Changes in Public Attitudes: A Comparison of Content Analysis of Media and Survey Results

Regents Of The University Of Michigan - Ann Arbor, Ann Arbor MI

Investigators

Abstract

In the mid-1940s, public opinion surveys began to gather trend data on racial attitudes in America. There, however, is no survey data or other reliable evidence to answer questions about attitudes prior to the mid-1940s. This study will address this by completing a systematic content analysis of newspapers and periodicals from 1930 until 1946 - the latter date the first year when even minimal national trend survey data became available on white racial attitudes. The investigator has developed an extensive set of categories and has used these to analyze a random sample of newspaper articles newspapers over some 35 years in order to determine what, if any, changes occurred that reflected or influenced white racial attitudes. A second research goal is to study attitudes in the years from 1946 to the mid 1960s. For these years, unlike the earlier 1930 to 1946 period, there are repeated national surveys to which the results of the content analysis can be compared. This enables the investigator to indirectly isolate the extent to which analyzing newspaper texts can contribute to understanding changes in attitudes between the earlier 1930 to 1946 period. Initial findings provide promising evidence of considerable change: for example, the context analysis of newspaper articles yielded a statistically reliable increase in the stories about black accomplishments as against reports of alleged black crimes. This study will examine critically other areas in which change has occurred, including shifts in attitudes about gender equality, and continue to develop the methodology that allows one to determine the association of findings from content analyses of newspaper articles and those based on available national surveys over the same period. Broader Impact The research has the potential to increase understanding of attitudes toward minorities during the Depression and World War II years. The research will also help to assess the value of content analysis of mass media in identifying attitudinal trends among the general public, thus complementing and supplementing the evidence obtained from sample surveys. The attempt to compare systematically two very different types of social science data should yield insights into the advantages and limitations of each, and contribute to an understanding of whether and how content analysis can shed light on attitudes when trustworthy survey data are not available or are suspect because of limitations (for example, low response rates). The results of the research will be disseminated through presentations and articles that cross disciplinary boundaries, reaching historians, sociologists, communications researchers, and others, and therefore adding to cross-disciplinary communication and awareness. In addition, the systematic comparison of findings from content analysis and survey data has the potential to be of value to graduate and undergraduate students of social science research methods, and to be of interest both to those studying social change generally and attitudes about different types of human rights, specifically.