Show simple item record

dc.contributor.authorScales, Monica Beckham
dc.description.abstractIn Part I of this dissertation, Self Categorization Theory (SCT) and the concept of prototype group health behavior were used to generate and test a model that examined the influence of in-group prototypes on adolescent risky health behavior. Participants (N = 325, mean age 14.92) were low income high school freshmen recruited from an urban and a rural school. The sample was approximately evenly split between male and female, and Black and White high school students. Independent measures included measures of group identity, perceived group behavior, perceived prototype group member behavior, race and gender. Dependent measures were self-reported behavior and behavior intent measures for four health risk behaviors (cigarettes, alcohol, marijuana & sex). The hypothesized model significantly predicted all four risk behaviors, and three of the four risk behavioral intentions indicating that the model had good predictive power. Importantly, neither race nor gender interacted with any of the concepts in the model, suggesting that the model worked well across subgroups. In Part II, a new way to assess identity accessibility was proposed. Attitude and social norm accessibility commonly are measured using reaction time tasks (valence of reaction, e.g., like/dislike, multiplied by how quickly one responds). This dissertation applied that measurement to social group identity. Using the same sample noted above, independent measures included attitude, social norm and group identity reaction time measures as well as a typical group identity task (included for a validity test). Results were disappointing: The social identity measure showed poor validity and, in hypothesis testing, only predicted 1 of 4 risk behaviors for adolescents. Other findings concerning attitude, family, and social norm accessibility were complicated. Attitude toward the behavior (Like/dislike), and social/family norm valence predicted intent to engage in a behavior and self-reported behavior. However the proposed reaction time * valence interactions were not significant for 6 of 8 outcomes with the exceptions of intent to smoke marijuana and intent to have sex. Limitations and problems with measurement are discussed. The discussion describes the utility of the concept of prototypes and social categorization theory in studies of health communication and adolescent risky health behaviors.
dc.subjectSelf-categorization theory
dc.subjectadolescent health
dc.subjectrisky health behaviors
dc.subjecthealth behaviors
dc.subjecthealth communication
dc.subjectreaction time measures
dc.subjectstructural equation modeling
dc.titleSelf categorization theory
dc.title.alternativepredicting adolescent health behavior
dc.description.departmentSpeech Communication
dc.description.majorSpeech Communication
dc.description.advisorJennifer Monahan
dc.description.committeeJennifer Monahan
dc.description.committeeJeff Springston
dc.description.committeeTina Harris
dc.description.committeeJerold Hale
dc.description.committeeVicki Freimuth

Files in this item


There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record