An exploratory field study was conducted to examine internal and external factors that influence adolescents' consumer behavior when serving as the family purchasing agents. Demographic, lifestyle, and marketing activities were examined to determine the influences that affect whether the adolescent will purchase the preferred family brands or other brands. Participating adolescents were sent by their parents to the grocery store on two separate occasions to purchase four preselected grocery items. The brands purchased were recorded and compared to the preferred brand names provided by the parents. While no statistical significance was found, occasional trends were observed. The analysis indicated that adolescents who experience a pluralistic family communication style will purchase products other than the preferred household brands. Adolescents who are exposed to television and radio tend to deviate more from the preferred family brands more often than do adolescents with less media exposure. Adolescents who work are more likely to go to the grocery store more often for their families than do nonworking adolescents. Also, adolescents seem to possess a price sensitivity to both high and low-involvement grocery items.

This research began with the Hill Inventory. Cognitive style preference variables were classified as one of following four types: Theoretical Codes, Qualitative Codes, Social-Cultural Codes or Reasoning Modalities. A consumer behavior perspective was then used to form an alternative structure for the Hill Inventory variables. The following three constructs were proposed: Evaluation Codes, Perceptual Codes, and Reasoning Modalities. The purpose of this research was to assess the structural form and measurement validity of the Hill Inventory. Specific steps taken to accomplish this objective included: developing confirmatory factor and structural equation models; using the LISREL software package to analyze the model specifications; and assessing the validity of the questions used to measure the variables. A descriptive research design was used to compare the model specifications. The research instrument consisted of eight statements for each of twenty-eight variables for a total of 224 questions. Five-point response choices were described by the words: often, sometimes, unsure, rarely, or never. The sample consisted of 285 student subjects in marketing classes at a large university. Data analysis began by comparing the distributions of the data to a normal case. Parameter estimates, root mean square residuals and squared multiple correlations then were obtained using the LISREL …

The problem examined in this study was the nature of compulsive purchasing behavior. Three proposed models depicting this behavior as a learned adaptive response to anxiety and/or depression, an addiction, and a personality disorder were introduced and discussed in Chapter I. Background information concerning the areas examined in the models was presented in Chapter II. The research methodology was discussed in Chapter III and the findings of the research presented in Chapter IV. A summary, conclusions, implications, and recommendations were presented in Chapter V.