The Impact of Intrinsic and Extrinsic Motivation on
Online Consumer Review Content
Master’s Thesis
Astrid Kappesser Student Number: 10696970
Graduate School of Communication Master’s program Communication Science
Track Persuasive Communication Dr. ir. P.W.J. Verlegh
Abstract
To reward or not to reward? This constitutes one essential question in current electronic word of mouth (eWOM) developments. Marketers increasingly offer consumers (economic) incen-tives in return for evaluating products and services. This has become a popular practice as recommendations by other consumers, for instance, have a stronger persuasive impact than traditional advertising. However, whether this strategy leads to the intended objective that consumers in fact write reviews that are perceived as helpful and credible, and thereby lead to the desired positive persuasive effect, is questionable. Relating to self-determination theory, the present study posits that rewarding consumers for writing a review results in formulating less elaborate and more extreme reviews. Additionally, the study considered a possible mod-erating effect of the review’s subject matter. An online experiment with 171 participants has been conducted to assess to what extent extrinsically motivated consumers differ from intrin-sically motivated ones in terms of formulating an online review. Results showed no signifi-cant differences for type of motivation, implying that rewarding consumers do not have a negative impact on online review content. Whether a product or service is reviewed was found to only have a minor influence and should be addressed further in future research. This paper contributes to eWOM literature by examining the relationship between motivation and online review content out of the sender’s perspective. Moreover, the present study adds to this specific field of research as it goes beyond service contexts and additionally relates to prod-ucts as type of review subject matter. Lastly, by analyzing the effectiveness of stimulating consumers for writing online reviews, this study provides useful insight for marketers.
Keywords: electronic word of mouth (eWOM), online review, motivation, intrinsic, extrinsic, reward, online review content, product, service, eWOM sender
Introduction
Due to the emergence of the internet, consumers are no longer bound to pass on their experiences with products and services within the limits of personal face-to-face communica-tion. They have the possibility to share their opinions online and make them available to a wide range of other consumers. Consumers may share their opinion on different platforms such as retailer websites (e.g. Amazon), independent review websites (e.g. Ciao, Yelp), per-sonal blogs, or directly on a brand’s website, and make their experiences visible to others who seek for information about a specific product or service. But why do people contribute to the-se platforms and share their opinion, what do they write in their reviews, and how does this affect other consumers? To gain more knowledge about these phenomena, recent research is increasingly focusing on word of mouth within computer-mediated communication. In partic-ular, as this so-called electronic word of mouth (eWOM) plays a crucial role in consumer’s decision-making processes as it is suggested to have a strong persuasive impact. This is due to consumer’s tendency to trust consumer-generated information more than marketer-generated information as the former is perceived as a non-commercial source and thus as more credible and unbiased.
While much research has focused on the receiver, the present study is interested in ex-amining eWOM from the sender’s point of view, as well as the content they produce. There-by, the present study follows Berger’s call to “more deeply examine how people say what they say” (2014, p. 30), and Sweeney, Soutar and Mazzarol’s (2012) demand to explore mes-sage details within online word of mouth. In particular, it is examined to what extent motiva-tions play a role on how consumers formulate an online review. This topic has scarcely been addressed in eWOM research even though different motivations lead to differences in online review content, and how a consumer describes a product or service in a review in turn has an impact on consumers who read the evaluation (Schellekens, Verlegh, & Smidts, 2010; Scholz & Dorner, 2013; Sweeney, Soutar, & Mazzarol, 2012; Willemsen, Neijens, Bronner, & de Ridder, 2011).
In terms of motivation, the present study focuses on differences between reviews that are written out of an intrinsic motivation, that means consumers are inherently interested to share their opinion with others, and reviews written out of an extrinsic motivation, that is to say they write the review because it results in a desired outcome, e.g. obtaining a reward for writing the review (Ryan & Deci, 2000). This is interesting for three reasons. First, receiving economic incentives is one of the major motives why consumers state their opinion about a product or service online (Hennig-Thurau, Gwinner, Walsh, & Gremler, 2004). Secondly,
stimulating consumers to write reviews by offering an incentive is getting a popular practice and review referral reward programs offered by marketers are increasingly used (Verlegh, Ryu, Tuk, & Feick, 2013; Xiao, Tang, & Wirtz, 2011).Therefore, consumers share their opin-ion on a product or service not only because they inherently tend to do so, but because they are stimulated by an external reward. Third, to what extent extrinsic and intrinsic motivations affect message content has, despite the above mentioned trends, not conclusively been ex-plored. But how effective is it, when marketers try to stimulate reviews by giving out re-wards? And does this strategy influence the way in which consumers write their reviews?
In pursuing these questions, this paper reacts on current eWOM developments and tries to shed light on the impact of economic incentives on online consumer reviews from the sender’s perspective. For this purpose, the study draws on self-determination theory to explain how intrinsic and extrinsic motivation might affect online review content. Moreover, as for-mer studies have focused on reviews in a service context, this paper extends eWOM research by additionally addressing products and examining a possible interaction effect of a review’s subject matter (see Figure 1). To assess the effect of motivation on the formulation of online consumer reviews in conjunction with subject matter, an experiment, using a 2 x 2 between-subject-design, was conducted.
Figure 1: Conceptual model:
Type of Motivation (intrinsic/extrinsic) Online Review Content Argumentation Review Subject Matter (product/service)
Logos, Ethos, Pathos Personal Experience
Detail Length Extremity
Theoretical Framework
The present study examines to what extent intrinsic and extrinsic motivations influence how reviewers formulate an online review about a product or service. First, a short overview of the research about electronic Word of Mouth is given. Secondly, the types of motivation, which are relevant for this study, are described and it is examined why they might impact consum-ers’ formulation of an online review. Afterwards, it is outlined to what extent specific content aspects of a review are assumed to be influenced by the type of motivation. Lastly, a moderat-ing role of the review’s subject matter, whether a product or service is reviewed, is discussed.
The Impact of Electronic Word of Mouth
Electronic Word of Mouth is conceptualized by Hennig-Thurau et al. as “any positive or negative statement made by potential, actual, or former customers about a product or com-pany, which is made available to a multitude of people and institutions via the Internet” (2004, p. 39). Statements about products or companies may for instance occur on online dis-cussion forums, social networking sites, consumer review sites, blogs, brand websites, or re-tailers’ websites (Hennig-Thurau et al., 2004; Lee & Youn, 2009). Online consumer-written reviews, which are in the focus of this study, constitute a part of eWOM communication (Bronner & de Hoog, 2010).
So far, within the growing body of eWOM literature, research has primarily concen-trated on effects of online reviews on other consumers. Findings within eWOM research show that online reviews have an impact on consumer evaluations such as attitude toward product and brand, and purchase intention (Bickart & Schindler, 2001; Doh & Hwang, 2009; Gupta & Harris, 2010; Lee & Youn, 2009; Zhang, Craciun, & Shin, 2010). Moreover, online reviews may impact the economical performance of a company and have an effect on its sales for in-stance (Chevalier & Mayzlin, 2006; Dellarocas, Zhang, & Awad, 2007; Floyd, Freling, Alhoqail, Cho, & Freling, 2014). Additionally, the valence of a review plays an important role, and for example, negative reviews are supposed to have a greater impact and weight on the reader of a review than positive reviews (Chevalier & Mayzlin, 2006).
But why do consumers post these online reviews in the first place? With regard to the sender’s viewpoint, research shows that consumers engage in electronic Word of Mouth for various reasons. For instance, they write online reviews for self-enhancement, for altruistic purposes and concern for others, for social interaction, or for receiving an incentive (King, Racherla, & Bush, 2014). Especially the latter motive is in the focus of this study as marketers increasingly reward consumers for writing a review (Verlegh et al., 2013; Xiao et al., 2011).
However, how this influences online recommendations written by consumers has not suffi-ciently been researched yet.
To what extent does the content of a review written by a consumer who engages in eWOM due to an altruistic interest to share his opinion in order to help others differ from the content of a review written by a consumer who gained an economic incentive to do so? Will their reviews differ in terms of how much they share of their personal experience or how many aspects of the review subject they address? Will it be longer, shorter, more or less ex-treme? As previous research showed, the underlying motive, why consumers write an online review, influences how they write it (Bronner & de Hoog, 2011; Yap, Soetarto, & Sweeney, 2013). Yap, Soetarto, and Sweeney’s study (2013), for example demonstrates that different motivations1 are associated with cognitive and affective characteristics of a message. De Vries’ (2012) results of a content analysis show that whether a consumer is intrinsically or extrinsically motivated to write a review affects the content of the review. Reviews written out of extrinsic motivation had lower argument diversity, provided less factual arguments, included more punctuation marks and were more extreme in terms of rating score.
The review’s content in turn has an impact on the consumer who reads it. For instance, whether a product evaluation is given in a more concrete or abstract way influences a receiv-er’s purchase intention and how a receiver draws inferences about the sendreceiv-er’s product atti-tude (Schellekens, Verlegh, & Smidts, 2010). Furthermore, whether a message shows greater argument quality, has a higher persuasive impact on eWOM receivers’ attitudes and inten-tions (Karmarkar & Tormala, 2010). The cognitive content, richness of content, and strength of delivery of a message have an impact on receivers’ attitude and behavior as well (Sweeney, Soutar, & Mazzarol, 2012).
Moreover, an online review’s content affects whether the reader perceives it to be ra-ther helpful or not what in turn influences the credibility and persuasiveness of the review (Scholz & Dorner, 2013; Willemsen et al., 2011). For marketers it is important that online consumer reviews are perceived as helpful because the perceived helpfulness of a review has an impact on consumers’ purchase decisions (Chen, Dhanasobhon, & Smith, 2008). Thus, in case rewarding consumers for writing online reviews would negatively affect how consumers formulate a review because it might lead to a review consisting of a low degree of credibility and persuasiveness, marketers might want to reconsider offering (monetary) incentives in return for an evaluation. For this purpose, it is important to understand why offering a reward
1
The authors focused on six motivations relating to Hennig-Thurau et al. (2004): positive self-enhancement, social benefits, advice seeking, concern for other consumers, helping the company, and venting negative feelings.
to consumers could result in negative online consumer writing behavior. Motivation literature, in particular self-determination theory, offers a possible explanation for comprehending the underlying mechanism.
Intrinsic and extrinsic motivation
Motives vary in terms of level, that is how much a person is motivated to act, and in terms of orientation, that means the type and nature of the motivation, the underlying goals why a person is motivated to act (Ryan & Deci, 2000). Ryan and Deci (2010) conceptualize within self-determination theory two types of motivations: intrinsic and extrinsic motivation. Intrinsic motivation refers to the notion that people do something because they consider it interesting or enjoyable. Extrinsic motivation on the contrary means that individuals act be-cause their action results in a separable outcome or they want to achieve a certain outcome. Extrinsic motivation is rather triggered by external factors, pressure, or rewards, whereas in-trinsic motivation is rather present due to performing the action itself and the joy or challenge in pursuing the action. For instance, when consumers write a review due to altruistic purposes or concern of others, it can be inferred that they do so out of an intrinsic motivation as this is a voluntary action without expecting something in return (Sundaram, Mitra, & Webster, 1998). In this case that means, consumers are inherently interested to write about a service or product they tested because they want to share their experiences with others in order to help them to make the best decision, or to warn them not to buy a certain product or to test the respective service (Hennig-Thurau et al., 2004). On the contrary, if consumers get an incentive to write the review, for example receiving a financial reward or benefit, they are considered to be ra-ther extrinsically motivated as the motivation to engage in eWOM was initiated by obtaining the reward.
A reward may enhance referral likelihood, especially in increasing referral to weak ties and for weaker brands, as Ryu and Feick (2007) show within their study of referral re-ward programs. In such programs companies aim to encourage consumers to recommend a product or service to others by rewarding them. However, the authors debate further that, with regard to self-perception theory, extrinsic motivated behavior is likely to lead to
overjustification, i.e. intrinsic motivation will be undermined because of the presence of an incentive, which triggers extrinsic motivation. This may ultimately lead to less interest in the activity and to a more negative performance (Deci, Koestner, & Ryan, 1999). Additionally, Ryan and Connell (1989) illustrate that extrinsic motivation negatively affects achievement behavior. Extrinsic motivation resulted in less interest, less value and less effort to conduct
the respective task. Intrinsic motivations, on the contrary, lead to more interest, more enjoy-ment, and more positive coping with the task. With respect to writing an online review, con-sumers who are rewarded to do so might show less interest and put less effort in writing, what should result in a less elaborated and less detailed review. As opposed to consumers who write an online review out of inner interest might provide a more elaborated and more critical review. Moreover, research suggests that intrinsically motivated consumers avoid giving an extreme review as they aim to help and convince others to make a good decision and therefore should provide a more balanced, reasonable, and comprehensible review to ensure the readers consider their review as helpful and not as extreme or biased, and in turn as less credible and less persuasive (Bronner & de Hoog, 2011; Yap et al., 2013). In contrast, extrinsically moti-vated reviewers are expected not to pay as much attention or care about the review’s content and how the message is perceived. Therefore, they should be more likely to write an extreme review, for instance by using intense language or give an extreme rating. Thus, with respect to online review writing behavior, it is inferred that extrinsically motivated consumers are less interested in how they write the review, put less effort in writing the review, and are more extreme with regard to communicating their opinion about a product or service than consum-ers who are intrinsically motivated. Consumconsum-ers who provide an evaluation of a product or service due to a reward are expected to give a more superficial recommendation, whereas consumers who do so due to their inner interest should focus more on the review’s content and how they elaborate it to help others with their experience.
But what exactly constitutes a ‘good’ and helpful review? Especially, considering that each consumer writes and structures an online review differently and reviews vary from sim-ple and short to long and elaborate ones (Park & Kim, 2008; Pollach, 2006). Reviewers may provide just factual information about a service or product or write a rather emotional story including their personal experience. They may claim they are an expert or say they are green-horn within the product or service category. So which aspects and characteristics of an online review make it more or less elaborate, and what depicts an extreme review? And to what ex-tent are they influenced by intrinsic and extrinsic motivation? Referring to previous (e)WOM literature, particularly regarding (e)WOM content analyses, some aspects can be distinguished which relate to how elaborate and extreme an online review is formulated.
Relating Motivation and eWOM
Within (e)WOM content analyses, researchers analyzed (e)WOM messages regarding charac-teristics such as argumentation quality, valence, Aristotle’s rhetorical tactics logos, pathos,
and ethos, textual elements, or language style (e.g De Vries, 2012; Li & Zhan, Otterbacher, 2011; Willemsen et al., 2011). Based on previous literature, the following aspects regarding online review content are discussed, and the present study will examine to what extent these are assumed to be affected by type of motivation:
Argumentation
Aristotle’s logos, pathos, ethos Personal experience
Detail Extremity Review length
These content aspects are in focus because previous research suggests them to have an effect on the reader and the persuasiveness of a message (e.g. Archak, Ghose, & Ipeirotis, 2011; Bronner & De Hoog, 2011; Li & Zhan, 2011; Kim & Gupta, 2012; Kozinets, de Valck, Wojnicki, & Wilner, 2010; Ludwig et al., 2013; Mudambi & Schuff, 2010; Scholz & Dorner, 2013; Söderlund & Rosengren, 2007; Willemsen et al., 2011). Thus, they constitute important components when it comes to formulating a review as they can make a recommendation of a product or service more or less convincing and influential. With regard to self-determination theory, it is suggested that providing a reward elicits extrinsic motivation what leads to nega-tive achievement behavior. In terms of online review writing behavior the overall assumption is thus that obtaining a reward for writing a recommendation leads to less effort and less in-terest in writing a review. That means that some content characteristics will be less elaborated upon, e.g. argumentation will be less thorough or less factual information about the product or about personal experience is included. Consequently, receiving a reward for writing a review is expected to have a negative impact on the different content aspects discussed in this paper, and eventually on the persuasive qualities of the review. This is because the specified charac-teristics are suggested to require a certain amount of effort and interest on part from the com-municator. Therefore they might likely be affected by motivation; i.e. by how much effort a reviewer aims to put in his review.
Moreover, this study aimed to gain information directly from part of the sender’s per-spective. Therefore it was crucial that such content aspects of a review were selected, which consumers are able to assess and give their opinion on how they formulate a review within the limits of a questionnaire.
Argumentation. Argumentation is a relevant indicator of a review’s elaborateness as a solid argumentation shows that a reviewer put effort and interest in providing a thoroughly supported and balanced evaluation of a product or service. The presence and diversity of ar-guments contribute to a message’s credibility and in turn to its persuasiveness (Willemsen et al., 2011). Furthermore, the higher the argument quality, the higher the persuasive impact on readers’ attitudes and intentions (Karmarkar & Tormala, 2010). Stating positive as well as negative arguments enhances the credibility of a message and influences consumer behavior (Park, Lee, & Han, 2007). When reviewers provide both sides of an argument and critically reflect pros and cons of the review’s subject, this enhances the messages’ persuasion and the likelihood that the reader adopts the reviewer’s opinion (Eisend, 2007). Giving information of both sides is suggested to require more effort and elaboration. Consumers who are extrinsical-ly motivated to formulate a review are, however, posited to show less effort and interest with regard to the formulation of their review and should thus provide less arguments and lower argument diversity. This leads to the following hypothesis:
H1: Intrinsically motivated eWOM providers formulate a more balanced review about a product or service than extrinsically motivated ones.
Aristotle’s logos, ethos, pathos. Aristotle constituted three fundamental communica-tion strategies, which add to the persuasiveness of a message: logos, ethos, pathos (Buttle, 1998). Logos means a communicator tries to persuade his audience based on logical appeals (Buttle, 1998). This relates to the rational and cognitive dimension of a review, referring to reliable and clear information (Sweeney et al., 2012; Yap et al., 2013). A reviewer appeals to the reader’s sense of reason and provides rather factual information about the reviewed sub-ject (Otterbacher, 2011). For example, a consumer, writing a review about a product, may argue that a specific smartphone is better than others by describing distinctive factual features of the phone, comparing them to features of other smartphones. Whether the eWOM sender formulates his review based on reason adds to the credibility and persuasiveness of a review (Otterbacher, 2011; Pollach, 2006). As researching and combining different rational and fac-tual features, is assumed to comprise effort, interest, and thoroughness. Since, according to self-determination theory, extrinsically motivated consumers should have less interest and put less effort into a task, it is posited that they would present less factual information in a review. Intrinsically motivated consumers on the other hand, should be more likely and interested to
formulate a convincing review consisting of verifiable and factual information, as they are ascribed to be more willingly to put effort in writing a review. Additionally, consumers, who are motivated to share their opinion due to inner and altruistic reasons have been found to be more likely to concentrate on factual aspects in a message as they consider this to be the best way to help others to make a good choice (Bronner & de Hoog, 2011; Yap et al., 2013). Thus, in terms of logos, the following is hypothesized:
H2a) Intrinsically motivated eWOM providers include more factual information in a review about a product or service than extrinsically motivated ones.
The second communication tactic, ethos, stands for the communicator’s claims to have expertise in the respective topic and relates to the communicator’s personal qualities (Buttle, 1998). This means the sender of a message tries to convince a reader by referring to his or her experience, expertise, or qualifications (Otterbacher, 2011). For instance, a consumer could affirm to be an expert with regards to a certain product or service category. Expertise claims are an important strategy in online reviews as they increase the perceived usefulness of the review because people tend to trust and adopt information given by others who seem to have expert knowledge more than persons who do not have expert knowledge (Racherla & Friske, 2012; Willemsen et al., 2011). Moreover, readers use such claims as a heuristic cue and pre-sume that the sender is very likely to depict accurate and credible information based on the knowledge and experience on the product or service under consideration (Willemsen, Neijens, & Bronner, 2011). Consumers, who share their opinion due to an intrinsic motivation, should be more likely to make use of expertise claims as they want to be helpful and therefore try to write a credible and persuasive review (Yap et al., 2013). For this purpose they might also refer to their qualifications and leverage their reputation. On contrast, consumers who write the review only due to receiving a reward, are, as outlined earlier, assumed to put less interest and effort in writing and trying to persuade other consumers. Consequently, regarding ethos it is assumed that:
H2b): Intrinsically motivated eWOM providers are more likely to refer to their expertise in an online review about a product or service than extrinsically motivated ones.
Lastly, the third tactic, pathos, refers to the communicator’s emotional appeals (Buttle, 1998). This relates to rather affective dimensions of a review (Sweeney et al., 2012; Yap et
al., 2013) and considers how strongly a reviewer appeals to the reader’s emotions, e.g. by describing a personal experience (Otterbacher, 2011). For instance, when a consumer evalu-ates a smartphone he/ she would not relate to technological facts about the device, but rather to how much the phone contributes to his/her personal life, e.g. how positive and satisfying it is to stay in contact with family and friends, taking pictures of the latest vacation or restaurant visit. Research by Söderlund and Rosengren (2007) implies that emotions are an important component when it comes to how word of mouth is delivered and received. The authors’ study showed that a WOM sender’s emotional state is assessed and used as a clue to form an attitude about a company, which in turn affects the receiver’s purchase intention. ‘Creating’ an emotional and personal story, which involves readers emotionally, is suggested to be only applied when the consumer has the utter interest to provide such information in order to make the review as helpful and convincing as possible. As extrinsically motivated behavior is posit-ed to result in less interest and less effort, it is therefore suggestposit-ed that consumers who write an online review due to receiving an incentive put less effort in deliberately creating and tell-ing an affective ‘story’ to get readers involved. Thus, intrinsically motivated consumers are assumed to be more likely to make us of pathos:
H2c): Intrinsically motivated eWOM providers rather try to get readers emotionally involved in their online review about a product or service than extrinsically motivated ones.
Personal experience. In online reviews consumers often not only provide product at-tributes, for example the price of a smartphone or menu in a restaurant, but as well refer to their personal experience with the respective products and services. They share personal in-formation, e.g. why or for whom they bought a specific product, visited a restaurant, or booked exactly this hotel for their holidays. One might additionally write about one’s encoun-ter with shop assistants, concierges, or waiencoun-ters, how the food in a restaurant tasted, how often one has been to that restaurant before, or has stayed in that holiday region or hotel before. Whether personal experience is provided in a review is important as the reviewer’s usage and experience with the subject matter affects source credibility, which in turn influences how helpful the reader perceives the review (Li & Zhan, 2011). This notion is also supported by Scholz and Dorner (2013) who outline that giving subjective experiences has an impact on a review’s perceived helpfulness. Hence, evidence presence constitutes a relevant aspect of an online consumer review. However, as already outlined, if a consumer is obtaining a reward for giving his/her opinion, this leads according to self-determination theory to less interest in
the particular task as well as putting less effort in conducting the task. An extrinsically moti-vated consumer might therefore demonstrate less commitment in giving additional personal experience. On the contrary, a consumer who is inherently highly motivated to make his or her recommendation as helpful as possible is more likely to reference to as much of his/her personal experience as possible, in order to give advice and back up his claims and opinion:
H3): Intrinsically motivated eWOM providers provide more personal experience in an online review about a product or service than extrinsically motivated ones.
Detail. How detailed and concrete a review is written is a further aspect of an online review. For example, the number of features discussed in a review is positively associated with its perceived helpfulness (Li & Zhan, 2011). Moreover, in particular cognitive aspects of a review are related to details and giving a concrete review (Sweeney et al., 2012). But con-sumers may not only elaborate in terms of factual information and give a concrete and de-tailed description on as many important aspects possible. As well, they may support their online evaluation by reflecting upon all their qualifications related to the subject matter, or write an extensive and emotional story about their product and service experience to enhance their recommendation. Writing a very detailed and concrete evaluation is proposed to require a lot of effort and interest in the topic from the sender’s side. As it is assumed that extrinsic motivation leads to putting less endeavor into a task, it is inferred that extrinsically motivated reviewers are less scrutinized and formulate a less detailed and concrete review than a person who does so due to inherent, altruistic purposes and gives a very elaborate recommendation.
H4: Intrinsically motivated eWOM providers formulate a more concrete and detailed review about a product or service than extrinsically motivated ones.
Extremity. Some reviewers express their opinion in a rather moderate tone and lan-guage, and yet others use rather ‘strong’ words when evaluating a product or service. The language used in an online review is regarded as an indicator of the extremity of a message (Li & Zhan, 2011). Intense language, which includes emotion-laden words and phrases (e.g. pleasant, excellent, extremely, disgusting), for example, shows whether a communicator holds an extreme viewpoint or not. Displaying an extreme position is posited to have a negative impact on how competent the author of a message is perceived, what in turn reduces the per-suasiveness of the message (Li & Zhan, 2011). How consumers rate a product or service also
implies whether a consumer has an extreme opinion about a subject matter. Usually, ratings from one to five stars are given in online reviews. One and five star ratings relate to extremity and three star ratings can be considered as moderate ones as they are supposed to present both sides of arguments (Mudambi & Schuff, 2010). As noted earlier, consumers, who write an online review due to intrinsic motivation and want to help others with their evaluation, would attempt to give a balanced review and refrain from expressing an extreme position so they will not be perceived as being not reliable or not credible (Bronner & de Hoog, 2011; Yap et al., 2013). Extrinsically motivated consumers, on the other hand, are suggested to pay less attention on what kind of language they use, or to avoid using intense language or extreme ratings, as their major aim is not to be perceived as credible or write a convincing review, but to receive the reward. Thus, in reviews which are written by reviewers who gained an eco-nomic incentive to provide their opinion, one should rather detect intense language and ex-treme rating scores than in intrinsically motivated reviews. Based on this premise, in terms of review extremity, the following two hypotheses are posited:
H5a): Intrinsically motivated eWOM providers use less intense language in an online review about a product or service than extrinsically motivated ones.
H5b): Intrinsically motivated eWOM providers rate a product or service with a more moder-ate score than extrinsically motivmoder-ated ones.
Review length. Lastly, it is assessed whether intrinsic and extrinsic motivations have an influence on the review’s length as some consumers write only bullet points or a few sen-tences, others, however, a very long review. A review’s length is related to the comprehen-siveness of a review, and a comprehensive review recommendation is rather perceived as helpful (Li & Zhan, 2011). Intrinsically motivated reviewers are assumed to provide a com-prehensive and helpful review (Bronner & de Hoog, 2011; Yap et al., 2012) as they aim to help others and are supposed to show more commitment and effort when contributing to eWOM. Extrinsically motivated consumers though are assumed to put less effort and are less interested in writing a recommendation as they only do so for a reward in return. Thus they are assumed to include less information and write a shorter review:
H6: Intrinsically motivated eWOM providers write longer reviews than extrinsically motivat-ed ones.
In terms of the above outlined characteristics, it is examined whether intrinsic and ex-trinsic motivation to write an online review affects how consumers compose their message – whether they would be more or less elaborate and more or less extreme when stating their opinion. Additionally, it is of interest to examine if this effect differs in terms of what con-sumers write about. However, previous studies on motivation and online consumer review content have focused on services as subject matter (Bronner & de Hoog, 2011, De Vries, 2012, Yap et al., 2013). In how far previous conclusions might also hold true or are different when not a service, but a product is reviewed, still has to be examined further. Thus, a possi-ble interaction effect of a review’s subject matter is discussed.
Moderating Role of Review Subject Matter. The latest purchased digital camera, smartphone, CD or book, a new restaurant, or the hotel of the last vacations – there are nu-merous subjects on which consumers give their opinion and recommend or not to other con-sumers. Even though there are countless different subjects consumers evaluate, these show some conceptual similarities and, on the most superordinated level, can be categorized into products and services (Anderson, 1998, Murray & Schlacter, 1990). Products relate to a phys-ical entity composed of tangible attributes, whereas services relate to intangibility, simultanei-ty of production and consumption, inseparabilisimultanei-ty, and non-standardization (Murray &
Schlacter, 1990). With regard to products, consumers may visualize and examine its physical attributes prior to the purchase. Services on the contrary are more based on experience which can be evaluated after purchase. Additionally, Murray and Schlacter show that services are found to have higher perceived risk than products.
Due to the conceptual differences of these two types of review subject matter, differ-ences between how a review is written and which message characteristics a reviewer consid-ers as more or less important to include, are expected. Willemsen and colleagues (2011) showed that how useful consumers perceive a review is related to the content characteristics which are presented in the review, and furthermore, to what type of product is reviewed. In terms of review valence, this relationship varies for search and experience products. For an experience product negatively valenced reviews were considered more useful, and positive valenced reviews were perceived as more useful for search products. This is supposed to be due to differences regarding the pre-purchase verifiability of the two types of products as ex-perience products are more difficult to be judged prior to purchase than search products, and therefore negative information has a stronger impact.
Additionally, for experiential products, Scholz and Dorner (2013) found that reviews are judged as more helpful when they contain the reviewer’s personal, subjective experience with the product as compard to objective product features. For utilitarian goods, it is the op-posite way. In these two studies, experiential products (e.g. running shoes, sunscreen, eau de toilette) versus search products/utilitarian products (e.g. DVD-player, digital camera,
smartphones) as review subjects were used. The conceptualizations of these product catego-ries examined in the studies by Willemsen et al. (2011) and Scholz and Dorner (2013) relate to the present study’s definition of products and services. Experience products and services conform conceptually with regard to pre-purchase verifiability, that is, they have to be experi-enced to form and opinion, and evaluation takes place after purchase or trial. This is contrary to (search) products, which comprise specific physical attributes which may be evaluated pri-or to purchase.
Findings by the above mentioned studies, that depending on what kind of subject mat-ter is reviewed, some content characmat-teristics are perceived as more or less helpful, are also interesting for the present study. This is because type of motivation, intrinsic or extrinsic, de-termines which content characteristics are included and to which degree they are elaborated upon in an online review. A negative effect of extrinsic motivation on writing behavior might be more detrimental depending on whether a product or service is reviewed since some con-tent aspects, which are considered as very helpful by a consumer in particular for this type of subject matter, might be missing or not sufficiently be elaborated upon.
For example, regarding products, consumers regard specific product attributes and a well-reasoned argumentation based on factual information as helpful (Scholz & Dorner, 2013). In case extrinsically motivated reviewers do not put a lot of effort in their review and do not elaborate thoroughly on certain features, this is even more harmful, when they do so when reviewing a product. If a service is evaluated, this might be less detrimental, as in this case providing detailed factual information is perceived as less important and helpful.
For services, however, personal experience is a more important content aspect (Scholz & Dorner, 2013). In return, in case extrinsically motivated reviewers do not sufficiently pro-vide their own personal experiences with the reviewed service, this might be more negative within a service than within a product context. Consequently, rewarding a consumer for writ-ing a reviewer can be more or less favorable dependwrit-ing on the type of subject matter, which is reviewed.
As noted earlier, former research regarding motivation and online consumer review content focused on services contexts (Bronner & de Hoog, 2011, De Vries, 2012, Yap et al.,
2013). Thus, the present paper extends eWOM research by further including products and accounting for possible differences between a review’s subject matter. Given the explorative nature of this issue, no hypotheses were formulated and a possible moderating role of subject matter is addressed within the following research question:
To what extent does intrinsic and extrinsic motivation influence how a con-sumer writes an online review, and in which way does an effect differ when the review is written about a product or a service?
Method Design
The present study was conducted to examine to what extent consumers’ motivation to write an online review influences how they formulate their review. A possible interaction ef-fect of the review’s subject matter, whether a product or service is the focus of the review, is additionally considered. A 2 (type of motivation: intrinsic vs. extrinsic) x 2 (type of subject: product vs. service) between-subject experiment was conducted to test the outlined assump-tions.
Sample
Data for this study was collected from German consumers2 of different social demo-graphic backgrounds who were recruited online by means of a snowball technique. Partici-pants were invited to the experiment by a link which was distributed via email and the social networking site Facebook3. This approach was chosen due to limited time and budget. More-over, the experiment was conducted online in order to make sure that participants who have an Internet access partake in the study and therefore might be familiar with online consumer reviews. No incentive to take part in the experiment was offered and participation was volun-tary.
2
As the study used a snowball technique and participants were approached by the author due to time and budget constraints, a higher number of participants was expected when data were collected within the author’s native country.Thus, the study based on consumers of the German market.
3
The survey link was displayed on the author’s Facebook profile, which consists of 266 Facebook contacts. At least 11 of these contacts shared the link on their own Facebook profile. Additionally, 41 persons were contacted by the author via email. Contacts on Facebook and email were not largely overlapping. To distribute the survey link via email as well, was in particular used to reach further possible participants who are not registered and encountered a social network site. To the author’s knowledge, the survey link was forwarded to others via email by at least 3 of the email contacts.
In total, 233 participants took part in the experiment. However, 43 did not give their informed consent to partake in the experiment or did not finish the questionnaire. The 190 respondents, who completed the experiment, were between 19 and 82 years old (M = 35.59, SD = 13.63), covered a range of occupations and educational levels, and 57.9% percent of the participants were female (see Appendix A, Table 1 for overview of demographics).
Fourteen participants were excluded as they failed the manipulation check. Additional-ly, as the study referred to smartphones in the product conditions, five respondents who indi-cated they never use a smartphone were excluded from further analyses.4 This resulted in a total N = 171 participants for the main analyses, of which n = 53 were randomly assigned to the intrinsic-service condition, n = 41 to the intrinsic-product, n = 48 to the extrinsic-service , and n = 29 to the extrinsic-product condition.
Procedure
The online questionnaire for the experiment was set up in the program Qualtrics (see Appendix B for questionnaire)5. Participants had to click on a survey link to be directed to the survey website. First, participants were thanked for their time, were informed that the study is about online consumer reviews, and were asked to give their consent to participate in the study. After giving their consent, participants were automatically randomly assigned to one of the four conditions.
Participants were asked to imagine themselves in a scenario inviting them to provide their opinion about a smartphone in the product conditions, and about a touristic accommoda-tion in the service condiaccommoda-tions. The product and service category was given but not a specific product/service type or brand in order to avoid an influence of previous attitudes and that some participants might not be acquainted with the respective type of product/service or a specific brand, as well as to make sure that participants had experience with the subject matter and were thus able to state their opinion about it.
As for the product condition, a smartphone was selected because smartphones are amongst the most often reviewed product categories (Scholz & Dorner, 2013). It is assumed that the majority of German consumers owns a smartphone and therefore have experience with this kind of product and may state and opinion about it. As for the service condition, touristic accommodations (e.g. hotels) are selected, as these are a frequent review subject
4 Analyses with and without these 19 respondents were conducted. Results of the additional analyses including all respond-ents showed no substantial differences (see Appendix C, Table 2).
ter and consumers first have to stay at an accommodation before making a judgment about it (Racherla & Friske, 2007).
After reading the scenario, participants were forwarded to the main part of the survey in which they responded to questions regarding the dependent variables and additional control variables, such as how often they write online consumer reviews, how much time they spend on the internet, or how often they use the product/service. Finally, participants replied to a number of demographic variables.
Stimuli
Type of motivation was manipulated by means of different scenarios, which were dis-played at the beginning of the questionnaire. Intrinsic motivation was evoked by appealing to the writer’s intention to share their experiences with other consumers and help them to make the best choice. Extrinsic motivation was triggered by offering an economic incentive (see Appendix C for scenarios).
Measures
Dependent Variables. As the outlined message characteristics predominantly have been assessed and described within the context of content analyses but not within the scope of an experimental design referring to eWOM providers self-reports, the measures for the study’s purpose mainly had to be created by the author. Measures for the dependent variables of this study based on previous content analyses (e.g. De Vries, Li & Zhan, Otterbacher, 2010, Scholz & Dorner, 2012, Willemsen et al., 2011) and were changed to fit a question-naire. For the dependent measures, participants had to indicate on a scale ranging from 1 be-ing totally disagree to 7 bebe-ing totally agree how much they agree with the statements which measured how they would formulate an online consumer review.
Argumentation. Whether a consumer writes a two-sided review and provides positive
as well as negative aspects of a product or service is assessed with the following items, which were created by the author: “I would describe only positive aspects of the smartphone/accommodation in my review”, “I would describe only negative aspects of the smartphone/accommodation in my review”, “I would describe both positive and negative as-pects of the smartphone/accommodation in my review” (De Vries, Otterbacher, 2010, Willemsen et al., 2011). A principal component analysis showed that the three items form a single uni-dimensional scale (EV = 1.77, R2 = .59). However, reliability of the scale was not satisfactory (α = 0.33). Thus, it was proceeded with treating the single questions as separate
dependent variables in the main analyses. The single questions were grouped under argumen-tation as bivariate Pearson correlation analysis showed that providing balanced arguments is significantly related to describing only positive arguments (R= -.696, p < .001) as well as to including only negative arguments (R= -.184, p = .016).
Aristotle’s logos, ethos, pathos. The present study examines to what extent consumers
apply the three communication tactics by Aristotle. As these three constitute strategies which are independently of one another, they were thus measured with three individual questions: “In my review about the smartphone/accommodation, I would emphasize on factual infor-mation” (logos), “In my review, I would try to show how well-experienced I am with smartphones/booking and staying in accommodations” (ethos), “In my review, I would try to get readers emotionally involved” (pathos) (Buttle, 1998; De Vries, Otterbacher, 2010, Sweeney et al., 2012).
Personal Experience. To see if consumers refer to and include their own personal
ex-perience with a product or service, the following question was formulated by the author: “In my review, I would provide information about my personal experience with the smartphone/accommodation” (Scholz & Dorner, 2013).
Detail. How detailed consumers write about feature-oriented information or their
per-sonal experience and elaborate their review was assessed with the question “I would provide a concrete and detailed description on as many important aspects of the smartphone/accommodation as possible” (Sweeney et al., 2012).
Extremity. Relating to Li and Zhan (2011), how extreme an online review is
formulat-ed is assessformulat-ed in form of language intensity by the questions “When writing my review about the smartphone, I would include words and phrases such as, (un)pleasant, excellent, extreme-ly, completeextreme-ly, best of all” in the product conditions and “When I write my review about the accommodation, I would include words and phrases such as (un)pleasant, excellent, disgust-ing, extremely, best of all”. Additionally, as rating score implies whether a consumer holds an extreme attitude or not (Mudambi & Schuff, 2010), the following question was included: “I would be likely to use an extreme score in my review (1 or 5 stars), instead of a more moder-ate score (2 or 4 stars)”. As an overall bivarimoder-ate correlation analysis showed, these two aspects are positively associated, R = .208, p = .006 (see Appendix A, Table 3 and 4 for correlations for individual categories).
Review Length. In terms of how short or long consumers would formulate their
ra-ther long review”. Additionally, respondents were asked to estimate how many sentences they would approximately write. (R = .268, p < .001).
Control variables. Several additional variables were included and measured. Partici-pants’ subjective knowledge with the product/service category is assessed with four 5-point bipolar-items (knowledgeable–not knowledgeable, competent–not competent, an expert–not an expert, and experienced–inexperienced, (α=.92) (Kim & Gupta, 2012). Involvement with the product/service category is measured by Zaichkowsky’s (1994) Personal Involvement Inventory, consisting of ten 9-point bipolar items, (α=.89). Participants’ usage of prod-uct/service is measured with a 5-point scale ranging from 1- never to 5- very often. Moreover, participants’ frequency of writing online reviews is assessed with a 5-point scale ranging from 1- ‘never’ to 5- ‘very often’. Additionally, participants are asked to indicate how many hours they spend on the Internet.
Manipulation Check. As a manipulation check, whether participants felt intrinsic and
extrinsic motivated, was measured separately for the two conditions by the two questions “I was motivated to write the review because I receive a reward for writing it” for the extrinsic, and “I was motivated to write the review because I wanted to share my experience and help others to make the best choice” for the intrinsic conditions on a 7-point Likert scale, with 1 being totally disagree and 7 being totally agree. Additionally, in the extrinsic conditions, re-spondents had to indicate whether the scenario said they would receive a reward or not. On a seven-point Likert scale ranging from 1 being totally disagree to 7 being totally agree, partic-ipants furthermore had to indicate how attractive they consider the reward mentioned in the scenario.
Results
The majority of the sample does not often write online reviews as 28.7% indicated they never write reviews, 43.9% rarely, and 21.1% sometimes. On average, participants spend 3.92 hours (SD = 3.17) on the Internet. Of the 70 participants, who were in the product condi-tion, the majority uses smartphones very often (61.4%) and often (24.3%). Participants who answered questions with regard to a smartphone indicated to have medium knowledge of this product category (M = 3.06, SD = 0.89). Furthermore, they were rather involved in the prod-uct category (M = 6.35, SD = 1.56). With regards to booking touristic accommodations, 22.8% of the service condition (n = 101) often, 46.5% sometimes, and 25.7% rarely book ac-commodations. Participants in this condition stated they have medium knowledge in terms of
booking and staying in touristic accommodations (M = 3.42, SD = .93) and were moderately involved in the service category (M = 5.84, SD = 1.19).
Manipulation and Randomization Check
Respondents who were assigned to the intrinsic scenarios indicated they felt rather motivated to write the review in order to share their experience and help others (M = 5.73, SD = 1.30). In the extrinsic group, however, participants responded less satisfactory to the study’s manipulation to write the review due to receiving a reward (M = 4.06, SD = 2.08). The offered incentives were in both conditions rated as rather moderately attractive (Mservice = 4.44,
SDservice = 1.64, Mproduct = 3.90, SDproduct = 1.86).
Randomization check was conducted to test whether the randomization created equal experimental groups in terms of participants’ age, gender, educational level, and employment status. A one-way ANOVA was run in order to test whether the four experimental groups differed significantly in means with regard to age. Results indicated that there were no signif-icant differences between the four groups in terms of age, F(3, 167) = .32, p = .814, η2 = .01. Furthermore, crosstabulation analyses indicated that there were no significant differences be-tween the four conditions regarding gender, χ2 (3) = .54, p = .909, educational level, χ2
(21) = 16.75, p = .726, and employment status χ2
(15) = 22.43, p = .097. As a result, the randomiza-tion was successful.
Covariates
Bivariate Pearson correlation analysis was conducted to check for alternative explana-tions. It was tested whether there are relationships between the dependent variables and the control variables internet usage, involvement with product/service category, usage of prod-uct/service, subjective knowledge of product/service category, and online review writing fre-quency (see Appendix A, Table 3 and 4 for the correlation matrix of the individual catego-ries). The main analyses were conducted once with and once without covariates for the re-spective dependent variables which correlated with one or more control variablesi. As effects did not change substantially when including covariates, results without the control variables were presented in the following (see Appendix A, Table 5 for effects including covariates).
Hypotheses Testing
Separate ANOVAs with the respective review characteristics as dependent variables and type of motivation and subject matter as fixed factors were conducted to test the
hypothe-ses. In addition, to see if the mean scores for the respective characteristics differ significantly in the four conditions, simple effects tests were conducted.
As depicted in Table 1, except for one dependent variable, all main effects of motiva-tion on the single review characteristics were insignificant. Only in terms of providing only negative arguments, a significant difference was found, F(1, 167) = 4.81, p = .030, η2 = .03. Both experimental groups agree not to present only negative arguments when evaluating a product or service, whereas the intrinsically motivated ones were slightly less likely to do so (M = 1.94, SD = 1.29) than the extrinsically motivated ones (M = 1.62, SD = 0.90). Accord-ingly, the posited hypotheses that intrinsically and extrinsically motivated consumers differ in terms of online review content, in particular regarding argumentation, Aristotle’s logos, ethos, pathos, referring to personal experience, providing a detailed review, extremity, and review length, were not confirmed.
Additionally, it was examined in how far the subject matter of the review plays a pos-sible moderating role. Significant interaction effects were reached for the aspect of providing only negative arguments, F(1, 167) = 4.33, p = .039, η2 = .025, and for the Aristotle’s persua-sive communication device pathos, F(1, 167) = 3.93, p = .049, η2 = .023.
As pairwise comparisons showed, the effect that extrinsically motivated respondents agree slightly more than the intrinsically motivated ones not to provide only negative argu-ments is stronger when a product is in focus of the review (p = .006) (see Table 2 for means and standard deviations). In terms of pathos, results showed that when evaluating a service, the intrinsic ones are slightly more likely to appeal to reader’s emotions (M = 3.38, SD = 1.64) than the extrinsic ones (M = 2.56, SD = 1.37) (p = .012). This is in line with the hypothesized direction that intrinsically motivated reviewers put more effort in their review and for this purpose also include more personal and emotional information.
Even though no further significant interaction effects were found, simple effect tests were conducted for all message characteristics due to the study’s explorative nature regarding a review’s subject matter. The results of the simple effect tests are presented in Table 2 and 3 for the two subject matters respectively. For including personal experience in a review, pair-wise comparisons approached towards significance in the product conditions between extrin-sically and intrinextrin-sically motivated respondents (p = .074). Participants in the extrinsic group were more likely to relate to their personal experience in a review about a smartphone (M = 6.24, SD = 0.95) than the ones in the intrinsic group (M = 5.63, SD = 1.43).
Effects of motivation and subject matter on online review content
F(1, 167) p η2
Experimental Group
Mean (SD)
Motivation Intrinsic Extrinsic
Argumentation - only positive .13 .717 .00 2.05 (1.54) 1.99 (1.35) - only negative 4.81 .030 .03 1.94 (1.29) 1.62 (0.90) - balanced .13 .248 .01 6.34 (1.37) 6.55 (0.68) Logos .46 .499 .00 5.47 (1.40) 5.58 (1.56) Ethos .68 .410 .00 2.34 (1.61) 2.19 (1.55) Pathos 1.49 .223 .00 3.05 (1.68) 2.66 (1.58) Personal experience 2.28 .133 .03 5.56 (1.41) 5.82 (1.39) Detail .28 .596 .00 4.67 (1.68) 4.81 (1.87) Intense language 1.63 .204 .01 4.47 (1.75) 4.17 (1.69) Rating score 1.84 .177 . 01 3.37 (1.93) 2.88 (1.78) Review length 2.10 .150 .01 2.94 (1.31) 3.22 (1.45)
Estimated number of sentences 1.75 .188 .01 9.05 (7.97) 10.99 (17.54)
Subject Matter Product Service
Argumentation - only positive .11 .740 .00 1.99 (1.51) 2.05 (1.42) - only negative .02 .649 .00 1.79 (1.20) 1.80 (1.10) - balanced .60 .441 .00 6.34 (1.31) 6.5 (0.97) Logos 3.30 .071 .02 5.76 (1.46) 5.36 (1.47) Ethos .13 .722 .00 2.36 (1.72) 2.22 (1.48) Pathos .88 .348 .01 2.71 (1.75) 2.99 (1.57) Personal experience 3.38 .068 .02 5.89 (1.28) 5.53 (1.47) Detail 1.10 .296 .01 4.54 (1.88) 4.86 (1.68) Intense language 20.78 .000 .11 3.64 (1.67) 4.81 (1.60) Rating score 2.23 .137 .01 3.40 (2.00) 2.98 (1.77) Review length .19 .665 .00 2.99 (1.37) 3.12 (1.39)
Estimated number of sentences 7.86 .006 .05 13.04 (19.45) 7.76 (4.67)
Motivation x Subject Matter
Argumentation - only positive .09 .766 .00 - only negative 4.33 .039 .03 - balanced .04 .841 .00 Logos .13 .725 .00 Ethos 1.94 .165 .01 Pathos 3.93 .049 .02 Personal experience 1.61 .207 .01 Detail .34 .562 .00 Intense language .80 .371 .01 Rating score 1.53 .218 .01 Review length .71 .401 .00
Table 2
Means, standard deviations, and p-values of simple effects tests: Product Intrinsic Mean (SD) Extrinsic Mean (SD) p Argumentation - only positive 2.05 (1.55) 1.90 (1.47) .669 - only negative 2.10 (1.37) 1.34 (0.72) .006 - balanced 6.24 (1.59) 6.48 (0.74) .381 Logos 5.66 (1.43) 5.90 (1.52) .506 Ethos 2.59 (1.83) 2.03 (1.52) .153 Pathos 2.63 (1.66) 2.83 (1.89) .623 Personal experience 5.63 (1.43) 6.24 (0.95) .074 Detail 4.41 (1.80) 4.72 (2.00) .472 Intense Language 3.68 (1.59) 3.59 (1.80) .807 Rating score 3.41 (2.00) 3.38 (2.04) .938 Review length 2.78 (1.35) 3.28 (1.36) .140 Estimated number of sentences 11.02 (10.72) 15.90 (27.46) .122
Table 3
Means, standard deviations, and p-values of simple effects tests for subject matter service Intrinsic Mean (SD) Extrinsic Mean (SD) p Argumentation - only positive 2.06 (1.55) 2.04 (1.29) .959 - only negative 1.81 (1.21) 1.79 (0.97) .930 - balanced 6.42 (1.18) 6.58 (0.65) .452 Logos 5.32 (1.38) 5.40 (1.57) .798 Ethos 2.15 (1.41) 2.29 (1.57) .656 Pathos 3.38 (1.64) 2.56 (1.37) .012 Personal experience 5.51 (1.41) 5.56 (1.56) .849 Detail 4.87 (1.57) 4.85 (1.81) .969 Intense Language 5.08 (1.64) 4.52 (1.53) .089 Rating score 3.34 (1.89) 2.58 (1.56) .043 Review length 3.06 (1.28) 3.19 (1.51) .634
Regarding service as the focal subject matter, additionally to pathos, which has been depicted above, significance was reached for rating scores (p = .043), and almost reached for using intense language (p = .089). The ones who are intrinsically motivated would rather give an extreme rating and use intense language in their review about a touristic accommodation (Mrating = 3.34, SDrating = 1.89, (Mlanguage = 5.08, SDlanguage = 1.64) than the extrinsically
moti-vated ones (Mrating = 2.58, SDrating = 1.56, Mlanguage = 4.52, SDlanguage = 1.53). Yet, this is not in
line with the hypothesized direction as it was assumed that the extrinsic ones would be more likely to depict an extreme position in their review.
Whether a product or service is in focus of a review plays a moderating role only to a small extent and has rather a direct influence. As the overview in Table 1 depicts, significant main results were found for subject matter on language intensity and for estimated number of sentences. Almost significant results of whether a product or service is reviewed, were found for Aristotle’s logos and for referring to one’s personal experience.
Conclusion and Discussion
The aim of the present paper was to shed further light on the impact of economic in-centives on online consumer reviews. It was assessed in how far extrinsic motivation com-pared to intrinsic motivation influences the content of an online review. The main question of this study concerned whether a consumer who writes an online review due to extrinsic moti-vation formulates the review less elaborate and more extreme than a consumer who shares his opinion in order to help others to make the best decision. Additionally, the paper aimed to contribute to eWOM research by accounting for different subject matters, that is, if a product or a service is in focus of the review. Based on self-determination theory and former research, it was posited that intrinsically motivated consumers provide a more balanced, informational, and personal, elaborate, and moderate review than extrinsically motivated ones. To test the outlined assumptions, an experiment was conducted in which type of motivation was manipu-lated and participants had to indicate how they would formulate a review with regard to cer-tain content characteristics.
The results of the experiment did not support the hypothesized effects for type of mo-tivation. There were no significant differences between extrinsically and intrinsically motivat-ed participants regarding argumentation, Aristotle’s logos, ethos, and pathos, personal experi-ence, detail and concreteness, extremity, and review length. It can thus be assumed that con-sumers do not differ in terms of formulating a review when the driving force stems from
in-trinsic or exin-trinsic motivation. In both cases, they seem to value the quality of the information they provide and would write a balanced review, and give concrete and detailed information on as many aspects as possible to the same extent. Furthermore, they would appeal to the readers’ sense of reason by giving factual information, include personal information, try to get readers emotionally involved, or claim to be an expert in the respective subject matter. More-over, extrinsically motivated reviewers are not necessarily more extreme than intrinsically motivated ones as they would not use more intense language or tend to give more extreme rating scores than intrinsically motivated ones. Additionally, when taking a review’s subject matter into consideration, a relation between motivation and online review content did not differ considerably for products and services. This means that intrinsic or extrinsic motivation to write a review does not evoke different effects for products or services.
Previous research suggests that motivation to write an online consumer review has an influence on the formulation of the review (Bronner & de Hoog, 2011; De Vries, 2012; Yap et al., 2013). In particular, in terms of extrinsic versus intrinsic motivation, De Vries’ (2012) found that extrinsic motivation negatively affected argument diversity, logos, rating score, and review length, resulting in more unbalanced reviews, providing less factual arguments, a more extreme rating, and a longer review. The present study, however, did not replicate these findings.
One possible explanation why the present study did not find an effect of type of moti-vation on review content could be due to the study’s methodological approach. As the study was not conducted within a real life context but within an experimental design using theoreti-cal scenarios, respondents might have felt less involved and intrinsitheoreti-cally/extrinsitheoreti-cally moti-vated as in a real life context. This limitation is further discussed in more detail in the previ-ous chapter. Moreover, social-desirability constitutes one additional possibility why the out-comes of the experiment are not in line with the postulated hypotheses. Respondents in the extrinsic condition could have indicated to provide a well elaborated review because they consider this to be the more favorable answers to the questions and the more desirable behav-ior in the given context, since well elaborated reviews are perceived as helpful, credible, and persuasive.
Additionally, other motives than the intended ones could have influenced participants’ response. For instance, self-enhancement might have affected respondents’ statements. This leads to providing more factual information presented in a persuasive and affective way (Yap et al., 2013), what could explain why the extrinsic group scored to the same degree on the questions concerning information quality and personal experience, although they were
ex-pected to provide a less factual and personal review. This implies that respondents might not exclusively have answered the questions based on one specific motivation, which could ex-plain the small or unexpected differences within the two groups.
Another explanation for the study’s results bases on the assumption that the extrinsi-cally motivated consumers might feel obliged to give an elaborate review as they consider themselves to be in a ‘quid pro quo’ situation. That is, for receiving a reward, they give a fa-vor in return by providing a qualitatively good review.
A different consideration, on the other hand, relates more to the intrinsic group. Intrin-sically motivated consumers are suggested to be more likely to focus on factual and cognitive aspects and provide an informative review as they are supposed to put more emphasis on writ-ing a helpful and persuasive recommendation due to their concern for others (Bronner & de Hoog, 2011; Yap et al., 2013). On contrast, the more a consumer wants to help others, the more affective and intense his behavior can be which leads to the affection being expressed in his or her actions (Jeffries, 1998). This might explain why in terms of extremity both groups did not differ significantly from each other. The intensity and affection an intrinsically moti-vated consumer feels could lead to a more affective and extreme review tonality, thus attenu-ating the posited assumption that extrinsically motivated reviewers are more likely to use in-tense language or an extreme rating score. This reasoning does also account for the effects found in the service groups, that the ones who are intrinsically motivated would rather use intense language and give an extreme rating score than the extrinsically motivated ones when evaluating a touristic accommodation.
To summarize, in the present study, the type of motivation did not affect online con-sumer review’s content and intrinsically motivated reviewers do not necessarily provide more elaborate and less extreme reviews than extrinsically motivated reviewers. A review’s subject matter seems to constitute an interesting factor in relation to the formulation of a review, which needs further attention.
Limitations and Future Research
Though the present study adds to eWOM research, it comes with limitations and offers various avenues for future research regarding motivation and online review content. First of all, the experiment did not take place in a real-life context. Participants did not evaluate a spe-cific product or service they actually purchased or tried and no actual monetary rewards were distributed due to budget constraints. The study’s manipulation was based on scenarios, which asked consumers to imagine a specific situation. However, this approach may account for
