Even in Modern Media, the Picture is Still the Same: A Content Analysis of Clipart Images - Statistical Data Included

Both psychological theory and empirical research have consistently indicated that human beings use media images to form cognitive schemas, and that these schemas can then have an effect on perceptions of ability and performance. Gender and ethnic biases are well documented in common media, such as television and picture books. This study examined images of human beings in two popular computer clipart packages, Microsoft Office 97 and Print Shop Ensemble III, to investigate whether this new medium would embody modern, egalitarian goals for gender and racial equality or would depict more traditional and differentiated views. As hypothesized, computer clipart was similar to other media, depicting Caucasian males more frequently and in more active/nonnurturant and desirable roles than any other group. Findings suggest that individuals using these programs to make business and educational materials more interesting and engaging may inadvertently activate maladaptive cognitive schemas.

Since cave dwellers began drawing pictures on walls, human beings have been interested in self-representation and have inferred meaning from the pictures that they create and view. Research has investigated the content and implications of images of people in specific print and televised media, and has consistently found gender biases and stereotypical messages. Recently, computer technology has increased the use of visual images in everyday media such as office bulletins, school flyers, memos, school presentations, and personal letters. The nature of these computer images, the messages they send, and the implications of such messages are unclear. In this study, we expanded upon previous research on common media images in our culture by examining the nature of human representations in popular computer programs.

Goffman's early analysis of print advertisements found gender-based discrepancies, with females depicted as shy, passive, and gentle, whereas males were depicted as dominant, and powerful (Goffman, 1976). Since that time, many researchers have investigated gender stereotypes by conducting content analyses of "common images" presented by popular media (e.g., cartoons, children's books, magazines, music videos, and television). Although individual studies each investigate a specific media domain, these content analyses yield similar findings, showing disparities between depictions of male and female characters. For example, research on books for preschool and early elementary-age children has found differences in both visual images and text descriptions based on characters' gender. Males generally appear in illustrations more often than females and are more likely to be cast in central roles and to be cited in the titles than females (McDonald, 1989; Tepper & Cassidy, 1999). Females tend to be portrayed in pas sive or nurturing roles or both (e.g., mother, nurse, or secretary), whereas males are portrayed in active/nonnurturant, productive or aggressive roles or both (e.g., baseball player, construction worker, or scientist; Crabb & Bielawski, 1994; McDonald, 1989; Tepper & Cassidy, 1999).

Research on diverse forms of television programming has found similar results. In televised music videos, men appear twice as often as women, and engage in significantly more aggressive and dominant behavior. Women in these videos are significantly more likely to be depicted as the passive object of sexual advances than men (Sommers-Flanagan, Sommers-Flanagan, & Davis, 1993). Televised commercials depict women as sexual targets and men as active and dominant, significantly more than the opposite (e.g., by showing women in an undressed state or revealing clothing, see Browne, 1998; Coltrane & Adams, 1997; Willemsen, 1998; Zebrowitz McArthur & Resko, 1975). Even in children's cartoons, gender stereotypes are demonstrated (e.g., Chu & McIntyre, 1995; Dines, 1995). Thompson and Zerbino's review of 20 years of cartoons demonstrates that changing gender roles and increasing societal awareness of gender biases have not influenced the disparities in the frequency and nature of representations of males and females (T hompson & Zerbino, 1995). The modern disparities in cartoons, television commercials, and music videos parallel Goffman's longstanding finding that print advertisements show women, but not men, in predominantly domestic and dependent roles (e.g., Goffman, 1976; Whipple and Courtney, 1985).

In fact, these inequities pervade educational and professional environments as well as entertainment media. Hogben and Waterman (1997) examined textbooks for introductory psychology and found that women and ethnic minorities were underrepresented in both text and illustrations in these books. Chappell (1996) examined educational software designed to teach mathematics in elementary school classrooms through the use of engaging stories and activities, and found that the majority of the characters depicted in the stories were male. Further, the proportion of female characters decreased as the intended grade level of the game player increased. Ogletree, Merritt, and Roberts (1994) found that even postage stamps were much more likely to depict men than women, and that men were more likely to be depicted in more active roles, less likely to be depicted in nurturing roles, and engaged in more diverse activities, while women were generally portrayed in a limited number of traditional, nurturing activities.

Postage stamps and television are certainly pervasive, and many children are exposed to educational software in school or home settings or both. However, individuals in modern society may be exposed to computerized images as frequently as to television, in both computer-based applications and printed material. Clipart graphics (the images contained in word processing programs, inserted into desktop published documents and visible on virtually every website) are currently used in most business and educational settings. For example, clipart has been incorporated into educational materials from kindergarten through college (Abramson, French, Huss, & Mundis, 1999; Chappell, 1996) and in psychological surveys for adults (Francis & Irwin, 1988; Marschalek, 1988). Use with younger children (preschool and early primary grades) is also quite widespread. Clipart has been used with children to foster creative learning (Clements, 1995; Sisson, Mayfield, & Entz, 1985), to facilitate story telling (Riding & Tite, 1985), f or play therapy (Kokish, 1994), and for diagnosing dyslexia (Karnes, 1982). Despite the ubiquitous presence of clipart in educational settings, no previous research has analyzed the content of these images.

Historical and recent research documenting pervasive and systematic depictions of females and males in stereotypical roles have led theorists to explore how these images may affect an individual's concept of one's social role and related rules for behavior which may limit the aspirations of some individuals, especially those who are depicted in a restricted range of activities and contexts. Goffman's discussion of individual consciousness suggests that our perception of reality is constructed through exposure to commercial symbolism (e.g., books, advertisements, cartoons, and television; Goffman, 1959). As a result, Goffman (1959) and others (e.g., Gordon & Chafetz, 1990) suggest that, in modern society, media depictions of men and women are one of the fundamental sources of our gender identity. The "cognitive miser" theory refines this view, proposing that humans rely on common images to create stereotypes, and then use those stereotypes as a decision-making aide in order to conserve mental resources (Macra e, Milne, & Bodenhausen, 1994; Sherman & Frost, 2000).