New Survey Method Gets Closer to the Truth on Controversial Topics

Contact: Jay Peters, 581-2355; George Manlove, 581-3756

ORONO — UMaine School of Social Work researchers are reporting positive results from a new type of opinion questionnaire that reveals more accurate responses to controversial, even “ugly” questions about social attitudes.

For sociologists wanting to know how people feel privately about racism, homophobia, sexuality and other controversial subjects, the new survey method will provide more and better results, according to Jay Peters, a researcher and lecturer in the School of Social Work.

In a study exploring possible benefits of a new way of asking questions on surveys, researchers Peters, Winston Turner of the UMaine School of Social Work faculty and Carey Nason, director of UMaine’s Safe Campus Project, surveyed nearly 700 undergraduates at a university in the Northeast about their attitudes towards sex, sex roles, fighting and drinking.

Half of the sample used a traditional two-answer “forced choice” method and half used a newer “phrase completion” survey, allowing much greater variation in answers. Researchers found that respondents using the latter survey were willing to acknowledge having negative attitudes that others, using the traditional survey, would not.

“In survey research on racism, homophobia and misogyny, people either refuse to answer or select the most politically correct answer,” Peters says.

With the new phrase-completion format, however, participants completed 55 percent more of the questions. Better yet, while 90 percent of participants using the forced choice format denied having the negative attitudes, only 47 percent did so when using the phrase-completion format.

Traditional “forced choice” surveys with only two-answer options were designed to prevent people from giving socially desirable answers. The research shows that traditional survey methods often fail to get at the truth and can lead to false conclusions, according to Peters.

When they used the phrase-completion format, which allows respondents to qualify answers in a 1 through 10 index, a bell curve profile emerged, Peters says, which showed people are not absolute in their approval or disapproval of some negative attitudes.

“This research indicates that a simple change in how questions are presented may result in a much better understanding of negative attitudes and ultimately how those attitudes related to behavior such as rape and hate crimes” says Peters.

Peters, Turner and Nason published the results of their study in the September 2007 issue of the journal “Social Work Research.” The article is available at

Peters intends to use the new model, introduced in 2003 but seldom used by social science researchers, for future social attitude studies. “Our thought is this is going to be a much better way of getting at their true feelings,” he says.