Introduction Manufacturers have recently been using comparative advertising for the apparent purpose of producing clearer and more favorable brand images. A number of empirical studies have measured the effects of manufacturer sponsored comparative advertising and identified some situations where it is more effective than noncomparative advertising. However, no study has addressed the feasibility of retail sponsored comparative advertising. Statement of the Problem The importance of store image to the retailer is well documented. Advertising plays an important role in the development and enhancement of store images. Despite this, very few guidelines exist to aid retailers in their search for advertising approaches which will produce favorable store images. Therefore, this study dealt with store image. It explored the degree to which store image was enhanced when a retailer employed comparative advertising. Methodology An after-only with control group experimental design was utilized to generate data measuring the relative effects of North Carolina department store sponsored comparative and noncomparative advertising. A preliminary study facilitated development of four mock comparative and noncomparative newspaper advertisements. Together, the ads stressed sponsoring store superior performance across seven store image attributes (easy to find items you want, friendly personnel, reasonable prices, good buys, good service, fast checkout, and low pressure salespeople). Comparative ads claimed sponsoring store superior performance on these seven attributes over three competing stores. Three dependent variables were used to measure comparative and noncomparative advertising effects. These variables were: (1) overall store image as measured by a multiattribute attitude model, (2) performance evaluations of the seven advertised attributes and (3) belief levels for claims made in experimental ads. The study hypothesized noncomparative, comparative and control group differences in these dependent variables. Analysis of variance was used in conjunction with the Scheffe multiple comparison test procedure to identify significant group differences. Results and Conclusions The results of hypotheses testing produced the following major findings: Hypothesis 1 No significant group differences in overall store images. Hypothesis 2 No significant group differences on: low pressure salespeople, good buys on products, good service, reasonable prices for value, and fast checkouts attributes. Comparative advertising significantly less effective than current advertising (control group scores) for "easy to find items you want" attribute. Noncomparative advertising significantly less effective than current advertising for "friendly personnel" attribute. Hypothesis 3 Noncomparative advertising significantly more effective than comparative advertising for 9 out of 28 ad claims and significantly more effective than current advertising on four claims. Comparative advertising significantly less effective than current advertising for three claims. Hypothesis 4 In general, noncomparative advertising was significantly more effective than comparative advertising promoting attributes and making claims aimed at specific segments as defined by age, sex, income and store loyalty. In a smaller number of situations, the noncomparative approach was significantly more effective than current advertising. The results of hypotheses testing clearly showed comparative advertising producing effects significantly less favorable than noncomparative or current advertising. Whether or not significant group differences existed, noncomparative advertising consistently produced effects more favorable than comparative or current advertising. Although it has limitations, this study strongly concludes that retailers should avoid the use of comparative advertising. / Ph. D.