New Admissions Exam Takes Unconventional Approach
Stacy Berg Dale is weary of the news media and what she suspects is jaded reporters' excessive interest in the "Lego test."
Yes, nine top-notch colleges and universities are now using an admissions exam that judges high school students on their ability to build a robot out of the colorful, interlocking plastic bricks that are so popular with children, said Ms. Dale, one of the researchers implementing the test. But no, the test isn't childish.
In fact, she said, the test may help predict college-success rates if used in combination with several other measures that are part of the exam, a complicated pilot project started last fall.
"The test is trying to capture traits that aren't normally captured by standardized-test scores and grades," said Ms. Dale, a research associate with the Andrew W. Mellon Foundation in New York City, the philanthropy that put up $1.9 million to finance the project. "We're looking for leadership, communication skills, and initiative."
If found valid by researchers, the two-part exam should be used to gauge college ability in conjunction with such traditional information as standardized-test scores, grades, and teacher recommendations, Ms. Dale said.
The test is based on criteria for predicting college success devised by the Posse Foundation, a New York City-based group that has helped poor and minority students graduate from top institutions over the past decade. Deborah Bial, the organization's founder, created the exam as a project for Harvard University's graduate school of education, where she is a doctoral candidate.
But Ms. Dale contends that nobody is really interested in the test's thoughtful particulars. All that people seem to want to talk about are the Danish-made toy building elements.
Ms. Dale said last week that she had been so bombarded by phone calls from reporters, who want to know exactly how the manipulation of Lego bricks can predict college success, that she hadn't gotten any work done in days.
The truth is, the Lego portion of the admissions assessment is a psychology experiment more than it is a brainteaser, Ms. Dale said.
Not Exactly Child's Play
Groups of eight to 10 students are asked to construct a robot within 10 minutes. Each team member is allowed one quick glance at a model Lego robot that is kept in another room. Only one member of the team is permitted in the room at a time.
"We're not looking to see how kids play with Legos, we're looking to see how they interact," Ms. Bial said.
Meanwhile, evaluators are watching the discussions to see which members of the group show qualities that will be of use in college. Students get points, for example, if they exhibit leadership or innovative thinking.
Similar forums require students to develop a product and a plan to market it, Ms. Dale said. In another session, students pull discussion topics from a brown bag and take turns leading group talks.
Students also participate in a 30-minute interview to demonstrate speaking skills and articulate personal goals.
"A lot of kids get missed by traditional measures," Ms. Dale said. The experimental test, she said, may help "broaden the way admissions officers think about students."
Moreover, this is one exam students actually enjoy taking, Ms. Bial added. While students know they are being scored, they also understand they are being tested on what they do best: expressing individualism.
"They leave elated," Ms. Bial said.
Researchers plucked some 700 mostly minority students from public schools in New York City and administered the test to them last fall. Those with the top 100 scores were offered a $3,500 renewable scholarship to attend any of the nine colleges and universities participating in the program.
Each of the institutions will accept up to 20 students, provided they meet admissions criteria.
The colleges and universities participating in the project are: Beloit College in Beloit, Wis.; Carleton College in Northfield, Minn.; Colorado College in Colorado Springs, Colo.; Grinnell College in Grinnell, Iowa; Macalester College in St. Paul, Minn.; Pennsylvania State University in University Park, Pa.; Rutgers University in Newark, N.J.; the University of Michigan in Ann Arbor; and Delaware State University in Dover.
The researchers will track students during their college careers, collecting grades, retention rates, and information about extracurricular activities, Ms. Bial said.
Administrators at participating colleges say they aren't yet ready to discard their current admissions practices.
Some, however, say they hope such an exam could eventually provide a suitable replacement for the use of race as an admissions factor—a practice that has faced growing legal and political challenges. Many admissions counselors say they simply want more clues as to students' strengths and weaknesses.
"We do not expect to throw out the tried-and-true admissions criteria," said Terry Swenson, the dean of admissions and financial aid at the 1,900-student Colorado College. "We are trying to find ways to identify characteristics that are essential to college. We're trying to get a sense of perseverance and motivation. Standardized tests don't do that."
While the researchers involved are hesitant to predict the results, they hope their findings will be similar to those enjoyed by the Posse Foundation. Each year, that organization sends 50 students from inner-city New York to highly competitive institutions, including Vanderbilt University in Nashville, Tenn., and Brandeis University in Waltham, Mass. The foundation reports that 90 percent of the participants graduate.
By contrast, only about 30 percent of all other students enrolled in college graduate in four years, the U.S. Department of Education reports. Only 14 percent of African-Americans and 11 percent of Hispanics earn a bachelor's degree within six years.
Striking a cautionary note, George F. Madaus, an education professor at Boston College and a leading testing expert, said that even if the "Lego test" proves valid, many institutions will be hesitant to use it because of time and money constraints. Such exams "are more costly and tend not to last very long, because they can't compete with technology, which is cheaper and faster," Mr. Madaus said.
But those points of contention will only be of concern in the distant future, said Ms. Dale of the Mellon Foundation.
First, it must be determined if the test works.
"Admissions officers know itis very hard to predict who can succeed in college," she said."If [the test can help] tap into that, this would be a breakthrough."
Vol. 19, Issue 23, Page 6