Journal of Interactive Advertising, Volume 4, Number 2,
The present study attempts to examine the effects of animated banner ads, as well as the moderating effects of involvement, on each stage of the hierarchy of effects model, and to explore the applicability of the hierarchy of effects model to the banner advertising environment through an online experiment. The results provide support for the notion that animated banner ads prompt better advertising effects than do static ads. Animated banner advertising has better attention-grabbing capabilities, and generates higher recall, more favorable Aad, and higher click-through intention than static ads. Furthermore, an individual’s product involvement moderates the effects of animated banner advertising on recall, Aad, and click-through intention. However, the study does not provide solid evidence of the feasibility of the traditional hierarchical model (Cognition Affect Behavior) in the online banner advertising environment. Several implications and limitations of these results are discussed, and future research is suggested.
While online advertising has grown dramatically during the past several years (Low 2000), attracting individuals’ attention and persuading them remains one of the critical issues for the practitioner. As competition for individuals’ limited attention is of concern in the online advertising environment, animation is one innovation widely used by practitioners (Sundar et al. 1999); such ads substitute for static ads (Cleland and Carmichael 1997). The increased use of animation in online advertising is based on the belief that dynamic images have superior attention-grabbing potential over static images (Beattie and Mitchell 1985; Heo, Sundar, and Chaturvedi 2001; Reeves and Nass 1996), thus enhancing the effectiveness of persuasion (Ellsworth and Ellsworth 1995).
For advertising scholars, the applicability of the traditional advertising theories to online advertising has been of great concern since the advent of online advertising. Traditional approaches remain quite relevant to the online advertising environment, because not only do the fundamental goals of online advertising tend to be similar to the goals of traditional advertising (Pavlou and Stewart 2000), but also the theoretical models developed for traditional advertising have successfully been applied to online advertising (Cho 1999; Rodgers and Thorson 2000).
The century-old advertising approach, the hierarchy of effects model, has received widespread attention from both the practitioner and academic communities as a specific description of the way advertising works, and in turn, as a basis for measuring the effects of advertising (Barry and Howard 1990; Weilbacher 2001). Because of its simplicity and logic, the hierarchy of effects model provides information on where advertising strategies should focus, and in turn provides for good advertising planning because the model acts as a conceptual tool to predict consumer behavior (Barry 2002). However, little academic research has dealt with the effects of animated banner advertising in terms of this well-known theoretical framework. Thus, the dearth of academic work in this area calls for further research.
The present study attempts to assess the effects of animated banner ads over static ads within the framework of the hierarchy of effects. For the empirical examinations, hypotheses were postulated and tested through an online experiment. Specifically, this experiment examines the effects of animated banner ads, as well as the moderating effects of involvement, on each stage of the hierarchy of effects model, and explores the applicability of the hierarchy of effects model to the banner advertising environment.
From the well-known AIDA (attention-interest-desire-action) model, which originated with St. Elmo Lewis in the late 1800s, to the recent ‘association model’ posited by Preston and Thorson (1984), hierarchy of effects models have been around in the advertising literature for more than a century. The traditional hierarchy framework asserts that consumers respond to advertising messages in a very ordered way. The frequently cited hierarchy model posited by Lavidge and Steiner (1961) suggests that consumers move over time through a variety of stair-step stages, beginning with product ‘unawareness’ to actual purchase. These researchers’ view of the advertising hierarchy is implicitly a causal relationship from cognition to affect, and from affect to conation. The recent ‘association model’ (Preston and Thorson 1984) supports the traditional hierarchy of effects framework, and focuses on a comprehensive advertising process that takes into consideration advertising research techniques (e.g., syndicated data, surveys, experimentation) and concepts consistent with behavioral intentions models.
While there is fundamental agreement regarding the importance of the three stages of the hierarchy among advertising researchers (Barry and Howard 1990), there has been significant discrepancy regarding the order of the three stages. For example, Krugman (1965) suggested a cognition-conation-affect sequence as an alternative model in low involvement situations. On the other hand, Zajonc and Markus (1982) suggest an affect-conation-cognition sequence, in which preferences do not require a cognitive basis, but instead are mainly affectively based. Ray et al. (1973) suggested another alternative sequence (i.e., conation-affect-cognition), in which consumers’ purchasing behavior comes first, attitudes are then formed to reinforce their choice, and selective learning follows to further support purchase decisions. The several alternatives to the original Lavidge and Steiner’s model (1961) suggest that advertising researchers have developed different hierarchical models for various consumer decision making situations, but agree on the importance of the three basic tenets (i.e., cognition, affect, and conation) of the hierarchy of effects model.
Conceptually, cognition has generally been viewed as "a system of beliefs structured into some kind of semantic network" (Holbrook and Batra 1987). On the other hand, affect is typically treated as feelings and emotions which are physiologically based or have some physiological component (Barry and Howard 1990; Peterson, Hoyer, and Wilson 1986). Finally, conation has usually been referred to as either intentions to perform a behavior or the performance of the actual behavior. However, criticism concerning the hierarchy of effects indicates that the operationalizations of each stage have been a problem among many researchers (Barry and Howard 1990). In other words, there does not appear to be a universally accepted means of distinguishing between cognition and affect.
With regard to this issue, we have followed Barry and Howard’s (1990) suggestions on the operationalizations of cognition, affect, and conation. They suggested memory, such as various recall, recognition, and key comprehension scores for the operationalization of cognition; attitude toward the ad (Aad), measured by a unidimensional bipolar continuum (Holbrook and Batra 1987; Homer 1990; MacKenzie and Lutz 1989; MacKenzie, Lutz, and Belch 1986) for the operationalization of affect; and finally behavioral intention and actual product purchase for the operationalization of conation (Barry and Howard 1990; Holbrook and Batra 1987). Based on Barry and Howard’s (1990) operationalizations, in this study, we measured recall and recognition for memory (cognition), Aad (affect), and click-through intention (conation) to assess the effects of animated banner advertising.
Animation is one of the unique innovational features of banner advertising, carrying moving images and graphics to simplify or enhance the presentation of persuasive messages (Ellsworth and Ellsworth 1995). Several technological developments including plug-ins, JAVA script, Flash, and streaming media have contributed to improving the design and interactivity of online banner advertising. Motion is often considered to be a critical component of animated banner ads (Reiber 1991), because most animated banner ads are a series of static images superimposed on one another to create an illusion of motion (Kalyanaraman and Oliver 2001). Researchers studying motion effects have suggested that motion elicits responses based on the actual image, per se, as well as on the implied relationships, such as a “slow moving” or “fast-moving image” (Rieber 1991; Sundar et al. 1999). The characteristic distinguishing animated banner ads (i.e., motion) from static ads is related to the effects of animated banner ads.
In the subsequent section, how the effects of animation in banner ads are related to each stage of the traditional hierarchy of effects model will be explored, and furthermore, we will propose the hypotheses for empirical tests based on the discussion.
Virtually all hierarchy-of-effect models assume attention responses as an antecedent to actual processing. The term “attention” refers to the amount of mental effort or cognitive capacity allocated to a task (Kahneman 1973), and the concept is considered to have both direction (i.e., the focus of mental effort) and intensity (i.e., the amount of mental effort focused in a particular direction) (MacKenzie, Lutz, and Belch 1986). Traditionally, one of the common means of attracting an individual’s attention is by creating a distinctive or unusual ad execution (Shimp 2000). Since animated banner ads are regarded as more distinctive and unusual than static ads, it is reasonable to suggest that animated banner ads may have better attention-getting potential than static ads. Furthermore, Reeves and Nass (1996) noted that “when objects or people in pictures move, attention will be higher than during segments with no motion” (p. 220). This suggests that an image with animation will be perceived as representing motion, relative to the static version of the same image, thus inducing greater attention in the online advertising environment. The discussion leads to the following hypothesis:
H1: An animated banner ad will have greater attention-getting capability than a static banner ad.
Memory plays a critical role in guiding an individual’s advertising perception process. Of the massive amounts of advertising information available on the Web, an individual can be selectively exposed to only a limited amount. Of the information to which the individual is exposed, only a relatively small amount is attended to and passed on to the systematic processing part of the brain for interpretation. Studies examining both visual and verbal stimuli suggest that distinctive stimuli are more likely to be remembered (Gati and Tversky 1987). Additionally, Childers and Houston (1984) noted that more recall occurs as the access to the features that are distinctive in the stimulus increases. Accordingly, a stimulating visual image on a calm background, or an animated object on a still background would be considered distinctive, and such distinctive images are theoretically presumed to develop unique memory traces, making them easier to locate in memory (Li and Bukovac 1999). Furthermore, in their ‘Flow of Effect Model,’ Watt and Welch (1983) noted that the increased attention as a result of using dynamic visual images may affect further information processing and an individual’s memory (i.e., recall or recognition). Thus, in addition to the increased attention-getting potential, animated banner ads are likely to result in better memory performance than are static ads. Thus:
H2: An animated banner ad will result in better memory than will a static banner ad.
Attitude toward the ad (Aad)
Historically, the construct, attitude toward the ad (Aad) has been conceptualized in different ways. In the unidimensional view, Aad is purely affect and not consisting of a cognitive or behavioral component (Holbrook and Batra 1987; Lutz, MacKenzie, and Belch 1983; MacKenzie, Lutz, and Belch 1986), while the proponents of the multidimensional viewpoint propose that Aad may consist of two (Batra and Ahtola 1991: hedonic and utilitarian; Shimp 1981: cognitive and affective) or three dimensions (Fishbein and Ajzen 1975: cognitive, affective, and behavioral). In this study, we are primarily concerned with the former tradition of defining Aad as a unidimensional bipolar construct, which consists solely of an affective dimension. Specifically, we believe that this practice is not compounded by cognitive and behavioral responses, and, in turn, it has exerted a potentially restrictive effect of Aad in our theoretical framework.
The importance of Aad has been studied extensively over the last few decades in advertising. Review of various literature revealed that one of most common sets of relationships is that Aad tends to have a strong direct impact on attitude toward the brand (Ab), which in turn tends to have a strong positive effect on purchase intention (i.e., Aad Ab PI). Furthermore, Aad has been considered an efficient indicator for measuring the effects of advertising.
Babin and Burns (1997) considered imagery as a mediator of eliciting stronger attitude formation for visual stimuli, because imagery is a process by which sensory information is represented in active memory (MacInnis and Price 1987). Thus, imagery incorporates sensory processing, resulting in greater impact on attitude formation (Babin and Burns 1997). In general, animated images contain more identifiable ad elements than do static images, thus provoking stronger visual imagery processing (Rossiter and Percy 1978; 1983), which further affects individuals’ attitude formation. Distinctive advertising cues such as pictures and motion trigger more vivid imagery that, in turn, generates more favorable attitudes toward the ad and the brand (Babin and Burns 1997). Furthermore, a single exposure to a banner ad without click-through generates favorable attitudes, and inflates the likelihood of inclusion of the brand into a consideration set (Briggs and Hollis 1997). Given the expectation that animated banner ads will result in stronger and positive attitudes than static ads, we propose the following hypothesis:H3: An animated banner ad will generate more favorable Aad than will a static banner ad.
Along with the belief that banner advertising is supposed to be more accountable than its traditional counterparts, one of the most frequently used indicators of ad effectiveness is the click-through rate. In the advertising industry, click-through rate is an important factor in online advertising, with many firms billing based on clicks generated rather than on the conventional cost-per-thousand exposures (CPM) model. Click-through refers to the process of clicking through a banner advertisement to the advertiser’s destination. In the hierarchical advertising model, the click-through means the behavioral response to an advertisement. Given the expectation that animated banner ads will result in higher attention, memory, and Aad, it is reasonable to expect that they would be more likely to initiate individuals’ behavioral response in the form of clicking on the animated banner than would static ads. This expectation leads to the following hypothesis:H4: An animated banner ad will have higher click-through intention than will a static banner ad.
Moderating Role of Involvement
Consumer researchers have employed several different conceptualizations and operationalizations of involvement (Muehling, Laczniak, and Andrews 1993). For instance, Batra and Ray (1983) suggested that most prior research has used the term involvement to describe one of two phenomena: involvement with a product class (Zaichkowsky 1985), or involvement with an advertising message (Greenwald and Leavitt 1984; Petty, Cacioppo, and Schumann 1983). In either case, personal relevance seems to be an important factor in determining individuals’ level of involvement with products and/or advertising messages (Petty and Cacioppo 1986). In an ad-processing context, researchers have found that the level of involvement is positively related to individuals’ cognitive engagement in the ad (Petty, Cacioppo, and Schumann 1983). Thus, individuals with higher product involvement pay more attention to advertising stimuli and spend more time processing advertisements than those with lower product involvement (Celsi and Olson 1988). For example, Gardner, Mitchell, and Russo (1985) have found that higher involvement increases memory for an advertising message, because higher involvement increases the accessibility of message details, which leads to better recall (Hawkins and Hoch 1992). Furthermore, in attitude formation and change process, those with high involvement are believed to elaborately process ad messages and form enduring attitudes toward the ad and brand (Petty, Cacioppo, and Schumann 1983).
Recently, the concept of involvement has been employed in banner ad effectiveness studies (e.g., Briggs and Hollis 1997; Cho and Leckenby 2000) as well, and found to affect individual’s click-through, and attitudes toward the banner ad and brand. For example, Briggs and Hollis (1997) found that high-involvement products were remembered better than low-involvement products. Furthermore, Cho and Leckenby (2000) noted that individuals with high product involvement are more likely to click through banner ads than are those with low product involvement, and higher click-through rates, in turn, lead to more favorable attitudes toward the banner ad and brand.
Based on the above discussion, we suggest that animated banner advertisements can affect each stage of the hierarchy of effects model, but the mechanism driving the effects is believed to be different under conditions of low versus high involvement (Greenwald and Leavitt 1984; Petty and Cacioppo 1986). Therefore, we suggest the following hypotheses:H5: Product involvement will play moderating roles in the effects of animated banner ads on (a) attention, (b) memory, (c) Aad, and (d) click-through intention.
A 2 (Level: animation vs. static) x 2 (Involvement: high vs. low) between subjects design was used in the study. A pre-test preceded the main study in order to select a proper product category for ad stimuli (banner ads). Books were selected as an appropriate product category, based on the results of the pre-test.
Product Category Selection. Two principal considerations guided the selection of a product category to be used in the study: The product category should 1) demonstrate a strong appearance in the online advertising, and 2) be appropriate for use as a stimulus for a population of available subjects. Students were determined to be appropriate subjects in the study because college students make up a significant proportion of the Internet population (GVU’ s 10th Survey 1998), and appealing to college students is of importance to the broader societal acceptance and potential success of online advertising (Davis 1999). Furthermore, Calder, Phillps, and Tybout (1981) supported the use of college students as subjects in consumer research when the objective of the study was theoretical in nature.
Informal in-depth interviews with 55 undergraduate students (21 male and 34 female) were conducted. The students were asked to list 1) top of the mind product categories when it comes to banner advertising, 2) appropriate product categories that well fit the banner advertising format, and 3) the most frequently encountered banner advertising while surfing the Web. The results of the in-depth interviews showed six product categories — books, credit cards, DVD rentals, online gambling, music CDs, and Web cams — are highly visible in the Web environment. Among the six product categories, the two most frequently listed product categories, as well as those listed by fewer than 10 percent of the participants, were eliminated to avoid ceiling and floor effects. Two product categories from the six categories were then selected. The categories included books and DVD rentals. We selected ‘books’ as an appropriate product category, because some students indicated that they do not own DVD players, so they were not interested in any ads about DVD rentals.
Stimulus Material. One target ad (i.e., ebooks.com) and two filler ads (filler ads A and B) were developed by a professional Web designer (see Appendix A). Each advertisement was designed to have two levels (i.e., static vs. animated) with the identical creative style in terms of the layout, the number of ad elements, and the size (550 x 100 pixels). However, three advertising cues (i.e., one visual and two ad messages) in the stimuli sets were differentiated in terms of motion. For the manipulation purpose, the animated banner ad contained three moving advertising cues, which looped one time each ten seconds, while the static one did not include any moving advertising cues. In addition to the creation of banner ads, three different online newspaper-type Websites (see Appendix B) were created to use as background Web sites for the banner ads. Three sports — tennis, figure skating, and golf — were selected as the main themes of each site to avoid subjects’ different responses to different Web site themes. In order to control for possible effects from the amount of information contained in the Web sites, each Web site contained the same number of stories and visuals. The target and filler banner ads were placed at the top of each site, and the types of Web sites and filler ads were counterbalanced to guarantee a balanced distribution of background Websites to the target ad, and prevent any order effects.
Sampling and Procedure. A total of 50 subjects (29 male and 21 female) were recruited from an introductory marketing class at a major southwestern university. Each subject was randomly assigned to one of two experimental conditions (static vs. animated ad). They were exposed to one target ad and the two filler ads, posted on one of the three Web sites as the orders of types of Web site and filler ads were properly counterbalanced.
Each subject was given online instructions that provided a fictitious study objective (i.e., Web site design evaluation) and the general online experiment procedure. By clicking the ‘Next’ button at the bottom of the instruction page, participants were subsequently exposed to the three different newspaper-type Web sites (i.e., two Web sites with filler ads and one Web site with a target ad). Each site remained on the screen for 45 seconds, the average duration of a page-viewing (Nielsen/NetRatings 2002), as the sites were automatically refreshed by the function of JAVA scripts. By manipulating the 45-second exposure to each site, we were able to generate in the experiment a condition more similar to the natural Web surfing environment, and furthermore, we provide an equal opportunity for all subjects to process the advertising information. The Web sites contained no active links to limit subjects to surfing only to the experimental sites. The site containing the target banner ad was preceded by one filler site and followed by the other filler site to limit primacy and recency effects. After being exposed to all three Web sites, subjects were directed to the questionnaire site and asked to answer a series of questions measuring Web site evaluations (i.e., fillers), level of product involvement (i.e., independent variable) as well as dependent variables (i.e., level of attention, recall, recognition, attitude, and click-through intention), and then were thanked.
Measures. There were two independent variables: the level of animation in banner ads (static vs. animated), and the product involvement (high vs. low product involvement). The level of animation was manipulated as described above. Personal product involvement (See Zaichkowsky 1985) was measured by a three-item, seven-point semantic differential scale. The items were anchored by "important/unimportant," "appealing/unappealing," and "interested/uninterested." The scores of the three items were averaged to obtain an index score of product involvement (Cronbach alpha = .95), and, using a median split, we divided the subjects into two groups (high vs. low product involvement groups).
There were four dependent variables of primary interest: level of attention, memory (measured by recall and recognition), attitude toward the banner ad, and click-through intention. The level of attention paid to the banner ads was measured by two items, which were modified from Duncan and Nelson’s (1985) measure: a seven-point scale anchored by "paid no attention" and "paid a lot of attention," and a seven-point Likert-type scale ("The banner ad was eye-catching") anchored by "strongly disagree" and "strongly agree." The scores of the scales were averaged to derive an index score of attention (Cronbach alpha = .88). In order to measure recall, a retrospective thought-listing procedure was used. Subjects were asked to list all of the brand names from banner ads they saw during the experiment. For the recognition measure, subjects were asked to select the banner ad they were exposed to during the experiment from among three choices including one target banner ad, and two additional banner ads that were not presented during the experiment. The designs of the three banner ads for the recognition measure were very similar. Both recall and recognition were coded as dichotomous variables (1 = yes and 0 = no). Then, Aad was measured on a four-item, seven-point semantic differential scale, which was borrowed from the prior research studies with the unidimensional view on Aad (See Homer 1990; MacKenzie and Lutz 1989; MacKenzie, Lutz, and Belch 1986). The items were anchored by " pleasant/unpleasant," " good/bad," " favorable/unfavorable," and " likable/unlikable." The scores of the four items were averaged to generate an index score of Aad (Cronbach alpha = .93). Finally, subjects indicated their click-through intention, measured by one seven-point Likert-type scale (" I would like to click-through the banner advertisement") anchored by " strongly disagree" and " strongly agree."
A MANOVA test was conducted with attention, Aad , and click-through intention as dependent variables to test Hypotheses 1, 3, 4, 5a, 5c, and 5d (see Table 1 for means and standard deviations and Table 2 for MANOVA results). Where necessary, a series of t-tests followed, as specified by the hypotheses. Two logistic regressions were also conducted to test Hypotheses 2 and 5b, since both recall and recognition were coded as dichotomous variables (1 = yes and 0 = no). Results regarding each of the hypotheses are presented in the subsequent section.
Table 1. Descriptive Statistics
As shown in Table 2, multivariate statistics (Wilks’ Lambda) for the animation level, product involvement, and the interaction (animation level x involvement) were significant at α=.05 level. Therefore, there were statistically significant effects of the animation level, product involvement, and the interaction between the animation level and product involvement on the three different dependent variables. Detailed relationships will be examined separately as they relate to the hypothesis in the following section.
Table 2. Effect of Animation and Product Involvement on Attention, Aad,
and Click-Through Intention (MANOVA)
** p < .05
*** p < .01
Hypothesis 1: Effect of Animation on Attention. Hypothesis 1 stated that subjects exposed to animated banner ads would pay more attention to the ad than those exposed to static ads. Consistent with the hypothesis, the analysis revealed a significant main effect of animation on individuals’ attention ( animation = 3.83, S.D. = .96, vs. static = 3.06, S.D. = 1.16, F (1, 46) = 6.60, p < .05). Therefore, the subjects exposed to animated banner ads paid more attention to the ad than those exposed to static ads.
Table 3. Effect of Animation and Product Involvement
on Recall (Logistic Regression)
** p < .05
*** p < .01
Table 4. Effect of Animation and Product Involvement on
Recognition (Logistic Regression)
* p < .10
** p < .05
*** p < .01Hypothesis 2: Effect of Animation on Memory. Hypothesis 2 expected that subjects exposed to animated banner ads would have better recall and recognition of the target ad (i.e., ebooks.com) than those exposed to static ads. The model assessing the probability of recall was statistically significant (χ2df=3 =16.06, p < .01), but not for recognition (χ2df=3 =5.26, p = .15). The results of logistic regressions showed a significant effect of animation on ad recall (b = -7.81, Wald χ2 = 9.15, p < .01). Therefore, the results indicated that the subjects exposed to animated banner ads had better ad recall than those exposed to static ads, while there was no significant effect of animation on ad recognition (b = -1.54, Wald χ2 = .63, p = .43) over static ads. Therefore, Hypothesis 2 was partially supported.
Hypothesis 3: Effect of Animation on Aad. Hypothesis 3 predicted a positive effect of animation on Aad. Consistent with the hypothesis, the results showed a significant main effect of animated banner ads on Aad over static ads (animation = 4.24, S.D = 1.07, vs. static = 3.42, S.D = 1.19, F (1, 46) = 6.87, p < .05). Therefore, subjects exposed to animated banner ads had more favorable Aad than those exposed to static ads, supporting Hypothesis 3.
Hypothesis 4: Effect of Animation on Click-Through Intention. Hypothesis 4 expected that those exposed to animated banner ads would have higher click-through intention than those exposed to static ads. As shown in Table 2, consistent with the hypothesis, the results showed the significant main effect of animation on click-through intention (animation = 4.07, S.D = 1.45, vs. static = 3.26, S.D = .96, F (1, 46) = 5.46, p < .05). Thus, subjects exposed to animated banner ads had higher click-through intention than those exposed to static ads.The results supported Hypothesis 4.
Hypothesis 5a: Moderating Effect of Involvement on Attention. Hypothesis 5a predicted that the level of product involvement moderates the effects of animation on attention. The results showed an insignificant Animation Level x Involvement interaction effect, F (1, 46) = 1.60, p = .21. However, we found a significant main effect of product involvement (high-involvement = 3.78, S.D = 1.05, vs. low-involvement = 3.11, S.D = 1.12, F (1, 46) = 4.88, p < .05), which indicated that the level of product involvement affects an individual’s attention independently. Therefore, Hypothesis 5a was rejected at the p = .05 probability level.
Hypothesis 5b: Moderating Effect of Involvement on Memory. Hypothesis 5b predicted that the level of product involvement moderates the effects of animation on memory. The analyses of logistic regressions revealed a significant interaction between animation level and product involvement on ad recall (b = -4.35, Wald χ2 = 8.19, p < .05), but not for an interaction effect on recognition (b = -1.64, Wald χ2 = 1.41, p = .24). The results showed a significant moderating effect of involvement on ad recall, indicating that the impact of animation on ad recall was greater under high (χ2 = 4.89, p < .05) rather than low involvement (χ2 = 3.95, p < .10). However, we were not able to find any moderating effect of involvement on ad recognition. Therefore, Hypothesis 5b was partially supported.
Hypothesis 5c: Moderating Effect of Involvement on Aad. Hypothesis 5c expected that the level of product involvement moderates the effects of animation on Aad. Consistent with the hypothesis, the results revealed a significant interaction effect between animation level and product involvement (F (1, 46) = 4.57, p < .05), indicating that animation had a significant impact on Aad only under high involvement (t (24) = 2.97, p < .01), but not under low involvement (t (24) = .43, p =.68). Therefore, Hypothesis 5c is strongly supported. However, there was no main effect of product involvement on Aad (high-involvement = 4.07, S.D. = 1.24, vs. low-involvement = 3.59, S.D. = 1.12, F (1, 46) = 2.14, p = .15).
Hypothesis 5d: Moderating Effect of Involvement on Click-Through Intention. Hypothesis 5d predicted that the level of product involvement moderates the effects of animation on click-through intention. The results showed a marginally significant interaction effect between animation level and product involvement (F = 3.09, p < .10), indicating that animation had a marginally significant impact on click-through intention under high involvement (t (24) = 1.76, p <.10), but not under low involvement (t (24) = .56, p = .58). There was no main effect of product involvement on click-through intention (high-involvement = 3.87, S.D. = 1.43, vs. low-involvement = 3.46, S.D. = 1.11, F = 1.24, p = .27). Therefore, marginal support for Hypothesis 5d was found.
In order to assess the applicability of the hierarchy of effects model to the banner advertising environment, one of the basic premises of the hierarchical model — causal influences — was tested through a series of regression analyses.
Table 5. Causal Relationships between Dependent Variables* p < .10
** p < .05
a. Logistic regressionAs Table 5 shows, the series of regression analyses revealed one significant causal relationship between attention and click-through intention, and three marginally significant causal relationships (attention click-through intention, and click-through intention Aad). The results do not provide strong evidence of the applicability of the traditional hierarchical model (CAB) to the banner advertising environment. However, interestingly, a marginally significant causal relationship between Aad and click-through intention illustrated the importance of affective responses rather than cognitive responses (recall and recognition) in predicting click-through intention (conation).
Discussion and Conclusions
Given the importance of online advertising, many advertisers try to attract consumers’ attention and to persuade them through various advertising executions. Animated banner advertising is one alternative to conventional static banner ads. This study empirically examined the effects of animated banner advertising within the framework of the hierarchy of effects model, and explored the applicability of the hierarchical advertising model to the banner advertising environment through an online experiment.
The results of the analyses offer support for the notion that animated banner ads prompt better advertising effects than do static ads. In other words, animated banner advertising has better attention-grabbing capabilities, and generates higher recall, more favorable Aad, and higher click-through intention than do static ads. Furthermore, an individual’s product involvement moderates the effects of animated banner advertising on recall, Aad, and click-through intention. However, the study does not provide solid evidence of the feasibility of applying the traditional hierarchical model in the banner advertising environment.
The findings in this study have several implications, as well as limitations. The main implication found here is that animated banner advertising is a better alternative to traditional static ads in terms of each stage of the hierarchy of effects model. However, we assume that this relationship between advertising effectiveness and animation (or motion) is subject to the phenomenon of the inverted U-shaped curve. At some point, too much animation or motion may reduce the advertising effectiveness due to the individual’s limited cognitive capacities or some negative affective responses (such as irritation or annoyance), even though those banner ads are eye-catching. Thus, additional research is needed to determine the underlying process and generalizability of this argument.
The results showed that the effect of product involvement on attention was independent from animation effects. This result indicates that, as an individual’ s product involvement increases, the level of attention to banner advertising will also rise, regardless of the level of animation in banner ads. In addition, the hypothesized effects of the animation and/or involvement on recognition did not find empirical support. It is believed that a recall measure is required to have relatively higher cognitive efforts than a recognition measure (Du Plessis 1994). The present study employed relatively short time intervals between the actual ad exposure and subjects recognition task (i.e., less than 10 minutes), which may have led to easily reminding subjects of a stimulus banner advertisement and have precluded any deeper cognitive processing, and consequently, we believe this may have overwhelmed the effects of animation or involvement on recognition measures.
This study could not provide solid evidence of the applicability of the hierarchy of effects model to the banner advertising environment. The traditional hierarchy of effects model suggests that advertising is essentially a ‘long-term’ process (Barry and Howard 1990), so that a causal influence between stages must occur over the long-run. However, because this study was conducted in the experimental setting with one-time ad exposure, it is not feasible to examine the long-term effects of the banner advertising, in which causal relationships between the three basic tenets of the hierarchy of effects model are expected to be identified.
The preceding results and interpretations are limited by the nature of our stimuli and respondents. The present study used one product category (e.g., books) as an advertising stimulus. However, using different product categories (e.g., highly emotional products or highly habitual products), one may find different results. The FCB planning model (Vaghn 1980; 1986) classified ‘books’ as a rationally-based, high involvement product. Therefore, the results may not hold for other product categories due to the effects of emotion and involvement engaged in banner advertising processing.
Although the choice of student subjects seemed appropriate for the study, the small convenience samples limit the generalizability of the findings to the general Internet population. Thus, future research with a larger and more diverse sample is imperative to expand the scope of the present study.
A series of regression analyses illustrated the importance of affective responses to banner ads when it comes to predicting behavioral responses. However, at this point, it would be premature to conclude that affective responses are more important than cognitive responses in the banner advertising environment, partly due to the aforementioned limitations in the study, and partly due to a marginally significant reciprocal relationship between Aad and click-through intention. Thus, future research is required to examine the underlying process of this phenomenon and the generalizability of our findings.
Babin, Laurie A. and Alvin C. Burns (1997), "Effects of Print Ad Pictures and Copy Containing Instructions to Imagine on Mental Imagery That Mediates Attitudes," Journal of Advertising, 26 (3), 33-44.
Barry, Thomas E. (2002), "In Defense of the Hierarchy of Effects: A Rejoinder to Weilbacher," Journal of Advertising Research, 42 (3), 44- 47.
—- and Daniel J. Howard (1990), "A Review and Critique of the Hierarchy of Effects in Advertising," International Journal of Advertising, 9 (2), 121-135.
Batra, Rajeev and Olli T. Ahtola (1991), "The Measurement and Role of Utilitarian and Hedonic Attitudes," Marketing Letters, 2 (2), 159-170.
Batra, Rajeev and Michael L. Ray (1983), "Advertising Situations: The Implications of Differential Involvement and Accompanying Affect Responses," in Information Processing Research in Advertising, Richard Jackson Harris, ed.,
: New Jersey Erlbaum Associates. Lawrence
Beattie, Ann E. and Andrew A. Mitchell (1985), "The Relationship Between Advertising Recall and Persuasion: An Experimental Investigation," in Psychological Processes and Advertising Effects: Theory, Research, and Application, Linda F. Alwitt and Andrew A. Mitchell, eds.,
: Hillsdale, NJ Erlbaum, Associates, 129-155. Lawrence
Briggs, Rex and Nigel Hollis (1997), "Advertising on the Web: Is There Response Before Click-Through," Journal of Advertising Research, 37 (2), 33-46.
Calder, Bobby J.,
W. Phillips, and Alice M. Tybout (1981), "Designing Research for Application," Journal of Consumer Research, 8 (September), 197-207. Lynn
Celsi, Richard L. and Jerry C. Olson (1988), "The Role of Involvement in Attention and Comprehension Processes," Journal of Consumer Research, 15 (September), 210-224.
Childers, Terry L. and Michael J. Houston (1984), "Conditions of a Picture-Superiority Effect on Consumer Memory," Journal of Consumer Research, 11 (September), 643-654.
Cho, Chang-Hoan (1999), "How Advertising Works on the WWW: Modified Elaboration Likelihood Model," Journal of Current Issues and Research in Advertising, 21 (1), 33-50.
—- and John D. Leckenby (2000), "Banner Clicking and Attitude Changes on the WWW," Proceedings of the 2000 Conference of
of Advertising, 230. American Academy
Cleland, K. and M. Carmichael (1997), "Banners that Move Make a Big Impression," Advertising Age, January 26-28.
Davis, Judy Foster (1999), "Effectiveness of Internet Advertising by Leading National Advertisers," in Advertising and the World Wide Web, David W. Schumann and Esther Thorson, eds., Mahwah, NJ: Lawrence Erlbaum Associates, 81-98.
Du Plessis, Erik (1994), "Recognition versus Recall," Journal of Advertising Research, 34 (3), 75-91.
Duncan, Clavin P. and James E. Nelson (1985), "Effects of Humor in a Radio Advertising Experiment," Journal of Advertising, 14 (2), 33-64.
Ellsworth, Jill H. and Matthew V. Ellsworth (1995), Marketing on the Internet: Multimedia Strategies for the World Wide Web, New York: John Wiley & Sons, Inc.
Fishbein, Martin and Icek Ajzen (1975), Belief, Attitude, Intention and Behavior: An Introduction to Theory and Research,
Reading, MA: Addison-Wesley Publishing Company.
Gati, Itamor and Amos Tversky (1987), "Recall of Common and Distinctive Features of Verbal and Pictorial Stimuli," Memory & Cognition, 15 (March), 97-100.
Gardner, Meryl P., Andrew A. Mitchell, and J. Edward Russo (1985), "Low Involvement Strategies for Processing Advertisements," Journal of Advertising, 14 (2), 4-13.
Greenwald, Anthony G. and Clark Leavitt (1984), "Audience Involvement in Advertising: Four Levels," Journal of Consumer Research, 11 (June), 581-592.
GVU’s 10th WWW User Survey (1998), <http://www.gvu.gatech.edu/user_surveys>.
Hawkins, Scott A. and Stephen J. Hoch (1992), "Low-Involvement Learning: Memory without Evaluation, Journal of Consumer Research," 19 (September), 212-225.
Heo, Nokon, S. Shyam Sundar, and Smita Chaturvedi (2001), "Wait! Why Is It Not Moving? Attractive and Distractive Ocular Responses to Web Ads,” Paper presented at the Annual Conference of the Association for Education in Journalism and Mass Communication,
. Washington, DC
Holbrook, Morris B. and Rajeev Batra (1987), "Assessing the Role of Emotions as Mediators of Consumer Responses to Advertising," Journal of Consumer Research, 14 (December), 404-420.
Homer, Pamela A. (1990), "The Mediating Role of Attitude toward the Ad: Some Additional Evidence," Journal of Marketing Research, 27 (February), 78-86.
Kahneman, Daniel (1973), Attention and Effort, Englewood, Cliffs, N.J.: Prentice-Hall.
Kalyanaraman, Sriram and Mary Beth Oliver (2001), "Technology or Tradition: Exploring Relative Persuasive Appeals of Animation, Endorser Credibility, and Argument Strength in Web Advertising," Paper presented at the Annual Conference of the Association of Education in Journalism and Mass Communication, Washington, DC.
Krugman, Herbert E. (1965), "The Impact of Television Advertising: Learning without Involvement," Public Opinion Quarterly, 29, 349-356.
Lavidge, Robert J. and Gary A. Steiner (1961), "A Model for Predictive Measurements of Advertising Effectiveness," Journal of Marketing, 25 (4), 59-62.
Li, Hairong and Janice L. Bukovac (1999), "Cognitive Impact of Banner Ad Characteristics: An Experimental Study," Journalism and Mass Communication Quarterly, 76 (2), 341-353.
Low, George S. (2000), "Correlates of Integrated Marketing Communications," Journal of Advertising Research, 40 (January/February), 27-39.
Lutz, Richard J., Scott B. MacKenzie, and George E. Belch (1983), "Attitude Toward the Ad as a Mediator of Advertising Effectiveness: Determinants and Consequences," In Advances in Consumer Research, Vol. 10, Richard P. Bagozzi and Alice M. Tybout, eds, Ann Arbor, MI: Association for Consumer Research, 532-539.
MacInnis, Deborah and Linda L. Price (1987), "The Role of Imagery in Information Processing: Review and Extensions," Journal of Consumer Research, 13 (March), 473-491.
MacKenzie, Scott B. and Richard J. Lutz (1989), "An Empirical Examination of the Structural Antecedents of Attitude Toward the Ad in an Advertising Pretesting Context," Journal of Marketing, 53 (April), 48-65
—-, —-, and George E. Belch (1986), "The Role of Attitude Toward the Ad as a Mediator of Advertising Effectiveness: A Test of Competing Explanations," Journal of Marketing Research, 23 (May), 130-143.
Muehling, Darrel D., Russell N. Laczniak, and J. Craig Andrews (1993), "Defining, Operationalizing, and Using Involvement in Advertising Research," in Journal of Current Issues and Research in Advertising, 15 (1), 21-58.
Nielsen/NetRatings (2002), <http://www.nielsen-netratings.com/hot_off_the_net.jsp> (accessed
Pavlou, Paul A. and David W. Stewart (2000), "Measuring the Effects and Effectiveness of Interactive Advertising: A Research Agenda," Journal of Interactive Advertising, 1 (1), <http://jiad.org/vol1/no1/Pavlou/>.
Peterson, Robert A., Wayne D. Hoyer, and William R. Wilson (1986), "Reflections on the Role of Affect in Consumer Behavior," in The Role of Affect in Consumer Behavior: Emerging Theories and Applications,
Lexington, MA: Lexington Books, 141-159.
Petty, Richard E. and John T. Cacioppo (1986), Communication and Persuasion: Central and Peripheral Routes to Attitude Change, New York, NY: Springer-Verlag.
—-, —-, and David Schumann (1983), "Central and Peripheral Routes to Advertising Effectiveness: The Moderating Role of Involvement," Journal of Consumer Research, 10 (September), 135-146.
Preston, Ivan L. and Esther Thorson (1984), "The Expanded Association Model: Keeping the Hierarchy Concept Alive," Journal of Advertising Research, 24, 59-65.
Ray, Michael L., Alan G. Sawyer, Michael L. Rothschild, Roger M. Heeler, Edward C. Strong, and Jerome B. Reed (1973), "Marketing Communication and the Hierarchy of Effects," In New Models for Mass Communication Research, Peter Clarke, ed., Beverly Hills, CA: Sage Publishing, 147-176.
Reeves, Byron and Clifford Nass (1996), The Media Equation: How People Treat Computers, Television, and New Media Like Real People and Places, Stanford, CA: CSLI Publications and Camsbridge University Press.
Rieber, L. (1991), "Animation, Incidental Learning, and Continuing Motivation," Journal of Educational Psychology, 83, 318-328.
Rodgers, Shelly and Esther Thorson (2000), "The Interactive Advertising Model: How Users Perceive and Process Online Ads," Journal of Interactive Advertising, 1 (1) <http://jiad.org/vol1/no1/Rodgers/>.
Rossiter, John R. and Larry Percy (1978), "Visual Imaging Ability as a Mediator of Advertising Response," in Advances in Consumer Research, H. Keith Hunt, ed., Ann Arbor, MI: Association for Consumer Research, 5, 621-629.
—- and —-(1983), "Visual Communication in Advertising," in Information Processing Research in Advertising, Richard Jackson Harris, ed., New Jersey: Lawrence Erlbaum Associates.
Shimp, Terrence A. (1981), "Attitude toward the Ad as a Mediator of Consumer Brand Choice," Journal of Advertising, 10 (2), 9-16.
—- (2000), Advertising and Promotion: Supplemental Aspects of Integrated Marketing Communications, Fort Worth, TX: The Dryden Press.
Sundar, S. Shyam, Sunetra Narayan, Rafael Obregon and Charu Uppal (1999), "Does Web Advertising Work? Memory for Print vs. Online Media," Journalism and Mass Communication Quarterly, 75 (4), 822-835.
Vaghn, Richard (1980), "How Advertising Works: A Planning Model," Journal of Advertising Research, 20 (5), 27-33.
—– (1986), "How Advertising Works: A Planning Model Revisited," Journal of Advertising Research, 26 (1), 57-66.
Watt, James H. and Alicia J. Welch (1983), "Effects of Static and Dynamic Complexity on Children’s Attention and Recall of Televised Instruction," In Children’s Understanding of Television, J. Bryant and D. R. Anderson, eds., New York, NY: Academic Press.
Weilbacher, William M. (2001), "Point of View: Does Advertising Cause a "Hierarchy of Effects?" Journal of Advertising Research, 41(6), 19-26.
Zaichkowsky, Judith L. (1985), "Measuring the Involvement Construct," Journal of Consumer Research, 12 (December), 341-352.
Zajonc, Robert B. and Hazel Markus (1982), "Affective and Cognitive Factors in Preferences," Journal of Consumer Research, 9 (September), 123-131.
Stimulus Material: Target and Filler Ads
Stimulus Material: Web Sites
Chan Yun Yoo is currently a doctoral candidate in the Department of Advertising at the University of Texas at Austin. His research focuses on interactive advertising, consumer behavior on the Web, advertising media planning, and agenda-setting effects of new media. His research appears in the International Journal of Advertising and several academic conferences.
Kihan Kim is a doctoral candidate in the Advertising Department at the University of Texas at Austin. He got his MA from Missouri School of Journalism in 2001. His research areas include source effects in advertising, branding, sponsorship communication, new communication technology, and agenda-setting studies.
Patricia A. Stout, Ph.D., is Professor of Advertising and John P. McGovern Regents Professor in Health and Medical Science Communication in the College of Communication at The University of Texas at Austin. She is a co-director of the Center for Public Health Promotion Research in the School of Nursing at The University of Texas at Austin. She joined UT-Austin in 1984 after receiving her Ph.D. in Communications from the University of Illinois at Urbana-Champaign. Prior to pursuing a doctoral degree, she worked as an advertising executive in Montana and assisted with communication and campaign design for the Minnesota Heart Health Program while at the University of Minnesota. As a visiting research professor at the U.S. Centers for Disease Control and Prevention in Atlanta with the National AIDS Information and Education Program, she conducted communication research on the “America Responds to AIDS” campaign. Dr. Stout has conducted experimental research related to health communication and emotion-based messages directed to the college-aged population on HIV prevention and drinking and driving. Her research focuses on viewer response to persuasive messages and advertising, with particular interest on individuals’ emotional response and with messages delivered online via the Web. She has served as an advisor to the Texas Department of Health on a three-year statewide public health awareness campaign, to the Texas Task Force on Obesity, to the Centers for Disease Control and Prevention Research Priorities Group, and to the NIMH on stigma and mental illness and suicide prevention. Dr. Stout has received funding from NIMH and the Hogg Foundation for research on the role of media in mental illness stigma.
Copyright © 2004 Journal of Interactive Advertising