, in addition to the standardized test they already use.
``Applications are becoming more and more useless, between grade inflation and people writing recommendations with the fear that they'll end up in court" if they say something negative, said Ray Diffley, Choate's director of admission. ``Admissions offices need to become more expert in assessment."
Sternberg says he has the data to prove that this approach works. He developed his own test, longer and more involved than the essay questions on Tufts' new application, but based on the same ideas. It was given to about 800 college students across the country.
Sternberg says he found smaller racial and economic differences in scores than on the SAT. He also found that by using his test along with the SAT, he could better predict a student's grades during the first year of college.
Some psychologists doubt Sternberg's theory that creativity and practical abilities are distinct aspects of a person's intelligence, while others say they worry that he risks creating another flawed test.
``If I really want to know if someone is creative, I involve them in a lengthy, open-ended project, like inventing a better mousetrap or writing a play," said Harvard professor Howard Gardner, who has his own theory of intelligence. ``If it's a Mickey Mouse task -- like, `What are the five uses of a paper clip?' -- it has no predictive value about whether you will ever create anything worth noting."
But some admissions deans are eager to see Tufts' results.
``If [Sternberg] is able to document that his measurements supersede economic background, that would be wonderful," said Marilee Jones, dean of admissions at the Massachusetts Institute of Technology.
Bombardieri can be reached at email@example.com.