C. Yoo, D MacInnis /Journal of Business Research 58(2005)1397-1406 by which brand attitudes are formed, depends on the nature meaningful or relevant to them. For example, recent work of the ad to which consumers are exposed. We begin by by MacInnis et al.(2002) shows that ads regarded describing the theoretical constructs that comprise our ideas. credible and meaningful are associated with increasing sales, even after controlling for other variables, including other evaluative thoughts 2. Constructs and hypotheses 2.1.2. Feelings We anticipate that informational vs emotional ad formats Studies have also focused on feelings(also known as create very different"routes"to persuasion, though each emotions)as predictors of ad attitudes( see review by Brown route depends critically on the evocation of a set of common et al.(1998). Feelings are defined as acute, transitory and responses. Fig. I identifies these constructs and summarizes specific affective experiences that occur as a result of some our ideas about the different routes to persuasion involved. experience(Holbrook and O'Shaughnessy, 1987) We define each construct used in the models below and then develop the hypotheses reflected in Fig. 1 2.1.3. Beliefs and ad attitudes(Aad) Many long-standing models of persuasion(e. g, Fishbein 2.1. Constructs and Ajzen, 1975; Lutz, 1975; Petty and Cacioppo, 1986; Petty and Wegener, 1993; Meyers-Levy and Malaviya, 2.1.1. Credibility and meaningfulness thoughts (hereafter 1999)hold that persuasion is contingent on a communica- credibility) tion's capacity to alter consumers' beliefs. Beliefs are Advertising research has long examined the role of defined as the strength with which a consumer associates evaluative responses or judgments of ads as predictors of a brand with a certain attribute or outcome(e.g, how brand attitudes(MacInnis et al., 2002; Batra and Ray, 1986; strongly they believe that a product has a whitening Holbrook and Batra, 1987; Edell and Burke, 1987; Homer ingredient or that it makes teeth 50% whiter than regular and Yoon, 1992; MacKenzie et al., 1986; MacKenzie and toothpaste brands). Notably, although beliefs are clearly Lutz, 1989; Stayman and Aaker, 1988). Indeed, a variety of linked to persuasion(e. g, Mick, 1992; Smith and Swinyard, scales of ad response have been developed to characterize 1982; Deighton, 1984), they are not the only predictor of consumers'evaluations of ads(e. g, Schlinger, 1979; Aaker brand attitudes (MacKenzie et al, 1986). Indeed, when and Norris 1982). Among the evaluations central to involvement in a purchase decision or product category is consumers'ad reactions, included are their assessments of low, consumers'brand attitudes may be more affected by the credibility of the ad and the extent to which it is their overall attitude toward (or global liking for) the ad tha a Emotional Ad Format HIb, H2b Credibility Negative H3 Belief Feelings b Informational Ad Format Aad Credibility HSb. H8b Feelings Fig. 1. Hypotheses and structural modeL.by which brand attitudes are formed, depends on the nature of the ad to which consumers are exposed. We begin by describing the theoretical constructs that comprise our ideas. 2. Constructs and hypotheses We anticipate that informational vs. emotional ad formats create very different ‘‘routes’’ to persuasion, though each route depends critically on the evocation of a set of common responses. Fig. 1 identifies these constructs and summarizes our ideas about the different routes to persuasion involved. We define each construct used in the models below and then develop the hypotheses reflected in Fig. 1. 2.1. Constructs 2.1.1. Credibility and meaningfulness thoughts (hereafter credibility) Advertising research has long examined the role of evaluative responses or judgments of ads as predictors of brand attitudes (MacInnis et al., 2002; Batra and Ray, 1986; Holbrook and Batra, 1987; Edell and Burke, 1987; Homer and Yoon, 1992; MacKenzie et al., 1986; MacKenzie and Lutz, 1989; Stayman and Aaker, 1988). Indeed, a variety of scales of ad response have been developed to characterize consumers’ evaluations of ads (e.g., Schlinger, 1979; Aaker and Norris 1982). Among the evaluations central to consumers’ ad reactions, included are their assessments of the credibility of the ad and the extent to which it is meaningful or relevant to them. For example, recent work by MacInnis et al. (2002) shows that ads regarded as credible and meaningful are associated with increasing sales, even after controlling for other variables, including other evaluative thoughts. 2.1.2. Feelings Studies have also focused on feelings (also known as emotions) as predictors of ad attitudes (see review by Brown et al. (1998)). Feelings are defined as acute, transitory and specific affective experiences that occur as a result of some experience (Holbrook and O’Shaughnessy, 1987). 2.1.3. Beliefs and ad attitudes (Aad) Many long-standing models of persuasion (e.g., Fishbein and Ajzen, 1975; Lutz, 1975; Petty and Cacioppo, 1986; Petty and Wegener, 1993; Meyers-Levy and Malaviya, 1999) hold that persuasion is contingent on a communica￾tion’s capacity to alter consumers’ beliefs. Beliefs are defined as the strength with which a consumer associates a brand with a certain attribute or outcome (e.g., how strongly they believe that a product has a whitening ingredient or that it makes teeth 50% whiter than regular toothpaste brands). Notably, although beliefs are clearly linked to persuasion (e.g., Mick, 1992; Smith and Swinyard, 1982; Deighton, 1984), they are not the only predictor of brand attitudes (MacKenzie et al., 1986). Indeed, when involvement in a purchase decision or product category is low, consumers’ brand attitudes may be more affected by their overall attitude toward (or global liking for) the ad than Fig. 1. Hypotheses and structural model. 1398 C. Yoo, D. MacInnis / Journal of Business Research 58 (2005) 1397 – 1406