• No results found

THE STIMULATING EFFECT OF MARKETER-GENERATED CONTENT ON SOCIAL MEDIA:

N/A
N/A
Protected

Academic year: 2021

Share "THE STIMULATING EFFECT OF MARKETER-GENERATED CONTENT ON SOCIAL MEDIA:"

Copied!
28
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

THE STIMULATING EFFECT OF

MARKETER-GENERATED CONTENT

ON SOCIAL MEDIA:

A DETAILED MESSAGE-LEVEL ANALYSIS

Master thesis, MSc, specialization Marketing Management University of Groningen, Faculty of Economics and Business

January 14, 2019

MAUREEN STUULEN

Student number: 3194868

Zuidbargerstraat 82 7812 AJ Emmen, The Netherlands

(2)

THE STIMULATING EFFECT OF

MARKETER-GENERATED CONTENT

ON SOCIAL MEDIA:

A DETAILED MESSAGE-LEVEL ANALYSIS

ABSTRACT

(3)

1.

INTRODUCTION

Marketing in the retail landscape has changed dramatically over the past two decades. Technological innovations and developments, such as the global penetration of the Internet, have given marketers the ability to reach customers through new touch points. These technological changes have caused social media to become a key aspect in shopper marketing (Lamberton & Stephen, 2016). Social media helps firms to engage with customers via online channels, and has the ability to create synergistic effects (Chang et al., 2015). It allows for interaction between customer and firm, via platforms such as Facebook, Instagram, Twitter, and YouTube. Due to its scales, it has the potential of sharing content amongst many users in only a short period of time. Many firms want to use this feature to its advantage, which causes a large growth in the worldwide social media marketing spending. The total spending on social media advertising has almost doubled from $16 billion in 2014 to $31 billion in 2016, and is still expected to grow towards approximately 48 billion U.S. dollars in 2021 (The Statistics Portal, 2017). Social media marketing is used for a variety of marketing objectives, including branding, customer relationship management, research, service, and sales promotions (Ashley & Tuten, 2015). Because of its enormous potential and innovativeness, social media marketing is widely studied in the past couple of years. Because of social media its interactive character, existing literature on social media marketing makes a distinction between user-generated content (UGC) and marketer-generated content (MGC). MGC has become a dominating social media marketing activity for firms, since it is relatively controllable compared to UGC (Ding et al., 2014). It refers to firm-initiated marketing communications via its official social media pages (Kumar et al., 2015). Despite the extensive use of social media and the increasing number of firms who use MGC, companies have not reached a clear understanding yet of how MGC affects its marketing performance (Wan & Ren, 2017). Whereas previous research has proven MGC to have a positive and significant effect on consumer’s behaviour (i.e. Kumar et al., 2015), the knowledge is only limited to broad concepts. Current knowledge simply lacks the details which would make the findings more practical and easier applicable for marketers. Comparing this existing knowledge on social media marketing to the knowledge on traditional marketing, leads us to the conclusion that traditional marketing knowledge is much deeper and developed. This is simply due to the fact that it exists for a longer period of time and is therefore more studied over the years.

(4)

A message is persuasive when the author is intended to change the reader’s behaviour or attitude (O’Keefe, 1990). However, persuasiveness is a perception of the consumer and therefore depends on the extent to which the consumer taste is shifted by a marketing communication (Ackerberg, 2001). The perception of whether a message is perceived as persuasive or not depends on functional matching of the arguments (Petty & Wegener, 1998). This means that the quality of arguments in a persuasive message depends on the attitudes of the individual. If the message arguments do not meet the personal traits of the recipient, it directly affects the validity perception of the message. Lavine & Snyder (1996) state that this message validity perception influences the post-message attitudes of the recipient. In practice, this clear distinction between informative and persuasive content is also likely to be applicable for social media marketing. Kumar et al. (2015), who study the influence of MGC in social media on key customer metrics, ask for future research that makes a distinction between the two types of messages. To our humble knowledge, the differential effects of these two types of content have not been investigated yet for social media MGC. The outcomes of the study should therefore lead to new academic knowledge. This knowledge can serve as the foundation for the creation of more practical insights for marketing managers, who want to optimize their returns on MGC. Therefore, we ask ourselves: Is there a significant difference between the effects of informative versus persuasive MGC? The study of Chang et al. (2015) is one of the few papers that goes into detail for the type of content. The researchers focus on persuasive MGC, and how it affects the behavioural intention of the consumer. More specifically, the study looks at how the different elements of persuasive content affect the consumer’s attitudes and beliefs, which in turn have a relationship to the consumer’s behavioural intentions. The aim of the current study is to find out whether there is a significant difference in consumer behaviour for informative and persuasive firm-generated messages. We therefore replicate the study of Chang et al. (2015), but adapts it to the situation in which respondents get exposed to informative content instead of persuasive content. The results of this study will be compared to the outcomes of Chang et al. (2015) in order to tell whether there is a difference in the behavioural intention of the consumer for informative and persuasive MGC.

(5)

2.

THEORETICAL FRAMEWORK

We look at the relationship between informative messages, the matching attitudes and beliefs, and the effect it has on the behavioural intention of consumers. Goal is to compare to existing literature on persuasive content (Chang et al., 2015), in order to tell whether we should make a distinction between the two types of messages in MGC. In order to make a reliable comparison, variables will be adopted from the persuasive research model.

2.1 Conceptual Model

A conceptual framework has been developed in order to make a graphical representation of the relationships as hypothesized in section 2.3-2.5. An important difference between the model of Chang et al. and the model as projected below, is the absence of the moderation effects. Chang et al. (2015) included the moderating role of user expertise and relative significance in their study. Based on the scarcity of resources for the current study, moderation effects have been left out from our research model.

Figure 1 Conceptual Model.

2.2 Central concepts

On the first level of the model, Chang et al. (2015) use three elements for persuasive messages; argument quality, post popularity, and post attractiveness. As this study focusses on informative messages instead of persuasive messages, the argument quality variable is not directly applicable. In order to keep this variable as closely related to the objective of this study and comparable to the variable of Chang et al. (2015), the variable is adjusted to information quality. With the term information quality, we refer to the accuracy, adequacy and credibility of information exchanged (Li & Lin, 2006).

(6)

Last, the term post attractiveness is used to describe the extent to which consumers perceive the post as admirable and appealing (Ahearne, Gruen, & Jarvis, 1999).

For the second layer of the model we look at the attitudes and beliefs of the consumer. With over 2 billion social media users worldwide (Statista, 2018), one of the main challenges for businesses is to stand out in the crowd, and stepping on the forefront of the consumer’s newsfeed. According to Chang et al. (2015) Facebook users mainly read posts because of two factors: usefulness and preference. Usefulness can be defined as the utility perception of the decision maker attributed to the information (Larcker & Lessig, 1980). Preferences are favourable feelings and interests for a specific product or service (Kim & Son, 2009).

Finally, we want to tell something about the behavioural intention of the consumer. Chang et al. (2015) look at the like and share intention as indicators of the consumer’s diffusion behaviour. The intention to like can be described as the human’s desire to show affiliation with a message by hitting the like button (Gerlitz & Helmond, 2012). The like function is designed as a shortcut to responding, in order to replace short and common comments on messages. Existing literature explains the share intention as the desire to click on the share button, in order to distribute a message along social contacts to invoke further social activities (Chen & Lee, 2013).

2.3 Informative messages

Information quality is introduced as the first element that determines the evaluation of the informative post. The current online landscape is associated with a high level of ambiguity. Credible information is therefore suggested to be of great importance for locating and relying on brands (Flanagin et al., 2014). The information as provided by the firm or marketer should therefore help the consumer in their decision-making process. This high need for information leads us to hypothesize the following: H1: Information quality has a positive effect on usefulness.

(7)

For post attractiveness, the level of admiration and appeal are both determined from visual elements of the posts. The visual elements help in eliciting the brand attitude (Yoon & Park, 2012). Hence, attractiveness may affect the individual’s preferences (Verhagen et al., 2012). This knowledge leads the us to hypothesize:

H3: Post attractiveness positively affects preference.

2.4 Beliefs & Attitude

Within the second layer of the model, we look at the evaluation of the informative post based on attitudes and beliefs. However, these attitudes and beliefs can also influence each other according to psychology (Feldman & Lynch, 1988). Whenever a post is perceived as useful, it is supposed to be favourable for the recipient. Therefore, we hypothesize:

H4: Usefulness positively influences the preference.

In relation to the third layer of the model, where the behavioural intention is being accessed, Bhattacherjee & Premkumar (2004) found evidence for a positive relationship between perceived usefulness and user intentions. More specifically, the Innovation diffusion theory explains how perceived usefulness regulates the diffusion intention of the recipients (Rogers, 1995). Whereas like and share behaviour can be perceived as the diffusion of a message, we hypothesize:

H5a: Usefulness positively affects the like intention. H5b: Usefulness positively affects the share intention.

From the consumer’s perspective, social media’s share button enables the reader to take content from across the Web and share it with Facebook friends, so the content can be re-shared and the most interesting posts get noticed by as many people as possible. However, this like and share behaviour only occurs when recipients have an interest in posts, and feel the desire for further engagement (Gerlitz & Helmond, 2013). Since personal preference is (amongst other factors) build upon the human’s cognitive influences (McClure et al., 2004), and therefore related to the individual’s personal interests, we hypothesize a positive connection between preference and the like and share intention:

H6a: Preference has a positive effect on the like intention. H6b: Preference has a positive effect on the share intention.

2.5 Behavioural Intention

(8)

towards the post (De Vries et al., 2012), and therefore it is most likely that liking a post increases the likelihood of a recipient sharing the content.

H7: Like intention positively influences the share intention.

3.

RESEARCH DESIGN

3.1 Data gathering

(9)

3.2 Use of measures

There are seven concepts employed in the survey: ‘information quality’, ‘post popularity’, ‘post attractiveness’, ‘usefulness’, ‘preference’, ‘like intention’, and ‘share intention’. In order to make these constructs measurable but also comparable, items were adopted from Chang et al. (2015). All items will be measured on a 5-point Likert scale ranging from strongly disagree to strongly agree, similar to the existing data for persuasive messages.

Information Quality

Even though Chang et al. (2015) use the construct ‘argument quality’ instead of ‘information quality’, the items used are universal for both the constructs. The items of Bhattacherjee & Sanford (2006) measure the perceived informativeness, helpfulness and value of the post in order to say something about the perceived quality. These items can be used for both measuring argument quality and information quality, as Bhattacherjee & Sanford measure quality as a broader concept.

Post popularity

For testing the respondent’s perceptions towards post popularity, items of He et al. (2009) are adopted. These researchers use the concepts of trustworthiness, reliability, and believability as tools to gain better knowledge about the respondent’s thoughts and linked perceptions to popular items.

Post attractiveness

Based on the study of Verhagen et al. (2012), post attractiveness is measured with the help of two items: attractiveness and the perceived appeal. Since the level of post attractiveness is fully subjective, it is important to provide the right questions and items in order for the recipient to explain its perception. Usefulness

Chang et al. (2015) adopt their items for measuring the usefulness from a study from Lu et al. (2008). Lu et al. look at the perceived usefulness of wireless mobile data services. In order to measure this variable, they look at three items: reduced time perception, increased quality perception, and the overall usefulness. These items can be adjusted to the context of our study, in which we want to find the perceived usefulness of the informative post.

Preference

(10)

Like Intention

The intention to like is a rather straightforward variable. However, it remains important to measure this concept by using various items. Chang et al. (2015) base their questioning on a study from Yi et al. (2006), and employ the following items: intention to press like and the anticipation of pressing like. Share Intention

Last, for measuring the share intention of the recipient, three items have been selected in order to tell something about the construct. These items are based upon a study from Lee & Ma (2012), who investigate the news sharing behaviour of the individual on social media. They use the following items for their survey questions: intention to share, expectation of sharing, and plan of sharing. These items are translated to this study’s context, in order to tell us something about the recipient’s share intention of informative MGC on social media.

3.3 Use of statistical techniques

Since the conceptual model is formed as a path model, with variables connected to each other as cause-effect relationships, it is most interesting to use a structural equation modelling approach. There are multiple reasons to choose for PLS-SEM in this case: 1) When the research purpose is to predict or explore, PLS is characterized as the most suitable technique in an early stage of theoretical development (Henseler, Ringle, & Sinkovics, 2009); 2) The approach has the ability to model multiple dependent and multiple independent variables (Garson, 2016); and 3) This approach looks further than just performing simple regression analyses, because it takes the latent constructs into account when predicting the indicators (Hair et al., 2011).

(11)

4.

RESULTS

4.1 Measurement model

The PLS-SEM model assessment is a two-step process, that first looks at the measurement model followed by the judgment of the structural model. In the measurement model composite reliability is tested as a measure of the constructs’ internal consistency. For the evaluation of the reflective measurement model, the validity assessment focusses on two elements; convergent validity and discriminant validity. The convergent validity is evaluated based on three criteria: 1) Factor loadings should exceed 0.55 (Comrey & Lee, 1992); 2) Cronbach’s Alpha should exceed 0.7 (Christmann & Van Aelst, 2006); and 3) the Average Variance Extracted (AVE) should exceed 0.5 (Fornell & Larcker, 1981). The results of the construct items can be found in table 1. The table shows that all the constructs conform the criteria, and we can conclude that the selected items are able to create reliable constructs.

Table 1 Adjusted convergent validity (after deletion of IQ_1)

CONSTRUCTS / ITEMS FL CA AVE

Information Quality (based on Bhattacherjee & Sanford, 2006) 0.874 0.781

IQ_2: This post is helpful to me. 0.934

IQ_3: This post is valuable to me. 0.831

Post Popularity (based on He, Qiao, & Wei, 2009) 0.918 0.796 PP_1: I think that posts with more people showing like & share behavior are trustworthy. 0.770

PP_2: I think that posts with more people showing like & share behavior are reliable. 0.837 PP_3: I think that posts with more people showing like & share behavior are believable. 1.046

Post Attractiveness (based on Verhagen et al., 2012) 0.878 0.788

PA_1: I perceive the post as attractive. 0.941

PA_2: I perceive the post as appealing. 0.832

Usefulness (based on Lu, Liu, Yu, & Wang, 2008) 0.826 0.618 US_1: This post can reduce the time spent on making my decision. 0.775

US_2: This post can increase the quality of my decision making. 0.658

US_3: Overall, I find this post useful. 0.922

Preference (based on Hsu & Lin, 2008) 0.942 0.844 PR_1: I feel pleasant when being informed about the newest mobile phones. 0.860

PR_2: I feel good when learning about the newest mobile phones. 0.970 PR_3: I like tob e informed about the newest mobile phones. 0.923

Like Intention (based on Yi, Jackson, Park, & Probst, 2006) 0.894 0.811

LI_1: I intend to press like on such a post. 0.929

LI_2: I anticipate that I will press like on such a post. 0.871

Share Intention (based on Lee & Ma, 2012) 0.973 0.924 SI_1: I intend to share such a post on my timeline. 0.989

SI_2: I expect to share such a post on my timeline. 0.931 SI_3: I plan to share such a post on my timeline. 0.963

(12)

Table 2 Adjusted discriminant validity (after deletion of IQ_1)

IQ PP PA US PR LI SI

Information Quality (IQ) 0.884

Post Popularity (PP) 0.332 0.892

Post Attractiveness (PA) 0.441 0.339 0.888

Usefulness (US) 0.719 0.476 0.552 0.786

Preference (PR) 0.586 0.455 0.540 0.496 0.919

Like Intention (LI) 0.570 0.352 0.354 0.502 0.583 0.901

Share Intention (SI) 0.503 0.227 0.101 0.404 0.488 0.703 0.961

After the adjustment of the items for information quality, the model shows good validity; the new measurement model fits for both convergent and discriminant validity. Now that the measurement model is qualified as good, we can move on to the structural model.

4.2 Structural model

The number of model fit indexes suitable for the judgement of the structural model is limited (Hair et al., 2017), because many fit measures imply restrictive assumptions on the residual covariances, which does not apply for PLS-SEM (Lohmöller, 1989). Therefore, we evaluate the overall fit of the model based on the indexes SRMR and NFI. The estimated model is used to obtain the covariance matrix. The Standardized Root Mean Square Residual (SRMR) has a cut-off value close to 0.08 (Hu & Bentler, 1999). The SRSM value of 0.081 is close the cut-off value of 0.08, and therefore adapts to the benchmark. The Normed Fit Index (NFI) value for the model is 0.820, which is below the minimum criterium of 0.90 which was set by Hair et al. (1998). However, Forza & Filippini (1998) suggest a NFI value greater than 0.8 for the model to be considered good. Based on this criterium, a value of 0.820 can be considered a good fit.

Another aspect in judging the quality of the data analysis is the R2 value. Figure 2 shows the estimated

model, with its R2 values and path coefficients. The predictive power of this model can be judged as

relatively weak, with the R2 value of both Preference and Like Intention being below the cut-off value

of 0.5 (Moore et al., 2013). With R2 values between 0.3 and 0.5, Moore et al. (2013) speak of weak or

low effect size. However, the judgement of a high or low R2 value depends on the research discipline.

When doing research in consumer behaviour, a R2 value of 0.2 can already be evaluated as a high R2

value (Hair et al., 2011). Since this study focusses on behaviour, we accept that the R2 values for the

two variables are below 0.5.

(13)

Figure 2 SmartPLS results for the structural model. k

Table 3 Results for the structural model.

H RELATIONSHIPS b T-STAT P-VALUE SUP/REJ

H1 Information Quality -> Usefulness 0.631 6.561 0.000 Supported H2a Post Popularity -> Usefulness 0.266 3.282 0.001 Supported H2b Post Popularity -> Preference 0.249 2.404 0.016 Supported H3 Post Attractiveness -> Preference 0.355 3.356 0.001 Supported

H4 Usefulness -> Preference 0.182 1.540 0.124 Rejected

H5a Usefulness -> Like Intention 0.281 2.534 0.011 Supported H5b Usefulness -> Share Intention 0.040 0.405 0.686 Rejected H6a Preference -> Like Intention 0.444 4.077 0.000 Supported H6b Preference -> Share Intention 0.105 0.961 0.337 Rejected H7 Like Intention -> Share Intention 0.622 5.287 0.000 Supported

Second, usefulness has a significant effect on the like intention of the consumer (b=0.281, p=0.011), whereas the effect of usefulness on the share intention is not significant (b=0.040, p=0.686). Also, the T-value of usefulness on preference (b=0.182, p=0.124) is below the level of 1.96 (Hair et al., 2011), and thus not significant. Preference positively affects the like intention of the consumer (b=0.444, p=0.000), but the relationship between preference and share intention is not significant (b=0.105, p=0.337).

(14)

5.

DISCUSSION

In the last couple of years social media marketing has taken over the marketing landscape. Nonetheless existing knowledge of social media marketing is much less detailed compared to traditional marketing. In traditional marketing the type of message is proven to have a differential effect for the behavioural effectiveness of the message (e.g. Marquez, 1977; Abernethy & Franke, 1996; Goh et al., 2013). This effect was not yet studied for social media marketing (Kumar et al., 2015), and therefore an interesting gap in modern literature. Existing data on persuasive MGC, combined with our data on informative MGC allowed us to find an answer to the question whether there is a significant difference between the behavioural effects of informative vs. persuasive MGC.

5.1 Summary of findings

From the obtained data, we find that the like intention for informative messages is directly stimulated in a positive manner by the usefulness perception and preference of the consumer. The positive significant Beta’s for both relationships support our hypotheses H5a and H6a. These results are in line with the results of Chang et al. (2015), and consequently equal to the outcomes for persuasive messages. However, interesting is the detail that the effect of usefulness is larger than the effect of preference for persuasive messages, whereas our results indicate preference to have the largest effect on the like intention for informative messages. The fact that these effects have switched for the variables may have been caused in the first layer of the model. In the first layer we already see that usefulness is built differently compared to the results of Chang et al. (2015). Even though the information quality and post popularity both have proven to be positive and significant in relation to usefulness, the information quality has a much larger effect compared to post popularity. In contrary, in the study of Chang et al. (2015) post popularity has a larger effect than the argument quality. This seems logical, as the usefulness perception for both types of content is likely to be different. The perceived usefulness of informative content is focussed on the utility perception of the decision maker attributed to the information (Larcker & Lessig, 1980). On the contrary, for persuasive content this usefulness evaluation is more focussed on the perceived benefit in terms of performance (Bhattacherjee & Sanford, 2006). These existing definitions explain why we experienced strong correlation in the first phase of the data analysis, and consequently why the effect between information quality and usefulness is relatively strong.

(15)

persuasive content evaluation might explain why the popularity of the content has a larger effect on the usefulness evaluation, compared to informative content. Popularity is often used as an indicator of the social norm (Cheung & Lee, 2010), and is easily accepted for forming a superficial opinion.

The perceived post attractiveness is also proven to have a positive significant relationship to preference. This means that we can accept our H3 hypothesis. The effect of the post attractiveness on the preference is comparable for informative and persuasive messages.

In contrast with the findings above, we did not find evidence that the share intention is positively affected by the usefulness perception and the preference of the consumers. Our lack of effect may occur due to confound variables which we did not account for. I.e. personal characteristics of the respondent may determine whether someone is more or less willing to share on Facebook (Kietzmann et al., 2011), and older adults have different motivations and usage patterns compared to younger adults (Brandtzaeg et al., 2010). However, since Chang et al. (2015) did find significant results for the share intention, it is also possible that a direct relationship between the variables simply does not exist in this case. Combining this knowledge with convincing evidence for a strong positive effect between the like intention and share intention, we can assume that sharing behaviour for informative posts only happens after the consumer has already liked the post. This assumption makes sense, because liking a post is perceived as a way of showing support for the content (Brandtzaeg & Haugstveit, 2014). The lack of complexity makes the button easy to use for the consumer. Sharing content on the contrary is a task that evolves with more thought, whereas there are privacy concerns involved such as social surveillance and social control (Brandtzaeg et al., 2010). The same researchers show us that these concerns cause social media users to be more aware of what they want to share via these channels. In that light it makes sense that consumers first like the post as a first form of acknowledgement, before actually sharing the content. Furthermore, we hypothesized that usefulness positively affects the preference. This was not supported by our results, which means that H4 cannot be accepted. This is in contrast with the results of Chang et al. (2015), who found a positive significant relationship between the two variables. Explanation of this dissimilarity might be again in the evaluation of usefulness, which we said differs for informative and persuasive content.

5.2 Theoretical implications

(16)

Results show that the information quality evaluation plays an essential role in getting the consumer to like and share informative content (see appendix B). These findings are interesting, especially because previous research by Chang et al. (2015) indicates that the post popularity is of greatest importance when creating behavioural intention to like and share for persuasive messages.

Our results indicate not only that the drivers for creating usefulness and preference perceptions differ for persuasive vs. informative content, but also that the behavioural outcomes vary for both the types of messages. There is no evidence found for a direct intention to share content, based on the usefulness perception and preference of the consumer. This share intention is only facilitated through the intention to like the post. This means that the overall like and share behaviour must be different for both the types of messages. However, qualitative research should be performed in order to come to more valuable conclusions to why or why not people intent to like and share such content.

5.3 Managerial implications

First, this research has proven that the usefulness and preference perceptions for informative MGC are fostered by information quality, post popularity, and post attractiveness. At first glance these results do not seem to differ from previous research on persuasive MGC. However, we know that the effects for informative and persuasive MGC differ, which makes the new information valuable in the planning and execution of MGC. For persuasive MGC post popularity was clearly of greatest importance in determining the further social activity (like and share intention). In contrary, the specific indirect effects of the three elements on the like intention are almost equal to each other for informative MGC. This implies that marketeers who have the goal to persuade consumers with its post, face the risk of all-or-nothing: the consumer’s behavioural intention of a persuasive post is mostly determined by the popularity of the post. This means that a number of people should first be triggered to like and share the content because of the quality and attractiveness of the post, in order for the post to be picked up by the larger audience. For informative post the role of post popularity is less crucial; the intention to like is mostly triggered by good information quality. It is therefore essential to create content which entails the right factual data, whether it is a simple message, movie or photograph.

(17)

5.4 Limitations and future research

Looking critically at the execution of this study, there are a number of limitations to bear in mind when evaluating the quality of this research. First, we compare our study to an existing study, in order to find differences between two types of messages. Even though we have tried to mirror the circumstances of the study of Chang et al (2015), there are a couple of factors that are not corresponding. The study of Chang et al. was executed in an Asian country, in the year 2015. We collected our data in a European country in 2018. Cultural paradoxes cause consumer behaviour to be region specific, and therefore non-generalizable on a global scale (de Mooij, 2005). This leads us to believe that comparing the data of both studies is simply not representative for finding the differences between informative and persuasive messages. Besides, consumer behaviour on social media is likely to have changed over the last couple of years. Especially privacy concerns have altered the consumer its behaviour on social media channels (Ketelaar & van Balen, 2018). These two aspects make that we doubt the comparability of both the studies, and our conclusions on how informative messages differ from persuasive messages. However, the outcomes allow us to speculate that there actually is a difference between the two types of messages. Future research could solve these problems and create more reliable outcomes, by doing homogeneous research that looks at persuasive and informative messages in the same region, at the same time. This would require a large research capacity, and thus a sizable sample. Such a research design was simply not realistic for the resources we had to execute this current study.

Next, the overall generalizability of our study is low due to the small sample size. Resource restrictions limit the options to create a sample on selection basis, so respondents had to be approached via online social networks. This makes the sample rather random, and therefore decreases the generalizability of the study.

(18)

6. REFERENCES

Abernety, A.M., & Franke, G.R. (1996). The Information Content of Advertising: A Meta-Analysis.

Journal of Advertising, 25(2): 1-17.

Ackerberg, D. (2001). Emperically Distinguishing Informative and Prestige Effects of Advertising.

RAND Journal of Economics, 32: 316-333.

Ahearne, M., Gruen, T., & Jarvis, C. (1999). If Looks Could Sell: Moderation and Mediation of the Attractiveness Effect on Salesperson Performance. International Journal of Research in Marketing, 16(4): 269-284.

Ashley, C., & Tuten, T. (2015) Creative Strategies in Social Media Marketing: An Exploratory Study of Branded Social Content and Consumer Engagement. Psychology & Marketing, 32(1): 15-27. Bhattacherjee, A., & Premkumar, G. (2004). Understanding changes in belief and attidude toward information technology usage: A theoretical model and longitudinal test. MIS Quarterly, 28(2): 229-254.

Bhattacherjee, A., & Sanford, C. (2006). Influence processes for information technology acceptance: An elaboration likelihood model. MIS Quarterly, 30(4): 805-825.

Berkman, H.W., & Gilson, C.G. (1978). Consumer Behaviour: Concepts and Strategies. Dickenson Publishing: Encino, California.

Bolton, R.N. (2013) Understanding Generation Y and their use of Social Media: A Review and Research Agenda. Journal of Service Management, 24(3): 245-267.

Brandtzaeg, P.B., & Haugstveit, I.M. (2014). Facebook likes: a study of liking practices for humanitarian causes. International Journal Web Based Communities, 10(3): 258-279.

Brandtzaeg, P.B., Lüders, M., & Skjetne, J.H. (2010). Too Many Facebook “Friends”? Content Sharing and Sociability Versus the Need for Privacy in Social Network Sites. International Journal

of Human-Computer Interaction, 26(11): 1006-1030.

Chang, Y-T., Yu, H., & Lu, H-P. (2015). Persuasive Messages, Popularity Cohesion, and Message Diffusion in Social Media Marketing. Journal of Business Research, 68(4): 777-782.

Chen, W., & Lee, K-H. (2013). Sharing, Liking, Commenting, and Distressed? The Pathway Between Facebook Interaction and Psychological Distress. Cyberpsychology, Behavior, and Social

Networking, 16(10): 728-734.

Cheung, C.M.K., & Lee, M.K.O. (2010). A theoretical Model of Intentional Social Action in Online Social Networks. Decision Support Systems, 49(1): 24-30.

Chin, W.W. (1998). The Partial Least Squares Approach to Structural Equation Modeling. Advances

in Hospitality and Leisure, 8(2): 295-336.

Ching, A.T., & Ishihara, M. (2012). Measuring the Informative and Persuasive Roles of Detailing on Prescribing Decisions. Management Science, 58(7):1374-1387

(19)

Comrey, A.L., & Lee, H.B. (1992). A first course in Factor Analysis. Vol2. Lawrence Earlbaum Associates.

Crandall, C.S., Eshleman, A., & O’Brien, L. (2002). Social Norms and the Expression and Surpression of Prejudice: The Struggle for Internalization. Journal of Personality and Psychology, 82(3): 359-378.

Dean, D.H. (1999). Brand Endorsement, Popularity, and Event Sponsorships as Advertising Cues Affecting Consumer Pre-Purchase Attitudes. Journal of Advertising, 28(3): 1-12.

De Mooij, M. (2005). Global Marketing and Advertising: Understanding Cultural Paradoxes. Thousand Oaks, CA: Sage Publishers.

De Vries, L., Gensler, S., & Leeflang, P.S.H. (2012). Popularity of Brand Posts on Brand Fan Pages: An Investigation of the Effects of Social Media Marketing. Journal of Interactive Marketing, 26: 83-91.

Dehghani, M., & Tumer, M. (2015). A research on effectiveness of Facebook advertising on enhancing purchase intention of consumers. Computers in Human Behavior, 49: 597-600. Ding, Y., Phang, C.W., Lu, X., Tan, C.H., & Sutanto, J. (2014) The role of marketer- and user-generated content in sustaining the growth of a social media brand community. 47th Hawaii

International Conference System Sciences (HICSS), 1785-1792.

Feldman, J.M., & Lynch, J.G. (1988). Self-Generated Validity and Other Effects of Measurement on Belief, Attitude, Intention, and Behavior. Journal of Applied Psychology, 73(3): 421-435.

Flanagin, A.J., Metzger, M.J., Pure, R., Markov, A., & Hartsell, E. (2014). Mitigating risk in

ecommerce transactions: Perceptions of information credibility and the role of user-generated ratings in product quality and purchase intention. Electronic Commerce Research, 14(1): 1-23.

Fornell, C., & Larcker, D.F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1): 39-50.

Forza, C., & Filippini, R. (1998). TQM impact on quality conformance and customer satisfaction: a causal model. International Journal of Production Economics, 55: 1-20.

Garson, G.D. (2016). Partial Least Squares: Regression and Structural Equation Models. Asheboro, N.C.: Statistical Associates Publishers.

Gerlitz, C., & Helmond, A. (2013). The like economy: Social buttons and the data-intensive web. New

Media & Society, 15(8): 1348-1365.

Goh, K-Y., Heng, C-S., & Lin, Z. (2013). Social Media Brand Community and Consumer Behavior: Quantifying the Relative Impact on User- and Marketer-Generated Content. Information Systems

research, 24(1): 88-107.

Grewal, D., Monroe, K.B., & Krishnan, R. (1998). The Effects of Price-Comparison Advertising on Buyers’ Perceptions of Acquisition Value, Transaction Value, and Behavioral Intentions. Journal of

Marketing, 62: 46-59.

(20)

Hair, J.F., Hollingsworth, C.L., Randolph, A.B., & Chong, A.Y.L. (2017). An Updated and Expanded Assessment of PLS-SEM in Information Systems Research. Industrial Management & Data Systems, 117(3): 442-458.

Hair, J.F., Ringle, C.M., & Sarstedt, M. (2011). PLS-SEM: Indeed a Silver Bullet. Journal of

Marketing Theory and Practice, 19(2): 139-152.

Hamid, M.R., Sami, W., & Mohmad Sidek, M.H. (2017). Discriminant Validity Assessment: Use of Fornell & Larcker criterion versus HTMT Criterion. Journal of Physics: Conference Series, 890: 1-5. Hartmann, P., & Apaolaza-Ibáñez, V. (2011). Consumer Attitude and Purchase Intention toward Green Energy Brands: The Roles of Psychological Benefits and Environmental Concern. Journal of

Business Research, 65: 1254-1263.

He, W., Qiao, Q., & Wei, K. K. (2009). Social relationship and its role in knowledge management system usage. Information & Management, 75(6): 55-71.

Henseler, J., Ringle, C.M., & Sinkovics, R.R. (2009). The use of partial least squares path modeling in international marketing. New Challenges to International Marketing Advances in International

Marketing, 20: 277-319.

Hoy, M.G., & Milne, G. (2010). Gender differences in privacy-related measures for young adult Facebook users. Journal of Interactive Advertising, 10(2): 28-45.

Hsu, C.L., & Lin, J.C.C. (2008). Acceptance of blog usage: The roles of technology acceptance, social influence and knowledge sharing motivation. Information & Management, 45(1): 65-74.

Hu, L., & Bentler, P.M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary

Journal, 6(1): 1-55.

Jamieson, L.F., & Bass, F.M. (1989). Adjusting stated intention measures to predict trial purchase of products: a comparison of models and methods. Journal of Marketing Research, 26: 336-345. Kamalski, J., Sanders, T., & Lentz, L.R. (2008). Coherence Marking, Prior Knowledge, and Comprehension of Persuasive Texts: Sorting Things Out. Discourse Processes, 45: 323-345. Ketelaar, P.E., & van Balen, M. (2018). The smartphone as your follower: The role of smartphone literacy in the relation between privacy concerns, attitude and behaviour towards phone-embedded tracking. Computers in Human Behavior, 78: 174-182.

(21)

Kumar, A.m, Bezawada, R., Rishika, R., Janakiraman, R., & Kannan, P.K. (2015). From Social to Sale: The Effects of Firm-Generated Content in Social Media on Customer Behavior. Journal of

Marketing, 80: 7-25.

Lamberton, C., & Stephen, A.T. (2016). Athematic Exploration of Digital, Social Media, and Mobile Marketing: Research Evolution from 2000 to 2015 and an Agenda for Future Inquiry. Journal of

Marketing, 80: 146-172.

Larcker, D.F., & Lessig, V.P. (1980). Perceived usefulness of information: a psychometric examination. Decision Sciences, 11: 121-134.

Lavine, H., & Snyder, M. (1996). Cognitive processing and the functional matching effect in

persuasion: The mediating role of subjective perceptions of message quality. Journal of Experimental

Social Psychology, 32(6): 580-604.

Lee, C.S., & Ma, L. (2012). News sharing in social media: The effect of gratifications and prior experience. Computers in Human Behavior, 28(2): 331-339.

Li, S., & Lin, B. (2006). Accessing information sharing and information quality in supply chain management. Decision Support Systems, 42: 1641-1656.

Lindgaard, G., Fernandes, G., Dudek, C., & Brown, J. (2006). Attention web designers: You have 50 miliseconds to make a good first impression. Behaviour & Information Technology, 25(2): 115-126. Lohmöller, J-B. (1989). Latent Variable Path Modeling with Partial Least Squares, Heidelberg: Physica.

Lu, J., Liu, C., Yu, C.S., & Wang, K. (2008). Determinants of accepting wireless mobile data services in China. Information & Management, 45(1): 52-64.

Marquez, F.T. (1977). Advertising Content: Persuasion, Information or Intimidation? Journalism &

Mass Communication Quarterly, 54(3): 482-491.

McClure, S.M., Li, J., Tomlin, D., Cypert, K.S., Montague, L.M., & Montague, P.R. (2004). Neural Correlates of Behavioral Preference for Culturally Familiar Drinks. Neuron, 44: 379-387.

Moore, D.S., Notz, W.I., & Flinger, M.A. (2013). The basic practice of statistics. 6th ed. New York:

W.H. Freeman and Company.

Mueller, D.C., & Stratmann, T. (1994). Informative and persuasive campaigning. Public Choice, 81: 55-77.

O’Keefe, D.J. (1990). Persuasion: Theory and Research. Vol 2. Newbury Park, California, US: Sage. Petty, R.E., & Wegener, D.T. (1998). Matching Versus Mismatching Attitude Functions: Implications for Scrutiny of Persuasive Messages. Personality & Social Psychology Bulletin, 24(3): 227-240. Pew Research Center (2018). Millennials stand out for their technology use, but older generations also embrace digital life. [online]. Via:

http://www.pewresearch.org/fact-tank/2018/05/02/millennials-stand-out-for-their-technology-use-but-older-generations-also-embrace-digital-life/ Accessed:

16/10/2018.

Pütter, M. (2017). The Impact of Social Media on Consumer Buying Intention. Journal of

(22)

Rose, A.J., Swenson, L.P., & Waller, E.M. (2004). Overt and Relational Aggression and Perceived Popularity: Differences in Concurrent and Prospective Relations. Developmental Psychology, 40(3): 378-387.

Sarstedt, M., Hair, J.F., Ringle, C.M., Thiele, K.O., & Gudergan, S.P. (2016). Estimation issues with PLS and CHSEM: Where the bias lies. Journal of Business Research, 69: 3998-4010.

Schivinski, B., & Dabrowski, D. (2016). The effect of social media communication on consumer perceptions of brands. Journal of Marketing Communications, 22(2): 189-214.

Schultz, W., Nolan, J., Cialdini, R., Goldstein, N., & Griskevicius, V. (2007). The Constructive, Destructive, and Reconstructive Power of Social Norms. Psychological Science, 18: 429-434. Shareef, M.A., Mukerji, B., Alryalat, M.A.A., Wright, A., & Dwivedi, Y.K. (2018). Advertisements on Facebook: Identifying the persuasive elements in the development of positive attitudes in

consumers. Journal of Retailing and Consumer Services, 43: 258-268.

Sherif, M., & Sherif, C.W. (1953). Groups in Harmony and Tension. New York: Harper.

Spears, N., & Singh, S.N. (2004). Measuring Attitude Toward the Brand and Purchase Intentions.

Journal of Current Issues and Research in Advertising, 26(2): 53-66.

Stapel, J. (1971). Sales effect of print ads. Journal of Advertising Research, 1(3): 32-36. Statista (2018) Number of Social Network Users Worldwide from 2010 to 2021. [online]. Via

https://www.statista.com/statistics/278414/number-of-worldwide-social-network-users/ Accessed:

05/11/2018.

Taylor, D.G., Lewin, J.E., & Strutton, D. (2011). Friends, Fans, and Followers: Do Ads Work on Social Networks? How Gender and Age Shape Receptivity. Journal of Advertising Research, 51(1): 258-275.

The Statistics Portal. (2017). Social Media Marketing – Statistics & Facts [online]. Via

https://www.statista.com/topics/1538/social-media-marketing/ Accessed: 09/10/2018.

Valverde, W., Lefevre, C., & Hoehling, A. (2013). eNonprofit Benchmarks Study: An Analysis of Online Messaging, Fundraising, Advocacy, Social Media and Mobile Metrics for Nonprofit Organizations. The Nonprofit Technology Network, New York.

Verhagen, T., Feldberg, F., van den Hooff, B., Meents, S., & Merikivi, J. (2012). Understanding users’ motivations to engage in virtual worlds: A multipurpose model and empirical testing. Computers in

Human Behavior, 28(2): 484-495.

Walker-Naylor, R., Poynor-Lamberton, C., & West, P.M. (2012). Beyond the “Like” Button: The Impact of Mere Virtual Presence of Brand Evaluations and Purchase Intentions in Social Media Setting. Journal of Marketing, 76: 105-120.

(23)

Yoon, S.J., & Park, J.E. (2012). Do Sensory Ad Appeals Influence Brand Attitude? Journal of

Business Research, 65(11), 1534-1542.

(24)

APPENDICES

APPENDIX A Dear participant,

Thank you for finding the time to participate in this study. By completing the survey, you help to gain new insights in Marketer Generated Content on social media.

Completing the survey will take approximately five minutes. Participating in this study is fully voluntary, which means that you can withdraw at any point. Please be assured that your participation is anonymously, and all the data will be handled with confidentiality.

If you have any questions concerning this survey or the study in general, feel free to email me at

m.l.stuulen@student.rug.nl

Please click the red button to begin the survey. Q1: Please select your age

o Under the age of 18 years o 18-24 years o 25-34 years o 35-44 years o 45-54 years o 55-64 years o 65 years or older Q2: Please select your gender

o Male o Female o Other

Q3: What is the highest level of education completed? o I have no diploma

o High School diploma

o Community College diploma (MBO)

o University of Professional Education diploma (HBO) o Bachelor degree (WO bachelor)

o Masters degree (WO master) o PhD or higher

Q4: How would you describe your English skills? o Basic communication skills

o Good command o Very good command o Fluent

o Native speaker

Q5: Do you own a Facebook account?

(25)

In the next section we will show you an informative video advertisement posted on Facebook. After the video, a number of questions will be asked to reflect on the advertisement.

Please click the red button to move on to the video.

(Video’s exposed are promotional videos of Apple and Samsung. Both videos present a new product, and show factual features of the new phone in a visually attractive manner. The videos are equal in subject, goal, message, and duration).

Now we ask you to reflect on the informative post you have just watched. Please select how much you agree with the following statements:

Q6: This post is informative to me. o Strongly agree

o Agree o Neutral o Disagree

o Strongly disagree Q7: This post is helpful to me.

o Strongly agree o Agree

o Neutral o Disagree

o Strongly disagree Q8: This post is valuable to me.

o Strongly agree o Agree

o Neutral o Disagree

o Strongly disagree

Q9: I perceive the post as attractive. o Strongly agree

o Agree o Neutral o Disagree

o Strongly disagree

Q10: I perceive the post as appealing. o Strongly agree

o Agree o Neutral o Disagree

o Strongly disagree

Q11: This post can reduce the time spent on making my decision, if I wanted to buy such a product. o Strongly agree

o Agree o Neutral o Disagree

(26)

Q12: This post can increase the quality of my decision making, if I wanted to buy such a product. o Strongly agree o Agree o Neutral o Disagree o Strongly disagree

Q13: Overall, I find this post useful. o Strongly agree

o Agree o Neutral o Disagree

o Strongly disagree

Q14: I feel pleasant when being informed about the newest mobile phones. o Strongly agree

o Agree o Neutral o Disagree

o Strongly disagree

Q15: I feel good when learning about the newest mobile phones. o Strongly agree

o Agree o Neutral o Disagree

o Strongly disagree

Q16: I like to be informed about the newest mobile phones. o Strongly agree

o Agree o Neutral o Disagree

o Strongly disagree

Q17: I intend to press like on such a post. o Strongly agree

o Agree o Neutral o Disagree

o Strongly disagree

Q18: I anticipate that I will press like on such a post. o Strongly agree

o Agree o Neutral o Disagree

o Strongly disagree

Q19: I intend to share such a post on my timeline. o Strongly agree

(27)

Q20: I expect to share such a post on my timeline. o Strongly agree o Agree o Neutral o Disagree o Strongly disagree

Q21: I plan to share such a post on my timeline. o Strongly agree

o Agree o Neutral o Disagree

o Strongly disagree

Q22: I think that posts with more people showing like & share behaviour are trustworthy. o Strongly agree

o Agree o Neutral o Disagree

o Strongly disagree

Q23: I think that posts with more people showing like & share behaviour are reliable. o Strongly agree

o Agree o Neutral o Disagree

o Strongly disagree

(28)

APPENDIX B

Specific indirect effects persuasive vs. informative messages

EFFECTS Persuasive Informative

Information/Argument Quality -> Usefulness -> Like Intention (0.30 * 0.44) = 0.132 (0.63 * 0.28) = 0.177 Post Popularity -> Usefulness -> Like Intention (0.59 * 0.44) = 0.260 (0.27 * 0.28) = 0.075 Post Popularity -> Preference -> Like Intention (0.29 * 0.34) = 0.099 (0.25 * 0.44) = 0.110 Post Attractiveness -> Preference -> Like Intention (0.34 * 0.34) = 0.116 (0.36 * 0.44) = 0.158

APPENDIX C

Comparison for Beta results persuasive vs. informative messages

RELATIONSHIPS Persuasive (Chang et al., 2015) Informative

Information Quality -> Usefulness b = 0.30* b = 0.63*

Post Popularity -> Usefulness b = 0.59* b = 0.27*

Post Popularity -> Preference b = 0.29* b = 0.25*

Post Attractiveness -> Preference b = 0.34* b = 0.36*

Usefulness -> Preference b = 0.32* b = 0.18*

Usefulness -> Like Intention b = 0.44* b = 0.28*

Usefulness -> Share Intention b = 0.21* b = 0.04*

Preference -> Like Intention b = 0.34* b = 0.44*

Preference -> Share Intention b = 0.12* b = 0.11*

Like Intention -> Share Intention b = 0.49* b = 0.62*

Referenties

GERELATEERDE DOCUMENTEN

Overigens kan het ook zijn dat een procedure over een irreële inschrijving niet leidt tot terzijdeleg- ging, maar tot een veroordeling tot heraan- besteding, als de

Chapter 2 Comparison of dynamic magnetic resonance defecography with evacuation of rectal contrast and conventional defecography for posterior pelvic floor compartment

Door slimme innovaties toe te passen kan, ofwel de levensduur verlengd worden, ofwel de introductie van nieuwe assets verbeterd worden (“smooth introduction”).. Van Dongen

Although the moderating effect of family functioning on the link between headache intensity and distress has not been explored previously, a few studies in the context of

studentenvereniging in de Lindenberg dus dan ga je ook sneller dat als je ziet dat iets in de Lindenberg is, ga ik sneller. Dat is hetzelfde met LUX of ik heb weleens een rondleiding

Still considering the vast selected range of rubber tread compounds for the present study, the prediction of actual tire lateral grip on the road with a solid test wheel within

This led to the development of human disease mimicking in vitro models advancing from 2D monocultures/cocultures to self-assembled 3D spheroids and patient-derived organoids;

This study focuses on investigating the reinforcing behavior of a TESPT modified lignin-based filler in a SSBR/BR blend in comparison to CB and silica/TESPT.. With mechanical