In "Methodological Issues in the Use of Published Cartoons as Data", researchers Ellen Giarelli of the School of Nursing's Center for Autism Epidemiology and Research and Lorraine Tulman, both of the University of Pennsylvania, explore the methodolocial approaches to analyzing images, especially cartoon images, as reflection of public opinion on issues of publich health. They cite HyperRESEARCH for its ability to catalog and code entire images or portions of images as a qualitative tool for image based source material.
Their paper discusses "When common understanding of a phenomenon is under investigation, mass media representation in general, and cartoon images in particular, are a useful source of data. Sample selection, data collection, and the analysis of constructed images differ from other kinds of data. Cartoonists may intend to stimulate multiple interpretations among readers. Uncovering these interpretations is essential to understanding public discourse of the phenomenon of interest. Semantic validation is used to assess the degree to which the meanings of text relative to their context are accurately represented. Analysis of the image and text of cartoons has the potential to yield important understanding of public discourse surrounding issues of the public's health and well-being." Appearing in the September 2003 issue of Qualitative Health Research (Vol 13, No. 7, pp 945-956), the article can be access online here.