media regulation

EthicsinsocialmediamarketingHowshouldsponsorshipinformationbedisclosedinonlineproductreviews.pdf

Home >English homework help >media regulation

Ethics in social media marketing How should sponsorship information be disclosed in online product reviews?

Wen-Chin Tsao and Tz-Chi Mau Department of Business Administration,

National Chin-Yi University of Technology, Taichung, Taiwan

Abstract Purpose – Consumer-generated online product reviews (OPRs) have become a crucial source of information for consumers; however, OPRs are increasingly being incentivized. The purpose of this paper is to find a method of sponsorship and disclosure that could be considered ethically sound. Design/methodology/approach – This study adopted a quasi-experimental approach to clarifying how the method of sponsorship impacts reader perceptions of OPRs in terms of helpfulness, credibility and purchase intention. Two experiments were performed on an online platform using data from 480 participants. Hypotheses were tested using analysis of covariance. Findings –Meaning under the premise that sponsorship information is disclosed and not withheld from the readers, Study 1 revealed that experiential sponsorship is the best sponsorship. Study 2 revealed that featuring reviewers with greater influence in the online community increases the positive influence of disclosing experiential sponsorship on OPR persuasiveness. Originality/value – The findings in this study provide rational incentives for firms to disclose sponsorship information, i.e. demonstrate high ethical standards in marketing. This was shown to create a win-win-win situation for consumers, firms and reviewers. Managerial implications for online marketing managers are also discussed. Keywords Online product reviews, Marketing ethics, Experience goods, Online community influence, Sponsored reviews, Sponsorship type Paper type Research paper

1. Introduction Electronic word-of-mouth (eWOM) is the sharing of relevant knowledge and opinions (positive or negative) about the products and/or services of a firm among potential or actual customers (Hennig-Thurau et al., 2004; Tsao, Hsieh, Shih and Lin, 2015; Tsao, Mau, Hsieh and Chen, 2015; Chen et al., 2017). In Web 2.0, consumers actively receive information through networks and also act as the main entity spreading information. This has prompted explosive growth in online forums, blogs and social media, which allow consumers to share their thoughts on products or services quickly and conveniently (Kuo, 2015; Kim and Song, 2018). Consumer-generated online product reviews (OPRs) are growing in popularity and importance ( Jonas, 2010; Stephen et al., 2012; Lee and Choeh, 2018). OPRs contain the opinions and ratings of users with regard to the attributes, performance and use of products. They are generally found on platforms that present an extensive range of information, such as blogs (Forrest and Cao, 2010; Tsao and Hsieh, 2015; Yu, 2017).

OPRs should in no way be connected with firms (to avoid ethical conflict); however, many studies have reported that firms can influence OPRs in an indirect manner. This calls into question the impartiality and credibility of OPRs (Chu, 2010). Many firms are operating under the belief that the voluntary disclosure of sponsorships would undermine the effectiveness of OPRs, prompting them to devise methods to hide the sponsorship information (Stephen et al., 2012). The concealment of sponsorship information is essentially Aslib Journal of Information

Management Vol. 71 No. 2, 2019

pp. 195-216 © Emerald Publishing Limited

2050-3806 DOI 10.1108/AJIM-04-2018-0080

Received 12 April 2018 Revised 8 June 2018

6 October 2018 Accepted 28 November 2018

The current issue and full text archive of this journal is available on Emerald Insight at: www.emeraldinsight.com/2050-3806.htm

The authors wish to thank the Dr Ming-Tsang Hsieh who is graduated from the National Taiwan University of Science and Technology for providing general affairs support for this research, and Cheng-Hao Tsai, CEO of Well Done International Cooperation, for providing practical consulting.

195

Ethics in social media marketing

an attempt to deceive consumers. It is clearly a violation of corporate ethics and is even illegal in many countries (Forrest and Cao, 2010; Fair Trade Commission, 2017).

Previous studies on the persuasiveness of eWOM and OPRs explored content properties (e.g. quality, quantity, focus or valence) (Park et al., 2007; Park and Kim, 2008; Zhang et al., 2010; Qiu et al., 2012; Tsao, Hsieh, Shih and Lin, 2015; Tsao, Mau, Hsieh and Chen, 2015; Chakraborty and Bhat, 2018), whereas others investigated product type (e.g. search, experience and credence goods) or eWOM platform (e.g. independent or corporate platform). Still, others attempted to enhance eWOM persuasiveness by separating and combining various product types and eWOM platform variables (Park and Lee, 2009; Tsao and Hsieh, 2015). Researchers have used the technology acceptance model to examine the influence of eWOM trust on consumer willingness to purchase tourism products (Chen et al., 2017). Other researchers have used big data to determine how eWOM impacts the growth of consumer finance (Shen et al., 2017). Still, others adopted used experiential models to verify how blog characteristics and experiential value induces purchase intentions in consumers (Wang, 2017; Lee, 2016). However, most previous studies were conducted from the perspective of the firm and how marketing communications and OPR design could benefit them. Very few studies touched on the issue of ethical business practices in the operation of OPRs. This issue is of particular importance in countries where sponsorship disclosure is regulated by law.

Very few studies have discussed the disclosure of sponsorship information by reviewers. Tsao, Hsieh, Shih and Lin (2015) and Tsao, Mau, Hsieh and Chen (2015) verified the deliberate withholding of sponsorship information could have a devastating impact on the persuasiveness of OPRs in the event that this situation became known. Hwang and Jeong (2016) investigated the relationship honesty in sponsorship disclosure on blogs (such as simple disclosure and “honest opinions” disclosure) and the credibility of the source. Lu et al. (2014) examined how various types of sponsorship influence consumer attitudes toward the recommendations; however, they focused only on the provision of financial incentives by sponsors. Kim and Song (2018) explored the influence of content sponsorship (organic vs sponsored) and content type (promotional vs experience centric) on the inference of manipulative intent, brand attitude and click intention among Twitter users. All of these studies discussed the influence of sponsorship type on OPRs or eWOM persuasiveness; however, none of them explored the reasons for favoring one strategy over another. Nor did any previous study take into consideration how the traits of the reviewer (such as community influence) could affect the outcomes of sponsorship disclosure. In this study, we employed non-disclosure as the control variable to identify the best sponsorship disclosure policy while taking into account the sponsorship type and the influence of reviewers in the community. In other words, this study sought to identify the types of sponsorship that would be most beneficial to the firm, the reviewer and the consumers.

In this study, we endeavored to provide firms with suggestions on dealing with sponsorship and disclosure issues. Our objective was to identify strategies that would comply with high ethical standards in marketing as well as relevant legal regulations without compromising the persuasiveness of their recommendations or the credibility of the source. Tsao, Hsieh, Shih and Lin (2015) and Tsao, Mau, Hsieh and Chen (2015) reported that disclosing sponsorship messages in blogs would significantly decrease the persuasiveness of OPRs by undermining the credibility of the blogger, resulting in a notable decrease in purchase intention. In this study, we conducted two sub-studies: first, an investigation into the type of sponsorship disclosure strategies best able to preserve the persuasiveness of OPRs; and second, an investigation into the degree to which the influence of the reviewer in the community could alter the outcomes of sponsorship disclosure. The most significant contribution of this study was the identification of rational incentives for firms to abide by marketing ethics.

196

AJIM 71,2

2. Theoretical background, and literature review Tsao, Hsieh, Shih and Lin (2015) and Tsao, Mau, Hsieh and Chen (2015) conducted an experiment using two scenarios with one group of participants. The participants first read a blog post with no disclosure of sponsorship information. The researchers then told them that the reviewer received financial sponsorship from the firm, and examined how this revelation affected the influence of the review. They found that a failure to disclose sponsorship information greatly undermined the persuasiveness of the OPR, including the helpfulness and credibility of the reviews as well as the purchase intention of the readers. However, that study failed to provide a solution to the problem of information disclosure. Lu et al. (2014) examined the sponsored recommendations of bloggers; however, they focused only on financial rewards. Sponsorship can also be provided using non-monetary methods (Chu, 2010; Stephen et al., 2012). Hwang and Jeong (2016) discussed the ethical implication of sponsorship disclosure (e.g. disclosure and “honest opinions” disclosure), rather than the types of sponsorship that are actually employed by firms. Kim and Song (2018) divided content sponsorship into two categories: organic (i.e. unpaid) and sponsored (i.e. paid). Their primary focus was on OPR content (promotional vs experience centric); however, they did not touch on the issue of how firms could be motivated to proactively disclose sponsorship. In this study, we divided the types of sponsorship into financial, non-financial and dual sponsorship. This study also adopted the evaluation index used in previous eWOM researches, using helpfulness, credibility and purchase intention to determine the persuasiveness of OPR (Li et al., 2013; Tsao and Hsieh, 2015; Tsao, Hsieh, Shih and Lin, 2015; Tsao, Mau, Hsieh and Chen, 2015). This study also examined sponsorship disclosure within the context of blogs, due to their growing popularity and influence in the online community (Fullwood et al., 2009).

2.1 Disclosure of sponsorship information Information disclosure is a term that was first used in corporate finance, wherein firms communicate with the public, stakeholders and their investors through the disclosure of financial information. In some cases, the disclosure of information is voluntary, whereas in other cases, it is mandated by the law (Brounen et al., 2001). The issue of information disclosure is becoming important in internet marketing as well. Community website posts that are remunerated by the manufacturer or distributor (in any way) are deemed to be sponsored (Hwang and Jeong, 2016). In the case of blogs, numerous countries including the USA, UK, Taiwan and South Korea have clear laws stipulating that any paid endorsements from firms related to reviewed products must be disclosed by the blogger (Stephen et al., 2012; Lu et al., 2014; Jensen and Yetgin, 2017; Fair Trade Commission, 2017). Regardless of the legal ramifications, ethical business practices dictate that firms be encouraged to disclose sponsorship information. Unfortunately, many firms are concerned that knowledge of sponsorship could alter the opinions of consumers. They fear that the review would not be considered honest or regard it simply as a crass marketing tool. There remains a dearth of research pertaining to the outcomes of sponsorship disclosure and the credibility of sponsored reviewers.

2.2 Online product reviews and sponsored reviews OPRs include opinions and experiences with products or services, which are shared via the internet to assist others in making purchase decisions (Stephen et al., 2012; Li et al., 2013; Kim and Song, 2018). OPRs are becoming a significant factor in the success of internet marketing (Kwon and Sung, 2012). Non-commercial OPRs that are involved with the source firm should be designated as sponsored reviews ( Jensen and Yetgin, 2017). Marketers use network analysis (communications traffic and content analysis) to identify key consumers who could be recruited as agents (Forrest and Cao, 2010). Agents (i.e. professional reviewers)

197

Ethics in social media marketing

are often active on major retailer websites (such as Amazon.com or Twitter) or social network platforms (such as blogs) and silently perform eWOM communication for other firms (Forrest and Cao, 2010; Chu, 2010; Floyd et al., 2014; Kim and Song, 2018). The main criticism aimed at this type of communication is the way it emulates conventional (i.e. objective) eWOM by failing to disclose sponsorship information. In fact, many firms actively seek to disguise or conceal their sponsorship, which in many cases is seen as an attempt to skirt the law.

2.3 OPR persuasiveness Researchers have sought to gauge the persuasiveness of eWOM, with a particular focus on the cognitive aspect of assessments, i.e. the perceived helpfulness and credibility of reviews (Li et al., 2013; Tsao and Hsieh, 2015; Tsao and Lin, 2017; Hwang and Jeong, 2016; Chakraborty and Bhat, 2018; Hajli, 2018). Other researchers have focused on the behavioral aspect, including purchase intention, booking intention, movie selection or adoption intention (Tsao, 2014; Tsao, Hsieh, Shih and Lin, 2015; Tsao, Mau, Hsieh and Chen, 2015; Tsao and Hsieh, 2015; Chakraborty and Bhat, 2018; Hajli, 2018; Tsao and Lin, 2017). Thus, this study also included helpfulness, credibility and purchase intention to describe the persuasiveness of OPR.

2.3.1 OPR helpfulness. Li et al. (2013) examined OPR helpfulness from the perspective of helping behavior, believing that with limited time and resources, consumers would use the internet to search for information and reduce uncertainty. OPRs from customers or experts can help potential consumers to make purchase decisions. When consumers come into contact with this kind of information, it is generally assumed to be a non-commercial endorsement from another consumer who simply wants to help others. Potential consumers, therefore, believe that OPRs will benefit them in making purchase decisions (Li et al., 2013; Pan and Zhang, 2011). The degree to which the content of an OPR informs the reader is a crucial factor in characterizing helpfulness (Stephen et al., 2012). Lee and Choeh (2018) reported that the helpfulness of a review can help reinforce the influence of review quantity, valence and length on movie box office results. OPRs that are more helpful tend to increase OPR adoption, thereby assisting consumers to decide which brand/product to purchase, i.e. guiding future consumption scenarios (Hajli, 2018).

2.3.2 OPR credibility. Credibility is widely cited by researchers in the assessment of information and information sources. It has recently been included in research models on the effects of eWOM and social WOM (Tsao and Hsieh, 2015; Tsao, Hsieh, Shih and Lin, 2015; Tsao, Mau, Hsieh and Chen, 2015; Chakraborty and Bhat, 2018; Hajli, 2018). Cheung et al. (2009) defined OPR credibility as the degree to which readers view the content of a recommendation/review to be believable, true or factual, whereas Qiu et al. (2012) defined it as the degree to which a review is considered true and valid. The identity and activeness of the reviewer are two of the clues that readers use to assess OPR credibility. In online communities where anonymity is possible, readers are more likely to trust and accept the views of reviewers whose real identities are more easily known and credit said reviewers with their contribution of knowledge (Ma and Agarwal, 2007). This shows that when the OPR source has more professional or notary certification (such as more identifying labels and certification marks or a high popularity index) or is named, the credibility of the OPR is higher.

2.3.3 Purchase intention. When a consumer shows personal action tendencies toward a certain product/brand, then it means that they have purchase intention (Spears and Singh, 2004). Spears and Singh (2004) explained purchase intention as a consumer’s conscious efforts and plans to purchase a product. Dodds et al. (1991) defined purchase intention as the

198

AJIM 71,2

likelihood that a consumer will purchase a product based on their attitude or assessment of said product after being stimulated by an external factor. Hence, greater likelihood implies a higher purchase intention (Tsao, 2014).

3. Study 1: influence of sponsorship type on OPR persuasiveness A review of the literature revealed that concealing sponsorship information can enhance the persuasiveness of OPRs (Tsao, Hsieh, Shih and Lin, 2015; Tsao, Mau, Hsieh and Chen, 2015); however, it is a direct violation of marketing ethics and is outright illegal in some countries. Thus, in sub-study 1, this study sought to identify the type of sponsorship best suited to the disclosure of sponsorship information.

3.1 Sponsorship type Customers are offered numerous types of incentive in exchange for OPRs; however, they can be broken down into two major categories: monetary and non-monetary. Monetary incentives include money, raffles, discounts, points that can be redeemed for free products, and partial refunds or rebates. Non-monetary incentives involve the provision of free products or services experiences (Stephen et al., 2012; Chu, 2010). Blogs are among the most popular OPRs because they encompass text, pictures, video files and links to other sites as well as responses from reviewers (i.e. high in richness and interactivity). The unique style of a blogger can have a profound impact on others and gain a considerable number of followers (Fullwood et al., 2009). In in-depth interviews with bloggers, Chu (2010) found that firms provide a variety of incentives to obtain recommendations for products. Based on the online influence of the blogger, the incentives could range from free products to gifts or even monetary rewards. Based on the classification of incentives and the characteristics of blogs, we divided sponsorship into three types: monetary sponsorship, experiential sponsorship (based on free trials of products) and dual sponsorship (a combination of the two).

3.2 Research hypotheses eWOM is characterized by actual use or experiences of consumers. As for their commercial nature, no transactions exist between the reviewers and firms, so OPRs are essentially different from commercial advertisements (Hennig-Thurau et al., 2004). These two characteristics also make OPRs more persuasive than conventional advertisements. However, in practice, eWOM, particularly OPRs, is showing a trend of being incentivized (Stephen et al., 2012; Chu, 2010; Floyd et al., 2014; Lu et al., 2014). In some countries, incentivized OPRs are considered another form of advertising and are similar to the testimonial commercials on television (Stephen et al., 2012).

In terms of experiential sponsorship, the blogger actually uses the product before writing an OPR. This means that the review contains personal experiences. Researchers have discovered that when consumers read OPRs, they pay attention to whether the blogger has actually experienced the product that they recommend, and this affects their perception of the authenticity and credibility of the review (Lin et al., 2011; Stephen et al., 2012). For some researchers, free products are seen as a type of reward. When experts or opinion leaders try out and endorse a product, they can make the product sound more interesting. It is also possible that receiving exclusive offers may enhance their engagement with the product (Oberhofer et al., 2014). In view of the above, this study infers that consumers tend to regard sponsored products as the firm trying to encourage consumer participation with the aim of more fully and accurately conveying product performance. Thus, it would be reasonable to expect that the persuasiveness of reviews based on experiential sponsorship should be on par with that of reviews for which sponsorship information is not disclosed.

199

Ethics in social media marketing

Tong et al. (2007) discovered that when reviewers accept economic rewards from a sponsor, it is harder for them to say what they really feel in their reviews. This type of review is likely to favor the firm, which is why they are referred to as professional fake reviews (Schneiderman, 2013). By paying, firms are seeking to control the reviewer directly and imperceptibly to produce messages that are favorable to their company/product. Readers who know that a recommendation was paid for tend to question the fairness and authenticity of the review process, which can lower their expectation of the quality of the product and may even negatively affect the persuasiveness of the review (Stephen et al., 2012). Thus, readers see paid reviews as professional fake reviews, even in cases where the product experience process is described. Thus, the persuasiveness of such reviews is considerably weaker than that of reviews with experiential sponsorship. Reviews with purely monetary sponsorship cannot be considered eWOM (i.e. they are no longer non-commercial and are no longer based on personal experience). From this, this study infers that this greatly undermines the persuasiveness of such reviews (Stephen et al., 2012; Schneiderman, 2013). Thus:

H1. Compared to non-disclosure, the disclosure of monetary or dual sponsorship exerts a significantly negative influence on (a) OPR helpfulness, (b) OPR credibility and (c) purchase intention, whereas the disclosure of experiential sponsorship does not have a significant influence.

3.3 Methods 3.3.1 Design and participants. Study 1 deals exclusively with the influence of sponsorship type (independent variable) on OPR persuasiveness (dependent variable), i.e. the other variables were excluded. The experimental design adopted in this study provided control over other factors. Thus, the researchers needed only to manipulate the independent variables (i.e. factors) and observe the resulting changes in the dependent variables. When an experiment cannot be conducted in a laboratory, or when the study samples cannot be randomly assigned, the experimental design is called a quasi-experimental design (Pimperl et al., 2017).

The participants took part in the experiment in places that were familiar to them and allowed them to be at ease, such as their offices, schools or homes, which made this a field experiment (Wu and Lin, 2001). Difficulty in obtaining a register of consumer populations precluded the random assignment of participants. Thus, in Study 1, we adopted a one-factor quasi-experimental design focusing on sponsorship type: monetary/experiential/dual/ non-disclosure with non-disclosure serving as a control. By manipulating these four scenarios, this study examined how OPR persuasiveness varied among the four scenarios. The participants were recruited on a community website. Participants were presented four hyperlinks, each of which led to one of four scenarios. The participants were free to choose from among the hyperlinks, i.e., they were not randomly assigned to the scenarios.

All of the participants reported reading online restaurant reviews in the past. A total of 180 participants completed the experiment, which yielded 160 valid samples. The statistics of the participants were as follows. Males 45.6 percent, between the age of 21 and 30 (80.0 percent), students (38.8 percent), working in the service industry (24.4 percent). The participants reported reading blogs every month as follows: 6–10 times per month (53.8 percent) and 1-5 times per month (28.1 percent). The most common types of blog were travel blogs (22.5 percent) and restaurant blogs (18.8 percent).

3.3.2 Development of stimuli and measures. Peng et al. (2011) observed that restaurant blogs are very common on community websites. The photos and vivid descriptions help readers to form a sense of presence and encourage them to give it a try. Thus, the study used

200

AJIM 71,2

blogs to represent OPRs in this study. Consumers wishing to buy experience or credence goods tend to rely more on the opinions of experts or experienced consumers as their source of information, in order to reduce uncertainty and perceived risk (Park and Lee, 2009; Tsao, Hsieh, Shih and Lin, 2015; Tsao, Mau, Hsieh and Chen, 2015). In previous studies, restaurants are widely regarded as representative of experience goods (Peng et al., 2013; Tsao, 2014; Tsao, Hsieh, Shih and Lin, 2015; Tsao, Mau, Hsieh and Chen, 2015). We, therefore, focused on restaurants in this study. The use of a known brand is the easiest way to enhance the authenticity of an experiment. However, the study sought to avoid the potential moderating effect of corporate awareness on the results. The research, therefore, selected a newly established restaurant (de reve café), as the subject of this experiment. This French-style café is modeled on the Eiffel Tower and offers mainly afternoon tea.

This study used an actual blog article for the experiment to enhance authenticity. Following a review of blogs with suitable photos and text, we selected the review of de reve café on the blog Farmer, listed on Holiday PaPaGo in Taiwan. The article described the location, facilities and menu of the restaurant, i.e. it was an informed review. It earned a recommendation from the reviewer. The article included 38 photos inserted throughout the text. To prevent interference from the reputation of the blogger, we omitted any mention of the blogger’s identity from the article.

The independent variable in this experiment was the type of sponsorship disclosed to the public (participants): monetary sponsorship, experiential sponsorship and dual sponsorship. Non-disclosure was used as a control group. The dependent variables were the various aspects of OPR persuasiveness, which were represented using three constructs: OPR helpfulness, OPR credibility and purchase intention. OPR helpfulness refers to the degree to which the OPR can help consumers understand the firm and product, provide guidance for future consumption scenarios, and assist in making judgments and decisions (Li et al., 2010; Pan and Zhang, 2011). This study used nine question items in three sub-constructs (perceived source credibility, perceived content diagnosticity and perceived vicarious expression) to measure OPR helpfulness. OPR credibility refers to the degree to which consumers perceive that the OPR content is true, valid and believable (Cheung et al., 2009; Qiu et al., 2012). This study designed four question items to measure OPR credibility. A total of 13 items of OPR helpfulness and OPR credibility were measured via responses to questionnaires using a seven-point Likert scale, ranging from 1 (strongly disagree) to 7 (strongly agree). Purchase intention is defined as the likelihood that a consumer will dine in this restaurant (Dodds et al., 1991). Three items were measured using a seven-point Likert scale, ranging from 1 (very low probability) to 7 (very high probability). Question content and related references are provided in the Appendix.

3.3.3 Experimental procedure. The implementation of this experiment and the collection of opinions were all performed online. Invitations to participate in the experiment were posted on various community websites with the aim of recruiting participants who had previously read restaurant reviews. Participants arbitrarily clicked on one hyperlink, which led to one of four scenarios, thereby initiating the experiment process (see Figure 1).

Step 1 Explanation of cooperation and

experiment

Step 2 Select URL

Step 3

60s 10s

1. Monetary sponsorship 2. Experiential sponsorship 4. Non-disclosure3. Dual sponsorship

Read review

Step 4 Disclosure of

sponsorship type

Step 5 Questionnaire

Step 6 provide

information for raffle

Figure 1. The comprehensive

experimental procedure in the Study 1

201

Ethics in social media marketing

Of special note is the manipulation of the disclosure statements provided each of the review scenarios in Step 4: experiential sponsorship (this review was solicited by the restaurant in exchange for a free meal); monetary sponsorship (this review was solicited by the restaurant in exchange for financial remuneration); dual sponsorship (this review was solicited by the restaurant in exchange for a free meal and financial remuneration. The control group did not receive a disclosure statement.

3.4 Results and discussion 3.4.1 Manipulation check and hypothesis testing. The validity of the samples was verified by manipulating two of the questions: first, confirming the type of sponsorship; and second, a negatively worded item related to attitude. This resulted in 160 valid samples. Cronbach’s α of OPR helpfulness, OPR credibility and purchase intention were 0.972, 0.971 and 0.931, respectively. This indicates a high degree of internal consistency among the four dependent constructs (Wortzel, 1979). These results demonstrate the success of the manipulations in this experiment and the suitability of the resulting experiment data for subsequent analysis.

The hypothesis was tested using one-way analysis of variance. According to the literature, gender and reading behavior (number and types of blog read each month) could influence OPR persuasiveness. Thus, we included both of these items as covariables in the model (InsightXplorer, 2014). As shown in Table I, the main effect of type of disclosed sponsorship revealed significant differences in OPR helpfulness (F(3, 153)¼ 58.643, po0.001), OPR credibility (F(3, 153)¼ 73.269, po0.001) and purchase intention (F(3, 153)¼ 27.360, po0.001). This is a clear indication that the type of disclosed sponsorship influences the persuasiveness of OPR.

Scheffe’s procedure is the most popular post hoc procedure, providing a high degree of flexibility and conservative results. Scheffe’s test was performed to further evaluate the hypothesis and explore differences between the mean of non-disclosure and the mean values of the other three types of sponsorship.

As shown in Table II, the mean of OPR helpfulness resulting from non-disclosure was significantly higher than those resulting from monetary sponsorship (Meannon-dis.¼ 4.969, Meanmonetary¼ 2.539, po0.001) and dual sponsorship (Meannon-dis.¼ 4.969, Meandual¼3.161, po0.001). However, we did not observe a significant difference between the mean non-disclosure values and those of experiential sponsorship (Meannon-dis.¼4.969, Meanexperiential¼ 4.629, po0.05). Thus, H1a is supported. The mean of OPR credibility resulting from non-disclosure was significantly higher than those of monetary sponsorship (Meannon-dis.¼ 5.981, Meanmonetary¼ 2.831, po0.001) and dual sponsorship (Meannon- dis.¼ 5.981, Meandual¼ 3.100, po0.001). Again, we did not observe a significant difference between the mean non-disclosure values and those of experiential sponsorship (Meannon- dis.¼ 5.981, Meanexperiential¼ 5.613, pW0.05). Thus,H1b is supported. The mean of purchase intention resulting from non-disclosure was significantly higher than those resulting from monetary sponsorship (Meannon-dis.¼ 5.383, Meabmonetary¼ 3.075, po0.001) and dual

OPR helpfulness OPR credibility Purchase intention Source/dependent variable F p F p F p

Gender 0.520 0.472 0.748 0.388 0.103 0.749 Number of blogs read 5.042 0.026 3.198 0.076 3.659 0.058 Type of blogs read 0.511 0.459 0.567 0.453 2.196 0.140 Type of sponsorship disclosed 58.643 0.000*** 73.269 0.000*** 27.360 0.000** Notes: *po0.05; **po0.01; ***po0.001

Table I. ANCOVA results

202

AJIM 71,2

sponsorship (Meannon-dis.¼ 5.383, Meandual¼ 3.592, po0.001). Once again, we did not observe a significant difference between the mean non-disclosure values and those of experiential sponsorship (Meannon-dis.¼ 5.383, Meanexperiential¼ 4.942, po0.548). Consequently, H1c is supported.

3.4.2 Discussion. Our objective in Study 1 was to identify the best type of sponsorship in cases where disclosure is promoted or expected. Experiential sponsorship was shown to be the best method of sponsorship due to the fact that the persuasiveness of these reviews does not diminish, even after being disclosed. Another significant finding of this study was that any type of sponsorship involving money had a negative influence on OPR persuasiveness, and the effects were most pronounced when using pure monetary sponsorship. The addition of experiential sponsorship (i.e. dual sponsorship) somewhat mitigated the negative effects of monetary rewards. This is a clear indication that in OPR communication, the experience users have with a product/service plays a crucial role in the positive conversion of communication or persuasiveness.

4. Study 2: application of online community influence on OPR management Opinion leaders and professional critics are generally experts in their fields; therefore, their knowledge, opinions or actions are valued highly by others (Fenner, 2012). Individuals in this position can have a profound impact (direct or indirect) on the purchase decisions of consumers, particularly when it comes to experience goods, restaurants, movies and tourist activities (Tsao, Hsieh, Shih and Lin, 2015; Tsao, Mau, Hsieh and Chen, 2015; Tsao, 2014). Their reviews also tend to receive strong responses from followers (Fenner, 2012; Tsao, 2014). Response indicators include ratings, the number of page views, comments, shares and awards. These accomplishments are commonly displayed on web pages as an indirect of the influence the reviewer has in the online community (Lu et al., 2014). These indicators can add to the fame and reputation of the reviewer and further influence the decisions and intentions of the followers (Pixnet, 2014). Study 2 extended the analysis in study 1 to include the influence of the online community in the research framework with the aim of finding an advantageous approach to sponsoring opinion leaders under the premise of full disclosure.

4.1 Influence of online communities The internet serves as a nexus of interpersonal relationships (Ridings and Gefen, 2004). Online communities have become prominent platforms allowing novices and experts alike to

Control group Sponsorship type Mean Mean difference (I–J) p-value

OPR helpfulness Non-disclosure (I) (M¼ 4.969) Experiential sponsorship ( J) 4.629 0.340 0.480

Monetary sponsorship ( J) 2.539 2.431 0.000*** Dual sponsorship ( J) 3.161 1.808 0.000***

OPR credibility Non-disclosure (I) (M¼ 5.981) Experiential sponsorship ( J) 5.613 0.369 0.615

Monetary sponsorship ( J) 2.831 3.150 0.000*** Dual sponsorship ( J) 3.100 2.881 0.000***

Purchase intention Non-disclosure (I) (M¼ 5.383) Experiential sponsorship ( J) 4.942 0.442 0.548

Monetary sponsorship ( J) 3.075 2.308 0.000*** Dual sponsorship ( J) 3.592 1.792 0.000***

Notes: *po0.05; **po0.01; ***po0.001

Table II. Scheffe’ test

results for OPR persuasiveness

203

Ethics in social media marketing

share information in an informal setting. The sharing of information had a profound impact on the way that online consumers collect product information and make purchase decisions (Lee et al., 2003; Pollster, 2012). The influence of prominent online community members has not gone unnoticed in the corporate world. Prominent internet personalities are increasingly being used by corporations to guide the behavioral intentions of those within their sphere of influence (Kozinets et al., 2010).

4.2 Research hypotheses Social influence refers to the altering of one’s opinions in response to the opinions of others. Researchers have defined two categories of influence: normative and informational (Cheung et al., 2009). Normative influence is the pressure of submission that an individual feels when wanting to meet the expectations of a group, such as agreeing with a reviewer’s recommendation or rating of a product. As can be seen, social influence takes place in community aggregations where OPRs exist. The indicators of response that reviewers display can also be considered a source of normative influence.

Many online consumers also share consumption information with others on their own websites or blogs (Hwang and Jeong, 2016). Bloggers often describe the characteristics and functions of a product based on personal experience and even rate the product as a consumer reference. They may be driven by an altruistic desire to help others and/or a selfish desire to gain attention from advertisers or other corporate entities (Li and Du, 2011). Many bloggers with strong community influence make an effort to provide professional content to assist readers in making purchase decisions. They also work hard to interact with their followers to build their credibility and the value of their OPRs. Websites that have been operating for an extended period are more easily found, and their articles are widely disseminated. This means that they have a greater chance of influencing consumer decisions. This study, therefore, infers that the OPRs of bloggers with greater influence in an online community possess stronger credibility and stronger influence in the purchase intention of their readers. Accordingly, the hypothesis is proposed:

H2. Influence in an online community has a positive impact on (a) OPR helpfulness, (b) OPR credibility and (c) purchase intention.

In evaluating the value of OPRs, consumers often take into account whether the reviewer has actually used the product or service in question. This kind of experience is valued far more highly than simply reposting the experiences of others or providing a simple description (Lin et al., 2011; Stephen et al., 2012). Experiential sponsorship is an incentive provided by a firm; however, many people regard free products of this kind as a form of trial sample, similar to those provided to consumers. As a result, consumer perceptions of experiential sponsorship differ somewhat from their perceptions of monetary sponsorship. The source of this difference lies in whether the description in the OPR is based on the personal experience of the reviewer, which is viewed as more authentic (Verlegh et al., 2013; Carr and Hayes, 2014).

The operation of blogs has recently been changing. In the past, many bloggers positioned their platforms as venues for the sharing of opinions on products and brands; however, this often turned into aimless chats in the comment sections. Popular fashion bloggers are increasingly establishing themselves as tastemakers, which involves strategically using their sense of taste to gain followers and their opinions to influence them. Many popular bloggers place a monetary value on their taste/opinions, and view free products as fair exchange for the service they provide (Fenner, 2012). Thus, the study believes that strong community influence can have a direct impact on the persuasiveness of OPRs (as asserted in H2). This study also believes that the positive influence of

204

AJIM 71,2

non-disclosure and experiential sponsorship on OPR persuasiveness mentioned in H1 will be augmented. Hence, the hypothesis is suggested:

H3. The influence of bloggers over an online community strengthens the positive influence of non-disclosure and disclosing experiential sponsorship on (a) OPR helpfulness, (b) OPR credibility, and (c) the purchase intention of readers.

4.3 Methods 4.3.1 Design and participants. Again, we adopted a quasi-experimental design for Study 2. A field experiment was similarly adopted, for which participants were openly recruited on a community website. In this study, we manipulated two factors: sponsorship type and the influence of online communities. The hypotheses were tested by conducting a 4 (sponsorship type: monetary/experiential/dual/non-disclosure) × 2 (online community influence: high/low) between-subjects factorial design. The participants selected one hyperlink from among eight, each of which led participants to one of eight experiment scenarios (i.e. non-random assignment).

All of the participants had previously read online reviews of restaurants. Invitations to participate in the experiment were posted on various community website platforms. A total of 357 participants completed the experiment, which resulted in 320 valid samples. The statistics of the participants were as follows. Males (45.9 percent), between the ages of 21 and 30 (79.1 percent), students (39.1 percent), working in the service industry (25.0 percent). The participants reported reading blogs every month as follows: 6–10 times per month (54.1 percent) and 1–5 times per month (27.8 percent). The most common types of blog were travel blogs (20.3 percent) and restaurant blogs (20.0 percent).

4.3.2 Development of stimuli and measures. In Study 2, we again employed the article that was used in Study 1. Again, the article mentioned the type of sponsorship that was disclosed; however, it also included the influence of online communities, which was defined as the degree of activity and searchability in the community website as perceived by consumers. Activity was measured in terms of update frequency, the number of page views and diffusion capacity. Searchability indicated the ease with which the website could be found and accessed (Lee et al., 2003; Pixnet, 2014). We then created profiles for two virtual bloggers who were tasked with manipulating the degree of online community influence, which was divided into high and low. To verify that the manipulations were suitable, we also designed four question items for this construct, based on a seven-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree) (see Appendix). The definition and measurement of OPR persuasiveness were the same as those in Study 1.

4.3.3 Experiment procedure. As shown in Figure 2, the experimental procedure of Study 2 was the same as that of Study 1. Note the manipulation of the “influence of the blogger on the online community” in Step 3. The situation involving high influence in the online community was illustrated by showing participants (in the blogger introduction) that the blog had been operating for 3,967 days, had 3,325 subscribers and 569 friends, included

Step 1 Explanation of cooperation and

experiment

Step 2 Select URL

Step 3

10s 10s60s

1. High 1. Monetary sponsorship 3. Dual sponsorship 4. Non-disclosure

2. Experiential sponsorship 2. Low

Online community influence of

blogger

Step 4 Read

review

Step 5 Disclosure of sponsorship

type

Step 6 Questionnaire

Step 7 Provide

information for raffle

Figure 2. The comprehensive

experimental procedure in the study 2

205

Ethics in social media marketing

3,158 articles, was ranked No. 1 in the cuisine category, had won seven medals, and had previously been visited 36,195,263 times. The situation involving low influence in the online community was illustrated by showing participants (in the blogger introduction) that the blog had been operating for 797 days, had 3 subscribers, 0 friends, 5 articles, 56 visits and no medals for accomplishments. The manipulation of sponsorship disclosure appears in Step 5 (Figure 2). In accordance with the scenario, the participants were shown the following sentences below the review: experiential sponsorship (this review was solicited by the restaurant in exchange for a free meal); monetary sponsorship (this review was solicited by the restaurant in exchange for financial remuneration); dual sponsorship (this review was solicited by the restaurant in exchange for a free meal and financial remuneration. The control group did not receive a disclosure statement.

4.4 Results and discussion 4.4.1 Manipulation check. Preliminary screening left 320 valid participants who correctly answered the question item related to the type of sponsorship. Cronbach’s α values for the four constructs were as follows: OPR helpfulness (0.964), OPR credibility (0.909), purchase intention (0.935) and influence in online community (0.982), indicating a high degree of internal consistency (Wortzel, 1979). With regard to the manipulation check for online community influence, the results indicated that the mean given to online community influence by the participants in the high online community influence group (160 participants) was significantly higher than those given by participants in the low online community influence group (160 participants) (Meanhigh¼ 5.968, Meanlow¼ 2.057; t¼ 31.554***; po0.001). These results indicate that manipulation of the two factors in this experiment was successful and found to be effective in the following study.

The two hypotheses were tested using 2× 2 analysis of covariance. As shown in Table III, the main effect of the type of disclosed sponsorship revealed significant differences in OPR helpfulness (F(3, 309)¼ 33.677, po0.001), OPR credibility (F(3, 309)¼ 36.613, po0.001), and purchase intention (F(3, 309)¼ 416.891, po0.001), The results for the main effects of the type of disclosed sponsorship on these three dependent variables were the same as those in the study 1. This demonstrates that the designs of the two studies provided sufficient stability and reliability.

4.4.2 Hypothesis 2 testing. Table III shows that the main effect of influence on the online community showed significant differences in OPR helpfulness (F(1, 309)¼ 117.492, po0.001), OPR credibility (F(1, 309)¼ 97.849, po0.001), and purchase intention (F(1, 309)¼ 66.475, po0.001). Further mean difference testing revealed significantly higher scores for these three variables in the group that encountered high influence in the online community, compared to the group that encountered low influence (see Table IV ). Thus,H2a–H2c are all supported. As indicated in the test results above, online community influence has a direct positive influence on OPR helpfulness, OPR credibility and purchase intention.

OPR helpfulness OPR credibility Purchase intention Source/dependent variable F p F p F p

Gender 1.470 0.226 2.414 0.121 0.026 0.873 Number of blogs read 3.051 0.082 1.887 0.171 4.655 0.032* Type of blogs read 3.109 0.079 1.717 0.191 0.007 0.934 Type of sponsorship disclosed (A) 33.667 0.000*** 36.613 0.000*** 416.891 0.000*** Online community influence (B) 117.492 0.000*** 97.849 0.000*** 66.475 0.000*** A×B 19.170 0.000*** 26.252 0.000*** 10.320 0.000*** Notes: *po0.05; **po0.01; ***po0.001

Table III. Two-way ANCOVA test results

206

AJIM 71,2

4.4.3 H3 testing. The interaction effects between the type of sponsorship disclosed and online community influence on OPR persuasiveness were examined. Table III shows that the interaction effect between the type of sponsorship disclosed and online community influence had a significant influence on OPR helpfulness (F(3, 309)¼ 19.170, po0.001), OPR credibility (F(3, 309)¼ 26.252, po0.001), and purchase intention (F(3, 309)¼ 10.320, po0.001). Comprehensive graphical analysis and mean testing was then performed.

As shown in Table V and Figure 3, the influence of experiential sponsorship (Meanexperiential, high¼ 4.629, Meanexperiential, low¼ 2.747, t¼ 10.832, po0.001), dual sponsorship (Meandual, high¼ 3.161, Meandual, low¼ 2.472, t¼ 2.674, po0.05) and non-disclosure (Meannon-dis., high¼ 4.969, Meannon-dis.,low¼ 2.767, t¼ 13.531, po0.001) on OPR helpfulness was significantly strengthened by high online community influence. However, this phenomenon was not apparent with monetary sponsorship (Meanmonetary,

high¼ 2.539, Meanmonetary, low¼ 2.438, t¼ 0.348, pW0.05). Hence, H3a is partially supported. As shown in Table V and Figure 4, the influence of experiential sponsorship

(Meanexperiential, high¼ 5.613, Meanexperiential, low¼ 3.094, t¼ 11.057, po0.001) and

OPR helpfulness OPR credibility Purchase intention Independent variable Level Mean SD Mean SD Mean SD

Online community influence High 3.825 1.390 4.381 1.876 4.248 1.643 Low 2.606 1.072 2.945 1.386 3.025 1.362

Mean difference 1.219 1.436 1.223

Independent-sample t-test (p-value) 7.786*** (0.000) 8.414*** (0.000) 7.248*** (0.000) Notes: *po0.05; **po0.01; ***po0.001

Table IV. The mean OPR

persuasiveness of online community

influence

Factor A Factor B OPR helpfulness OPR credibility Purchase intention Sponsorship type Community influence Mean (SD) Mean (SD) Mean (SD)

Monetary sponsorship High 2.539 (1.384) 2.831 (1.607) 3.075 (1.552) Low 2.438 (1.217) 2.769 (1.418) 2.792 (1.519)

Experiential sponsorship High 4.629 (0.515) 5.613 (0.790) 4.942 (0.515) Low 2.747 (0.971) 3.094 (1.205) 2.900 (0.971)

Dual sponsorship High 3.161 (1.156) 3.100 (1.458) 3.592 (1.497) Low 2.472 (1.148) 2.863 (1.517) 3.083 (1.434)

Non-disclosure High 4.969 (0.457) 5.981 (0.835) 5.383 (1.202) Low 2.767 (0.923) 3.056 (1.411) 3.325 (1.246)

Table V. The mean OPR

persuasiveness of the two factors

Monetary

O PR

s h el

pf ul

ne ss

2.438

2.539 2.747 2.472 2.767

4.969

High

Online community influence

Low

4.629 3.161

Experiential Dual Non

Figure 3. Interaction effect

between sponsorship type and community

influence on OPR helpfulness

207

Ethics in social media marketing

non-disclosure (Meannon-dis., high¼ 5.981, Meannon-dis., low¼ 3.056, t¼ 11.282, po0.001) on OPR credibility was strengthened by high online community influence, but this phenomenon was not apparent with dual sponsorship (Meandual, high¼ 3.100, Meandual, low¼ 2.863, t¼ 0.714, pW0.05) or monetary sponsorship (Meanmonetary, high¼ 2.831, Meanmonetary, low¼ 2.769, t¼ 0.184, pW0.05). As a result, H3b is supported.

As shown in Table V and Figure 5, the influence of experiential sponsorship (Meanexperiential, high¼ 4.942, Meanexperiential, low¼ 2.900, t¼ 7.823, po0.001) and non- disclosure (Meannon-dis., high¼ 5.383, Meannon-dis., low¼ 3.325, t¼ 7.519, po0.001) on purchase intention is strengthened by high online community influence. Similarly, this phenomenon was not apparent with dual sponsorship (Meandual, high¼ 3.592, Meandual, low¼ 3.083, t¼ 1.551, pW0.05) or monetary sponsorship (Meanmonetary, high¼ 3.075, Meanmonetary, low¼ 2.792, t¼ 0.825, pW0.05). Consequently, H3c is supported.

4.4.4 Discussion. The establishment of H2a–H2c shows that a positive relationship exists between online community influence and OPR persuasiveness, which means that if a reviewer’s website has higher update frequencies, page views and diffusion capacity and can easily be found via search engines, then it will enhance the reviewer’s influence on reader perceptions of their OPRs and the subsequent behavioral intentions of the readers. Furthermore, the support or partial support of H3a–H3c indicates that online community influence plays a strengthening role on persuasiveness with non-disclosure and experiential sponsorship. On the premise that the sponsorship information is disclosed, then offering experiential sponsorship to reviewers with high community influence will result in the best OPR persuasiveness.

5. Discussion and conclusions 5.1 Conclusions In response to the importance of marketing ethics, this study conducted two experiments to clarify the differences between eWOM and advertising, and to disclose the sponsorship type. This study proposed a means by which incentivized OPRs can be turned into

Monetary

O PR

s c re

di bi

lit y

2.769 2.863

2.831

3.094

3.100

3.056

5.981 5.613

High

Online community influence

Low

Experiential Dual Non

Figure 4. Interaction effect between sponsorship type and community influence on OPR credibility

Monetary

Pu rc

ha se

in te

nt io

2.792 2.900

3.075

3.083 3.325

5.383

3.592 4.942

High

Online community influence

Low

Experiential Dual Non

Figure 5. Interaction effect between sponsorship type and community influence on purchase intention

208

AJIM 71,2

testimonial advertising and provide the same communication effects. The results of this study could assist firms in formulating morally acceptable communication strategies in accordance with current internet trends.

The results of Study 1 indicate that disclosing information on sponsorship involving money (purely monetary or dual sponsorship) has an adverse impact while disclosing that the review received experiential sponsorship did not have a significant negative effect. For this reason, experiential sponsorship is the best choice for sponsorship disclosure. This study verified that the personal experiences of the reviewer regarding the product are crucial to whether consumers accept and believe the review (Lin et al., 2011). Regardless of the financial incentives that a free product implies, recommendations made by reviewers that have actually experienced using the product seem to make consumers more understanding of the firm’s motives behind the sponsorship than those that were simply paid for with money.

Study 1 found that monetary rewards have an adverse impact on the helpfulness and credibility of the review, especially the latter. This supports the findings of previous studies (Tsao, Hsieh, Shih and Lin, 2015; Tsao, Mau, Hsieh and Chen, 2015; Stephen et al., 2012). However, unlike previous research, this study used non-disclosure as a control group to compare with the three types of sponsorship one-by-one and confirmed that experiential sponsorship is the most beneficial for honest disclosure. The persuasiveness of this type of sponsorship is close to that of non-disclosure, which makes a compelling reason for firms to engage in honest disclosure. This study also proposed three types of financial and non-financial sponsorship types that could explain why Lu et al. (2014) found that the influence of sponsorship type on consumer attitudes toward sponsored recommendations were not significant in their study, i.e. they only considered financial incentives. Furthermore, the studies conducted by Stephen et al. (2012), Lu et al. (2014) and Hwang and Jeong (2016) only used single variables to measure the effects or persuasiveness of reviews. In contrast, this study interpreted OPR persuasiveness using three constructs: helpfulness, credibility and purchase intention, which gives a more comprehensive understanding of the effects of disclosure. Furthermore, while Kim and Song (2018) mentioned sponsorship type in their study, they only addressed whether the sponsorship was paid or unpaid and overlooked non-monetary sponsorship methods. These are what set this study apart from past studies regarding its uniqueness, and contributions.

The results of Study 2 indicate that consumers perceived higher degrees of OPR helpfulness and OPR credibility in the articles written by bloggers with strong influence in the online community compared to those written by bloggers with less influence. Their purchase intention is significantly increased as well. Sponsored cooperation with influential reviewers is, thus, likely to have greater persuasiveness. The interaction effects also revealed that the influence of non-disclosure and the disclosure of experiential sponsorship on persuasiveness are strengthened by the online community influence of the endorser. A deeper implication is that offering experiential sponsorship to reviewers with strong influence in the online community produces the best disclosure effects. However, if the sponsorship involves money, influence in the online community does not have any positive effects.

Blogs allow the individual to express their opinions and feelings regarding people, matters and things. With their distinctive styles in writing and photography, they vary in charm or appeal (Leshed and Kaye, 2006; Li and Du, 2011). Thus, the personal traits of the bloggers are also crucial to whether the disclosure of sponsorship information influences their reviews. With the exception of Tsao and Hsieh (2015) in their discussion on the types of eWOM platforms, no other researcher has examined the community influence of reviewers on sponsorship disclosure. Hwang and Jeong (2016) discovered that skepticism interferes with the influence of the type of sponsorship disclosed on source credibility. In all fairness, skepticism is a personal trait of the consumer and cannot be controlled by

209

Ethics in social media marketing

reviewers or firms. However, firms can choose the reviewers they want to sponsor, so the findings of this study are still feasible in practice. This also gives this study more practical value compared to previous studies.

5.2 Marketing implications and contributions The implications of our findings in this study are two-fold: first, our primary objective was to minimize the negative effect of disclosing information pertaining to sponsorship. It appears that exchanging free products for consumer reviews is the sponsorship method that is most likely to be accepted by consumers. Free products provide an indirect financial benefit to the recipient; however, consumption psychology has revealed that consumers are willing to overlook this if the items are ostensibly means as a trial sample (Lin et al., 2011). In the minds of consumers, this is sufficient to retain the personal experience of eWOM and dilute the commercial overtones. Second, this study discovered that the influence of reviewers in the online community moderates the effect of disclosure. This study also determined that inviting popular bloggers to experience products free of charge before endorsing them is the most effective method of disclosing sponsorship information. This approach makes it possible for firms to provide bloggers with detailed product information, thereby enhancing the quality and helpfulness of the review. More importantly, consumers infer from this a willingness on the part of the firm to make their sponsorship public. This implies that the firm has confidence in their product, which makes the consumers less suspicious of the sponsorship process. These effects are particularly pronounced when the endorsements are from influential bloggers.

The most notable contribution of this study is our provision of an effective method for the management of OPRs, which abides by ethical marketing standards. It is hoped that our results in this study will provide firms with a rational incentive to follow ethical marketing standards and the law. This could create a win-win-win situation for firms, consumers and reviewers, rather than forcing them to operate in the gray zone by manipulating OPRs and potentially cheating consumers. Finally, this study fills previous gaps in the research on the relationship between eWOM/OPRs and ethics. It also positions the issue of ethical business practices as an issue of importance issue in the field of eWOM.

5.3 Limitations and directions for further research Despite the rigor of the experiment method, it was not without limitations. First, based on the manipulation requirements of the experimental design, blogs were the only platform explored in this study, i.e. we did not take into account other types of community website. Second, the experiments in this study were conducted via the internet, which proved highly efficient but raises a number of concerns in terms of effectiveness. Our primary concern is that many of the respondents may have used cell phones or tablets to complete our questionnaire, which would have raised browsing issues that could affect the test results. To increase the validity of our results, we offered incentives to gain genuine cooperation. Each experiment also included a strict screening process to eliminate invalid samples. Third, due to the difficulty of performing random sampling of internet populations, this study randomly altered the sequence of hyperlinks used in the experiment scenarios with the aim of overcoming preference bias and to make sampling process as random as possible. Fourth, the gender ratio in this study is nearly 1:1, and the main age group was between 20 and 30 years of age. The sample characteristics in this study are compatible with the recent studies based on internet samples (Lu et al., 2014; Tsao, Hsieh, Shih and Lin, 2015; Tsao, Mau, Hsieh and Chen, 2015; Tsao et al., 2016). The 2016 internet user behavior survey in Taiwan found that individuals in the 20–30 year bracket, students and service employees are the groups with the highest network utilization (Taiwan Network Information Center, 2016). This supports the representativeness of the sample in this study.

210

AJIM 71,2

Based on the limitations listed above, this study proposes the following suggestions for future research. First, OPRs could be presented in increasingly diverse ways, such as the currently popular unboxing videos (Mau et al., 2013). Future studies could also include other internet platforms, such as Facebook, Twitter, Weibo or YouTube (Kim and Song, 2018). Second, future studies could examine different types of products and include an in-depth analysis of the relationship between product type and the method of sponsorship (Tsao and Hsieh, 2015). It might also prove effective if researchers were to include experts or endorsers in the research (Kim and Song, 2018) or consider the location in which sponsorship information is disclosed ( Jensen and Yetgin, 2017).

References

Brounen, D., Schweitzer, M. and Cools, T. (2001), “Information transparency pays: evidence from European Property Shares”, Real Estate Finance, Vol. 18 No. 2, pp. 39-49.

Carr, C.T. and Hayes, R.A. (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, Vol. 14 No. 1, pp. 38-50.

Chakraborty, U. and Bhat, S. (2018), “The effects of credible online reviews on brand equity dimensions and its consequence on consumer behavior”, Journal of Promotion Management, Vol. 24 No. 1, pp. 57-82.

Chen, Y.C., Xiao, Y.W. and Lin, Y.C. (2017), “The impact of trust of eWOM on consumer purchase intentions towards tour products of travel social networks: an extended perspective of technology acceptance model”, Journal of Tourism and Leisure Management, Vol. 5 No. 1, pp. 28-41.

Cheung, M.Y., Luo, C., Sia, C.L. and Chen, H. (2009), “Credibility of electronic word-of-mouth: informational and normative determinants of on-line consumer recommendations”, International Journal of Electronic Commerce, Vol. 13 No. 4, pp. 9-38.

Chu, C.W. (2010), “The secrets behind bloggers: to uncover the veil of product placement”, available at: http://handle.ncl.edu.tw/11296/ndltd/57740537957344806460 (accessed January 23, 2016).

Dodds, W.B., Monroe, K.B. and Grewal, D. (1991), “Effect of price, brand and store information on buyer’s product evaluations”, Journal of Marketing Research, Vol. 28 No. 3, pp. 307-319.

Fair Trade Commission (2017), “Fair trade commission's explanation towards testimonial advertisings”, available at: www.ftc.gov.tw/internet/main/doc/docDetail.aspx?uid=165&docid=13021 (accessed April 19, 2017).

Fenner, J. (2012), “How bloggers’ influence can benefit fashion brands”, Apparel Magazine, Vol. 53 No. 5, pp. 4-5.

Floyd, K., Freling, R., Alhoqail, S., Cho, H.Y. and Freling, T. (2014), “How online product reviews affect retail sales: a meta-analysis”, Journal of Retailing, Vol. 90 No. 2, pp. 217-232.

Forrest, E. and Cao, Y. (2010), “Opinions, recommendations and endorsements: the new regulatory framework for social media”, Journal of Business and Policy Research, Vol. 5 No. 2, pp. 88-89.

Fullwood, C., Sheehan, N. and Nicholls, W. (2009), “Blog function revisited: a content analysis of MySpace blogs”, Cyberpsychology & Behavior, Vol. 12 No. 6, pp. 685-689.

Hajli, N. (2018), “Ethical environment in the online communities by information credibility: a social media perspective”, Journal of Business Ethics, Vol. 149 No. 4, pp. 799-810.

Hennig-Thurau, T., Gwinner, K.P., Walsh, G. and Gremler, D.D. (2004), “Electronic word-of-mouth via consumer-opinion platforms: what motivates consumers to articulate themselves on the Internet?”, Journal of Interactive Marketing, Vol. 18 No. 1, pp. 38-52.

Hwang, Y. and Jeong, S.H. (2016), “ ‘This is a sponsored blog post, but all opinions are my own’: the effects of sponsorship disclosure on responses to sponsored blog posts”, Computers in Human Behavior, Vol. 62, pp. 528-535.

211

Ethics in social media marketing

http://handle.ncl.edu.tw/11296/ndltd/57740537957344806460

www.ftc.gov.tw/internet/main/doc/docDetail.aspx?uid=165&docid=13021

InsightXplorer (2014), “Blog survey and observation the habits of blog users”, available at: www.dma. org.tw/upload/ResourceReport/20140416020724842.pdf (accessed April 15, 2015).

Jensen, M.L. and Yetgin, E. (2017), “Prominence and interpretation of online conflict of interest disclosures”, MIS Quarterly, Vol. 41 No. 2, pp. 629-643.

Jonas, J.R.O. (2010), “Source credibility of company-produced and user-generated content on the internet: an exploratory study on the Filipino Youth”, Philippine Management Review, Vol. 17, pp. 121-132.

Kuo, C. (2015), “Change of shopping behavior in Taiwanese consumers: examining the AISAS model online consumer behavior”, Chinese Journal of Communication Research, No. 27, pp. 139-165.

Kim, M. and Song, D. (2018), “When brand-related UGC induces effectiveness on social media: the role of content sponsorship and content type”, International Journal of Advertising, Vol. 37 No. 1, pp. 105-124.

Kozinets, R.V., De Valck, K., Wojnicki, A.C. and Wilner, S.J. (2010), “Networked narratives: understanding word-of-mouth marketing in online communities”, Journal of Marketing, Vol. 74 No. 2, pp. 71-89.

Kwon, O. and Sung, Y. (2012), “Shifting selves and product reviews: How the effects of product reviews vary depending on the self-views and self-regulatory goals of consumers”, International Journal of Electronic Commerce, Vol. 17 No. 1, pp. 59-82.

Lee, F.S.L., Vogel, D. and Limayem, M. (2003), “Virtual community informatics: a review and research agenda”, Journal of Information Technology Theory and Application, Vol. 5 No. 1, pp. 47-61.

Lee, K.Y. (2016), “Applying gourmet blogs marketing strategy in experiential marketing, perceived value, trust, and purchase intention-case of the Tainan local delicacy”, unpublished thesis, Department of Hospitality Management at the Southern Taiwan University of Science and Technology, Tainan.

Lee, S. and Choeh, J.Y. (2018), “The interactive impact of online word-of-mouth and review helpfulness on box office revenue”, Management Decision, Vol. 56 No. 4, pp. 849-866.

Leshed, G. and Kaye, J.J. (2006), “Understanding how bloggers feel: recognizing affect in blog posts”, CHI’06 extended abstracts on Human Factors in Computing Systems, ACM, Montreal, pp. 1019-1024.

Li, F. and Du, T.C. (2011), “Who is talking? An ontology-based opinion leader identification framework for word-of-mouth marketing in online social blogs”, Decision Support Systems, Vol. 51 No. 1, pp. 190-197.

Li, M., Huang, L., Tan, C.H. and Wei, K.K. (2013), “Helpfulness of online product reviews as seen by consumers: source and content features”, International Journal of Electronic Commerce, Vol. 17 No. 4, pp. 101-136.

Li, M.X., Tan, C.H., Wei, K.K. and Wang, K.L. (2010), “Where to place product review? An information search process perspective”, in Lacity, M., March, S. and Niederman, F. (Eds), Proceedings of the 31st Annual International Conference on Information Systems, ACM, Singapore, p. 60.

Lin, C.Y., Tsai, W.R. and Kuo, W.Y. (2011), “Reality of blogger endorsement article on consumer purchasing behavior”, Proceedings of the 17th Conference on Information Management and Practice, Chia Nan University of Pharmacy and Science, ACM, Tainan, December 10.

Lu, L.C., Chang, W.P. and Chang, H.H. (2014), “Consumer attitudes toward blogger’s sponsored recommendations and purchase intention: the effect of sponsorship type, product type, and brand awareness”, Computers in Human Behavior, Vol. 34, pp. 258-266.

Ma, M. and Agarwal, R. (2007), “Through a glass darkly: information technology design, identity verification, and knowledge contribution in online communities”, Information Systems Research, Vol. 18 No. 1, pp. 42-67.

Mau, G., Steinmann, S., Brückner, R., Schramm-Klein, H. and Schuhen, M. (2013), “Shared first product moments: the impact of unboxing videos on the evaluation of new products”, Proceedings of the AMA Winter Educators’ Conference, pp. 21-22.

212

AJIM 71,2

www.dma.org.tw/upload/ResourceReport/20140416020724842.pdf

Oberhofer, M., Füller, J. and Hofmann, V. (2014), “Tryvertising-what makes consumers share product innovations with others?”, Advances in Consumer Research, Vol. 42, pp. 348-351.

Pan, Y. and Zhang, J.Q. (2011), “Born unequal: a study of the helpfulness of user-generated product reviews”, Journal of Retailing, Vol. 87 No. 4, pp. 598-612.

Park, C. and Lee, T.M. (2009), “Information direction, website reputation and eWOM effect: a moderating role of product type”, Journal of Business Research, Vol. 62 No. 1, pp. 61-67.

Park, D.H. and Kim, S. (2008), “The effects of consumer knowledge on message processing of electronic word-of-mouth via online consumer reviews”, Electronic Commerce Research and Applications, Vol. 7 No. 4, pp. 399-410.

Park, D.H., Lee, J. and Han, I. (2007), “The effect of on-line consumer reviews on consumer purchasing intention: the moderating role of involvement”, International Journal of Electronic Commerce, Vol. 11 No. 4, pp. 125-148.

Peng, K.F., Lai, C.J. and Chen, M.H. (2011), “The effects of blog type and social presence on blog viewers”, Electronic Commerce Studies, Vol. 9 No. 4, pp. 459-482.

Peng, L., Cui, G. and Li, C. (2013), “The comparative impact of critics and consumers applying the generalisability theory to online movie ratings”, International Journal of Market Research, Vol. 55 No. 3, pp. 413-436.

Pimperl, A., Schulte, T., Mühlbacher, A., Rosenmöller, M., Busse, R., Groene, O., Rodriguez, H.P. and Hildebrandt, H. (2017), “Evaluating the impact of an accountable care organization on population health: the quasi-experimental design of the German Gesundes Kinzigtal”, Population Health Management, Vol. 20 No. 3, pp. 239-248.

Pixnet (2014), “GOLDDOT”, available at: http://2014golddot.events.pixnet.net/blogger.php (accessed April 6, 2016).

Pollster (2012), “FB fan group is the most effective propaganda channel for the internet (in Chinese)”, available at: www.pollster.com.tw/Aboutlook/lookview_item.aspx?ms_sn=1422 (accessed June 13, 2015).

Qiu, L., Pang, J. and Lim, K.H. (2012), “Effects of conflicting aggregated rating on eWOM review credibility and diagnosticity: the moderating role of review valence”, Decision Support Systems, Vol. 54 No. 1, pp. 631-643.

Ridings, C.M. and Gefen, D. (2004), “Virtual community attraction: why people hang out online”, Journal of Computer-Mediated Communication, Vol. 10 No. 1, available at: https://doi.org/10. 1111/j.1083-6101.2004.tb00229.x

Schneiderman, A.G. (2013), “A.G. Schneiderman announces agreement with 19 companies to stop writing fake online reviews and pay more than $350,000 in fines”, available at: www.ag.ny.gov/ press-release/ag-schneiderman-announces-agreement-19-companies-stop-writing-fake-online- reviews-and (accessed June 22, 2015).

Shen, C.H., Lai, H.J. and Hwang, T.Y. (2017), “The effect of word of mouth from Internet media on business of financial consumer in banking”, Cross-Sreait Banking and Finance, Vol. 5 No. 1, pp. 23-41.

Spears, N. and Singh, S.N. (2004), “Measuring attitude toward the brand and purchase intentions”, Journal of Current Issues & Research in Advertising, Vol. 26 No. 2, pp. 53-66.

Stephen, A.T., Bart, Y., Du Plessis, C. and Goncalves, D. (2012), “Does paying for online product reviews pay off? The effects of monetary incentives on consumers’ product evaluations”, INSEAD Working Papers Collection, Vol. 96, pp. 1-44.

Taiwan Network Information Center (2016), available at: www.twnic.net.tw/download/200307/20160 922e.pdf (accessed 19 April 2017).

Tong, Y., Wang, X. and Teo, H.H. (2007), “Understanding the intention of information contribution to online feedback systems from social exchange and motivation crowding perspectives”, System Sciences, Proceedings of the 40th Hawaii International Conference, p. 28.

Tsao, W.C. (2014), “Which type of online review is more persuasive? The influence of consumer reviews and critic ratings on moviegoers”, Electronic Commerce Research Journal, Vol. 14 No. 4, pp. 559-583.

213

Ethics in social media marketing

http://2014golddot.events.pixnet.net/blogger.php

www.pollster.com.tw/Aboutlook/lookview_item.aspx?ms_sn=1422

https://doi.org/10.1111/j.1083-6101.2004.tb00229.x

www.ag.ny.gov/press-release/ag-schneiderman-announces-agreement-19-companies-stop-writing-fake-online-reviews-and

www.twnic.net.tw/download/200307/20160922e.pdf

Tsao, W.C. and Hsieh, M.T. (2015), “eWOM persuasiveness: do eWOM platforms and product type matter?”, Electronic Commerce Research, Vol. 15 No. 4, pp. 509-541.

Tsao, W.C. and Lin, C.H. (2017), “Exploring the effect of the type of unboxing articles on the helpfulness-In the view of consumers’ purchase involvement”, Ling Tung Journal, No. 42, pp. 57-87.

Tsao, W.C., Hsieh, M.T. and Lin, M.T. (2016), “Intensifying online loyalty! The power of website quality and the perceived value of consumer/seller relationship”, Industrial Management & Data Systems, Vol. 116 No. 9, pp. 1987-2010.

Tsao, W.C., Hsieh, M.T., Shih, L.W. and Lin, T.M. (2015), “Compliance with eWOM: the influence of hotel reviews on booking intention from the perspective of consumer conformity”, International Journal of Hospitality Management, Vol. 46, pp. 99-111.

Tsao, W.C., Mau, T.C., Hsieh, M.H. and Chen, J.J. (2015), “Honesty or cheat? The impact of the disclosure of sponsorship information on consumer behavior”, Proceedings of the 16th Electronic Business Management Conference, Da-Yeh University of Information Management, Changhua, May 29.

Verlegh, P.W., Ryu, G., Tuk, M.A. and Feick, L. (2013), “Receiver responses to rewarded referrals: the motive inferences framework”, Journal of the Academy of Marketing Science, Vol. 41 No. 6, pp. 669-682.

Wang, Z.H. (2017), “The relationships among travel blog characteristics, travel blog browsing experiences, experience satisfaction, and travel intention”, unpublished thesis, Department of Business Administration at the Tamkang University, New Taipei City.

Wortzel, L. (1979), Multivariate Analysis, Prentice Hall, New Jersey, NJ.

Wu, W.E. and Lin, C.H. (2001), Business Research Methods, Wha-Tai Publishing, Taipei.

Yu, S.C. (2017), “Mining aboriginal cultural tourism on the internet information using destination image: a study of blogs for Fu-Hsing District, Taoyuan City”, Journal of National United University, Vol. 14 No. 1, pp. 29-42.

Zhang, J.Q., Craciun, G. and Shin, D. (2010), “When does electronic word-of-mouth matter? A study of consumer product reviews”, Journal of Business Research, Vol. 63 No. 12, pp. 1336-1341.