Presidential Address Polling as Political Science and Polling as Journalism

The mass media and the survey research profession share many attitudes and beliefs about what they do and their contributions to a functioning democracy. There are, however, some cultural differences that create a source of conflict. Our work is based more in science, theirs more in art, although each is, of course, a blend of the two. As one who has worked in both fields, I shall offer some notes, and perhaps the beginnings of a theoretical perspective, on some of the sources of conflict. Several years ago, a popular young instructor at my university was denied tenure by the department of political science. His publications, the department chair explained, were more like journalism than political science. We in the school of journalism tried not to take that personally. But I understood the dilemma. In my newspaper career 1 was often accused of writing stuff that was more like political science than journalism. One case stands out in my memory with particular clarity. Jimmy Carter was challenging President Jerry Ford for the presidency in 1976, and the pundits were writing about Carter's religiosity and how it was tapping a conservative strain in the electorate and helping his cause. Then one of the major polls published a simple crosstab: it showed that Carter did neither better nor worse among people with strong religious feeling. After that the punditry focused on the failure of Carter's religiosity to sway the voters. That happened to be the one presidential election year when KnightRidder Newspapers had its own national poll, and 1 saw a chance to challenge the pundits. Remembering Morris Rosenberg's elegant work on The Logic of Survey Analysis, I checked to see if age was working as a suppressor variable here. Carter had a strong appeal for young people, but the young are less religious. Sure enough, those young, non-churchgoing Carter fans were masking the religion effect. When age was held constant, religion effect appeared. Revealing it took nothing more complicated than a three-way table.