No abstract

MacKenzie, Lutz, and Belch (1986) test four advertising attitude models and find that the Dual Mediation Hypothesis is the best. This research proposes an extended model within an online context, using intentions to return (Ir) to a Web site versus purchase intentions, with a direct path between attitudes toward the Web site (Asite) and Ir. This path is hypothesized as Web sites contain informative or entertaining content that attracts subsequent visits, and Ir depends on other non-brand-related factors such as security, ease of use, transactional capabilities, etc. Data from visitors to three actual Web sites––digital cameras, watches, and a charity––demonstrate significant relationships between Asite and Ir. In support of this perspective, when Asite was decomposed into its claim and non-claim components, the non-claim component had a significant effect on Ir for all three sites. Implications for online researchers and advertisers are discussed.

The relationship between attitude toward the advertisement (A ad ) and intentions to buy (I b ) is a controversial one. The present research examines the potential for a direct relationship between A ad and I b within an on-line advertising context, substituting A site (attitude toward the Web site) for A ad . The article replicates previous findings with respect to the four competing A ad models they tested. The article then predicts and finds a significant A site → I b path based on an opportunity and motivation perspective. In terms of opportunity, Web sites contain nonproduct information that is independent of traditional measures of brand attitude (A b ), such as security, ease of use, transactional capabilities, which are likely to affect I b . Further, on-line consumers are motivated to process this information when they visit a Web site because they cannot directly examine the products they are considering. Evidence for the opportunity and motivation perspective is provided by the decomposition of A site into its claim and nonclaim components and assessment of the A site → I b path across levels of motivation to process. Implications for on-line advertisers are discussed. © 2005 Wiley Periodicals, Inc. Psychology & Marketing, Vol. 22(4) Estimates are that between 60% and 80% of all U.S. adults regularly go on-line, with 30% to 50% making purchases (ABC News Poll, 2003;Pastore, 2001). The ascendance and obvious importance of on-line marketing and advertising has led researchers to examine some of the essential differences between on-line and off-line media (Karson & Korgaonkar, 2001;Yoh, Damhorst, Sapp, & Laczniak, 2003). Foremost among these differences are the richness of the Internet as a medium (Bruner & Kumar, 2000;Peterson, Balasubramanian, & Bronnenberg, 1997), the potential for interactive and personalized on-line experiences (cf. Evans & Wurster, 2000), and the ability of on-line visitors to control what they see (Mohammed, Fisher, Jaworski, & Cahill, 2001).These and other differences between the on-line and off-line advertising media raise questions about the appropriateness of using existing advertising models to model consumers' on-line responses. Consequently, the first objective of the present study is to undertake an on-line replication of MacKenzie, Lutz, and Belch (1986). The authors found that the dual mediation hypothesis (DMH) was the best fitting of the four attitude models they tested with print ads. The present research substitutes A site (attitude toward the site) for the traditional A ad measure to reflect the potential for nonproduct information such as site navigation, security, and other design features to provide information that affects the attitudes and intentions of on-line consumers. The second research objective is to propose and test an extended dual mediation hypothesis (EDMH). The EDMH, which includes a direct link from A site to buying intentions (I b ) in addition to the traditional DMH paths, is proposed because of the strong potential for nonproduct informat...

No abstract

Given the Similarities and potential differences between the Internet and other mass media, this research investigates the applicability of the Elaboration Likelihood Model (ELM) in this new domain. The moderating effect of involvement on both arguments and peripheral cues is tested, as is the influence of peripheral cues under higher involvement. Although the results of one experiment cannot be interpreted as conclusive, major tenets of the ELM were not supported. This study gives reason to believe that lnternet advertising may be processed differently than "traditional I' advertising. Specifically, involvement (both manipulated, and self-reported) has little moderating effect for arguments or peripheral cues on attitudes and intentions. Additionally, peripheral cues havelittle influence, regardless of involvement. On thebasis of the results theoretical and practical implications are provided, as well as important directions for future research.