When Discussions Change Minds

Study finds campus dialogue with racially diverse participants altered student views -- but conversation among those with differing views didn't.
April 11, 2007

A paper presented Tuesday at the American Educational Research Association annual meeting finds that a "meaningful dialogue" involving participants of multiple races tends to change the attitudes of college students. But, strikingly, it also finds that diversity of opinion in such a dialogue does not trigger students to reevaluate their existing beliefs. While the authors suggest this may stem from the fact that many study participants holding minority viewpoints likely kept their thoughts to themselves during the dialogue -- so diversity of opinion wasn't salient -- they point out that such a scenario is akin to that of a classroom discussion.

"Given the current legal context" in which several states prohibit affirmative action, "this finding is particularly noteworthy," the authors write.

"Opponents of racial diversity suggest that other forms of diversity, such as socioeconomic and ideological diversity, should take precedence over race,” write Nicholas A. Bowman and Joshua Gottlieb, graduate students at the University of Michigan, in their paper, “The Impact of Racial/Ethnic Structural Diversity and Opinion Diversity on Reconsidering One's Beliefs about College Access."

“Intriguingly this study shows that opinion diversity within a single-session dialogue does not substantially contribute to people's questioning their own beliefs.”

In the study, Bowman and Gottlieb ran 42 discussions with 671 adults of all ages (some college students, some not) on the issue of, “Who is college for?” Before each conversation began, the 10 to 20 participants gathered in a circle -- where they could clearly observe the racial diversity of the discussion group . They completed a pre-test questionnaire, indicating their level of agreement with such statements as “Only the best and brightest high school students should go to college,” “Race, ethnicity and gender should be considered in college admissions,” and “High tuition prevents many people from going to college.”

Following the pre-test, participants discussed three approaches to understanding college access, one suggesting that college should be for those who are willing to work hard for it, another that college should be for the most academically qualified and the last suggesting that college should be for everyone who wants to attend.

After the discussion, participants completed the same seven-item questionnaire distributed for the pre-test, and Bowman and Gottlieb determined the magnitude of attitudinal change. Racial/ethnic diversity was calculated based on the proportion of minority-group participants in the discussion, while diversity of opinions was determined by calculating the standard deviations for the pre-test answers provided by each participant in a group. The total discussion, including time to answer the questionnaires, lasted about 90 minutes.

Bowman and Gottlieb find no effect of opinion diversity or racial/ethnic diversity on attitudinal change among older adults not in college, suggesting, they write, that some of the benefits of exposure to diverse people and perspectives may not occur beyond certain developmental stages -- consistent with earlier findings that traditional college-aged students may be particularly primed for cognitive growth stemming from contact with diverse peers.

However, among college students, they find a positive relationship between racial/ethnic diversity within the discussion group -- independent of opinion diversity -- and attitude change: “In other words, meaningful interactions with a diverse group of people cause students from all racial/ethnic backgrounds to reconsider their current beliefs.” The authors add that the effect is greatest for students of color, and argue that further analyses ruled out the conclusion that students simply adopted more socially desirable views after the discussions.

Bowman and Gottlieb find that, on the other hand, there is no significant relationship between opinion diversity and attitude change. The authors suggest that perhaps this results from the fact that the format used for the study did not highlight opinion diversity. They refer to a 2004 study that found that opinion diversity did affect complex thinking, and note that in the 2004 study, an associate of the researcher strongly voiced the minority opinion within a group of four (as opposed to 10 to 20), whereas in their study, no one was required to speak in the dialogue and so participants with minority viewpoints may have stayed quiet.

“In fact," Bowman and Gottlieb write, "this silencing of minority opinions may accurately reflect group discussions in college classrooms.”

“This is not to say that opinion diversity is useless. It's just to say in a context in which people might not voice their opinions, it doesn’t have an impact on students reconsidering their beliefs,” Bowman, a Ph.D. candidate in higher education and social psychology at Michigan, said in a phone interview Tuesday.

“Essentially, you can’t substitute opinion diversity for racial diversity and expect to see the same effect,” he said -- at least on this one particular measure, designed to reflect the classroom setting. “Whether it be race-based affirmative action, per se, there needs to be a way to ensure not only the presence of racially diverse peers but also meaningful interactions across [races].”

“Currently, affirmative action seems to be the most meaningful way to accomplish this, but of course it’s becoming a less plausible way in some universities,” Bowman added. Michigan became the latest state to ban race-based preferences in public university admissions and hiring when state voters approved a ballot measure in November.


Back to Top