Crunching numbers, quantifying attitudes

Coan and Chen said they are trying to contribute meaningful data to the campus dialogue as co-chairs of the USG’s new Analysis of Princetonian Attitude Committee (APAC).

The creation of APAC is one of the first accomplishments of USG president Connor Diemand-Yauman ’10, representing an attempt to improve the USG’s responsiveness to student concerns. The committee, Coan said, will address a range of projects aimed at improving USG operations and informing the broader campus community.

“When those within the USG approach the administration with ideas or concerns, we’re expected to present logical arguments and creative solutions that are well thought out and grounded in the reality of our undergraduate experience,” Diemand-Yauman said in an e-mail, adding that he believes statistical analysis is one of the most effective ways to gain an understanding of campus life.

He added that he, Chen and Coan formed APAC “to ensure that the USG is addressing issues that really matter to undergraduates and to prevent the USG from being led by unsubstantiated claims or anecdotal evidence.”

Coan said he became interested in the project after listening to conversations about University policy that were based on assumptions about student opinion, rather than data.

“There’s an assumption the administration is trying to kill the eating clubs, but recent eating club membership data don’t support that position,” Coan noted as one example.

Coan explained that the Committee on Background and Opportunity (COMBO) survey was a good first step toward quantifying student opinions, but he added that more information is still needed. “There are incredibly interesting things in the COMBO survey,” he said. “I’m very excited to try to find out the answers to some of the questions that remain.”

APAC’s first two projects will both build on COMBO, which found that students in higher socioeconomic classes were more likely to be happy at Princeton, join eating clubs and purchase all of the required materials for their courses, Coan added. Coan is also a contributing columnist for The Daily Princetonian.

Coan said that COMBO has not yet been properly analyzed. The original analysis, conducted by Vice Provost for Institutional Research Jed Marsh, grouped students by self-reported family social class. Of those surveyed, 51 percent of respondents self-identified as upper-middle class, though those respondents also reported family incomes ranging from less than $25,000 to more than $1,000,000. “There’s this incredible bunching that makes it very difficult to identify differences among people,” Coan said. Accordingly, Coan added, the committee will re-analyze the data by grouping respondents based on their family incomes.

Coan said that Marsh will not be involved in the re-analysis of COMBO data or the committee’s other projects, as he is “extremely busy.” Marsh was unavailable for comment.

Edward Freeland GS ’92, the assistant director of the Wilson School’s Survey Research Center, said in an e-mail that he is skeptical that the data need to be re-analyzed. “We need to stop quibbling about response rates and whether income is a better measure of social class than self-reported social class,” he said, adding that income and self-reported class will likely “end up being highly correlated.”

The committee also plans to conduct a follow-up survey, COMBO II, this March, Coan said, adding that he thinks changes like the expansion of financial aid and the introduction of the four-year residential colleges may have shifted student opinions since the survey was first conducted in spring 2007. In addition to repeating many of the questions from the first survey, COMBO II will also gauge student opinion on potential University actions that could level differences across socioeconomic classes, Coan explained.

Get the best of the ‘Prince’ delivered straight to your inbox. Subscribe now »

Coan said that the survey, which the committee has yet to finalize, will likely ask students who chose not to join an eating club whether or not they would have done so had financial aid policies been different. The survey may ask these students if they would have joined clubs had no financial disparity existed between dining hall and club contracts, or if incremental changes like financial aid for sophomore fees and social fees would have influenced their decisions, he added.

Freeland said he thought conducting COMBO II might be more useful than re-analyzing the results from COMBO. “If there is any evidence that differences in social class are related to students’ sense of attachment to Princeton, then we need to move ahead to a discussion of what to do about that,” he said, adding that “if the next COMBO survey is to have any value, it should focus on this question.”

The committee is recruiting students to analyze survey data, Chen added, and it is also attempting to convince statistics professors to incorporate USG data analysis into students’ final projects.

“We hope that releasing to students more campus life data substantiated with rigorous analysis would spark informed discussions within the campus community and student organizations,” Chen said.