The effects of dispositional and situational attributions on evaluative responses to noncelebrity testimonial advertising
MetadataShow full item record
Testimonial advertising has been widely used to convey brand recommendations using an endorser or a group of endorsers on behalf of the advertiser. This type of advertising appeal involves a complex context in which consumer responses to the message are likely influenced by many perceptual and motivational factors. While a number of studies have examined endorsement effects over the years, relatively little attention has been paid to how consumers process such testimonial messages from a theory-driven point of view. This research attempts to examine causal associations between consumer perceptions generated by the endorser and the message’s persuasive effects, using an attributional approach. The present study is concerned with how consumers make causal inferences regarding the endorser’s motivation to support the brand – specifically, dispositional versus situational attributions – and how such inferences affect advertising effectiveness. Based on correspondent inference theory and additional theoretical perspectives such as the source credibility model and the elaboration likelihood model, it is hypothesized that consumers will generate predictable patterns of attributional responses to testimonial messages according to the level of source credibility and their personal involvement (product involvement and ad involvement). Further, it is posited that these responses, in turn, will influence the consumers’ evaluations of the advertisement and the brand being endorsed. Mediating roles of endorser-generated attributions in the causal links from source credibility and personal involvement to the selected indicators of persuasion are also proposed for hypothesis testing, as well as the effects of correspondence bias, a psychological trait derived from the correspondent inference theory. Two-stage randomized field experiments were administered to attain reliability and generalizability of findings, by using replicated measurements and a consumer panel sample for the second experiment. After three pretests designed to develop and refine the experimental instruments, Experiment I and Experiment II were conducted with 231 college students and 356 consumer panelists, respectively. Results support the differential effects of dispositional and situational attributions on the persuasiveness of noncelebrity testimonial advertisements. That is, while participants generated situational attributions to a greater extent than dispositional attributions, advertising effectiveness and the impacts of source credibility and personal involvement on persuasion were better predicted by dispositional attributions than situational attributions. An additional investigation of correspondence bias also confirmed its positive effect on consumer responses to testimonial messages, consistent with the effects of dispositional attributions. Based upon the empirical findings from the two experiments, theoretical and managerial implications, as well as directions for future research, are discussed.