In Blogs We Trust
The effect of disclosure presence on the product attitude, purchase intention and attitude towards the blog, mediated by perceived credibility and moderated by interest
in the subject.
Master Thesis Laurien Visschedijk
10003648
Persuasive Communication Graduate School of Communication
University of Amsterdam June 2016
Supervised by Dr. L. M. Willemsen
Abstract
This study examines under what circumstances disclosures have a negative effect on consumers’ attitudes towards blogs, a promoted product and purchase intention, and whether perceived credibility is mediating these effects. An online experiment (N = 105) showed that a blog was less likely to instigate positive consumer responses when it contained a disclosure (vs. no disclosure). As predicted, these effects were mediated by perceived credibility. Finally, interest in the topic of the blog was found to moderate the indirect effects of disclosures on consumer responses. For consumers who were interest in the topic of the blog, there was a negative indirect effect on the consumer responses. However, for consumers who were less interested in the topic of the blog, no significant results were found. These findings may offer an explanation for the inconsistent findings on the effects of disclosures as reported in the literature.
Key words: Disclosure, blogs, perceived credibility, interest in the subject, attitude, purchase
Introduction
In less than ten years, blogs have become a popular means of communication (Kerr, Mortimer, Dickinson & Waller, 2012; Nielsen, 2012). This is reflected by their ubiquity as well as their diversity. The number of blogs worldwide has increased from 36 million in 2006 to 181 million in 2011 (Nielsen, 2012), and can be found on any product and or subject such as food, beauty, lifestyle, technology, etcetera. What they have in common is that they allow people to share their opinions, reflections and or experiences on a topic of interest, in the form of an online personal journal, which is frequently updated, and arranged in reversed chronological order (Chu & Kamal, 2013).
People do not only like publishing blogs, they like reading blogs as well. According to Zhu and Tan (2007), reading blogs is one of the most popular time spending activities on the Internet. This is often attributed by their perceived credibility. Blogs are seen as a trustworthy source of information because in general, bloggers do not have any promotional intentions for posting their opinions and experiences. Given their perceived credibility, bloggers are often seen as important influencers (Belch & Belch, 2015). Indeed, research shows that consumers who read about the positive experience of a blogger with a product or service, are more likely to have a positive attitude towards the product and an increase in the purchase intention (Hsu et al., 2013; Wu & Wang, 2011).
Blogs have not remained unnoticed amongst marketers. To capitalize on the popularity and perceived credibility of blogs, marketers are increasingly sponsoring blog posts (Zhu & Tan, 2007). It is not uncommon for marketers to provides bloggers a monetary reward, free products or another incentive in exchange for a positive blog about their brand and or products. For example, in exchange for a reward, food bloggers write positive blogs about certain kitchen products, fashion bloggers recommend certain web shops and travel bloggers share their positive experiences with a travel destination.
Since 2014 the Dutch Reclamecode Social Media (RSM) urges bloggers to reveal the commercial nature of promoted blogs (Reclamecode, n.d.). More specifically, bloggers who receive an incentive to promote a particular product or service are asked to state that the content of the blog has been sponsored by a company. This is often done by means of a disclosure: a warning sign alerting consumers that the message contains sponsored content and thus is unbiased in nature (Boerman, Van Reijmersdal & Neijens, 2015). The reader can be warned by means of a textual message on the top of a blog, for example with the tag ‘sponsored content’ or by means of a specific hashtag such as #ads or #promo (Watson, 2013). By showing such tags, the RSM aims to prevent consumers from being deceived. When it is not clear for the reader that the message has persuasive intentions, consumers are not able to guard themselves against undesired persuasion which is considered as unethical (Sprague & Wells, 2010).
Until now, the RSM only offers guidelines regarding the use of disclosures. This means that it is not obligatory to disclose the promotional nature of sponsored blogs, but bloggers may decide for themselves whether they want to adhere to the guidelines or not (Reclamecode, n.d.). Hence, some bloggers are not making use of disclosures as they fear that this will deteriorate its effectiveness (Sprague & Wells, 2010). When readers discover that a blogger is recommending a product, not because of a genuine liking for that product but because of commercial motivations (i.e., an incentive from the company), this could impair the credibility of the source (Abendroth & Heyman, 2013).
Other bloggers want to be honest with their followers and are upfront about the promotional nature of some blog entrees. These bloggers feel that transparency about product promotions may lead to a preservation or increase, rather than a decline in credibility, especially when the blog is valued by its readers. This would explain why certain blog posts seem unaffected by the use of disclosures. An example is the website Beautylab, which
attracts more than 750.000 unique visitors each month. The website publishes blogs on beauty and lifestyle, of which some are promotional in nature and marked as such by means of the tag ‘sponsored’. As demonstrated by readers’ reactions towards a promoted blog on Christmas outfits in which the retailer SaleOnly.nl was featured and promoted (Beautylab, 2015), people may not care if a blog is sponsored or not, as long as it offers interesting content. One of the readers even reacted very positively: “I did not know this web shop but I will definitely visit it!”.
The aim of this study is threefold: (1) gain more insight into the effects of disclosures on the product attitude, purchase intention, and the attitude towards the blog, (2) examine the mediating role of perceived credibility, and (3) examine the moderating role of blog interest on the effects of disclosures. Studying the moderating role of blog interest seems imperative as it may determine the effectiveness of sponsorship disclosures. Blogs are particularly valued for their credibility, and hence, disclosures may in general have a negative effect the sincerity of the blog and therefore on the credibility. However, this may not be the case for consumers who are highly interested in the topic of the blog. As a result of their intrinsic liking for the topic of the blog, they might not care if the content of the message is sponsored or not. Thus, people who are interested in the topic of discussion may react differently towards a disclosure than people who are less interested in the topic.
By addressing these aims, this study contributes to the literature on electronic word-of-mouth (eWOM) marketing, as it provides more insight into the conditional effects of disclosures, and how they affect the persuasiveness of sponsored blog posts. So far, little research has focused on the influence of the characteristics of the reader. Gaining insight into the moderating role of sender characteristics such as blog interest does justice to the characteristics of blogs. A blog can be seen as a pull medium as the receiver actively selects the message he or she wants to read (Moorman, Neijens & Smit, 2011). Because of this, it is
likely that readers have at least some interest in the topic. Hence, it is important to investigate whether those readers will react differently to the persuasive attempt than readers who are not very interested in the topic.
This study has implications for marketers and policy makers as well. From the marketers’ perspective it is paramount to understand the persuasive effects of disclosures. As stated earlier, it is not obligatory for companies to use disclosures. If the current research shows that the use of a disclosure has negative effects on the product attitude, purchase intentions and attitude towards the blog, both bloggers and marketers might need to reconsider the use of blogs to promote products and services. From the policy maker’s perspective, this study provides more insight into the potential of disclosures to protect consumers from unfair persuasion. It could help them decide whether they keep the guidelines like they are now or whether they need to change them to actual legislation.
Theoretical framework The Credibility of Blogs
When blogs were introduced in the early days of the 21st century, they were heralded as a credible source of information. Credibility refers to the acceptance of the message as a valid source of information and consists of two dimensions: perceived expertise and perceived trustworthiness (Ohanian, 1990). While perceived expertise refers to the degree to which a source is believed to be knowledgeable enough to make valid assertions, perceived trustworthiness refers to the degree to which a source is believed to be motivated enough to make valid assertions. Based on these grounds, bloggers were seen as credible sources as they derive from consumers who have gained knowledge on a topic based on the first-experience (expertise), that is shared because of a genuine interest in the topic at hand (trust).
In these early days, communication research indeed found blogs to be a credible source of information (Johnson & Kaye, 2004; Lu, Chang & Chang, 2014; Schmallegger & Carson, 2008). According to Johnson and Kaye (2004), weblog users judged blogs as more credible compared to the traditional sources as radio, television, and the newspaper due to their independency. Furthermore, consumers tend to trust a message more when it is written by fellow consumers (Lu et al., 2014). Because blogs are often written by consumers instead of corporations, blogs are perceived as credible. As for travel blogs, it often contains information which is hard to assess prior to the purchase or consumption (Schmallegger & Carson, 2008). Therefore, people who are searching for information about a destination rely more on other consumers because of their objectivity.
Nowadays, the credibility of blogs is heavily debated due to the increase of the use of blogs by companies (Mack, Blose & Pan, 2008). Other companies invite bloggers to write a blog post for them in exchange for a monetary reward, free products or discounts (Verlegh, Ryu, Tuk & Feick, 2013). Bloggers are seen as important social influencers. People care about the opinions of peers (Hsu et al., 2013), especially because peers are perceived as more credible than marketers. Thus when a peer shares a positive evaluation about a certain product and this evaluation is believed to be true, this is likely to influence the individual sin in a way that it is beneficial for the company that is selling this product (Hsu et al., 2013).
As increasingly more blogs are used to promote products, it is relevant to examine whether, and if so, under which circumstances blogs are still regarded as a credible source of information. The question seems imperative, as credibility is a key factor in driving persuasion. A large body of literature shows that in order to change the opinions and behaviors of consumers, a message has to be perceived as believable (Chu & Kamal, 2008; Cosenza, Solomon & Kwon, 2015; Kaye, 2010; Wu & Wang, 2011). This is also confirmed by research on blogging.
Chu and Kamal (2008) found that consumers develop positive feelings when they are exposed to a blog which they perceive as credible. Those positive feelings will, in its turn, result in a positive attitude towards the blog. Similar findings have been reported on product attitudes. When a consumer believes the message, the reader is more likely to adopt the same attitude as expressed in the blog post (Wu & Wang, 2011). Likewise, Van Hemelen Smits and Verlegh (2013) demonstrated that if the consumer perceives the message as (less) credible, this would have a (negative) positive effect on product attitude. Finally, previous research also found a positive link between message credibility and purchase intentions. When the message is perceived as credible, it is more likely that the reader of the message will buy the promoted product because the reader will take the recommendation of a credible blogger into account during decision making (Hsu et al., 2013).
The Effect of Disclosures on Perceived Credibility and Brand Responses
The main goal of a disclosure is to inform the audience that the content of the message has commercial intentions. According to the RSM it is not ethical to make an attempt to persuade consumers without informing them about the commercial nature of a message (Reclamecode, n.d.). When consumers are not aware of the commercial nature of a message, consumers might be persuaded into buying a product or service which they otherwise might avoid. Hence, the RSM urges marketers and bloggers to be honest about the commercial nature of sponsored blog posts so that consumers can guard themselves against unfair persuasion.
Although disclosures are meant to decrease the persuasive impact of the messages, previous research on the effects of disclosures has found mixed effects. While some studies report negative effects of disclosures on consumer responses (Boerman, Van Reijmersdal & Neijens, 2012; Carr & Hayes, 2014; Wojdynski & Evans, 2015; Zhu & Tan, 2007), others
report positive effects (Abendroth & Heyman, 2013; Lu et al., 2014), or no effects at all (Liljander, Gummerus & Söderlund, 2014; Lu et al., 2014; Shiv, Edell & Payne, 1997).
The negative effects of disclosures on perceived credibility are explained based on the Persiasion Knowledge Model (PKM) and the cognitive dissonance theory. According to the PKM of Friestad and Wright (1994), people develop knowledge about persuasion techniques and use this knowledge in their reaction to the persuasion attempts. When people see a disclosure, the persuasion knowledge is more likely to be activated which could influence how they react to the message. Boerman et al. (2012) make a distinction between conceptual persuasion knowledge and attitudinal persuasion knowledge. Conceptual persuasion knowledge concerns the understanding of commercial intentions and attitudinal persuasion knowledge entails the presence of critical feelings which is similar to low levels of perceived credibility. In line with PKM, disclosures are successful in stimulating more critical feelings towards promoted messages. The audience is more skeptical towards the content of the message as reflected by low scores of credibility when they are exposed to disclosures (Boerman et al., 2012; Zhu & Tan, 2007). In a similar vein, Wojdysnki and Evans (2015) found that when consumers were exposed to a disclosure and recognized the message as being commercial, they assessed the message as less credible then when the audience did not perceive the message as being commercial.
Carr and Hayes (2014) found a negative effect of disclosures on perceived credibility as well. These effect were explained by cognitive dissonance theory. Cognitive dissonance refers to the psychological tension that people are likely to experience when new information is challenging existing beliefs (Festinger, 1962). Thus, when a disclosure informs consumers about the commercial nature of a blog, this might challenge initial perceptions about the blog post as a true and sincere recommendation. In line with these assumptions, Carr and Hayes (2014) found that, as a result of disclosure exposure, consumers perceive a blog as less
credible and more biased which results in the feeling they have that the blog is misleading in nature.
In contrast to previous research discussed above, other studies found no effects at all. Amongst these studies belongs the research of Shiv et al. (1997) who found no significant effects. They claimed that this was due to the way consumers process a message separately from a disclosure. Although a disclosure, by itself, was not appreciated by consumers, it did not have an impact on consumer responses. In a similar vein, Lu et al. (2014) concluded that even when consumers are aware of the promotional intentions of the blog, they are not affected by it as most of them already know that the content is sponsored. Furthermore, Liljander et al. (2014) found no significant effect of disclosure presence on the perceived credibility and mixed effects on consumer responses. This might be due to the knowledge consumers have about disclosures and filter the information. Moreover, when a company decided to sponsor a blog post, this is a sign of popularity and a good reputation of the blog (Liljander et al., 2014). They investigated the effect using an existing blog, this might have influenced the results as the participants may already had an attitude towards the blog.
These findings show that disclosures may not always hurt the credibility of the messages. Interestingly, there are also studies showing a positive effect of disclosures on perceived credibility. According to Abendroth and Heyman (2013) it is in the advantage of bloggers to be honest about the commercial nature of sponsored blog posts. They found that a disclosure had a positive effect on attitude towards the brand and purchase intention in comparison with no disclosure. Thus, when the audience was informed that the message contained sponsored content, they were more inclined to have positive feelings towards the brand and have stronger intentions to buy the promoted product. When a company decided to advertise the product by using a consumer to persuade other consumers, those other consumers think that they company stands behind the product and they have faith in the
product (Abendroth & Heyman, 2013). However, they investigated the effect between two peers who already knew each other. The results might differ in a context where the blogger and the reader are not familiar with each other. Lu et al. (2014) reported similar effects, although they were explained by a different mechanism. In their study, the authors found that when the audience is informed that the content was sponsored, they are less skeptical because they know they are not being misled. In this situation, consumers were more sensitive for persuasion. Attitudes were formed in line with the positive message and, as a result, persuasion was still found to occur (Lu et al., 2014).
Because the blogger and the reader of the blog are not familiar with each other, it is not likely that a disclosure will have a positive effect on the consumer responses. Furthermore, the readers do not have any prior knowledge of the blog and therefore will not base their opinion on this knowledge but will solely base it on the blog post there were exposed to.
H1: A disclosure (no disclosure) has a negative (positive) effect on (a) product attitude, (b) purchase intentions, and (c) blog attitude.
H2: H1a-H1c is mediated by perceived credibility.
Interest in the subject.
Previous research found contradictory results regarding the effect of disclosures on perceived credibility, product attitude, purchase intention and blog attitude. This may be due to the consumers’ level of interest towards the subject. Interest in the subject refers to a positive orientation towards the subject which is valued by the consumer (Ainley, 2012). Research demonstrates that subject interest might influence the way the consumer processes the given information and consequently their reactions towards this information (cf. Laurent & Kapferer, 1985).
Armstrong and McAdams (2011) found that the blog users’ intention predicts to what extent he or she trust the blog. Moreover, they concluded that people who read blogs because they are interested in getting more information, evaluate blog differently and trust them more than people who read blogs for entertainment reason. Namely, people who read blogs for information reasons are especially interested in the accuracy and the arguments which support or reject their current opinion. People who read blog for entertainment reasons want to be amused and care less whether the content is credible or not (Armstrong & McAdams, 2011). Based on these results, it can be stated that people who are interested in the subject perceives the message as more credible than people who are not interested when a disclosure indicated that the message is sponsored. Furthermore, Ballantine and Yeung (2015) found that when consumers are highly involved with the subject or the brand (a concept closely related to interest, see Laurent & Kapferer, 1985), it is more likely that they will constitute more positive feelings towards the content of the blog, irrespective of whether it is sponsored or not. They stated that this is the case because highly involved consumers are more interested in the product and or the brand itself (Ballantine & Yeung, 2015). In a similar vein, Celsi and Olson (1988) found that the more a consumer is interested in the subject and perceives the source as personally relevant, the more likely it is that he or she will devote attention to information about the product and be persuaded by it.
Based on these studies, it is hypothesized that interest in the subject serves as a moderator of the effect of disclosure on perceived credibility, such that when consumers are exposed to a disclosure but highly interested in the subject, the negative effect will decay. This because highly interested consumers are positively oriented towards the message itself and may pay not care whether the content of the blog is sponsored.
H3: The negative indirect effect of the presence of a disclosure on (a) product attitude, (b) purchase intentions and (c) blog attitude through credibility will deteriorate when consumers
are highly interested in the subject of the blog than when they are less interested in the subject.
The conceptual model can be found in figure 1.
Figure 1. Conceptual model
Method Design and Sample
To test the hypotheses, a 2 (disclosure vs. no disclosure) x 2 (low vs. high interest in the subject) factorial between-subjects experimental design was used. In total, 105 respondents participated in the online experiment. The sample existed of mostly females (68.6%, n = 72), aged between 19 to 80 years old (M = 33.88, SD = 15.79), who had enjoyed vocational education or higher (83.8% HBO or higher, n = 88)
Stimuli
All the participants were exposed to a fictional travel blog “World of Wonders”. In the blog, the blogger shared a post about his/her positive experiences with a yoga retreat in Morocco and explained the program (see appendix). This subject was chosen because both men and women may vary in the degree to which they are interested in yoga. Furthermore, travel blogs are often sponsored by for example a travel agency or by a local institution. Additionally,
abundance of Instagram pictures showing how both men and women who are engaged in this activity (De Wever, 2013). The blogs were similar in storyline and layout across conditions, except for the manipulation of disclosure type.
To manipulate disclosure type, half of the participants were exposed to a disclosure stating the post was “Sponsored” (n = 56). The other half were exposed to the same blog without a disclosure (n = 49). The disclosure was clearly shown on top of the blog post, right below the title (see appendix). The moderator, interest in the subject, is a quasi-experimental variable. Hence, this variable was not manipulated, but measured instead (Novaco, Kliewer & Broquet, 1991).
To test whether the manipulation was successful, a pre-test was conducted amongst 22 participants (n = 22), aged 18 years to 58 years (M = 27.23, SD = 9.63). The participants were exposed to either the blog with or without a disclosure. After exposure participants indicated, on a scale from 1 to 7, to what extent they agreed with three statements: “The blog is sponsored”, “The blog has commercial intentions” and “The blogger received a reward to write the blog” (Carr & Hayes, 2014). These three items were found to form a reliable, one-dimensional scale (Cronbach’s = .96, EV = 2.78, R2 = .93). Hence, these three items were averaged to form a scale. As intended, participants who were exposed to the blog with a disclosure attributed more promotional intentions to that blog (M = 5.97, SD = .85) than participants who were exposed to the blog that showed no disclosure (M = 2.14, SD = 1.36), t (20) = 7.71, p < .001, CI [2.80, 4.86]. Thus, the manipulation was successful.
Procedure
Participants were invited via email and social media to take part in an experiment on traveling. They were additionally asked to share the invitation with others, using a snowballing technique. No reward was provided for participation. When participants accepted
the invitation, and clicked on the added link, they were forwarded to the online experiment. The experiment was conducted online as it mimics the natural environment in which blogs are shown. After giving informed consent, participants were randomly exposed to the stimulus material, which depicted a fictitious travel blog, either with or without the disclosure. Exposure was logged to ascertain how much time was spent on reading the blog. After this, participants had to answer a series of question which tapped: the dependent variables, perceived message credibility and subject interest. The last variable was asked through three questions. Of the 105 participants, 50 were not interested in the subject (n = 50) and the other 55 were interested in the subject (n = 55). Also some questions were asked in order to control some confounding variables. Finally, the participants had to answer some question about their demographic characteristics.
Measures
Product attitude. Product attitude stands for the overall evaluation of the participant of the accommodation: a yoga retreat in Marocco (cf. Lee, Park & Han, 2008). In line with previous research, attitude was measured through six bipolar adjectives on a 7-point scale. These adjectives were bad – good, dislike - like, and unpleasant – pleasant, negative – positive, unattractive – attractive, and hospitable – not hospitable (Heide & Grønhaug, 2009; Mitchell, 1986). These six items were found to form a reliable, one-dimensional scale (Cronbach’s = .93, EV = 4.53, R2 = .76). On average, participants had a neutral to positive attitude towards the product (M = 4.85, SD = 1.26).
Purchase Intention. Because the blog post contains the experience of the blogger of a yoga retreat, purchase intention will from now on will be referred to booking intention. Booking intention is defined as the likelihood that the potential visitor will book the trip (Tsao, Hsieh, Shih & Lin, 2015), in this case the intention to book the yoga retreat. The
intention to book thid trip is measured by using a 3-item scale, ranging from very 1 (not at all) to 7 (very much). The questions were: “If I went on a holiday to Morocco, it is probable that I will book this yoga retreat”, “If I went on a holiday to Morocco, it is likely that I will book this yoga retreat”, and “If I went on a holiday to Morocco I am willing to book this yoga retreat” (Tsao et al., 2015). These three items were found to form a reliable, one-dimensional scale (Cronbach’s = .96, EV = 2.79, R2 = .93). On average, participants had a negative to neutral intention to book the yoga retreat (M = 3.51, SD = 1.94).
Blog attitude. Blog attitude stands for the overall evaluation of the participants of the blog the World of Wonders. This attitude was measured by means of the same bipolar adjectives as used for product attitude, apart from the adjective hospitable – not hospitable. These five items were found to form a reliable, one-dimensional scale (Cronbach’s = .94, EV = 4.07, R2 = .81). On average, participants had a neutral to positive attitude towards the blog (M = 4.57, SD = 1.35).
Perceived message credibility. The perceived message credibility is the extent to which the participant believes the message (Park, Wang, Yao & Kang, 2011). This was measured through three bipolar adjectives on a 7-point scale: untrustworthy – trustworthy, not genuine – genuine, and unbelievable – believable (Park et al., 2011). These three items were found to form a reliable, one-dimensional scale (Cronbach’s = .94, EV = 2.67, R2 = .89). On average, participants perceived the message as neutral to credible (M = 4.81, SD = 1.63).
Subject interest. Subject interest is a quasi-experimental variable, such that it was measured rather than manipulated. Subject interest refers to the extent to which the participant is interested in the topic of the blog. This variable was measured through three statements where the participants had to indicate to which extent they agreed with the following: “I am interested in yoga”, “I like yoga” and “I like to read about yoga” (Boerman et al., 2012). These three items were found to form a reliable, one-dimensional scale (Cronbach’s = .95,
EV = 2.72, R2 = .91). On average, participants had a neutral interest in yoga (M = 4.01, SD = 1.90). Participants who scored below 4 on the scale were recoded as less interested and participants who scored above 4 on the scale were recoded as highly interested.
Confounding variables. In order to control for the influence of confounding variables, some extra questions were added to the questionnaire. The participants were asked how often they post blogs themselves and how often they read blogs per month. Also they were asked whether they have ever visited Morocco and whether they have the intention of doing this (again) in the future. Finally, some questions were asked regarding the demographic characteristics, namely their gender, age, and education level.
Manipulation check. To test whether the participants noticed the manipulation, the presence of the disclosure, the same three-item scale was used as in the pre-test. The three items were found to form a reliable, one-dimensional scale (Cronbach’s = .97, EV = 2.84, R2 = .95).
Results Manipulation check
In order to check whether the manipulation was successful, an independent t-test was conducted. The participants who were exposed to the blog post with a disclosure thought that the blog contained more promotional intentions (M = 5.76, SD = 1.48) than the participants who were exposed to the blog post without a disclosure (M = 2.97, SD = 1.89). This difference is significant t (90.64) = 8.34, p < .001, CI [2.13, 3.46]. Hence, the manipulation was successful.
In order to exclude alternate explanations, Pearson correlation analyses were performed. Table 1 presents an overview of the results. It shows that product attitude positively relates with gender (r = .19, p = .048). The dependent variable booking intention, positively correlates with gender (r = .22, p = .028), the intention to visit Morocco (r = .31, p = .001) and the degree of reading blogs (r = .30, p = .002) and negatively correlates with age (r = -.36, p = 0.002). As for the variable blog attitude, it positively correlates with the frequency of reading blogs (r = 0.27, p = 0.005). Thus, these variables were entered as covariates in subsequent analyses.
Product attitude Booking intention Blog attitude Perceived credibility Gender .19* .22* .22 .07 Age -.14 -.36** -.02 .03 Education .04 -.11 -.01 .02 Visiting intention .18 .31** .08 .11 Writing blogs -.03 -.06 .02 .03 Reading blogs .18 .30** .27** .13 Travel History -.06 -.11 -.07 -.10 Note: N = 105
* Correlation is significant at 0.05 level ** Correlation is significant at 0.01 level
Table 1. Bivariate correlations matrix
The Effect of Disclosures
H1a predicted that a disclosure (no disclosure) had a negative (positive) effect on the product attitude. To test H1a, an ANCOVA was conducted with presence of a disclosure as independent variable, product attitude as the dependent variable and gender as the covariate. The results yielded a significant, mediocre effect for disclosure presence F (1, 101) = 16.86, p
positive attitude towards the product (M = 4.37, SD = 1.14) than the participants who were exposed to the blog without a disclosure (M = 5.39, SD = 1.18). This means that H1a is accepted.
H1b predicted that a disclosure (no disclosure) had a negative (positive) effect on the booking intention. To test H1b, an ANCOVA was conducted with disclosure type as an independent variable, booking intention as a dependent variable and gender, age, intention to visit Morocco and blog reading frequency as covariates. A significant, small effect was found for the effect of the presence of a disclosure on the booking intention when controlled for gender, age, intention to visit Morocco and the blog reading frequency, F (1, 98) = 4.63, p = .034, 2 = .03. Participants who were exposed to the blog with a disclosure were less inclined to book the retreat (M = 3.05, SD = 1.61) than the participants who were exposed to the blog without a disclosure (M = 4.03, SD = 2.16). Thus, H1b is accepted.
H1c predicted that the presence of a disclosure (no disclosure) had a negative (positive) influence on attitude towards the blog. To test H1c, an ANCOVA was conducted with disclosure presence as an independent variable, attitude towards the blog as dependent variable and blog reading frequency as a covariate. The results showed a significant, mediocre to strong effect of the presence of a disclosure on the attitude towards the blog when controlled for the frequency of blog reading, F (1, 102) = 24.05, p < .001, 2 = .18. Participants who were exposed to the blog with a disclosure had a less positive attitude towards the blog (M = 4.03, SD = 1.22) than the participants who were exposed to the blog without a disclosure (M = 5.18, SD = 1.25). Thus, H1c is accepted.
The Mediating Role of Credibility
To test whether perceived credibility mediated the relation between the presence or absence of a disclosure (from now referred to as disclosure) and (H2a) product attitude, (H2b) booking
intention and (H2c) blog attitude, a regression analysis was conducted. According to Baron and Kenny (1986), the effect between the mediator and the dependent variable should be significant as well as the effect between the independent and the dependent variable, furthermore the significant relationship between the independent variable and the dependent variable should disappear when it is controlled for the mediator. To test whether this was the case multiple linear regressions were conducted.
H2a predicted that the relationship between a disclosure and product attitude is mediated by perceived message credibility. With regard to a disclosure and the product attitude under the influence of the perceived credibility, the regression model was found to be significant, F (3, 102) = 24.99, p < .001, R2 = .43. A disclosure is a significant predictor of the perceived credibility, b = 1.64, t = 5.92, p < .001 and for the product attitude, b = 1.02, t = 4.54, p < 001. The perceived credibility is also found to be a significant predictor of the product attitude, b = .48, t = 8.10, p < .001. The absence of a disclosure resulted in more credibility which in turn, resulted in more positive attitudes towards the product However, the relationship between the presence of a disclosure and product attitude controlled for the influence of perceived credibility, is not significant, b = .32, t = 1.43, p =.155. This supports the presence of full mediation. Indeed, a Sobel’s test shows that mediation was significant. Sobel’s Z = 4.26, p < .001. Therefore, H2a is being accepted. Figure 2a provides an overview of the results.
H2b predicted that the relationship between a disclosure and booking intention is mediated by perceived message credibility. Regarding a disclosure and the booking intention under the influence of the perceived credibility, the regression model was found to be significantly predict the data, F (6, 98) = 12.72, p < .001, R2 = .44. A disclosure is a significant predictor for the perceived credibility, b = 1.64, t = 5.92, p < .001 and for booking intention, b = .92, t = 2.76, p < .001. Perceived credibility is also a significant predictor of the booking intention, b = .58, t = 6.30, p < .001. The absence of a disclosure resulted in more credibility, which in
turn, resulted in a higher booking intention. However, the relationship between the presence of a disclosure and booking intention becomes non-significant when controlled for the influence of perceived credibility, b = -.04, t = -.09, p = .928. These results support the full mediation. This was also confirmed by a Sobel test z = 4.00, p < .001. Thus, H2b is being accepted. Figure 2b provides an overview of the results.
H2c predicted that the relationship between a disclosure and attitude towards the blog is mediated by perceived message credibility. The regression model was found to be significant,
F (3, 101) = 33.04, p < .001, R2 = .50. A disclosure is a significant predictor for both the perceived credibility, b = 1.64, t = 5.92, p < .001 and the blog attitude, b = 1.13, t = 4.90, p < .001. The perceived credibility is also a significant predictor of the blog attitude, b = .53, t = 8.97, p < .001. The absence of a disclosure resulted in more credibility, which in turn, resulted in more positive attitudes towards the blog. The relationship between the presence of a disclosure and blog attitude becomes non-significant when controlled for the influence of perceived credibility, b = .35, t = 1.58, p = .117. These results support the presence of full mediation. Indeed, a Sobel’s test shows that mediation was significant. Sobel’s Z = 4.52, p < .001. Thus, H2c is being accepted. Figure 2c provides an overview of the results.
Figure 2b. Mediation-analysis H2b with b-values
Figure 2c. Mediation-analysis H2c with b-values
The Moderating Role of Blog Interest
H3 predicted that the negative indirect effect of the presence of a disclosure on (a) product attitude, (b) booking intention and (c) attitude towards the blog through credibility will deteriorate when consumers are highly interested in the subject of the blog than when they are less interested in the subject. In order to test this hypothesis, a mediated moderation analysis using the PROCESS tool was conducted. The indirect effect is contingent upon interest, such that there is only a significant negative indirect effect when interest is high. This applies to the product attitude, b = .99, boot SE = .17, 95% CI [1.38, .68], the booking intention, b = -1.35, boot SE = .32, 95% CI [-2.05, -.76], and the attitude towards the blog, b = -1.11, boot SE = .20, 95% CI [-1.52, -.75]. Therefore, H3 is being rejected.
The aim of is this study is threefold: (1) gain more insight into the effects of disclosures on the product attitude, purchase intention, and the attitude towards the blog, (2) examine the mediating role of perceived credibility, and (3) examine the moderating role of blog interest on the perceived credibility. This was measured via an online experiment where half of the participants were exposed to a blog post with a disclosure and the other half to a blog post without a disclosure. The blog post was about the positive experiences of a yoga retreat in Morocco through the eyes of the blogger.
As predicted, the presence of a disclosure had a significant negative effect on the product attitude, booking intention, and attitude towards the blog. Participants who saw the blog with a disclosure had less favorable attitudes towards the product and the blog and were less inclined to book the yoga retreat. However, participants who were exposed to the disclosure condition, still had a positive attitude towards the product and the blog, it just was not as positive as the participants in the no disclosure condition. This could be the case because the message has been perceived as less credible and therefore participants were more skeptical towards the message but because of the fact that bloggers are seen as important influencers and consumers care about what other peers say and think, they still take their opinion into account during attitude formation (Belch & Belch, 2015; Litvin, Goldsmith & Pan, 2008).
The role of the perceived credibility is also being tested via a mediation analysis. It was predicted that participants perceived the message as more credible when they were exposed to the blog post without the disclosure than when they were exposed to the blog post with the disclosure. Furthermore, it was predicted that when the message was perceived as credible, this would have a positive influence on the product attitude, booking intention and attitude towards the blog. These hypotheses were also confirmed. It was indeed the case that participants in the disclosure condition perceived the message as less credible and had less
favorable feelings afterwards. This result is in line with some of the previous research. Consumers often visit blogs to obtain objective information from an independent source (Johnson & Kaye, 2004). Because of their independence, blogs are seen as credible sources of information which makes them more credible (Cosenza et al., 2015). A blog post without a disclosure could provoke less skeptical feelings towards the message which will result in a bigger chance that the consumers will adopt the bloggers’ opinions. However, it is important to keep in mind that the attitudes also can be based on the message itself and not on potential commercial intentions (Lu et al., 2014).
The final aim of this study was to test whether the interest in the subject played a moderating role in the relationship between the presence of a disclosure and the perceived message credibility. The results showed a negative indirect effect of disclosure presence and high interest in the topic of the blog on the perceived credibility and consumer responses. This is the opposite of what was expected. It could be explained by the Motivation, Opportunity and Ability (MOA) theory. Motivation is defined as the readiness and desire to process the given information, this is similar to the interest in the subject because it determines how motivated someone is in processing the message (Celsi & Olson, 1988; Gruen, Osomonbekov & Czaplewski, 2007). When the consumer is motivated to read the message, it is more likely that he or she will notice the disclosure and will take this disclosure into account during the attitude formation.
Previous research found mixed results regarding the effect of disclosures. This study contributes in a way that it gives more insight to the circumstances in which consumers react positively or negatively towards disclosures. A disclosure does not have to lead towards negative consumer responses, although consumers react significantly different towards the blog with and without the disclosure, the results were still positive. Furthermore, it depends
on different factors such as the level of interest the consumer has with the topic of the message.
Second, as mentioned before, blogging has becoming more and more popular throughout the years. Not only the amount of blogs increase but also the amount of blog readers. Because the blogosphere is a quick growing phenomenon it is important to keep investigating its effectiveness. Results can differ in a period of time because it is possible the readers will develop more knowledge about blogs and their intentions.
Consumers who visit a blog are often interested in the subject. It is unlikely that someone read a blog about travelling when he or she does not have the intention to travel nor care about the depicted destination. Results show that consumers will perceive the message as less credible when they are highly interested in the subject and exposed to a blog that contains a disclosure. Marketers and bloggers can decide if they want to change the place of the disclosure or make it less striking, hoping that they will not notice the disclosure or give it less attention. Furthermore, the first hypotheses showed that there is not a big difference in attitudes and behavioral intentions. Although the message has been sponsored, the attitudes towards the product and the blog still remain positive. Thus, when the RSM decided it is obligatory to warn consumers for the commercial intentions, a blog is still a good platform to use as a promotional platform.
Although this study offers insights into the effects of disclosures under different circumstances, it also comes with limitations. Interest in the topic of the blog is difficult to investigate. Underlying mechanisms could explain the difference between the highly interested participants and the less interested participants. For example, attention, to test this mechanism the online experiment is not the best research method, by using eye tracking it is easier to ascertain whether highly interested consumers pay more attention to the disclosure than less interested consumers.
Future research can focus on comparable platforms and test whether the found effects also applies to for example video blogs (vlogs) and tutorials or for sponsored messages on social media sites such as Facebook. Vlogs are different from blogs on many levels. There are both audio and visual aspects where consumers can base their opinion not only on the content of the message but also on the facial expressions of the person in the video. According to the cognitive theory of multimedia learning by Mayer and Moreno (2003) it is better to stimulate two senses, visual and verbal, because the message will be processed better. Hence, due to the way of processing, the results can differ from the results which are found in this study towards blogs.
In this study a fictional travel blog was used as stimulus. Previous research distinguished two types of goods; search goods and experience goods. The yoga retreat can be seen as an experience good because it is hard to evaluate the product prior to the purchase. It could be the case that consumer will react differently towards search goods, which are goods that are easy to evaluate before the actual purchase (Weathers, Sharma & Wood, 2007). Future research can investigate if consumers react differently to a sponsored blog post about a search good than to a sponsored blog post about an experience good.
Literature
Abendroth, L. J., & Heyman, J. E. (2013). Honesty is the best policy: The effects of
disclosure in word-of-mouth marketing. Journal of Marketing Communications, 19(4), 245-257.
Ainley, M (2012). Students’ interest and engagement in classroom activities. In S. Christenson, A. Reschly & C. Wylie (Eds.), Handbook of Research on Student
Engagement (pp. 283-302). New York: Springer.
Armstrong, C. L., & McAdams, M. J. (2011). Blogging the time away? Young adults’ motivations for blog use. Atlantic Journal of Communication, 19(2), 113-128. Ballantine, P. W., & Yeung, C. A. (2015). The effects of review valence in organic versus
sponsored blog sites on perceived credibility, brand attitudes, and behavioural intentions. Marketing Intelligence & Planning, 33(4), 508-521.
Baron, R., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychology research: conceptual, strategic, and statistical consideration. Journal of
Personality and Social Psychology, 51(6), 1173-1182.
Beautylab (2015). Kerst outfit inspiratie. Retrieved from http://www.beautylab.nl/kerst-outfit-inspiratie-win-e1000-belastingvrij/
Belch, G. E., & Belch, M. A. (2015). Advertising and promotion: An integrated marketing
communication perspective (10th ed.) Columbus, OH: McGraw Hill Education. Boerman, S. C., Van Reijmersdal, E. A., & Neijens, P. C. (2012). Sponsorship disclosure:
Effects of duration on persuasion knowledge and brand responses. Journal of
Communication, 62, 1047-1064.
Boerman, S. C., Van Reijmersdal, E. A., & Neijens, P. C. (2015). Using eye tracking to understand the effects of brand placement disclosure types in television programs.
Journal of Advertising, 44(3), 196-207.
Carr, C. T., & Hayes, R. T. (2014). The effect of disclosure of third-party influence on an opinion leader’s credibility and electronic word-of-mouth in two-step flow. Journal of
Interactive Advertising, 14(1), 38-50.
Celsi, R. L., & Olson, J. C. (1988). The role of involvement in attention and comprehension processes. Journal of Consumer Research, 15(2), 210-224.
Chu, S., & Kamal, S (2013). The effect of perceived blogger credibility and argument quality on message elaboration and brand attitudes. Journal of Interactive Advertising, 8(2), 26-37.
Cosenza, T. R., Solomon, M. R., & Kwon, W. (2015). Credibility in the blogosphere: A study of measurement and influence of wine blogs as an information source. Journal of
Consumer Behaviour, 14, 71-91.
De Wever, R. (2013, December 17). Je yoga-poses op Instagram zetten, dan heb je het niet begrepen. Trouw. Retrieved from
http://www.trouw.nl/tr/nl/5091/Religie/article/detail/3563744/2013/12/17/Je-yoga-poses-op-Instagram-zetten-dan-heb-je-het-niet-begrepen.dhtml
Friestad, M., & Wright, P. (1994). The persuasion knowledge model: How people cope with persuasion attempts. Journal of Consumer Research, 21(1), 1-31.
Gruen, T. W., Osmonbekov, T., & Czaplewski, A. J. (2007). Customer-to-customer exchange: Its MOA antecedents and its impact on value creation and loyalty. Journal of the
Academy of Marketing Science, 35(4), 537-549.
Heide, M., & Grønhaug, K. (2009). Key factors in guests’ perception of hotel atmosphere.
Cornell Hospitality Quarterly, 50(1), 29-43.
Hsu, C., Lin, J. C., & Chiang, H. (2013). The effect of blogger recommendations on customers’ online shopping intentions. Internet Research, 23(1), 69-88.
Johnson, T. J., & Kaye, B. K. (2004). Wag the blog: How reliance on traditional media and the internet influence credibility perceptions of weblogs among blog users. Journalism
& Mass Communication Quarterly, 81(3), 622-624.
Kaye, B. K. (2010). Going to the blogs: Towards the development of a uses and gratifications measurement scale for blogs. Atlantic Journal of Communication, 18(4), 194-210. Kerr, G., Mortimer, K., Dickinson, S., & Waller, D. S. (2012). Buy, boycott or blog:
Exploring online consumer power to share, discuss and distribute controversial advertising messages. European Journal of Marketing, 46(3), 387-405.
Laurent, G., & Kapferer, J. (1985). Measuring consumer involvement profiles. Journal of
Marketing Research, 22(1), 41-53.
Lee, J., Park, D., & Han, I. (2008). The effect of negative online consumer reviews on product attitude: An information processing view. Electronic Commerce Research and
Applications, 7, 341-352.
Liljander, V., Gummerus, J., & Söderlund, M. (2014). Young consumers’ responses to suspected covert and overt blog marketing. Internet Research, 25(4), 610-632.
Litvin, S. P., Goldsmith, R. E., & Pan, B. (2008). Electronic word-of-mouth in hospitality and tourism management. Tourism Management, 29, 458-468.
Lu, L., Chang, W., & Chang, H. (2014). Consumer attitude toward blogger’s sponsored recommendations and purchase intentions: The effect of sponsorship type, product type, and brand awareness. Computers in Human Behavior, 34, 258-266.
Mack, R. W., Blose, J. E., & Pan, B. (2008). Believe it or not: Credibility of blogs in tourism.
Journal of Vacation Marketing, 14(2), 133-144.
Mayer, R. E., & Moreno, R. (2003). Nine ways to reduce cognitive load in multimedia learning. Educational Psychologist, 38(1), 43-52.
Mitchell, A. A. (1986). The effect of verbal and visual components of advertisements on brand attitudes and attitude toward the advertisement. Journal of Consumer Research,
13(1), 12-24.
Moorman, M., Neijens, P. C., & Smit, E. G. (2011). Advertising media selection and planning. In: J. Sheth & N. Malhotra (eds.) Wiley Internation Encyclopedia of
Marketing, 4: Advertising and Integrated Communication, 22-29
Nielsen (2012). Buzz in the blogosphere millions more bloggers and blog readers. Retrieved from http://www.nielsen.com/us/en/insights/news/2012/buzz-in-the-blogosphere- millions-more-bloggers-and-blog-readers.html.
Novaco, R. W., Kliewer, W., & Broquet, A. (1991). Home environmental consequences of commute travel impedance. American Journal of Community Psychology, 19(6), 881-909.
Ohanian, R. (1990). Construction and validation of a scale to measure celebrity endorsers’ perceived expertise, trustworthiness, and attractiveness. Journal of Advertising, 19(3), 39-52.
Park, C., Wang, Y., Yao, Y., & Kang, Y. R. (2011). Factors influencing eWOM effects: Using experience, credibility, and susceptibility. International Journal of Social
Science and Humanity, 1(1), 74-79.
Reclamecode (n.d.). Reclamecode social media. Retrieved from
https://www.reclamecode.nl/nrc/pagina.asp?paginaID=289%20&deel=2 Schmallegger, D., & Carson, D. (2008). Blogs in tourism: Changing approaches to
information exchange. Journal of Vacation Marketing, 14(2), 99-110.
Shiv, B., Edell, J. A., & Payne, J. W. (1997). Factors affecting the impact of negatively and positively framed ad messages. Journal of Consumer Research, 24(3), 285-294. Sprague, R., & Wells, M. E. (2010). Regulating online buzz marketing: Untangling a web of
deceit. American Business Law Journal, 47(3), 415-454.
Tsao, W., Hsieh, M., Shih, L., & Lin, T. M. Y. (2015). Compliance with eWOM: The influence of hotel reviews on booking intention from the perspective of consumer conformity. International Journal of Hospitality Management, 46, 99-111.
Van Hemelen, N., Smits, T., & Verlegh, P. C. (2013). Geloofwaardigheid van e-WOM: De invloed van reviewpercepties of de persuasieve impact van online reviews. Tijdschrift
voor de Communicatiewetenschap, 41(4), 332-345.
Verlegh, P. W. J., Ryu, G., Tuk, M. A., & Feick, L. (2013). Receiver responses to rewarded referrals: the motive inferences framework. Journal of the Academy of Marketing
Science, 41(6), 669-682.
Weathers, D., Sharma, S., & Wood, S. L. (2007). Effects of online communication practices on consumer perceptions of performance uncertainty for search and experience goods.
Journal of Retailing, 83(4), 393-401.
Watson, C. (2013). Best practices: Blogger disclosure examples and tips. Retrieved from http://bloggylaw.com/blogger-disclosure-examples-tips/
Wojdynski, B. W., & Evans, N. J. (2015). Going native: Effects of disclosure position and language on the recognition and evaluation of online native advertising. Journal of
Advertising, 0(0), 1-12.
Wu, P. C. S., & Wang, Y. (2011). The influences of electronic word-of-mouth message appeal and message source credibility on brand attitude. Journal of Marketing and
Logistics, 23(4), 448-472.
Zhu, J. Y., & Tan, B. C. Y. (2007). Effectiveness of blog advertising: Impact of
communicator expertise, advertising intent, and product involvement. Twenty Eighth
International Conference on Information Systems. Montreal, Canada.
Appendix Stimuli blog post without disclosure
