Political knowledge is a central variable in most models of political behavior. For example, less knowledgeable people are less likely to vote and more likely to rely on heuristics in their vote choice. After a detailed analysis of hundreds of survey questions, Delli-Carpini and Keeter (1996) recommended a five-item knowledge scale to measure political knowledge. However, this standard scale is based entirely on measures of verbal knowledge. To a considerable extent, however, political information is visual. Graber (1988, 2001) has argued for a long time that ignoring visual information biases our estimates of political knowledge.
I propose to examine the relationship between visual and verbal political knowledge. If the two are largely independent, traditional measures of verbal political knowledge do not adequately reflect the concept we intend to measure. Moreover, I analyze whether predictors of political knowledge are related in similar ways to verbal and visual knowledge and to what extent the well-known effect of verbal knowledge on vote likelihood extends to visual knowledge.
Data for this study is collected through an original representative survey of the U.S. voting age population. Using a web-based survey instrument allows me to include visual stimuli in the questionnaire. In particular, the survey tests visual recognition (who is the person?), visual identification (which of the four persons is Dick Gephardt?), and visually-aided verbal knowledge questions (respondents are asked which job is held by Dick Cheney while a photo of Cheney appears on the screen). Random assignment to the same question with or without visuals provides tight experimental control and allows me to determine the exact effect of including visuals.
To my knowledge, this is the first time that visual political knowledge is assessed for a representative sample of the population. In addition to answering substantive questions on the relationship between visual and verbal knowledge, this study illustrates how web-based surveys can take advantage of the medium in ways not feasible in phone or face-to-face interviews.