The effectiveness of cross
Master’s Thesis
Graduate School of Communication
Master’s programme Communication Science Supervisor: dr. H. A. M. Voorveld
June 26, 2013
Wanted: brand status
The effectiveness of cross-media versus single-medium advertising
campaigns
K
School of Communication
Master’s programme Communication Science, Persuasive Communication track Voorveld
medium advertising
Kelly P. J. de Ruiter 10389687
2
Abstract
The aim of this study was to provide an understanding of the effectiveness of cross-media versus single-medium advertising campaigns at a cognitive, affective, and conative consumer response level and to better understand the (moderating) role of brand status in this relationship as well. Based on three theoretical explanations for synergy (encoding variability theory, repetition variation theory, and multiple source perceptions), customer-based brand equity, media engagement, and on the two processing routes of the elaboration likelihood model, it was expected that cross-media advertising would result in more positive consumer responses and that brand status would moderate this relationship. Using an online experiment, 110 participants were randomly exposed to either a single-medium (repeated TV-commercial) or cross-media (TV-commercial & website) advertising campaign for one of two familiar brands that differed in brand status. The results indicated that there was no difference between a cross-media or single-medium campaign in terms of generating more positive consumer responses and additionally, brand status did not have a moderating role in this relationship either. Rather, the study showed that brands with a strong status (compared to those with a weak status) generated more (direct) positive consumer responses, particularly on an affective and conative level. The current study therefore showed that cross-media effects do not always occur and that brand status itself should be considered to have a ‘wanted’ status: it appeared to be an important element in determining the effectiveness of advertising campaigns and it is something campaigns should take into account.
3 Introduction
Cross-media campaigns are very popular: worldwide, numerous companies and brands have included several types of media within one advertising campaign (Chatterjee, 2012). The main reason why cross-media campaigns are implemented is because such campaigns can create synergy, which refers to the idea that “the combined effect of multiple [media] activities exceeds the sum of their individual effects” (Naik & Raman, 2003, p. 375). Indeed, several studies have concluded that cross-media campaigns are more persuasive and more effective in generating positive consumer responses than single-medium campaigns (Chang & Thorson, 2004; Dijkstra, 2002; Naik & Raman, 2003; Voorveld, Neijens, & Smit, 2011).
Despite these similar results, there is still a lot to uncover in the realm of cross-media campaigns, for example when “effects are [most] likely to occur” (Voorveld, Neijens, & Smit, 2012, p. 204). It is quite possible that cross-media campaigns work differently for different types of brands and therefore one factor that could be taken into account is brand status (Wang, 2009, 2011). Brand status refers to “[s]trong and weak brands [which] are similar in that they both have a high market penetration rate and are familiar to consumers, but very
different in terms of market position and favorability” (Keller, 1998, in: Dahlén & Lange, 2005,
p. 474). This difference in favorability and market position can be seen in the fact that consumers are more likely to enjoy, notice and like information about and advertisements for strong brands than for weak brands (Dahlén & Lange, 2005; Rice & Bennett, 1998) and accordingly, different consumer responses may be expected as well.
Even though brands are a fundamental part of cross-media studies, these studies typically use unfamiliar or fictitious brands (e.g. Chang & Thorson, 2004; Chatterjee, 2012; Voorveld et al., 2011, 2012) to eliminate the influence of previous brand exposures, knowledge or evaluations (Blair & Innis, 1996; Yang, 2004). Consumers however typically come across advertising campaigns for familiar and existing brands (Kent, 2002) and while some studies did use a familiar and an unfamiliar brand (Stammerjohan, Wood, Chang, & Thorson, 2005) – so far, no cross-media study seems to have focused on or compared the effects of brands with a strong or weak status.
4
Brand status is expected to moderate cross-media effects, in particular due to the difference in market position and favorability of strong and weak brands. In addition to cross-media’s opportunity to create synergy, cross-media advertising campaigns are likely to be processed differently than single-medium campaigns - respectively, centrally and peripherally (Chang & Thorson, 2004). Together with one’s degree of media engagement in an advertising campaign’s media (Wang, 2009) and one’s personal relevance or interest in either a strong or weak brand (Dahlén & Lange, 2004), the difference in processing is likely to lead to different cross-media effects for strong and weak brands. More specifically, a cross-media campaign should be most beneficial when a strong brand is advertised while for a weak brand, a single-medium advertising campaign is expected to be more beneficial instead (Wang, 2009).
This study will take a television advertisement and a website into account to test the single-medium and cross-media effects, in particular as advertising campaigns often combine these two types of media (Chang & Thorson, 2004). Furthermore, the hierarchy of effects needs to be taken into account if one wants to determine a campaign’s effectiveness (Lavidge & Steiner, 1961). This hierarchy considers the three steps that consumers go through “from [...] initial [ad] exposure to […] purchase decision” (Voorveld et al., 2011, p. 81). Consumers’ first step is at a cognitive level (brand memory), the second at an affective level (attitude towards both the brand and the advertisement), and the third and final step is at a conative level (purchase intention) (Dijkstra, Buijtels, & Van Raaij, 2005; Lavidge & Steiner, 1961). Taking this all into account leads to the following two research questions:
1a) In what way does cross-media (TV-web) vs. single medium (TV) advertising influence consumers’ cognitive, affective, and conative responses towards the advertising? 1b) To what extent is this relationship moderated by a strong vs. weak brand status?
By testing cross-media and single-medium effects for a brand with a strong or weak status, the applicability of results from previous cross-media studies can be examined. Moreover, this study’s results could also have implications in terms of whether brand status is a factor
5 that advertisers need to take into account when deciding between using a cross-media or a single-medium advertising campaign.
Theoretical framework
The synergy effect of cross-media campaigns: three theoretical explanations
Synergy is likely to be the reason for the enhanced effectiveness of cross-media over single-medium campaigns (Chang & Thorson, 2004; Dijkstra, 2002; Dijkstra et al., 2005; Naik & Raman, 2003) and there are three underlying theoretical mechanisms that might explain why cross-media advertising could result in synergy: the encoding variability theory, the repetition variation theory, and multiple source perceptions.
Encoding variability theory
A first theory that might explain synergy is the encoding variability theory. This theory argues that when consumers are exposed to the same message in different types of media, this message is encoded into people’s memory in different ways (Chatterjee, 2012; Stammerjohan et al., 2005; Unnava & Burnkrant, 1991). For instance, information taken from the two media of print and television will form two different retrieval paths due to differences in pacing and opportunities for detailed information (Dijkstra et al., 2005), compared to one retrieval path if for example only the print medium was used (Unnava & Burnkrant, 1991). Using two media will then result “in a stronger, clearer, [and] more accessible information network in the brain” (Stammerjohan et al., 2005, p. 56) which would make (accurate) memory retrieval both more likely and easier (Chatterjee, 2012; Stammerjohan et al., 2005; Unnava & Burnkrant, 1991). Hence, cross-media advertising should result in more (accurate) memory than single-medium advertising (Unnava & Burnkrant, 1991). Unnava & Burnkrant’s (1991, p. 413) study supports this notion, in particular as this study showed that “varied ad executions enhance memory for brand name over repeated same-ad executions” which makes the comparison of varied ad executions to cross-media and repeated same-ad executions to single-medium likely. Furthermore, as “attitudes are thought to be summaries
6
of memory-based judgments” (Tavasolli & Lee, 2003 in Stammerjohan et al., 2005, p. 57), both brand memory and affective elements (attitude towards the brand and advertising material itself) should be enhanced when exposed to a cross-media advertising campaign compared to a single-medium campaign (Stammerjohan et al., 2005).
The encoding variability theory is related to the principles of forward encoding and priming. In forward encoding, an advertisement in a first medium (e.g. TV) can prime consumers’ “attention to, […] interest in, and […] curiosity for” (Voorveld et al., 2011, p. 70) subsequent advertisements in a second, different medium (e.g. website) (Dijkstra, 2002; Edell & Keller, 1989; Voorveld et al., 2011). Memory traces created with exposure to the first ad are more easily accessible and thus more likely to be used when exposed to the second ad - making encoding of the second ad easier as well (Dijkstra, 2002). Dijkstra (2002) and Voorveld et al. (2011) indeed showed that people were more interested in a second ad after they have been previously exposed to a first advertisement in a different medium. All in all, based on the encoding variability theory, forward encoding and priming, there should be more positive effects on both a cognitive and affective level for a cross-media than for a single-medium campaign.
Repetition variation theory
A second theory that can be used to explain synergy is repetition variation theory. According to this theory, people will have more positive responses towards advertising when they are exposed to a message in different media instead of being repetitively exposed to the same advertising message in the same medium (Chatterjee, 2012; Stammerjohan et al., 2005). Messages in media like television or on the Internet are rather complex and so they need to be repeated to ‘wear-in’ among consumers (Dijkstra, 2002). Repeating an advertising message will therefore enhance its effectiveness (Stammerjohan et al., 2005), particularly on a cognitive level (Dijkstra, 2002). However after continued repetitions of a message, ‘wear-out’ and tedium are likely to occur, causing people to resist or get bored with the message - decreasing its effectiveness (Campbell & Keller, 2003). Furthermore, when people realize
7 that they have already seen a certain advertising message in a specific medium before, they will have less interest in the same ad and thus pay less attention to it as well (Voorveld et al., 2011). However, when the repetition is varied (e.g. in a cross-media campaign where the same message is incorporated in several different media outlets), less ‘wear-out’ is expected (Chatterjee, 2012; Dijkstra, 2002). Based on repetition variation theory, cross-media campaigns should have a more positive effect than single-medium campaigns on both cognitive (Dijkstra, 2002) and affective (Stammerjohan et al., 2005) consumer responses.
Multiple source perceptions
A third and last theory that can be used to explain synergy is multiple source perceptions. Consumers consider different media to be independent and different sources of information (Bronner, Neijens, & Van Raaij, 2003; Dijkstra, 2002) and accordingly, these different media are perceived to be more persuasive, credible and convincing (Bronner et al., 2003; Chang & Thorson, 2004; Dijkstra, 2002; Voorveld et al., 2011). This enhanced credibility in turn results in an increased interest in advertisements and subsequently more positive attitudes towards the advertised brand and towards the advertising material (Chatterjee, 2012; Stammerjohan et al., 2005). Since attitudes are a good indicator for intentions (Wang, 2007), it is very likely that the consumers’ perceived enhanced credibility of using different types of media will lead to higher purchase intentions (Chang & Thorson, 2004). Therefore, based on multiple source perceptions and the notion of credibility, there should be more positive effects on both an affective and conative level for cross-media versus single-medium campaigns.
Empirical support for cross-media vs. single-medium campaign effects
The predictions from these three underlying theoretical mechanisms are empirically supported. Studies have shown that cross-media campaigns resulted in more positive effects on a cognitive level as such campaigns increased brand memory (Chatterjee, 2012; Edell & Keller, 1989; Tang, Newton, & Wang, 2007; Unnava & Burnkrant, 1991), on an affective level as attitudes were more positive towards the advertised brand (Tang et al., 2007; Wang,
8
2007, 2011; Wang & Nelson, 2006) and the advertising itself (Tang et al., 2007; Wang, 2011; Wang & Nelson, 2006), and it also resulted in more positive effects on a conative level as purchase intentions increased (Tang et al., 2007; Wang & Nelson, 2006). Based on the aforementioned theoretical explanations and the results from previous studies, it is therefore hypothesized that:
H1: A cross-media campaign will generate more positive effects than a single-medium campaign on a) a cognitive level (brand memory), b) an affective level (attitude towards brand and advertising material) and c) a conative level (purchase intention).
Brand status and its direct effects on consumer responses
Consumer responses are likely to be different for different types of brands (Broniarczyk & Gershoff, 2003) which can be explained by both Keller’s (1993) consumer-based brand equity model (CBBE) and by the elaboration likelihood model (ELM) (Petty, Cacioppo, & Schumann, 1983). Keller’s (1993) CBBE model should come to mind when thinking of brands and it can be “defined as the differential effect of brand knowledge on consumer response to the marketing of the brand” (Keller, 1993, p. 8). Consumers’ brand knowledge has to do with their feelings towards or experiences with the brand in question (Keller, 2003) and so the higher one’s brand equity, the more favorable his or her feelings and subsequent responses towards the brand and its advertising will be (Esch, Langner, Schmitt, & Geus, 2006; Hoeffler & Keller, 2003; Keller, 2003). In addition, the notion of brand equity is also related to one’s associative network of memory (Hoeffler & Keller, 2003; Keller, 2003) which explicates that “a brand will have a node in memory that may be associated with a variety of other nodes” (Hoeffler & Keller, 2003, p. 422). The stronger these associations, the higher one’s brand equity, and the more likely consumers are to recognize and recall brand related information and elements (Keller, 2003). As strong brands have a wider associative network in consumers’ memory than weak brands (Wang & Muehling, 2012), memory for strong brands is likely to be more enhanced (Hoeffler & Keller, 2003).
9 Furthermore, consumers also tend to be more “motivated to process information about strong brands” (Dahlén & Rosengren, 2005, p. 156) than information about weak brands (Dahlén & Rosengren, 2005; Kent & Allen, 1994), causing “strong brands [to] draw attention to the ad, [and] weak brands [to] draw attention from the ad” (Dahlén & Lange, 2005, p. 476). This difference in motivation to process information could be explained by ELM’s two processing routes (Petty et al., 1983). In the central route, an issue or message is considered to be personally relevant and as a result, one is willing to invest cognitive effort to scrutinize all presented elements while in the peripheral route, an issue is not personally relevant and so consumers will focus on heuristic cues such as the source’s credibility instead (Chang & Thorson, 2004; Levy & Nebenzahl, 2008; Petty et al., 1983). Studies have indeed shown that consumers tend to focus primarily on included brand information in advertisements for strong brands (Leigh, 1992) while heuristic cues were more important in advertisements for weak brands (Dahlén & Lange, 2004; Leigh, 1992) – supporting the idea that strong brands are processed via the central route and weak brands via the peripheral route.
This difference in information processing for strong and weak brands also has an influence on consumers’ responses. More specifically, strong brands tend to generate more positive consumer responses than weak brands (Wang & Muehling, 2012). This has to do with the fact that when consumers are exposed to advertising for a strong brand, their attitudes are more persistent and stronger (Petty et al., 1983; Wang, 2009) and also more predictive of purchase intentions than when they are exposed to advertising for a weak brand (Laroche, Kim, & Zhou, 1996, Mackenzie & Spreng, 1992; Petty et al., 1983). Based on both the CBBE and ELM models, the following can be hypothesized:
H2: Strong brands (vs. weak brands) will generate more positive direct consumer responses on a) a cognitive level (brand memory), b) an affective level (attitude towards brand and advertising material) and c) a conative level (purchase intention).
10
The moderating role of brand status in the relationship between advertising campaigns and consumer responses
The three theoretical explanations for synergy (encoding variability theory, repetition variation theory, and multiple source perceptions) have been used to explain the effect of a different kind of advertising campaign on consumer responses. However before explaining the moderating role of brand status, an even deeper understanding of cross-media and single-medium effects will be necessary.
In addition to the aforementioned notion of synergy and its effects on consumer responses, processing differences between cross-media and single-medium advertising campaigns also tend to result in different responses (Chang & Thorson, 2004; Wang, 2009). Chang and Thorson (2004) more specifically describe how cross-media and single-medium advertising are processed. They state that cross-media advertising “could result in greater perceived message diversity and higher message credibility” which in turn “acts as a motivator to enhance processing” (Chang & Thorson, 2004, p. 77). This message credibility can then lead to purchase intentions in two ways: firstly, it can affect the number of positive thoughts and subsequently, brand attitudes and purchase intensions. Secondly, the message credibility can also transfer to ads and brands, generating more positive thoughts and attitudes towards the brand and eventually also higher purchase intentions (Chang & Thorson, 2004). In single-medium advertising, people tend to focus particularly on heuristic cues (such as source credibility) to form their attitudes (Chang & Thorson, 2004). This source credibility then affects both ad and brand credibility which will eventually lead to more positive thoughts, more positive brand attitudes and higher purchase intentions (Chang & Thorson, 2004). Hence, it seems likely that consumers are more motivated to pay attention to a media campaign than to pay attention when a same message is repeated: cross-media advertising is therefore processed centrally whereas single-medium advertising is processed peripherally instead (Chang & Thorson, 2004).
Taking the moderating role of brand status into account, advertising for strong brands is prone to benefit from a cross-media over a single-medium campaign especially as the
11 characteristics of the media involved (television and websites) are likely to contribute to these enhanced effects (Wang, 2009). As consumers think of strong brands as something personally relevant or interesting, this will result in a willingness to process information about such brands more in-depth and thus centrally (Dahlén & Lange, 2004; Leigh, 1992). A TV commercial is externally paced (Bronner & Neijens, 2006; Dijkstra et al., 2005) and it tends to appeal to people’s sensory modes which not only makes it a medium that consumers are generally very interested in but also one that is good in grabbing people’s attention and not good for documenting information (Chang & Thorson, 2004; Dijkstra et al., 2005). A website is however an interactive medium (Liu & Shrum, 2009) that is internally paced and therefore a good medium to inform people (Bronner & Neijens, 2006; Dijkstra et al., 2005). A website thus requires cognitive effort while a TV commercial does not (Dijkstra et al., 2005; Voorveld et al., 2012) and therefore only when one considers a brand to be personally relevant or interesting, will one be interested in and willing to actively interact and engage with a website (Dijkstra et al., 2005; Liu & Shrum, 2009).
This interest in both a website and a television commercial in turn will result in “stronger media engagement” and subsequently, a cross-media campaign will have additional benefits over a single-medium campaign when a strong brand is advertised (Wang, 2009, p. 870). Wang (2009) indeed showed that when an item was personally relevant, consumers experienced higher media engagement (and subsequently, more positive consumer responses) in a cross-media than in a single-medium campaign. Based on this reason, it is hypothesized that:
H3a: The effect of a cross-media campaign on brand memory, attitude towards the brand and advertising material, and purchase intention is moderated by brand status, such that when a brand is perceived to have a strong brand status, a cross-media campaign has more positive effects than a single-medium campaign on people’s (i) cognitive, (ii) affective, and (iii) conative responses.
12
In the case of advertising for weak brands, a cross-media campaign is not likely to be more beneficial than a single-medium campaign (Wang, 2009), in particular as consumers will typically not consider a weak brand to be personally relevant or interesting (Dahlén & Lange, 2004; Leigh, 1992). As a result of this (general) disinterest in weak brands, consumers “may disregard advertising messages [for such brands that are] deemed unnecessary or redundant” (Wang, 2009, p. 860). Hence, consumers are not likely to be motivated to process messages about weak brands since such “messages do not matter to them” and as such, there would be no media engagement and also no need for a cross-media campaign (Wang, 2009, p. 869). Instead, a single-medium campaign would be more effective when advertising weak brands (Wang, 2009). Television is a good medium for products or brands that do not require (a lot of) cognitive effort (Chang & Thorson, 2004; Dijkstra et al., 2005) or when items are processed peripherally, such as when a brand is not perceived to be personally relevant or interesting (Buchholz & Smith, 1991; Dijkstra et al., 2005). Indeed, when an item was not considered to be personally relevant, consumers in single-medium advertising experienced higher media engagement than cross-media advertising (Wang, 2009). It is therefore hypothesized that:
H3b: The effect of a cross-media campaign on brand memory, attitude towards the brand and advertising material, and purchase intention is moderated by brand status, such that when a brand is perceived to have a weak brand status, a single-medium campaign has more positive effects than a cross-media campaign on people’s (i) cognitive, (ii) affective, and (iii) conative responses.
13
Figure 1 Conceptual model
Method
Research design and participants
An online experiment was conducted to test the proposed hypotheses. The study had a two (cross-media [TV-web] vs. single-medium [TV-TV]) by two (strong vs. weak brand status) between-subjects design and participants were randomly assigned to one of these four conditions. Current students at Dutch higher educational institutions (HBO and universities) were approached both via e-mail and social networking sites such as LinkedIn and Facebook and they were asked to participate in a research that studied how people handle advertising campaigns for a familiar brand. Participation was completely voluntary. In total, 110 participants completed the online experiment with a mean age of 22.21 (SD = 2.37), 72.7% studied at a university-level and 69.1% was female.
Pre-test and experimental stimuli
Comparable to Dahlén and Lange (2005), one strong and one weak brand were identified with the Best Global Brands 2012 list (Interbrand, 2013) – respectively, Coca-Cola (nr. 1) and
Pepsi (nr. 22). The product category of soft drinks was chosen based on the fact that
students regularly drink soft drinks (Harvard School of Public Health, 2012), making it very likely that they have had experience with Coca-Cola and Pepsi which is why Gibson (2008) also used these two brands in his study.
H2 H3
H4
H1 Consumer responses
Cognitive (brand memory) Affective (attitude towards brand
and advertising material) Conative (purchase intention) Cross-media campaign
vs.
Single-medium campaign
Strong brand status vs.
14
A pre-test was conducted to see whether the two brands of Coca-Cola and Pepsi were perceived to have a different (strong or weak) brand status among a subsample of the target participants (N = 12, M = 24.58, SD = 3.60, 50% female). Brand status was measured using eight items (prestigious, status, distinctive, sophisticated, high esteem, success, wealth, exclusive) on seven-point Likert scales (1: strongly disagree to 7: strongly agree) (O’Cass & Choy, 2008). Participants were asked to indicate to what extent they agreed with each statement, an example of this is ‘I think Pepsi is distinctive’. Together, these eight items proved to be reliable (α = .84; M = 4.36, SD = 1.04). The results showed that Coca-Cola (MCoca-Cola = 4.74, SD = 0.97) had a marginally significant higher brand status than Pepsi (MPepsi = 3.98, SD = 1.01; t(22) = 1.88, p = .073, 95% CI [-0.08,1.60]). None of the
participants in the pre-test had seen the TV commercials of Coca-Cola and Pepsi before. In order to remove any pre-experimental bias in terms of familiarity with a specific advertisement (Blair & Innis, 1996), ads for both brands were taken from the non-Dutch market (Australia for Pepsi and the U.S. for Coca-Cola). The included advertisements were chosen based on the comparability to one another in terms of content and duration: both ads lasted for a little over 1 minute and in both ads, animals chase after (and end up with) a bottle of cola. Screenshots of these two commercials can be found in Appendix A.
A filler clip was included after exposure to the TV commercial to ensure a certain degree of distraction or engagement so that participants did not continue to process the commercial (Cary & Reder, 2003; Purchase, Cohen, & James, 1996). The 2 minute clip showed animals acting funny and crazy. Finally, the last type of stimulus material was a website of Coca-Cola or Pepsi, either in English (UK version) or in Dutch. The websites of
Pepsi and Coca-Cola included the same information and sections in either the Dutch or
English version, only the layout differed. The Pepsi websites both included sections such as products, news, about, contact, and Beyoncé. Similarly, the websites by Coca-Cola both also included sections such as health, environment, FAQ, about us, coming together, and brands (see Appendix A for screenshots of the websites).
15 The cross-media condition included exposure to a TV commercial and then to a brand’s website while in the single-medium condition, participants were exposed to the same TV commercial twice. Exposure to first a TV commercial and then a website proved to be “effective for both high- and low-/moderate [product involvement]” (Voorveld et al., 2012, p. 212) and involvement is related to one’s personal relevance or interest (Petty et al., 1983). Since in this research, the level of personal relevance or interest in a brand was likely to vary per participant, the order of first a TV commercial and then a website (or the same TV commercial) would be effective irrespective of one’s (brand) relevance or interest, which is why this order was chosen in this study.
Procedure
Before the start of the online experiment, participants were asked to choose the language (Dutch or English) in which they wanted to receive the questions and material (particularly the website as the TV commercials were not language-specific). They were also asked to read and sign a consent form. This form explicated that their participation in the study was on a voluntary basis; that they could stop at any time for no particular reason; that answers would be treated completely anonymously; and that they understood what the research was about. After giving their consent, participants were randomly assigned to one for the four experimental conditions: Coca-Cola (CC) single-medium (28), CC cross-media (28), Pepsi (P) single-medium (28), and P cross-media (26).
In all conditions, participants were first exposed to a television commercial and then were shown the filler clip and they were asked to watch each clip only once. After watching the filler clip, participants in the single-medium condition were exposed to the same television commercial while participants in the cross-media condition were exposed to the brand’s website (comparable to e.g. Chang & Thorson, 2004; Voorveld et al., 2011, Wang, 2009). Participants were asked to browse the website as they would normally do, “for as long […] as they liked” (Dahlén & Lange, 2005, p. 480). After this second exposure to either new (cross-media) or repeating (single-medium) material, participants were asked to fill out a
16
questionnaire with questions pertaining to the three levels of consumer responses and some control variables. On average it took participants 12 minutes to complete the questionnaire and the questionnaire can be found in appendix B.
Measures
Brand memory
Two types of brand memory were measured: free recall and corrected aided recall. Free
recall (Moorman, Neijens, & Smit, 2007) was measured by asking participants “What [brand
was or what] brands were advertised in the [material that you have seen] earlier?” Please “write down all the [brand(s) you can] recall” (Dahlen & Lange, 2005, p. 480) while giving no cues. The answering option was open-ended and answers were recoded into 1 when the correct brand mentioned and 0 when incorrect or when multiple brands mentioned (97.3% correct).
For corrected aided recall, a list of existing brands was provided to the participants and included the advertised brand plus four other brands from the same product category of soft drinks that did not appear in the advertisements (Dr. Pepper. Sprite, Fanta) (Moorman et al., 2007). For each brand, participants were asked to indicate if they recalled seeing advertising for that brand (yes/no). Yes-answers for the advertised brand were recoded as 1 (correct) and all other answers were recoded as 0 (incorrect) (85.5% correct). If participants had ticked another brand or brands in addition to the correct advertised brand, the answer was recoded as incorrect.
Attitude towards the TV commercial and towards the website
Attitude towards the TV commercial and attitude towards the website were both measured using Chang and Thorson’s (2004) four item seven-point semantic differential scale (not interesting/interesting, bad/good, not appealing/appealing, and not likable/likable). Participants were asked to rate the TV commercial in the single-medium condition (or TV commercial and website in the cross-media condition) they had just seen by indicating for
17 each item what best reflected their view. Each type of attitude was reliable (Attitude towards the TV commercial: α = .91; M = 4.59, SD = 1.26; Attitude towards the website: α = .99; M = 2.28, SD = 2.46) and for the analyses, they were combined into the variable of 'Attitude towards the advertising material' (α = .84; M = 4.62, SD = 1.13).
Attitude towards the brand
Attitude towards the brand was measured with the same four items as the attitude towards the TV commercial and towards the website. Participants were asked to rate either Pepsi or
Coca-Cola overall as a brand and together, the items proved to be reliable (α = .93; M =
4.71, SD = 1.39).
Purchase intention
Purchase intention was measured with a two-item seven-point Likert scale (1: very unlikely to 7: very likely) and asked how likely it was for participants, if they were to buy soft drinks on their next shopping trip, that they would buy the advertised brand and how likely it was that they would recommend this brand to a friend (Voorveld et al., 2011). There was a significant strong correlation between the two items (r = 0.77, p = .000, M = 3.25, SD = 1.92).
Brand status
To check whether brand status was successfully manipulated, brand status was measured using the same measure as in the pre-test. A principal components analysis (PCA) showed that the 8 items loaded on two factors. Both factors had an eigenvalue above 1 (respectively
EV = 4.16; R² = .52 and EV = 1.15; R² = .14), however the scree plot showed a clear point of
inflexion after the first component. Therefore, only the first component was extracted which was reliable and included the items ‘successful’, ‘wealthy’, ‘distinctive’, ‘prestigious’, ‘high esteem’ and ‘sophisticated’ (α = .85; M = 4.69, SD = 1.17).
18
Background and other measures
In addition to the aforementioned dependent variables, some background measures such as participants’ age, gender, study, and educational level (HBO/WO) were also asked. Furthermore, three other variables (product category experience, product involvement, ad familiarity) were measured as well.
Product category experience was measured by asking participants to respond to
three statements such as ‘I have a lot of experience with cola’ on seven-point Likert scales (1: strongly disagree to 7: strongly agree) (Voorveld et al., 2011) and together, the items proved to be reliable (α = .90; M = 4.22, SD = 1.57).
Product involvement was measured with a two-item seven-point semantic differential
scale: very unimportant/very important and very uninteresting/very interesting (Voorveld et al., 2012). Participants were asked to indicate for each item what best reflected their view. There was a significant, strong correlation between the two items (r = 0.66, p = .000, M = 3.42, SD = 1.32).
Finally, ad familiarity was measured by asking participants “whether they had seen the TV commercial […] before” and the answering options were 0 (no) and 1 (yes) (Voorveld et al., 2011, p. 76).
Results
Manipulation check
An independent samples t-test showed that, as expected, Coca-Cola scored significantly higher on brand status than Pepsi (MCoca-Cola = 5.38, SD = 0.89; MPepsi = 3.97, SD = 0.97,
t(108) = -8.00, p = .000, 95% CI [-1.77,-1.07]): the manipulation of brand status was thus successful.
To remove any pre-experimental bias, ads were selected that should have been unfamiliar to participants. Almost all participants (90%) were indeed unfamiliar with the TV advertisement they were exposed to. Ad familiarity did not correlate with any of the
19 dependent variables nor did the control variables of study, gender, and educational level. There was however a significant, weak negative correlation between age and ‘Free recall’ (r
= -.20, p = .038), significant weak correlations between product involvement and both
‘Attitude towards the brand’ (r = .20, p = .035) and ‘Purchase intention’ (r = .33, p = .000); and also significant weak correlations between product category experience and both ‘Attitude towards the brand’ (r = .25, p = .008) and ‘Purchase intention’ (r = .28, p = .003).
A one-way ANOVA was conducted to assess whether these three control variables were equally distributed over the four experimental conditions. There were no statistically significant differences between the four experimental conditions in product involvement scores (Levene’s F(3, 106) = 0.17, p = .919), age (Welch’s F(3, 57.216) = 0.54, p = .660) or product category experience scores (Welch’s F(3, 58.490) = 1.11, p = .354). Any differences in results could thus only be due to exposure to the experimental material and therefore, no control variables were included when testing the hypotheses.
Cognitive consumer responses
To test how a cross-media vs. single-medium campaign influenced consumers’ cognitive responses and how brand status moderated this relationship, a logistic regression was performed for each type of brand memory (free recall and corrected aided recall). The two logistic regressions both included type of campaign, type of brand status, type of brand memory, as well as the interaction between type of campaign and type of brand status.
Free recall: The logistic regression for ‘Free recall’ was not statistically significant,
χ2(3) = 4.49, p = .213. Furthermore, as can be seen in Table 1, neither type of campaign (p = .998), type of brand status (p = .998), or the interaction between these two variables (p = .997) were significant predictors of correct ‘free recall’. Hypothesis 1a, 2a, 3a-i and 3b-i were all not supported for free recall.
Corrected aided recall: The logistic regression for ‘corrected aided recall’ was
statistically significant, χ2(3) = 12.70, p = .005. The model explained 19% (Nagelkerke R2) of the variance in corrected aided recall. Table 2 shows that neither type of campaign (p =
20
.957), type of brand status (p = .118), or the interaction between these two variables (p = .656) were significant predictors of correct ‘corrected aided recall’. Since the model itself was significant and as type of brand status showed a tendency to be significant, this element was explored nonetheless. When a campaign included a strong brand, participants were 0.17 times less likely to have correct ‘corrected aided recall’ as opposed to a single-medium campaign (b* = -1.77, p = .118, OR = 0.17, 95% CI [0.02, 1.57]). All predictors were however not significant and so hypotheses 1a, 2a, 3a-i and 3b-i were all not supported for corrected aided recall.
Affective consumer responses
To test how a cross-media vs. single-medium campaign influenced consumers’ affective responses and how brand status moderated this relationship, two separate ANOVA’s with type of campaign and type of brand status as independent variables and either attitude towards the advertising material or attitude towards the brand as the dependent variable were conducted.
Attitude towards the advertising material: As can be seen in Table 3, there were no
significant main effects of type of campaign (F(1, 106) = 0.09, p = .765, η² = 0.00) and type of brand status (F(1, 106) = 0.94, p = .333, η² = 0.01) on attitude towards the advertising material. There also was no significant interaction between type of campaign and type of brand status on attitude towards the material, F(1, 106) = 0.05, p = .828, η² = 0.00. Hypotheses 1b, 2b, 3a-ii and 3b-ii were thus all not supported for attitude towards the advertising material.
Attitude towards the brand: Table 4 shows that there also was no significant main
effect of type of campaign (F(1, 106) = 0.89, p = .347, η² = 0.01) nor was there a significant interaction between type of campaign and type of brand status on attitude towards the brand,
F(1, 106) = 0.18, p = .669, η² = 0.00. There was however a significant main effect for type of
brand status on attitude towards the brand, F (1, 106) = 23.82, p = .000, η² = 0.18. When exposed to advertising for a strong brand, participants had a more positive attitude towards
21 the brand (Mstrong = 5.29, SD = 1.31) than when exposed to advertising for a weak brand (Mweak = 4.11, SD = 1.21). Hypothesis 2b was thus supported while hypotheses 1b, 3a-ii and 3b-ii were not supported for attitude towards the brand.
Conative consumer responses
To test how a cross-media vs. single-medium campaign influenced consumers’ conative responses and how brand status moderated this relationship, an ANOVA with type of campaign and type of brand status as independent variables and purchase intention as dependent variable was conducted.
As can be seen in Table 5, there was no significant main effect of type of campaign (F(1, 106) = 0.03, p = .871, η² = 0.00) nor was there a significant interaction between type of campaign and type of brand status on purchase intention, F(1, 106) = 0.57, p = .453, η² = 0.01. There was however a significant main effect for type of brand status on purchase intention, F (1, 106) = 25.97, p = .000, η² = 0.20. When exposed to advertising for a strong brand, participants had higher purchase intentions (Mstrong = 4.08, SD = 1.97) than when they were exposed to advertising for a weak brand (Mweak = 2.39, SD = 1.45). Hypothesis 2c was thus supported for purchase intention while hypotheses 1c, 3a-iii and 3b-iii were not supported for purchase intention.
Conclusion and Discussion
The aim of this study was to provide an understanding of the effectiveness of cross-media versus single-medium advertising campaigns at a cognitive, affective, and conative consumer response level and to better understand the (moderating) role of brand status in this relationship as well. The results indicated that there was no difference between a cross-media or single-medium campaign in terms of generating more positive consumer responses and additionally, brand status did not have a moderating role in this relationship either. Rather, the results showed that brands with a strong status (vs. a weak status) generated more positive consumer responses, particularly on an affective and conative level.
22
Participants exposed to advertising for a strong brand had a more positive attitude towards the brand as well as higher purchase intentions than those exposed to advertising for a weak brand. These results are comparable to earlier results by Laroche et al. (1996) and Wang (2009). Both consumer-based brand equity and ELM can explain these results. Consumer-based brand equity refers to consumers’ brand knowledge and their feelings towards or experiences with a particular brand (Keller, 2003) and the higher one’s brand equity, the more favorable one’s responses will be (Esch et al., 2006; Hoeffler & Keller, 2003; Keller, 2003). Given consumer’s more favorable responses for strong over weak brands, the current study probably indeed indicated that consumers’ brand equity was indeed higher for a strong brand than for a weak brand, thereby supporting the notion of customer-based brand equity. Furthermore, it was mentioned that strong brands would be processed via ELM’s central route and weak brands via its peripheral route (Dahlén & Lange, 2004; Leigh, 1992). Consumers would then be more “motivated to process information about strong brands” (Dahlén & Rosengren, 2005, p. 156), making them notice and enjoy advertising more for that type of brand as well (Dahlén & Lange, 2005; Rice & Bennett, 1998). Given the more positive consumer responses for strong over weak brands, it may be argued that consumers might indeed have processed strong brands more centrally and weak brands more peripherally.
While participants did have more positive attitudes towards the brand, they did not have more positive attitudes towards the advertising material when they were exposed to advertising for a strong brand. Leigh (1992) argued that consumers have the tendency to particularly focus on brand information and this study’s results for attitude towards the advertising material indeed seemed to support this notion. The differences in attitudes could imply that the evaluation of advertising material is independent of the evaluation of a brand or that attitude towards the brand and brand status are possibly related items but additional research will be necessary to test this. In addition, a brand with a strong status did not result in more correct brand memory than a brand with a weak status. This study included two familiar brands and while consumers were not familiar with the ads used, the high degree of
23 familiarity with both brands probably caused brand memory to not differ with brand status. Unlike the claims made by Hoeffler and Keller (2003) and Wang and Muehling (2012), it seems like here the strong brand did not have a wider (associative) network in consumers’ memory (Hoeffler & Keller, 2003; Keller, 2003) than the weak brand, probably as both brands were already very familiar to consumers.
Contrary to expectations, a cross-media campaign did not generate more positive cognitive, affective, and conative consumer responses than a single-medium campaign. While previous research did find more positive consumer responses (e.g. Chatterjee, 2012; Edell & Keller, 1989; Tang et al., 2007; Unnava & Burnkrant, 1991; Wang, 2007, 2011; Wang & Nelson, 2006), in the current study it appeared that the three underlying theoretical mechanisms for a possible synergy effect (encoding variability theory, repetition variation theory, and multiple source perceptions) were most likely not present as there was no difference between a cross-media or single-medium campaign in terms of generating more positive consumer responses. Consumer responses were probably not more positive in the cross-media situation than in the single-medium situation as participants were exposed to a brand’s corporate website. It could be argued that using such a website does not generate cross-media effects as it does not enable consumers to recognize (or prime) elements from the TV-commercial on the subsequent website (which would make synergy more likely), something which is possible with an accompanying website to a TV-commercial.
Brand status did not moderate the relationship between advertising campaigns and consumer responses: the effectiveness of a cross-media or single-medium campaign did not differ for a strong or weak brand. It was expected that for strong brands, a cross-media campaign would be more engaging, personally relevant and therefore more beneficial for consumer responses than medium messages whereas for weak brands, a single-medium campaign instead would result in more positive consumer responses than a cross-media campaign (Chang & Thorson, 2004; Wang, 2009). However, the suggested difference in enhanced media engagement and personal relevance did not seem to play a role. These results differed from Wang (2009) who found that cross-media advertising for a personally
24
relevant item did result in more positive consumer responses and similarly, that single-medium advertising for a personally irrelevant item also resulted in more positive consumer responses. A possible explanation for the result for strong brand advertising could be that the consumers in the current study did not consider the included corporate website to be personally relevant or interesting and as a result, media engagement was not enhanced. This in turn, could have led to no difference in consumer responses for either a cross-media or single-medium advertising campaign for a strong brand.
Limitations and future research
This study has some limitations. Firstly, participants in the cross-media situation were exposed to a brand’s corporate website. At the time of this research, no accompanying websites for commercials were available for both brands. Future research could therefore decide to create an own accompanying website, or preferably use an existing accompanying website. Future research may even want to work together with the brand(s) in question and time a study accordingly with the release of a new cross-media advertising campaign.
Secondly, this study used two brands within the same product category. It is recommended that other studies use different brands and products (and so also different websites and TV commercials) to see if there are similar results which would enhance generalization of the results. While the pre-test and manipulation check did show a significant difference in brand status for the two brands used here (numbers 1 and 22 in the Best Global Brands list), future research could decide to compare a strong to a much weaker familiar brand (e.g. number 1 to number 100) to possibly find bigger cross-media or single-medium effects or bigger differences between the two.
In addition, it was suggested that consumers might evaluate advertising material and brands independent of one another or that attitude towards the brand and brand status are related items. Particularly the latter seems to be a logical explanation as customer’s brand equity was probably indeed higher for a strong than for a weak brand (Keller, 2003) since consumer’s responses were most favorable for a strong brand. This in turn also implies that
25 consumer’s knowledge about, feelings towards, and experiences with the strong brand were probably more positive than with the weak brand - and so one’s attitude towards the brand is likely to be related to a brand’s status. Future research should study and explore this link more in-depth.
Practical implications
While other studies did find cross-media effects, the current study did not. Cross-media advertising therefore does not always seem to work better than single-medium advertising and advertisers should keep this in mind. Nonetheless, the results of this study did show some practical implications. For instance, the results showed that consumers’ cognitive responses for the two included familiar brands did not depend on brand status. Hence, brand memory is not something familiar brands should be focused on, especially as consumers are already likely to be familiar with the brand and its elements. Instead, brand status did affect consumers’ affective and conative responses and so these responses should be of main concern for advertisers. Before creating an advertising campaign, advertisers should therefore first find out whether their brand is perceived to have a strong or weak brand status among their target audience, in particular as advertising for a brand with a strong status is likely to result in a more positive attitude towards the brand and higher purchase intentions than advertising for a brand with a weak status. Knowing a brand’s status therefore enables one to form expectations about a possible increase in attitude towards the brand and higher purchase intentions.
Furthermore, as in this study participants appeared to be uninterested in a brand’s corporate website, advertisers could decide to refrain from referring to the corporate website in a cross-media campaign. Instead, they may want to create and design an accompanying website for a specific advertisement to increase the likelihood for synergy and possible subsequent cross-media effects.
26
to be an important element in determining the effectiveness of advertising campaigns and it is something advertising campaigns will need to take into account.
References
Blair, M.E. & Innis, D. E. (1996). The effects of product knowledge on the evaluation of warranteed brands. Psychology & Marketing, 13(5), 445–456.
Broniarczyk, S. M. & Gershoff, A. D. (2003). The reciprocal effects of brand equity and trivial attributes. Journal of Marketing Research, 40(2), 161-175.
Bronner, A.E., Neijens, P. & Van Raaij, W. F. (2003). Multimediacampagnes: populair maar weinig onderzocht. In Ontwikkelingen in het marktonderzoek. Jaarboek MarktOnderzoekAssociatie (MOA), ed. A.E. Bronner, 25–38. Haarlem: de Vrieseborch.
Bronner, F. & Neijens, P. (2006). Audience experiences of media context and embedded advertising: A comparison of eight media. International Journal of Market Research,
48(1), 81-100.
Buchholz, L. M. & Smith, R. E. (1991). The role of consumer involvement in determining cognitive response to broadcast advertising. Journal of Advertising, 20(1), 4-17. Campbell, M. C. & Keller, K. L. (2003). Brand familiarity and advertising repetition effects.
Journal of Consumer Research, 30(2), 292-304.
Cary, M. & Reder, L. M. (2003). A dual-process account of the list-length and strength-based mirror effects in recognition. Journal of Memory and Language, 49(2), 231-248. Chang, Y. & Thorson, E. (2004). Television and web advertising synergies. Journal of
Advertising, 33(2), 75-84.
Chatterjee, P. (2012). The role of varying information quantity in ads on immediate and enduring cross-media synergies. Journal of Marketing Communications, 18(3), 217-240.
Dahlén, M. & Lange, F. (2004). To challenge or not to challenge: Ad-brand incongruency and brand familiarity. Journal of Marketing Theory and Practice, 12(3), 20-35.
27 Dahlén, M. & Lange, F. (2005). Advertising weak and strong brands: Who gains?
Psychology & Marketing, 22(6), 473-488.
Dahlén, M. & Rosengren, S. (2005). Brands affect slogans affect brands? Competitive interference, brand equity and the brand-slogan link. Brand Management, 12(3), 151-164.
Dijkstra, M. (2002). An experimental investigation of synergy effects in multiple-media
advertising campaigns. PhD dissertation, Tilburg University. The Netherlands.
Dijkstra, M., Buijtels, H. E. J. J. M. & Van Raaij, W. F. (2005). Separate and joint effects of medium type on consumer responses: A comparison of television, print, and the Internet. Journal of Business Research, 58(3), 377-386.
Edell, J. A. & Keller, K. L. (1989). The information processing of coordinated media campaigns. Journal of Marketing Research, 26(2), 149-163.
Esch, F. R., Langner, T., Schmitt, B. H. & Geus, P. (2006). Are brands forever? How brand knowledge and relationships affect current and future purchases. Journal of Product
& Brand Management, 15(2), 98-105.
Gibson, B. (2008). Can evaluative conditioning change attitudes toward mature brands? New evidence from the implicit association test. Journal of Consumer Research, 35(1), 178-188.
Harvard School of Public Health. (2012). Fact sheet: Sugary drink supersizing and the obesity epidemic. Retrieved 14 April 2013 from
http://www.hsph.harvard.edu/nutritionsource/files/2012/10/sugary-drinks-and-obesity-fact-sheet-june-2012-the-nutrition-source.pdf
Hoeffler, S. & Keller, K. L. (2003). The marketing advantages of strong brands. Brand
Management, 10(6), 421-445.
Interbrand. (2013). Best global brands 2012: The definitive guide to the 100 Best Global
Brands. Retrieved 12 March 2013 from
http://interbrand.com/Libraries/Branding_Studies/Best_Global_Brands_2012.sflb.ashx ?download=true
28
Keller, K. L. (1993). Conceptualizing, measuring and managing customer-based brand equity. Journal of Marketing, 57(1), 1-22.
Keller, K. L. (2003). Strategic brand management: Building, measuring, and managing brand
equity. 2nd edition. Prentice Hall/Pearson Educuation, New Jersey.
Kent, R. J. (2002). The effects of media-source cues in ad recall tests. Journal of Current
Issues and Research in Advertising, 24(1), 1-9.
Kent, R. J. & Allen, C. T. (1994). Competitive interference effects in consumer memory for advertising: The role of brand familiarity. Journal of Marketing, 58(3), 97-105.
Laroche, M., Kim, C. & Zhou, L. (1996). Brand familiarity and confidence as determinants of purchase intention: An empirical test in a multiple brand context. Journal of Business
Research, 37(2), 115-120.
Lavidge, R. L. & Steiner, G. A. (1961). A model for predictive measurements of advertising effectiveness. Journal of Marketing, 25(6), 59-62.
Leigh, J. H. (1992). Modality congruence, multiple resource theory and intermedia broadcast comparisons: An elaboration. Journal of Advertising, 21(2), 55-62.
Levy, S. & Nebenzahl, I. D. (2008). The influence of product involvement on consumers’ interactive processes in interactive television. Marketing Letters, 19(1), 65-77. Liu, Y. & Shrum, L. J. (2009). A dual-process model of interactivity effects. Journal of
Advertising, 38(2), 53–68.
Mackenzie, S. B. & Spreng, R. A. (1992). How does motivation moderate the impact of central and peripheral processing on brand attitudes and intentions? Journal of
Consumer Research, 18(4), 519-529.
Moorman, M., Neijens, P. C. & Smit, E. G. (2007). The effects of program involvement on commercial exposure and recall in a naturalistic setting. Journal of Advertising, 36(1), 121-137.
Naik, P. A. & Raman, K. (2003). Understanding the impact of synergy in multimedia communications. Journal of Marketing Research, 40(4), 375-388.
29 fashion clothing and perceived brand status. Journal of Product & Brand
Management, 17(5), 341-352.
Petty, R. E., Cacioppo, J. T. & Schumann, D. (1983). Central and peripheral routes to advertising effectiveness: The moderating role of involvement. Journal of Consumer
Research, 10(2), 135-146.
Purchase, H. C., Cohen, R. F. & James, M. (1996). Validating graph drawing aesthetics. In
Graph drawing (pp. 435-446). Berlin Heidelberg, Germany: Springer.
Rice, B. & Bennett, R. (1998). The relationship between brand usage and advertising
tracking measurements: International findings. Journal of Advertising Research,
38(3), 58-66.
Stammerjohan, C., Wood, C. M., Chang, Y. & Thorson, E. (2005). An empirical investigation of the interaction between publicity, advertising, and previous brand attitudes and knowledge. Journal of Advertising, 34(4), 55-67.
Tang, T., Newton, G. D. & Wang, X. (2007). Does synergy work? An examination of cross- promotion effects. International Journal on Media Management, 9(4), 127-134. Unnava, H. R. & Burnkrant, R. E. (1991). Effects of repeating varied ad executions or brand
name memory. Journal of Marketing Research, 28(4), 406-416.
Voorveld, H. A. M., Neijens, P. C. & Smit, E. G. (2011). Opening the black box:
Understanding cross-media effects. Journal of Marketing Communications, 17(2), 69-85.
Voorveld, H. A. M., Neijens, P. C. & Smit, E. G. (2012). The interacting role of media
sequence and product involvement in cross-media campaigns. Journal of Marketing
Communications, 18(3), 203-216.
Wang, A. (2007). Branding over mobile and internet advertising: The cross-media effect.
International Journal of Mobile Marketing, 2(1), 34-42.
Wang, A. (2009). Cross-channel integration of advertising: Does personal involvement matter? Management Research News, 32(9), 858-873.
30
Management, 17(3), 275-290.
Wang, A. & Muehling, D. D. (2012). The moderating influence of brand status and source confirmation on third-party endorsement effects in advertising. International Journal of
Advertising, 31(3), 605-622.
Wang, S. A. & Nelson, R. A. (2006). The effects of identical versus varied advertising and publicity messages on consumer response. Journal of Marketing Communications,
12(2), 109-123.
Yang, K.C.C. (2004). Effects of consumer motives on search behavior using internet advertising. CyberPsychology & Behavior, 7(4), 430-442.
31 Tables
Table 1
Logistic regression analysis predicting likelihood of correct ‘free recall’
Brand memory: Free recall
b* SE p Odds 95% CI for Odds
Variables Lower Upper
Type of campaign 17.91 7882.49 0.998 59832401.43 0.00 . Brand status 17.91 7595.76 0.998 59832401.40 0.00 . Interaction -36.55 10946.65 0.997 0.00 0.00 . Nagelkerke R2 0.18 Model X2 4.49 Table 2
Logistic regression analysis predicting likelihood of correct ‘corrected aided recall’
Brand memory: Corrected aided recall
b* SE p Odds 95% CI for Odds
Variables Lower Upper
Type of campaign -0.08 1.44 0.957 0.93 0.06 15.61 Brand status -1.77 1.13 0.118 0.17 0.02 1.57 Interaction -0.70 1.58 0.656 0.50 0.02 10.88 Nagelkerke R2 0.19 Model X2 12.70 Table 3
Results of a two-factor analysis of variance (N = 110) for the dependent variable ‘attitude towards the advertising material’.
Sum of Squares df Mean Square F p η² Campaign 0.12 1 0.12 0.09 0.765 0.00 Brand status 1.23 1 1.23 0.94 0.333 0.01
Campaign * Brand status 0.06 1 0.06 0.05 0.828 0.00
Error 138.01 106 1.30
Total 2490.09 110
Table 4
Results of a two-factor analysis of variance (N = 110) for the dependent variable ‘attitude towards the brand’.
Sum of Squares df Mean Square F p η² Campaign 1.44 1 1.44 0.89 0.347 0.01 Brand status 38.38 1 38.38 23.82 0.000 0.18
Campaign * Brand status 0.30 1 0.30 0.18 0.669 0.00
Error 170.76 106 1.61
32 Table 5
Results of a two-factor analysis of variance (N = 110) for the dependent variable ‘purchase intention’. Sum of Squares df Mean Square F p η² Campaign 0.08 1 0.08 0.03 0.871 0.00 Brand status 79.12 1 79.12 25.97 0.000 0.20
Campaign * Brand status 1.73 1 1.74 0.57 0.453 0.01
Error 322.93 106 3.05
33 Appendix A – Experimental material
• TV commercial Coca-Cola – Catch (reading order: from left to right)
34
35 • Website Pepsi (top: English, bottom: Dutch)
36
Appendix B – Experiment questionnaire
Informed consent form “Advertising campaigns and a familiar brand”
I hereby declare that the information about the nature and method of this research, as put forward in the invitation for this research, is clear to me.
I completely voluntarily agree to participate in this research. I reserve the right to withdraw this agreement without needing to give a reason for this. I realize that I can stop participating in the research at any time.
If any of my research results are used in scientific publications, or otherwise made public, this will be completely anonymous. My personal details will not be provided to third parties without my explicit consent.
If I want more information, now or in the future, I can contact Kelly de Ruiter (06 4846 7706, kelly.deruiter@student.uva.n). For any complaints about this research, I can contact the member of the ‘Commissie Ethiek’ (Ethics Committee) on behalf of ASCoR, Kloveniersburgwal 48, 1012 CX Amsterdam; 020- 525 3680; ascor-secr-fmg@uva.nl.
√ I understand the above and agree to participate in this research
Questionnaire
Below the questionnaire for the cross-media condition is shown. The single-medium condition questionnaire was the same, except that it did not include the questions ‘attitude towards the website’ and ‘ad familiarity with the website’ and it has a different second exposure (TV commercial instead of website).
- What brand was or what brands were advertised in the material that you have seen earlier? Please write down all brand(s) you can recall (open answer).
- Please indicate whether you recall having seen advertising for the following brands:
Fanta [yes – no]
Dr. Pepper [yes – no]
Coca-Cola [yes – no]
Sprite [yes – no]
Pepsi [yes – no]
- You have just seen a TV commercial [website] for <brand>. How would you rate this commercial [website]? Please indicate for each scale below what best reflects your view.
Not interesting Interesting Bad Good Not appealing Appealing Not likable Likable
- How would you rate <brand>, overall as a brand? Please indicate for each scale below what best reflects your view.
Not interesting Interesting Bad Good Not appealing Appealing Not likable Likable
37 - How likely is it, if you were to buy soft drinks on your next shopping trip, that you will
buy <brand>? Very unlikely Very likely
- How likely are you to recommend <brand> to a friend? .
Very unlikely
Very likely
- Please indicate to what extent you agree with the following statements:
Strongly disagree
Strongly agree
I have a lot of experience with cola Compared to the average person, I have
a lot of knowledge about cola
I describe myself as someone who is very
familiar with cola
- I would describe soft drinks as ...
Very unimportant Very important Very uninteresting Very interesting
- Have you seen the TV commercial before? no
yes
- Please indicate to what extent you agree with the following statements: I think <brand> …… Strongly disagree Strongly agree ... is prestigious … is distinctive … is sophisticated … has high esteem … is successful … is wealthy … is exclusive … signals status
- What is your gender? Male
38
- What is your age? (open answer)
- What is your study?
Communication Science Media (or related)
Arts or social sciences (e.g. history or psychology)) PABO
Engineering or technical (or related)
Business (incl. marketing, IBA, economics etc.) Other, namely:
- What is the educational level of your study? HBO
