A Study on the Salience of Forewarnings in Native Advertising
Name: Lotte Heleen Lummen Student number: 10618074
‘Master Thesis’
Graduate School of Communication
Master’s program Communication Science, Persuasive Communication. Supervisor: Drs. D. Muntinga
2
Abstract
This study investigates whether the saliency of forewarnings affects peoples overall brand evaluation and ad attitude mediated by perceived persuasive intent. By conducting a 2 (highly salient vs. less salient forewarnings) * 2 (familiar vs. unfamiliar brands) factorial design. This study also investigates whether brand familiarity moderates the effect of saliency in
advertorials on perceived persuasive intent. The effects for saliency on forewarnings do not significantly affect people’s overall brand credibility and ad attitude. The direction of those dependent variables is as expected, for advertorials with highly salient forewarnings brand credibility is higher and ad attitude is lower, but this is not significant. Brand attitude however is significantly different. Advertorials with a highly salient forewarning lead to a significant higher brand attitude. The moderator brand familiarity moderates significantly on the
perceived persuasive intent. Highly salient forewarnings in combination with familiar brands lead to a higher level of perceived persuasive intent. As this study is a first step in to
researching the terms and conditions of forewarnings instead of the presence or absence of forewarnings, a lot more research is needed. Regarding the practical implications for legislators and advertisers, this study indicates that highly salient forewarnings can lead to a lower brand evaluation. Keywords: Forewarning, Advertorial, Saliency, Perceived Persuasive Intent, Brand familiarity.
Warning the consumer before tyring to persuade them
The rapid growth of the internet has caused an explosive grow in various forms of digital marketing. In 2012 the European advertising market showed an increase of 14.7 billion Euros
3
(AdEx Benchmark, 2012). From all the existing advertising, 25.6 percent consists of internet advertising and an increasing percentage thereof is native advertising which includes the appearance and other characteristics from editorial material (Bong-Hyun, 1995).
With this rapid growth of advertisement forms, researchers and practitioners study what works best in order to keep up with this growth. Online advertisement such as banners cause irritations and eventually avoidance of the advertising (Dreze & Hussher, 2003). Therefore brands and advertisers try to come up with different forms of online advertising such as online advertorials; an implicit, native form of marketing. An example is a blogger writing on a new favorite clothing item and implicitly naming a brand. Implicitness prevents people from creating an instant opinion or attitude and makes people focus on the form of the content first (Zhou, 2013).
Legislators try to increase the persuasive awareness of consumers by inserting
forewarnings into those advertorials. Forewarnings are warning statements that claim a certain right (Leon, Rotunda, Sutton & Schlossman, 2003). They try to show consumers that the advertorial includes a persuasive attempt so that the perceived persuasive intent increases and people realize it is marketing. This results in a fair and honest evaluation of the advertorial since the consumer is aware of the persuasive attempt. But it could also cause a process of counter arguing resulting in negative brand evaluation (Friestad & Wright, 1994). Increasing the perceived persuasive intent is necessary since the distinction between commercial messages and editorial messages keeps getting smaller. When people are able to recognize such an advertorial as a commercial message, they are then able to activate and create a defense (d’Astous & Chartier, 2000).
Forewarnings are necessary in order to create legal internet advertising. The Dutch Commercial Code (‘’De Nederlandse Reclame code,’’ 2009) states that advertising, online
4
advertising being no exception, cannot be deceiving. Deceiving advertising is advertising with false, unclear or ambiguous information leading to decisions by the consumer that would not have happened otherwise. The Federal Trade Commission (‘’About the FTC,’’ 2013), the international commercial code, also supports this.
Most, rather old, studies looked at the presence or absence of forewarnings in sponsored content but none of them looked at the terms and conditions of forewarnings (Dekker & Van Reijmersdal 2010; Kiesler & Kiesler, 1964; McGuire & Papageorgis, 1962; Wei, Fischer & Main, 2008; Wood & Quin, 2003). Recent research looked at the disclosure time of forewarnings in sponsored video content, focusing more on the terms and conditions of the forewarning. They wanted to know if the level of perceived persuasive intent
influenced the brand evaluation (Boerman, Reijmersdal & Neijnens, 2014). This research will shift the focus of video content in to web content and specifically the upcoming form:
advertorials since more research is needed towards the consumer response towards
advertorials (Cameron & Ju-Pak, 2000; Dahlèn & Edinius, 2007; Wei et al., 2008). Besides the fact that advertorials are upcoming, the agreements surrounding advertising are still changing. The right usage of forewarnings is important in order to create the most effective brand evaluation.
This study will focus on whether the salience of a forewarning will affect the perceived persuasive intent and eventually the brand evaluation. As forewarnings are obligatory it is important to see whether the terms and conditions of a forewarning plays a role. In addition, this study will also check whether the role of brand familiarity plays a significant moderating role in the persuasiveness of forewarnings in advertorials. Since brand familiarity evokes a process of falling back on previous information and associations
5
something. This could enhance the perceived persuasive intent (Machleit & Wilson, 1988).
Theoretical Background
A new form of native marketing: Advertorials
As previous research focused mainly on forewarnings (warning statements) in TV programs, this study will shift this focus in to native marketing and especially advertorials (Boerman et al., 2014). An advertorial is a form of advertisement that uses the appearance and other characteristics of editorial material. Native marketing and thus an advertorial, has the intentional goal to blend in with editorial content (Bong-Hyun, 1995). A form where the reader does not immediately recognizes the content as advertised but sees it as journalistic article, or web content which includes opinions (Zhou, 2013). An advertorial is a form of sponsorship advertising. Advertorials can spread marketing objectives by associative means with editorial content by focusing on the content of the medium and not the persuasive attempt.
The persuasion knowledge model
The Persuasion Knowledge Model (PKM) is used to explain the perceived persuasive intent that people experience during persuasive attempts. Persuasiveness refers to the marketing strategy performed, being effective. Marketing communication is effective when people perceive a message as favorable and the influence on the consumer leads to purchases or the intention to buy (Leon et al., 2003).
Research on how people perceive and process certain forms of advertising shows that the perceived persuasive intent of a message influences consumers cause it leads to a more
6
skeptical and biased way of evaluating brands. (Friestadt & Wright, 1994). Perceived persuasive intent is also known as the general knowledge gained and developed about how, why and when a certain message tries to influence them (Boerman et al., 2014). According to the PKM people follow the ‘’if, then‘’ procedure. If consumers are aware that they are part of an persuasive attempt, via for instance a forewarning, they then have to manage a response to this attempt which is usually negative. Resulting in a more negative attitude towards the brand and sender (Friestadt & Wright, 1994). This results in a more systematic and biased way of processing the attempt which means that people will be more critical and skeptical towards the message presented and thus leading to lower brand evaluations, buying intentions and or actual purchases (Janssen, Fennis, & Pruyn, 2010).
The PKM also states that people need to be aware of the persuasion attempt in order to activate persuasion knowledge. Previous research shows that people learn how to become aware of the persuasive attempt early on in childhood. (John, 1999). The following constructs are important for this research: 1) comprehension of the intent of advertising, 2)
understanding the tactics of advertisement and 3) diagnose biased and deceptive
advertisements. Advertorials, as will be explained later on, are harder to distinguish from online editorial content. This means that the constructs learned in childhood do not
immediately come up and this leads to less activation of the persuasion knowledge and thus less critical and skeptical attitudes towards the brand.
Persuasion knowledge can be divided in to two concepts: conceptual and attitudinal persuasion knowledge. Conceptual persuasion knowledge refers to the cognitive dimension that includes the recognition of advertising, the source and audience (Boerman, Van
Reijmersdal & Neijnens, 2012). This study will focus on the attitudinal aspect of forewarnings in advertorials. This means that the attitude towards forewarnings can affect the recognition of
7
advertising. The attitudinal aspect includes critical attitudes, such as dislike and skepticism applied to a certain persuasive attempt. The attitudinal aspect leads to certain brand evolutions resulting in buying intentions and purchases, which is most important for practice.
The process of forewarnings in a persuasive attempt
A forewarning is defined as a warning statement that claims a certain right. Warning statements claim that the presented content includes advertisement (Leon et al., 2003). As forewarnings are obligatory they may also lead to less effectiveness of the persuasive attempt. Forewarning an audience creates resistance and stimulates counter arguing causing the persuasive attempt to be less successful. This process of counter arguing lets people focus on the persuasive attempt instead of the characteristics of the product and brand. People do not like to be persuaded they want to form their own opinions.(Leon et al., 2003; McGuire & Papageorgis, 1962; Wood & Quin, 2003).
Previous studies focused mainly on the presence or absence of forewarnings and confirmed the idea that people who were forewarned were less persuaded than people who were not (Kiesler & Kiesler, 1964; McGuire & Papageorgis, 1962; Wei et al, 2008; Wood & Quin, 2003).There is little known about the terms and conditions of forewarnings and how they work. It is well known that forewarning an audience leads to more resistance and counter arguing but we do not know if the terms and conditions of a forewarning have anything to do with this outcome. I therefore look at the salience as part of the terms and conditions of forewarnings instead of the presence or absence of forewarnings. A highly salient forewarning will probably create more resistance and counter arguing since the focus is more on the persuasive attempt than with a less salient forewarning.
8 Effects of attitude towards advertorials
The attitude towards the advertorial is defined as a predisposition that leads to a favorable or unfavorable evaluation of the advertorial and the characteristics that the advertorial includes. The attitude towards the advertisement leads to a better brand image and eventually increases buying intentions and purchase intentions (Alwitt & Mitchell, 1985).
Shen and Chen (2007) state that the context where the information of the
advertisement is placed, can determine the individual interpretation by the receiver. This means that unfavorable attributes, such as a forewarning, of the presented content can also create an overall unfavorable evaluation of the advertisement. (Shen & Chen, 2007; Chattyopadhyay & Nedundgadi, 1992).
Forewarnings activate a cognitive process of counter arguing and persuasion knowledge, people immediately could get critical and skeptical towards the advertisement. They have the feeling that the sender is trying to persuade in to something they might not agree with (Wood & Quin, 2003). Therefore it is hypothesized:
H1a. A highly salient forewarning in an advertorial leads to a lower ad attitude than a less salient forewarning.
Effects of attitude towards the brand
The attitude towards a brand is one of the most important indicators for the value of brand equity. Brand attitude is known as the overall evaluation of a certain brand by an individual (Liu, Li, & Mizerski, 2012). Individuals’ perception of the brand is a good and reliable indicator for buying intentions and actual purchases of a product.
Forewarnings may lead to less effective persuasive attempts and thus less brand attitude. It leads to more resistance and reduced persuasion since the persuasion knowledge is
9
activated. People therefore feel pressured into liking the advertorial (Boerman et al., 2012). When the forewarning is highly salient people will feel more pressured since the focus is even more on the persuasive attempt. The process of counter arguing gets activated quicker and this can lead to disliking the brand. The brand should not focus on the impending or threatening appeal such as the forewarning because it leads to negative associations with the brand. (Wood & Quin, 2003). Thus it is hypothesized:
H1b. A highly salient forewarning in an advertorial leads to lower levels of brand attitude
than a less salient forewarning.
Effects of brand credibility
Brand credibility is the believability of the product position information included in the brand presented. This is defined as the brand doing what they promised on a consistent base (Erdem, Swait & Louviere, 2002). Brand credibility includes two important elements, trustworthiness and expertise. Trustworthiness refers to the element that a brand must be willing to give what is promised and expertise refers to the capability of giving the promised good. When brand credibility is high, it significantly increases the buying intention and overall brand evaluation (Sweeny & Swait, 2008).
Brand credibility adds value to the brand since it shows an honest and sincere character which is very important to people (Sweeny & Swait, 2008). This value causes people to feel the need and urge to associate themselves with the brand and purchase it. When there is no credibility there will be intention to buy the product (Erdem & Swait, 2004). A forewarning decreases the thought that the product presented in the advertorial is biased since they are stating immediately what the advertorial is about. The brand and the advertisement are being honest to the customer by warning them. By using a highly salient
10
forewarning the advertisement seems even less biased since the focus is on being honest and warning people. In order to create a credible brand image, the thoughts about the
advertisement cannot be biased and this can be prevented by using a salient forewarning (Wood & Quin, 2003). Based upon this information the following hypothesis is stated:
H1c. A highly salient forewarning in an advertorial leads to higher levels of brand credibility than a less salient forewarning.
The mediating effect of perceived persuasive intent
This study will focus on the attitudinal aspect of forewarnings in advertorials. Hence I will mention the attitudinal dimension of persuasion knowledge resulting in perceived persuasive intent. As was said earlier, forewarnings can activate a cognitive process of counter arguing and persuasion knowledge, people could get critical and skeptical towards the advertisement (Boerman et al., 2014). With this said it is known that the persuasion knowledge, the level of which the consumer perceives the sender is persuading them, is getting bigger, the consumer might not agree with those statements being said. This means that a highly salient forewarning leads to a higher activated persuasion knowledge and thus perceived persuasive intent. This perceived persuasive intent leads to lower brand evaluations (Wood & Quin, 2003).The salience of a forewarning hereby is important cause in this way, people are more aware of this persuasive attempt and the persuasion knowledge (how, why and when) gets activated. Based on the PKM, this study posits that the level of salience of a forewarning influences the level of perceived persuasive intent, which in turn influences brand evaluations. It is therefore
11 H2. Persuasion knowledge negatively mediates the impact of highly salient forewarnings on a) ad attitude, b) brand attitude and c) brand credibility such that that people know that they are being influenced by the sender and thus create a more negative overall
attitude towards the brand
The moderating effect of brand familiarity
Brand familiarity can be seen as an uni-dimensional construct that is directly linked to the amount of time that a person spends on processing and thinking about a brand. It is the knowledge known by the consumer about a certain brand (Baker, Hutchinson, Moore, & Nedungadi, 1986). Brand familiarity evokes associative feelings and existing brand beliefs in the memory of an individual.
As explained earlier the PKM predicts the persuasion knowledge by the consumers. This response is activated by previous associative, general persuasion knowledge. Brand familiarity also falls back on associations and previous knowledge but now towards the brand and not the persuasive attempt. When people are confronted with familiar brands they
immediately fall back on previous information and associations such as the knowledge that the brand is commercial and tries to sell something (Machleit & Wilson, 1988). When viewers see and recognize a brand the perceived persuasive intent therefore also gets higher. For unfamiliar brands there are no existing associations and beliefs just that the overall commercial goal of selling something. Thus, it is hypothesized:
H3. The effect of salient forewarnings on perceived persuasive intent is moderated by brand familiarity. When brand familiarity is high, a forewarning leads to a higher level of persuasion knowledge. When brand familiarity is low, a forewarning leads to a lower level of persuasion knowledge. No differences are expected for less salient
12
forewarnings.
Figure 1 shows the theoretical model. It includes all the hypotheses and variables, advertorials with a highly salient or less salient forewarning, persuasion knowledge and brand evaluation.
Method
Research design & Pretest
To test the hypotheses in a controlled setting, this study performed an experimental study with an online questionnaire. This study uses a 2 (highly salient vs. less salient forewarnings in advertorials) x 2 (low vs. high brand familiarity) factorial design.
Before this experimental study took place 20 people were pretested in order to see if our measures and stimuli were sufficient. This pretest asked participants to answer a question on brand familiarity. The participants were asked to indicate how familiar they were with both shown brands. A 7 point congruity Likert scale was used in order to confirm our brand choices. Response categories ranged from 1 very unfamiliar to 7 very familiar. A t-test
13
showed that the advertorial with the familiar brand Heinz rated significantly higher on being familiar (Mfamiliarity=5.89; SD=1.76) than the unfamiliar brand Saucony (Mfamiliarity=1.42; SD=.67, t(20)=-8.10, p = <.000). Thus this means our manipulation succeeded in the pre test.
Procedure & Participants
Participants had to imagine themselves surfing on the web for a recipe for dinner. The participants had to read the advertorial as they would normally do on a website taking as much time as they needed. After looking at the advertorial several subsequent questions on the control variables (product involvement, website knowledge and website use frequency) were asked. After these questions the participants were asked about the manipulation checks and the brand evaluation (ad attitude, brand attitude and brand credibility) followed by some general information (gender, age and education and occupation).
We asked the participants (n=135) to voluntarily and anonymously fill out a
questionnaire via social media including a link that led them to Qualtrics. 71,9% were female and 28,1% were male with an average age of 30,74 years old (SD=12,57). All participants were randomly assigned to one of the four conditions. Figure 2 shows how the participants were divided over these four conditions. Appendix A shows the characteristics of the total respondents and appendix B the respondents per group, after some general information and the informed consent question the actual survey started.
Highly salient Less salient
Familiar N = 33 N = 31
Unfamiliar N = 34 N = 33
Figure 2. Participants divided into four conditions.
Null participants were removed because of some missing answers in the questionnaire. This eventually resulted in 135participants that fully answered the questionnaire. A total of 131
14
participants eventually took place in this experiment. This resulted in four experimental groups showed in figure 2. For further characteristics of these groups see appendix B.
Materials
Salience. In accordance with the regulations by the Dutch Commercial Code (‘’De
Nederlandse Reclame code,’’ 2009) and The Federal Trade Commission (‘’About the FTC,’’ 2013) the forewarnings were placed in the beginning. The advertorial showed the word ‘’advertisement’’ right after the title, the size was dependent on the condition. The highly salient forewarning used capital letters and a bold, underlined letter type whereas the less salient version used a rather small letter type without any capital letters or underscore.
Brand familiarity. The participants were asked to read an advertorial on a website that
included an Italian fish stew recipe (see appendix C). A little cover story explained them what the website is about. This recipe included several ingredients whereby the ingredient tomato ketchup was our main focus. In order to manipulate brand familiarity we used the existing brand Heinz Tomato Ketchup and the fictitious brand Saucony Tomato Ketchup. Both brands are in the same category Food (sauces). New content was created so that people did not know that the material included sponsored content. We coded those two in to binary groups of 0 (unfamiliar brand) and 1 (familiar brand).
Advertorials. The advertorial was placed in an existing website of for food
(‘’Foodrepublic,’’ 2010), to create a real setting. This is a website for food lovers who want to cook good in order to live right. This is also explained in the cover story. In order to
15
Two advertorials included a highly salient forewarning and the other two advertorials included a less salient forewarning. Two groups were randomly exposed to the advertorial with the highly salient forewarning and two groups were randomly exposed to advertorials with a less salient forewarning brand. We coded those two in to binary groups of 0 (less salient forewarning) and 1 (highly salient forewarning). See appendix C for the materials used.
Measures
Attitude towards the advertorial. This study uses three items which all included a seven
point semantic differential scale. The following items were included favorable/unfavorable, pleasant/ unpleasant, good/bad adopted from Dahlèn, Friberg and Nilsson (2009). These items were randomly exposed. Response categories ranged from 1 very unpleasant to 7 very
pleasant, 1 very bad to 7 very good and 1 very unfavorable to 7 very favorable. A principal component analysis showed that these items load on one factor (EV = 2.4; R2 = .80). The
orthogonal (Varimax) rotation was selected for these three components. A reliability analysis showed that the reliability is high enough to keep all the items (α = .87; M = 3.8, SD = 1.5).
Brand attitude. In order to measure brand attitude we used three items on a seven point
semantic differential scale. We included the following items satisfactory/unsatisfactory, negative/positive and good/bad adopted from Dahlèn et al., (2009). These items were randomly exposed and in different orders. A principal component analysis showed that these items load on one factor (EV = 2.4 ; R2 = .80). The orthogonal (Varimax) rotation was
selected for these three components. A reliability analysis showed that the reliability is high enough to keep all the items (α = .87; M = 4.4, SD = 1.3).
16 Brand credibility. For the variable brand credibility we again chose three items on a seven
point semantic differential scale. The following items were included: biased/unbiased, believable/ unbelievable and trustworthy/untrustworthy adopted from Dahlèn et al., (2009) and Spry, Pappu & Cornwell, B. (2012). The answer categories ranged from, 1 very biased to 7 very unbiased, 1 very unbelievable to 7 very believable and 1 very untrustworthy to 7 very trustworthy. These items were randomly exposed and in different orders. A principal
component analysis showed that only two items load on one factor, unbelievable/believable and untrustworthy/trustworthy (EV = 1.6; R2 = .80). The orthogonal (Varimax) rotation was
selected for these three components. A reliability analysis showed that the reliability is high enough to keep these two items (α = .76; M = 4.05, SD = .63). Since the biased/unbiased scale was not reliable enough and did not load on one factor we removed it.
Persuasion knowledge. In order to measure the persuasion knowledge we asked the
participants one item “How much do you agree with the following statement? ‘’The dinner tip recipe at was advertising’’ adopted from Boerman et al., (2014). By answering on a seven point scale the participants answered how much they agreed with this statement, whereby 1 is strongly disagree and 7 is strongly agree (M = 5.27, SD = 1.82).
Brand familiarity. As was earlier discussed brand familiarity is expected to moderate the
relationship between saliency and persuasion knowledge. In order to check whether the participants were familiar with the brand we asked them: Please indicate how familiar you are with the brand ‘’X’’ (whereby 1 is completely unfamiliar and 7 is completely familiar (Mheinz= 6.42, SD =.99; Msaucony = 1.31, SD = .74) adopted from Baker (1986).
17 Control Variables. This study measured several control variables to check whether any
effects found were not a result by possible differences in the experimental groups. The participants were asked about general information on age, gender, employment, education, product involvement and website involvement. To see the exact control variables that were used see appendix E.
Results
Randomization
As 135 participants took place in this study, the differences per experimental group were not significant which resulted in four almost equal groups: gender F(1) = .08, p= .826), age F(1) = .02, p = .877), employment F(1) = 1.63, p = .243) and education F(1) = .004, p= .951). See appendix B for further means and characteristics per group.
Manipulation checks
The participants were asked to indicate how familiar they were with both shown brands. A 7 point congruity Likert scale was used in order to see how familiar the participants were with the brands. Response categories ranged from 1 very unfamiliar to 7 very familiar. A t-test showed that the advertorial with the familiar brand Heinz rated significantly higher on being familiar (Mfamiliarity = 6.42; SD = .99) than the unfamiliar brand Saucony (Mfamiliarity = 1.31; SD = .74, t(129) = -33.53, p = <.001). Thus this means our manipulation succeeded. Therefore we excluded 4 participants from further analysis that were unfamiliar with Heinz (<5) or familiar with Saucony ( >3) resulting in 131 participants.
18 Control variables
A bivariate correlation analyses was conducted to see whether there is a need to control for the variables gender, employment, education and age. All dependent variables are normally dis-tributed (see appendix D). The bivariate correlation analyses shows that there are no
significant correlations between the dependent variables (brand attitude, brand credibility, ad attitude) and the control variables (gender, age, employment status and education, ketchup liking, recipe liking, ketchup use, website use frequency and website familiarity)). Therefore, we do not have to control for these variables. Appendix E shows the exact numbers of the bivariate correlation analyses per dependent variable. Appendix F shows the exact numbers of the bivariate correlation analyses per dependent variable and all the control variables. These results imply that the several groups and the specific differences referring to the variables are not because of the differences in the background variables. Therefore we will not control for these variables.
Effects of Forewarnings on…
Effects of forewarnings on attitude towards the advertorial.
Hypothesis one (H1a) stated that a highly salient forewarning leads to lower levels of ad atti-tude than less salient forewarnings. A t-test showed that the highly salient advertorial (M high-lysalient = 3.88, SD = 1.51) did not significantly show a lower or higher level of ad
attitude than the less salient advertorial (Mlesssalient = 3.70, SD = 1,50, t(129) = -.69, p = .491). Thus we reject the first hypothesis. The direction of the add attitude however confirms the direction that is stated in the first hypothesis. A highly salient forewarning leads to lower levels of ad attitude than less salient forewarnings, however this is not significant.
19
Effects of Forewarnings on Brand Attitude.
The second hypothesis (H1b) stated that highly salient forewarnings in advertorials lead to lower levels of brand attitude than less salient forewarnings. The highly salient advertorial (Mhighlysalient = 4.68, SD = 1.27) had a significantly higher level of ad attitude than the less salient advertorial (Mlesssalient = 4.19, SD = 1.38, t(129) = -2.12, p = .036). Thus we reject hypothesis 1b. The highly salient forewarning shows a more positive attitude towards the brand than does the less salient forewarning. This means that the direction in which the brand attitude goes is different than expected.
Effects of Forewarnings on Brand Credibility.
The third hypothesis (H1c) stated that highly salient forewarnings in advertorials lead to higher levels of brand credibility than less salient forewarnings. The highly salient advertorial (Mhighlysalient = 4.21, SD = 1.31) did not significantly show a higher level of brand credibility than the less salient advertorial (Mlesssalient = 4.02, SD = 1,24, t(129) = -.89, p = .380). Therefore we reject hypothesis 1c. The direction of the brand credibility however, confirms the direction that is stated in hypothesis 1c. Highly salient forewarnings lead to higher levels of brand credibility than less salient forewarnings, but this is not significant.
Mediating Effects of Perceived Persuasive Intent
Our second hypothesis stated that Persuasion knowledge negatively mediates the impact of highly salient forewarnings on a) ad attitude, b) brand attitude and c) brand credibility such that people know that they are being influenced by the sender and thus create a more negative overall attitude towards the brand. In order to see whether a mediation effect of perceived persuasive intent took place, according to Baron and Kenny (1986), three regressions need to
20
be performed.
The first step towards a mediation effect is to check whether the independent variable (highly vs. less salient forewarnings) significantly interacts on the dependent variables (brand attitude, brand credibility and ad attitude). As was shown in the chapter Effects of
Forewarnings on…, there was no significant effect of the independent variable saliency of the forewarning on the dependent variables ad attitude and brand credibility. But saliency did show a significant effect on brand attitude. A regression analyses confirms this for the variables: ad attitude; b* = -.06, p = .491), brand credibility; b* = .08, p = .380 and brand attitude; b* = -.18, p = .036.
Step two includes checking whether the independent variable (saliency) significantly acts upon the mediator (perceived persuasive intent). The regression analysis shows that the saliency of the forewarning is not a significant predictor of perceived persuasive intent;
b* = -.13; p = .146. The less salient forewarning leads to a higher level of perceived
persuasive intent, however this is not significant (Mhighlysalient = 5.05, SD = 1.88;
Mlesssalient = 5.51, SD=1.73).
The third and last step includes a regression of perceived persuasive intent and saliency on the dependent variables (ad attitude, brand attitude, brand credibility). The saliency of the advertorial and the perceived persuasive intent were not a significant predictor of ad attitude; b* = .16, p = .067, brand attitude; b* = .02, p = .796 and brand credibility; b* = .006, p = .944.
In order to state that there is a mediation effect, all three tests need to be significant but the Sobel test (Preacher, 2014) should also confirm this. A Sobel test confirms the fact that none of the dependent variables are significantly mediated by perceived persuasive intent and saliency of the advertorial: ad attitude S = .06 (SE = .34; p = .952), brand attitude S = .02 (SE
21
= .18; p = .986), and brand credibility S = .04 (SE = .20; p = .969) In this case none of the above tests are significant and therefore we reject hypothesis three.
Moderating Effects of Brand Familiarity
Our third hypotheses stated that the effect of highly salient forewarnings on perceived persuasive intent is moderated by brand familiarity. The effect of a forewarning on perceived persuasive intent is moderated by brand familiarity. When brand familiarity is high, a
forewarning leads to a higher level of perceived persuasive intent than a less salient forewarning. No differences are expected for less salient forewarnings. In order to test our third hypothesis an analysis of variance to see whether the interaction effects did occur. This test shows us that there is a significant moderation effect, F(1,127) = 7.25, p = .008. This means that the effect of brand familiarity on perceived persuasive intent is different for different levels of saliency. Figure 3 shows what this interaction looks like. The differences in perceived persuasive intent are much higher for highly salient forewarnings (Mfamiliar = 5.64, SD = 1.58 Munfamiliar = 4,45, SD = 1.99) than for less salient forewarnings (Mfamiliar = 5.26 SD = 1.86, Munfamiliar = 5,74; SD = 1.60) as was expected.
Next to the interaction effect we also looked at the main effects. This test showed that there is no significant main effect of saliency F(1,127) = 2.15, p = .146 on perceived
persuasive intent (Mhighlysalient 5.05, SD = 1.88, Mlesssalient = 4.51, SD = 1.73).
Within the highly salient forewarning group, the perceived persuasive intent is higher for a familiar brand (Mhighlysalient-familiar = 5.64, SD = 1.58; Mlesssalient-familiar = 5.26, SD = 1.87. Those means indicate that advertorials with highly salient forewarnings and familiar brands lead to a higher level of perceived persuasive intent, however this is not significant. Expected was, that there are no differences for the unfamiliar brand on perceived persuasive intent but
22 4,25 4,5 4,75 5 5,25 5,5 5,75 6
Less Salient Highly Salient
Saucony Heinz
the results show otherwise. Within the less salient forewarning group the perceived persuasive intent is higher for unfamiliar brands (M = 5.74, SD = 1.60 than for familiar brands
(M = 5.26, SD = 1.86). For advertorials with an unfamiliar brand the less salient forewarnings leads to a higher level of perceived persuasive intent, however this is not significant. Thus we partially accept hypothesis three confirming the moderation effect.
Discussion
As forewarnings are obligatory in every form of advertisement, advertorials being no exception for that rule, this research contributes to the world of terms and conditions
(saliency) for forewarnings in advertorials. Because more research is necessary on consumer responses towards advertorials and the brands in it (Cameron & Ju-Pak, 2000; Dahlèn & Edin-ius, 2007). This research compares how the saliency of a forewarning in advertorials
23
by perceived persuasive intent. Additionally, the moderating effect of brand familiarity was examined. This study contributes to existing knowledge by shifting the focus from presence or absence of forewarnings, as previous research had predominantly done (Kiesler & Kiesler, 1964; McGuire & Papageorgis, 1962: Wei et al., 2008; Wood & Quin, 2003), to the terms and conditions of forewarnings, especially saliency. The findings thereby challenge previous literature in the way that it has a different outlook on forewarnings.
Forewarnings and brand responses.
The primary goal of forewarnings is to inform the consumer about sponsored content and also raise awareness of the advertising goal and the perceived persuasive intent thereof (Boerman et al, 2014). Theory shows that there is a presumption that the saliency of forewarnings in advertorials could affect people’s ad attitude, brand attitude and brand credibility (Wood & Quin, 2003; Dahlèn & Edinius, 2007).
First, the brand credibility did not significant differ for a highly vs. less salient forewarning but in line with the proposed literature the findings indicate the expected direction, indicating that brand credibility is higher for salient forewarnings than for less salient forewarnings. Presuming that being honest and trustworthy by showing highly salient forewarnings can lead to better level of brandcredibility confirming previous literature on ad attitude (Sweeny & Swait, 2008).
Second, the attitude towards the advertorial did not significantly led to a more unfavorable evaluation of the advertorial with the highly salient forewarning (Alwitt et al., 1985). As was expected, a highly salient forewarning activates the persuasion knowledge, which in turn leads to more critical, negative attitudes towards an ad (Shen & Chen, 2007; Chattyopadhyay & Nedundgadi, 1992). Those non-significant results could be due to the fact
24
that this study mentions the forewarning in the beginning and Dekker and Van Reijmersdal (2010) state that naming the forewarning at the end leads to a more positive ad attitude and higher level of brand credibility.
Third, the effect of forewarnings on brand attitude however was confirmed.
Highly salient forewarnings lead to a higher level of brand attitude. But the direction of this effect is unexpected. As was hypothesized earlier, highly salient forewarnings lead to lower levels of brand attitude. As was stated in previous research, forewarnings lead to more resistance and reduced persuasion. This means that people feel pressured into buying something and this should have led to a lower levels of brand attitude (Boerman et al., 2012; Liu, Li, & Mizerski, 2012; Boerman et al., 2013). This research contradicts these previous studies resulting in a higher brand attitude for highly salient forewarnings. This result indicates that different processes and elements occur and matter concerning the attitude towards advertorials (Dahlèn & Edinius, 2007). It could also mean that for advertorials, different elements are important such as honesty and trustworthiness concerning the intention of the sender leading to a higher brand attitude. In this way it almost comparable with brand credibility. As was said earlier, being honest adds value to the brand since it shows a sincere character which is very important to people (Sweeny & Swait, 2008). Further research towards brand attitude and advertorials is needed in order to give a well-based answer.
Perceived persuasive intent in combination with brand responses
The results show no mediated effect of persuasive intent on the brand responses. In addition, in line with our expectations based on the PKM (Friestad & Wright, 1994), the results do indicate that highly salient forewarnings can lead to a higher perceived persuasive intent increasing the awareness of the persuasive attempt. Forewarnings activate persuasion
25
knowledge which causes a process of counter arguing and skepticism towards the brand (Janssen, Fennis, & Pruyn, 2010). This mediated effect, however, was not confirmed. This could be due to the fact that did not test the conceptual persuasion knowledge, referring to the cognitive dimension focusing on the source and the understanding of the recognition of the advertising (Boerman, Reijmersdal & Neijnens, 2012). This study only looked at the attitudinal persuasion knowledge focusing mainly on the consumer and their beliefs and attitudes. Not examining whether the consumer was aware and did understood the meaning of the forewarning, that it indicates a persuasive attempt, and that it was created by the brand. Therefore the process of counter arguing was not developed enough, leading to less significant results regarding the mediation effect (Dekker & Van Reijmersdal, 2010).
Brand familiarity and advertorials
Finally, the effect of saliency on perceived persuasive intent differs for different levels of brand familiarity. Highly salient forewarnings in combination with familiar brands lead to a higher level of perceived persuasive intent. Theoretically, as was explained earlier, the higher the perceived persuasive intent the quicker it leads to skepticism and disliking (Boerman et al., 2012). Familiar brands activate previous knowledge on a brand, for a familiar brand this means they immediately know it is a commercial sender with the goal of persuading them. For an unfamiliar brand they also know there is a commercial goal but with less previous
knowledge on the why and how, leading to less perceived persuasive intent. This study confirms the fact that a similar effect occurs, familiar brands lead to higher levels of perceived persuasive intent, when using advertorials and the consumer response towards it (Machleit & Wilson, 1988).
26 Managerial Implications
This study holds important implications for organizations and companies when deciding on the medium choice in combination with the type of forewarning. Unfamiliar brands are less likely to cause skepticism and disliking since there are no existing associations and beliefs yet. This means that the forewarning in an advertorial is less important to the viewer for an
unfamiliar brand (Machleit & Wilson, 1988). So in order to use advertorials as a form of advertisement it is better to do this with unfamiliar brands, because a forewarning for a familiar brand leads to more perceived persuasive intent (Boerman et al., 2013).
This study also indicates that a less salient forewarning can lead to higher brand credibility and higher ad attitude. Brands could use this information when creating an advertorial and decide which element is important for them as a brand, ad attitude or brand credibility. For current legislation this research provides information indicating that size does matter. It rules out the fact that the size of the forewarning is significantly important but a highly salient forewarning leads to higher perceived persuasive intent acknowledging the fact that the consumer is aware of the persuasive attempt. This results in a fair and honest evalua-tion of the advertorial
Limitations and further research
This study shows an insight in the world of types of forewarnings. As this study is an important step into a difficult area, we also need to be cautious when generalizing the results because of the following limitations.
First, the context in which this study tested the effects needs to be taken into account. The results come from stimuli with this sponsored content, these brands, for these consumers and these forewarnings. More research needs to be performed in order the effects of different
27
types of forewarnings, with different features, in other native marketing such as advertorials. This could also lead to more significant results (Boerman et al., 2014).
Second, This study did not check whether the place of the forewarning interferes with the results. As Dekker and Van Reijmersdal (2010) state that naming the forewarning at the end leads to a more positive ad attitude and higher level of brand credibility. This research could have included the place of the forewarning as an element in order to gain more detailed information on the terms and conditions of forewarnings in advertorials.
Third, this study only included sponsored content in which the brand plays a very minimal role, the editorial content plays a key role in the advertorials. Further research is needed to study the effects of forewarning for other types of sponsored content, for instance traditional brand placements (Boerman et al., 2014).
Notwithstanding the limitations, this research shows that highly salient forewarnings can lead to higher levels of brand credibility and lower levels of ad attitude. It is also important to recognize that unfamiliar brands lead to a lower perceived persuasive intent. Which means their persuasion knowledge gets activated less. This prevents people from being skeptical towards and annoyed by the advertisement. Theoretically, the findings of this research provide new insights in the use of forewarnings in advertorials and the perceived persuasive intent. The findings show that this is a first step towards looking at the terms and conditions of forewarnings instead of the presence or absence of forewarnings. With noticing the absence of significant results, but the expected directions, the results also emphasize that more research is needed in order to give legislators useful practical implications. Therefore, based on the outcomes of this study, extended research towards the terms and conditions of forewarnings is needed.
28
References
About the FTC (2013)Retrieved February 12, 2014 from: https://www.ftc.gov/about-ftc
AdEx Benchmark (2012). AdEx Benchamark 2012 reveals outstanding year for online
advertising as it surpasses newspaper advertising spend for the first time. Retrieved
February 10, 2014 from: https://iabeurope.eu.
Alwitt, L. F., & Mitchell, A. A. (1985). Psychological processes and advertising effects: theory, research, and applications. Journal of Advertising, 15(3), 56-67 doi: 10.1080/00913367.1986.10673019.
d’Astous, A., & Chartier, F. (2000). A study of factors affecting consumer evaluations and memory of product placements in movies. Journal of Current Issues and Research in
Advertising, 22(2), 31–40. doi: 10.1080/10641734.2000.10505106.
Baron, R.M., & Kenny, D.A., (1986). The Moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. Journal
of Personality and Social Psychology, 51(6), 1173-1182. doi: 10.1037/0022-
3514.51.6.1173
Baker, J., Hutchinson, W., Moore, D. & Nedungadi, P. (1986). Brand Familiarity and Advertising: Effects on the Evoked Set and Brand Preference", Advances in
Consumer Research, 13, 637-642.
Boerman, S.C., Reijmersdal, E.A., and Neijnens, P.C., (2012). Sponsorship Disclosure: Effects of duration on persuasion knowledge and brand responses. Journal of
Communication 62, 1047-1064. doi: 10.1111/j.1460-2466.2012.01677.x.
Boerman, S.C., Reijmersdal, E.A., and Neijnens, P.C., (2014). Effects of
sponsorship disclosure timing on the processing of sponsored content: A study on the effectiveness of European disclosure regulations. Journal of Psychology &
29
Marketing, 31(3), 214–224. doi: 10.1002/mar.20688
Bong-Hyun, K., (1995). The Effectiveness of Camouflaged Advertising format on audiences: with Special Focus on Advertorial Advertising. University of Alabama, Department of Mass Communication, Alabama.
Cameron, G.T., & Ju-Pak, K.H., (2000). Information Pollution? Labeling and Format of Advertorials. Newspaper Research Journal, 21(1) 65-76.
Chattopadhyay, A., & Nedungadi, P. (1992). Does Attitude Towards the Ad Endure? The Moderating Effects of Attention and Delay. Journal of Consumer Research, 19, 26-33 Dahlèn, M. (2005). The Medium as a contextual cue. Effects of Creative Media Choice. Journal of Advertising, 34 (3), 89-98. doi: 10.1086//209283.
Dahlèn, M., & Edinius, M., (2007). When is advertising advertising? Comparing responses to non-traditional and traditional advertising media. Journal of Current Issues &
Research in Advertising, 29(1) 33-45. doi: 10.1080/10641734.2007.10505206.
Dahlèn, M., Friberg, L., & Nilsson, E. (2009). Long live creative media choice. The medium as a persistent brand cue. Journal of Advertising, 38(2), 121-129. doi:
10.2753/JOA0091-3367380208
De Nederlandse Reclame Code (2009). Retrieved February 10, 2014 from: https://www.reclamecode.nl/nrc/
Dekker, K., & Van Reijmersdal, E.A. (2010). Waarschuwingen, beroemdheden en brand placement: De effecten van type waarschuwing en geloofwaardigheid op
kijkerreacties. Disclosures, Celebrities and brand placement: The effects of disclosure type and credibility on viewer responses. Tijdschrift voor Communicatiewetenschap, 38(4), 320-337. doi: 10.1080
30 Interactive Marketing, 17(4), 8-23. doi: 10.1002/dir.10063.
Erdem, T., & Swait, J., (2004). Brand Credibility, Brand Consideration, and Choice. Journal
of Consumer Research, 31(1), 191-198. doi: 10.1086/383434.
Erdem, T., Swait, J., & Louviere, J. (2002). The impact of brand credibility on consumer price sensitivity. International Journal of Research in Marketing, (19)1, 1-19. doi:
10.1016/S0167-8116(01)00048-9.
Foodrepublic (2010). Retrieved February 10, 2014 from http://www.foodrepublic.com. Friestad, M., & Wright, P. (1994). The Persuasion Knowledge Model: How People Cope With Persuasion Attempts. Journal of Consumer Research, 21, 1–31. doi:
10.1086/209380
John, D. (1999). Consumer Socialization of Children: A Retrospective Look at
Twenty-Five Years of Research. Journal of consumer Research, 26, 183-213. doi: 10.1086/209559
Janssen, L., Fennis, B. M., & Pruyn, A. T. H. (2010). Forewarned is Forearmed: Conserving Self-Control Strength to Resist Influence. Journal of Experimental Social Psychology, 46, 911-921. Doi: 10.1016/j.jesp.2010.06.008
Kiesler, C.A., & Kiesler, S.B. (1964). Role of Forewarning in Persuasive Communications.
The Journal of Abnormal and Social Psychology, 68(5), 574-549. doi:
10.1037/h0042145.
Leon, D.T., Rotunda, R.J., Sutton, M.A., & Schlossman, C. (2003). Internet forewarning effects on ratings of attraction. Computers in Human Behavior, 19(1), 39-57. doi: 10.1016/S0747-5632(02)00017-1.
Liu, F., Li, J., Mizersiki, D., Huangting, S. (2012). Self-congruity, brand attitude,
31
922-937. doi: 10.1108/03090561211230098.
Machleit, K.A. & Wilson, R.D. (1988). Emotional Feelings and Attitude toward the
Advertisement: The Roles of Brand Familiarity and Repetition. Journal of Advertising, 17, (3) 27-35. doi: 10.1080/00913367.1988.10673121.
McGuire, W.J., & Papageorgis, D., (1962). Effectiveness of forewarning in developing resistance to persuasion. Public Opinion Quarterly 26(), 24-34. doi: 10.1086/267068. Shen & Chen (2007) Contextual Priming and Applicability: Implications for Ad Attitude And Brand Evaluations. Journal of Advertising, 36(1), 69-80. doi: 10.2753/JOA0091- 3367360105
Spry, A., Pappu, R., Cornwell, B. (2012). Celebrity endorsement, brand credibility and brand equity. European Journal of Marketing 45(6), 882-909. doi:
10.1108/03090561111119958
Sweeney, J., & Swait, J., (2008). The effects of brand credibility on customer loyalty. Journal
of Retailing and Consumer Services, 15(3), 179-193. doi:
10.1016/j.jretconser.2007.04.001,
Wei, M., Fischer, E., & Main, K.J. (2008). An examination of the effects of activating persuasion knowledge on consumer response to brands engaging in covert marketing.
Journal of Public Policy & Marketing, 27(1), 34-44. doi: 10.1509/jppm.27.1.34.
Wood, W., & Quin, J.M., (2003). Forewarned and forearmed? Two meta-Analytic syntheses of forewarnings of influence appeals. Psychological Bulletin, 129(1), 119-138. doi: 10.1037//0033-2909.129.1.119.
Zhou, S. (2012) ‘Advertorials’: A Genre-Based Analysis of an Emerging Hybridized Genre.
32
Appendices
Appendix A. Characteristics of total respondents A1. Gender
33
A3. Employement
34
Appendix B. Characteristics per group
Mean SD N
Gender Unfamiliar less salient 1.30 .47 34
Unfamiliar Highly Salient 1.21 .42 34
Familiar less salient 1.29 .46 33
Familiar Highly salient 1.24 .44 31
Age Unfamiliar less salient 28.58 11.42 34
Unfamiliar Highly Salient 32.45 12.80 34
Familiar less salient 29.32 10.90 33
Familiar Highly salient 32.52 14.80 31
Employment Unfamiliar less salient 2.97 1.53 34
Unfamiliar Highly Salient 2.45 1.66 34
Familiar less salient 2.87 1.61 33
Familiar Highly salient 3.06 1.50 31
Education Unfamiliar less salient 12.94 1.41 34
Unfamiliar Highly Salient 13.09 .98 34
Familiar less salient 12.97 1.28 33
35
Appendix C. Stimuli
C1. Condition 1 Highly salient forewarning – Familiar brand.
36
C3. Condition 3 Less salient forewarning – Familiar brand.
37 Appendix D. Q-Q Plots. D1. Ad attitude D2. Brand attitude D3. Brand credibility
38
Appendix E. Bivariate correlation analyses control variables on dependent variables
Brand attitude Brand credibility Ad attitude
Gender Pearson Correlation .06 .10 -.05
Sig. (2-tailed) .480 .236 .553
N 131 131 131
Age Pearson Correlation .08 .03 .02
Sig. (2-tailed) .365 .741 .794
N 130 130 130
Employment Pearson Correlation -.02 .09 .16
Sig. (2-tailed) .797 .305 .062
N 131 131 131
Education Pearson Correlation -.07 .03 .13
Sig. (2-tailed) .412 .750 .144
N 131 131 131
Ketchupliking Pearson Correlation -.05 -.01 .09
Sig. (2-tailed) .599 .890 .336
N 131 131 131
Ketchupuse Pearson Correlation .05 .03 -.04
Sig. (2-tailed) .577 .699 .672
N 131 131 131
Website use Pearson Correlation -.05 .02 .08
Sig. (2-tailed) .558 .850 .376
N 131 131 131
Websitefamiliarity Pearson Correlation -.15 -.08 .06
Sig. (2-tailed) .085 .363 .528
N 131 131 131
**. Correlation is significant at the 0.01 level (2-tailed) *. Correlation is significant at the 0.05 level (2-tailed)
39
Appendix F. Bivariate correlation analyses all control variables
Gender age Employment Education
Gender Pearson Correlation 1 .180* .12 -.06
Sig. (2-tailed) .04 .17 .54
N 131 130 131 131
Age Pearson Correlation .18* 1 -.15 -.18*
Sig. (2-tailed) 0.4 .09 .05
N 130 130 130 130
Employment Pearson Correlation .12 -.15 1 -.06
Sig. (2-tailed) .17 .09 .51
N 131 130 131 131
Education Pearson Correlation -.06 -.78* -.06 1
Sig. (2-tailed) .54 .05 .51
N 131 130 131 131
Ketchup use Pearson Correlation -.07 .16 -.14 -.04
Sig. (2-tailed) .43 .07 .12 .67
N 131 130 131 131
Ketchup liking Pearson Correlation .01 -.19* .01 .08
Sig. (2-tailed) .89 .03 .96 .39
N 131 130 131 131
Website use Pearson Correlation -.22* -.10 .01 .22*
Sig. (2-tailed) .01 .24 .95 .01
N 131 130 131 131
Website familiarity Pearson Correlation -.16 -.42 .22* -.01
Sig. (2-tailed) .06 .63 .01 .88
N 131 130 131 131
**. Correlation is significant at the 0.01 level (2-tailed) *. Correlation is significant at the 0.05 level (2-tailed)
