2 ONLINE CONSUMER REVIEWS , CONSENSUS AND INDEPENDENT PLATFORMS.
THE INFLUENCE OF ONLINE CONSUMER REVIEWS ON THE CONSUMER’S ATTITUDE TOWARD A PRODUCT
By
Jos Keuning
University of Groningen Faculty of Economics and Business
3
Management Summary
Nowadays, consumers are more and more internet oriented in finding information about products or services. Consumers can post and read about consumer’s previous experiences with the product or service. These online consumer reviews (OCR’s) can be posted either on dependent platforms (company controlled) or on independent platforms (controlled by third parties). The OCR’s on independent platforms are perceived more trustworthy and credible, because the OCR’s are perceived to be more objective and the writers of the OCR’s are perceived to be unbiased. However, in the field of OCR’s there is little research on what effect these OCR’s have on the consumer’s attitude toward a product.
The current research will consist of two analyses to investigate the effect of OCR’s on the consumer’s attitude toward a product. The first analysis is used to investigate the effect of OCR’s on two independent platforms versus one independent platform on the consumer’s attitude toward a product. The moderators that are expected to strengthen the effect are product involvement and product knowledge. The second analysis is used to investigate the effect of inter consensus in OCR’s on two independent platforms on the consumer’s attitude toward a product. Also in this analysis are the moderators product involvement and product knowledge expected to strengthen the effect of the OCR’s.
4
Preface
This thesis study is the final piece in order to complete my Master of Marketing Management at the University of Groningen. The first months I was also following three courses next to the busy schedule of working on this paper. The second period of the semester, I had to follow only one course next to this process, and therefore, the last period I could focus quite good on my thesis. I did not have to think long about choosing this research topic. I was already very interested in this topic, I always read about trends in the field of eWOM. I have my own company, Woodt Sunglasses, and the main marketing practices are via social media. It is interesting to see what posts are liked better, what influence online consumer reviews (OCR’s) have, how to respond on negative or positive reviews of consumers etc. The topic is rather new, so there is much room for research, however still the topic eWOM is still quite broad. It took a while before I knew what I wanted to research, because a lot of things took my interest. But then I found articles about how independent platforms are perceived more credible and articles that discussed that highly involved consumers actively seek for information about products. However, there was little research on independent platforms and the effect of OCR’s on attitude. At that point I knew that I wanted to research what effect OCR’s on independent platforms have on the consumer’s attitude. I am glad that I could do my thesis on the topic that I most desired. At first, I would like to thank my supervisor, Liane Voerman, for stimulating feedback sessions and feedback by e-mail. I also want to thank my fellow group mates, for support and inspiration. Finally, I could not finish my analysis without the help of 197 participants that filled in the questionnaire.
5
Table of Content
1. Introduction ... 7
1.1 eWOM ... 7
1.2 Online Consumer Reviews (OCR’s) ... 9
1.3 OCR’s on a single platform versus OCR’s on two platforms ... 10
1.4 Consensus ... 11
1.5 Attitude toward a product ... 12
1.6 Elaboration Likelihood Model (ELM) ... 12
1.7 Product Involvement and Product Knowledge ... 13
1.8 Problem statement and research questions ... 13
2. Literature review ... 15
2.1 Multiple OCR’s on multiple independent platforms ... 15
2.1.1 Mere exposure effect of OCR’s on attitude ... 15
2.2 Consensus ... 17
2.2.1 Intra- and inter-consensus ... 17
2.2.2 Consensus and attitude ... 17
2.3 Elaboration Likelihood Model ... 19
2.3.1 Consumer’s product involvement ... 19
2.3.2 Consumer’s product knowledge ... 21
2.4 Conceptual model ... 24 3. Methodology ... 25 3.1 Research Design ... 26 3.1.1 Participants ... 26 3.1.2 Experimental design ... 27 3.2 Procedure ... 29 3.3 Measurement of concepts ... 29
3.3.1 Manipulation of seeing OCR’s on one or two independent platforms ... 29
3.3.2 Manipulation of inter consensus among independent platforms ... 32
3.3.3 Measuring the dependent variable attitude ... 34
3.3.4 Measuring product involvement ... 34
6
3.3.6 Overview of the measurements of concepts ... 36
3.4 Descriptives ... 37
3.4.1 Basic descriptives ... 37
3.4.2 Means of attitude toward the product and intention ... 38
3.5 Homogeneity of slopes ... 40
3.5.1 OCR’s on 2vs1 independent platform (analysis 1) ... 40
3.5.2 Inter consensus (analysis 2) ... 41
3.6 Plan of analysis ... 41
4. Results ... 42
4.1 OCR’s on independent platforms (2 vs 1) ... 42
4.2 Inter Consensus ... 43
5. Discussion ... 50
5.1 Conclusions ... 50
5.1.1 Research question one ... 50
5.1.2 Research question two ... 51
5.2 Managerial implications ... 54
5.2.1 Managerial implications of analysis 1 ... 54
5.2.1 Managerial implications of analysis 2 ... 54
5.3 Limitations and future research ... 55
Literature list ... 57
Appendix A ... 63
7
1. Introduction
The first thing that I do when I want to see a movie is checking the average rating of the movie on IMDb (Internet Movie Database). When the average star rating is below a 6, my intention of going to see that particular movie becomes very low. On the other hand, when the average star rating is a 7,5 or higher, I will definitely watch that movie. I am highly involved in movies and have moderate to high knowledge about movies, and therefore, I can make quite good assumptions about how likeable the movie will be for me. Still, I am influenced by the average online consumer rating on IMDb. I am not the only one, I see this happening in my direct environment. Almost all of my friends have the mobile application of IMDb, and will always bring up the average rating of the movie on IMDb in conversations about movies. I find this a very interesting development, and for that reason I am going to research the effect of eWOM (electronic word-of-mouth) on the consumer’s attitude toward a product.
1.1 eWOM
Word-of-mouth (WOM) has always been a an influential factor in the consumer marketplace, and this has been widely accepted by researchers (e.g. Sen & Lerman, 2007; Jansen, et al., 2009; Shu-Chuan & Yoojung, 2011). Word-of-mouth (WOM) communication can be defined as "oral, person-to-person communication between a receiver and a communicator whom the receiver perceives as noncommercial, regarding a brand, a product, a service or a provider" (Arndt 1967, p. 5). WOM plays a significant role in customer buying decisions (Richins & Root-Shaffer, 1988).
8 eWOM differentiates strongly from traditional WOM communications in several ways. First, the communication network in eWOM is global. eWOM has provided consumers with the ability to post their opinions and experiences on the internet (Armstrong and Hagel, 1996). More contributors and audiences are involved, and the reach of eWOM extends beyond direct personal connections to the Internet world. Second, eWOM eliminates the restrictions on time and location, consumers can have anytime, anywhere and at anyplace access to internet (Kaplan & Haenlein, 2010). Users are allowed to read and compare archived reviews of the product/ service they are interested in. The convenience of easy accessibility makes eWOM attractive to Internet users; as a result, it has become a favorite source for consumer advice (Man Yee, et al., 2009). eWOM may be less personal because it is not face-to-face, but it has a stronger influence, as a result of that it is immediate, is accessible by others, and has a significant reach (Hennig-Thurau et al., 2004). Consumers gained more power, and firms have been put somewhat to the sidelines as being observers, having neither the knowledge nor the chance to adjust customer provided information (Kaplan & Haenlein, 2010).
eWOM has been changing consumers’ behavior due to the growth of Internet usage: Consumers generally make offline decisions on the basis of online information, consumers tend to rely on the opinions of others when making purchase decisions or other decisions about matters like what stocks to invest in or which movie to watch (Dellarocas, 2003). In the future eWOM is expected to become even more important, because new social networking apps will become more widespread (Jansen, et al., 2009).
9 information is credible, the consumer has no reason not to trust the organization. Empirical evidence shows that trust predicts willingness to follow platform advice (McKnight & Kacmar, 2006).
1.2 Online Consumer Reviews (OCR’s)
The general form of eWOM conversations are OCR’s providing product evaluations from the perspective of the customer (Schindler & Bickart, 2005). An OCR can be defined as “new information presented from the perspective of consumers who have purchased and used the product. It includes their experiences, evaluations and opinions” (Park et al., 2007). Typically, an OCR is written to either recommend or discourage others from buying the product. Reviews offer positive arguments in support of the product or negative opinions against it (Sen & Lerman, 2007). Usually consumers find it difficult to make purchase decisions based on information provided by sellers when buying products from an online retail market. Therefore, consumers look for detailed product information through searching for OCR’s, i.e. consumer-oriented information. This consumer-oriented information is helpful in making purchase decisions because it provides indirect experiences of products (Park & Lee, 2008). An increasing number of consumers make use of product information from the eWOM network to make purchase decisions (Man Yee, et al., 2009). When making use of product information, consumers pay more attention to negative information than to positive information. Negative information tends to be perceived as more informative than positive or neutral information. (Herr et al., 1991)
10 very useful indicator for consumers to assume if the product is popular, and it will increase their purchasing intention (Park, et al., 2007). An interesting finding of Lee et al. (2008) is that negative online consumer reviews have a greater influence on the purchasing intention than positive online consumer reviews. 24% of the consumers make use of these online consumer systems and reviews before buying a product, and therefore, high product ratings on independent platforms can increase sales (Sridhar & Srinivasan, 2012). Similarly, other researchers found that there is significant evidence of the influence of online consumer reviews on sales (Chevalier & Mayzlin, 2006; Yong, 2006; Park, et al., 2007; Jansen, et al., 2009). Therefore, the direct effect of eWOM on purchase intention will not be researched. However, the direct of eWOM on the attitude will be researched, in the next section will the concept of attitude be discussed.
1.3 OCR’s on a single platform versus OCR’s on two platforms
Consumers can post reviews either on dependent platforms, platforms of the brand itself, or on independent platforms, platforms which are owned by a third party (Sridhar & Srinivasan, 2012; Meuter, 2013 ; Verlegh, et al., 2013). The main difference between those two types of platforms is the trait of independence. Research has shown that independent platforms are perceived more credible compared to company-controlled platforms (Sridhar & Srinivasan, 2012; Meuter et al., 2013). Independent platforms are also called third-party sources, these platforms are not company-controlled and are owned by a third party to let consumers review independently without having a vested interest in recommending a product or brand (Yubo, & Jinhong, 2008). Retailer tends to focus on the positive product attributes while hiding negative aspects. User generated content does not have this bias because independent consumers are objective and honest when assessing the product’s benefits and disadvantages. Therefore, information created by a consumer has a higher credibility than content created by the retailer (Dellarocas, 2003).
11 perceived credibility of OCR’s of known relationships on Facebook and anonymous OCR’s from strangers on an independent platform. Likewise, De Maeyer & Estelami (2011) find that the context of OCR’s significantly influences the consumers perceptions. With the context they imply that reviews on independent platforms have a stronger influence, compared to dependent platforms. Reviews on independent platforms are perceived as objective. Other researchers agree that the persuasive impact of online consumer reviews is therefore, often attributed to the perceived non-commercial nature of the reviewer. Consumers are believed to have no vested interest in recommending a product or brand, and their implied independence makes their reviews more credible and consequently more useful than marketer-generated information (Bickart & Schindler, 2001; Herr et al., 1991).
For that reason, consumers are more likely to buy a product that is shown on an independent platform compared to a dependent platform (e.g. Sridhar & Srinivasan, 2012; Ahuis, 2013; Meuter, 2013).
The focus in this thesis will be on the independent platforms. There is no research on the different effect of consumers viewing two independent platforms versus consumers who view a single independent platform, on the consumer’s attitude toward a product. Fennis & Stroebe (2010) argue that repetition of information increases the consumer’s liking of a certain product. Therefore, it might be possible that there is a significant difference between the effect of OCR’s on 2 independent platforms and the effect of OCR’s on a single independent platform on the consumer’s attitude.
1.4 Consensus
Consensus in OCR’s leads to an increase of the possibility of information adoption (Zhang & Watts, 2003). According to Kelley (1967) consensus can be defined as the extent to which a person's responses to a certain stimulus on a particular occasion are similar to others' responses. Consensus information basically concerns with the positive or negative experiences of the majority of
consumers in the past, OCR’s and overall ratings of OCR’s (Benedicktus, et al., 1986).
A high consensus occurs if the person responds to the stimulus in the same way as others, while a low consensus results from a disagreement between the person's responses and others'. The overall direction of eWOM messages (positive or negative) affects the customer’s response, because
12 consensus in reviews represents the degree of agreement between two or more users regarding a product or its performance (Chiou & Cheng, 2003). Therefore, the eWOM messages with higher
consensus can be more persuasive and powerful than messages with lower consensus. Also, when the OCR is significantly different from the average rating, the OCR will be perceived as less helpful and believable (Chiou & Cheng, 2003).
1.5 Attitude toward a product
OCR’s have a persuasive influence on consumers’ attitudes toward a product and/or brand, however negative OCR’s have a stronger influence on the consumer’s attitude compared to positive OCR’s (Man Yee, et al., 2009; Lee, et al., 2007). Robert Zajonc (1968) found that repeated exposure to a stimulus favorably increases a person’s attitude and liking toward the product or brand, what is called the mere exposure effect (Fennis & Stroebe, 2010). An attitude can be defined as an acquired internal state that influences the choice of personal action toward some class of things, persons, or product (Gagné, 1984). However, an attitude cannot be observed directly, inferences must be made from one or another kind of observable behavior. In addition, the relation between reported attitudes and definite behavior is seldom found to be a close one (Gagné, 1984). The definition used in this research is the definition of Petty & Wegener (1997), they state that attitudes can be viewed as a summary of evaluations of objects (e.g. oneself, other people, products) along a dimension that ranges from positive to negative. While Gagné (1984) states that all we are able to say about attitudes is that attitudes influence behavior, infer Petty & Wegener (1997) that favorable attitudes can lead to brand and product preference.
1.6 Elaboration Likelihood Model (ELM)
non-13 content cues. Involvement is associated with the motivation to process the information, and prior knowledge (expertise) is associated with the ability to process the information.
1.7 Product Involvement and Product Knowledge
Richards & Root-Shaffer (1988) state that researchers widely agree that product involvement and prior knowledge about the product have a strong influence on word-of-mouth effects. Therefore, it is interesting to research if product involvement and prior knowledge have an moderating effect on the effect of adding an independent platform and the consensus on the consumer’s attitude toward the product. Involvement can be defined as an internal state of arousal with intensity, direction, and persistence properties of an individual (Andrews, et al., 1990). In this research the focus will be on product involvement. Product involvement can be defined as a person's perceived relevance of the product based on inherent interests, needs, and values (Zaichkowsky, 1985). Consumers differ in the extent of their decision process and their search for information, this depends on their level of involvement(Laurent & Kapferer, 1985). Thus, high-involvement consumers might search for more information and might visit more platforms or channels to gain more product knowledge. Findings of Sridhar & Srinivasan (2012) indicate that consumers who influence others, are more likely to be influenced themselves by other consumers, this influence is dependent on their experience with the product.
According to Rao & Monroe (1988) prior product knowledge is defined to encompass the amount of accurate information held in memory about the product and alternatives as well as the consumer’s self-perception of this prior product knowledge.
1.8 Problem statement and research questions
14 consumer. Knowing how a consumer’s attitude is influenced by OCR’s on different independent platforms can further enhance the marketer’s understanding of the effect OCR’s have on consumers.
The purpose of this paper is twofold:
(1) To investigate whether the effect of OCR’s is greater on the consumer’s attitude toward a product, when a consumer views OCR’s on two independent platforms compared to OCR’s on a single independent platform. Secondly, if product involvement and product knowledge have an influence on the effect of OCR’s on the consumer’s attitude toward the product.
15
2. Literature review
In order to research the effects of online consumer reviews on the consumer’s attitude toward a product, the involved variables are discussed in this chapter. First, a few insights on independent platforms are discussed. Second, the variable consensus explained. The independent variable independent platforms follows and the fourth section discusses the independent variable consensus. The dependent variable attitude is described in the fifth section. In the sixth and seventh sections will continue with the moderators product involvement and product knowledge. The type of product will be discussed in the part that follows and in the last section the conceptual model and hypotheses will be discussed.
2.1 Multiple OCR’s on multiple independent platforms
Research provided significant evidence that OCR’s on independent platforms have a greater influence on the consumer’s attitude toward a product compared to OCR’s on dependent platforms (e.g. Sridhar & Srinivasan, 2012; Ahuis, 2013, Meuter, 2013). Mere exposure of OCR’s could lead to an increase of linking of a certain product. Therefore, in this research the focus is on independent platforms, and if the effect of OCR’s on a second independent platform viewed by the consumer has a greater influence than OCR’s on a single independent platform viewed by the consumer.
2.1.1 Mere exposure effect of OCR’s on attitude
16 However, not all researchers agree that the volume, and repeated exposure to OCR’s about a product will influence the consumer’s attitude toward the product. Findings of Duan et al. (2008) show that consumers are capable of judging the quality of a product from online consumer reviews, without being influenced by the volume and ratings of these online consumer reviews. Results of Meuter (2013) also show that a higher volume of OCR’s does not lead to favorable perceptions by customers. His results indicate that six positive recommendations from known Facebook friends did not lead to more favorable perceptions than three positive recommendations from three Facebook friends. This was also the case for the other independent platform, Yelp.com. It did not matter whether a consumer views three or more positive reviews within one independent platform for the perception of the service.
Yet, other researchers do find that review quantity has a significant influence on the consumer’s attitude, even when the arguments of the OCR’s are weak (Lee, et al., 2008;Man Yee et al., 2009). A study by Vermeulen & Seegers (2008) shows that mere exposure to OCR’s improves the average probability for consumers to consider buying a product. The degree of positivity or negativity of the OCR’s did not matter, because all OCR’s created awareness and familiarity with the product. Even though negative reviews lower consumer attitudes toward the reviewed product, enhanced product awareness compensates for this effect. This gives supportive claims to why volume of OCR’s are likely to influence the consumer’s attitude toward a product. The quantity of reviews will researched in the sense of comparing the influence of OCR’s on two independent platforms compared to the influence of OCR’s on a single independent platform. This is backed up by recent research of Kim et al. (2013) who show that the volume of OCR’s influences consumers in decision making, when consumers read and hear more information about the product the consumer is more likely to buy the product.
The above leads to the following hypothesis:
17
2.2 Consensus
2.2.1 Intra- and inter-consensus
In this research there is a distinction made between inter consensus (consensus among two types of independent platforms) and intra consensus (consensus on one particular platform). The distinction will make the direction of this research more clear, the inter consensus among OCR’s on two independent platforms. However, there is little written about these types of consensus. Olsen & Shorrock (2010) use something similar in their research, they use inter consensus as comparing different coders between two incident reports. When the two incident reports show consistent behavior in one or more categories then there is inter consensus among the two coders. For intra consensus Olsen & Shorrock (2010) tested one individual over time, to test if there are (in)consistencies for one individual coder over time. Comparably, if the OCR’s within a platform are consistent with each other, we use the term intra-consensus, while consistency between reviews on two different platforms occurs, we label this inter-consensus. Although these types of consensus are different and might have different effects, most literature is concerned with intra-consensus, so if OCR’s are placed on a platform and are consistent with other OCR’s or the aggregated rating on the same platform. Therefore, in this section theory on both types of consensus will be discussed.
2.2.2 Consensus and attitude
18 consistent information as an evidence of the truth. Consensus in provided information is therefore, an important influencer of consumers attitudes toward a product.
Positive reviews can be helpful in promoting positive attitudes toward products or services.
When a OCR is inconsistent with the prior knowledge and experience of the consumer, the OCR will have a negative influence on the information adoption of the consumer. In contrast, when a OCR is consistent with the prior knowledge and experience of the consumer, the OCR will deliver a positive influence on the information adoption of the consumer (Zhang and Watts, 2003)(Man Yee et al., 2009). A few negative reviews within the mass of positive messages are not critically harmful (Sun-Jae & Jang-Sun, 2009). Research by Meuter (2013) shows that consensus in OCR’s on an independent platform has an influence on consumers’ attitudes toward a product or service. Therefore, it would be interesting to find out if inter consensus among the two independent platforms, has an effect on the consumers attitude toward a product. There are four possible outcomes when using two independent platforms. The intra-consensus on platform 1 can either be positive or negative, this holds as well for platform 2. Therefore, the inter-consensus can be positive or negative, and there can also be no consensus when the intra-consensus on platform 1 is negative and platform 2 positive or the other way around (see figure 1).
19 The focus in this research is on the inter consensus between two types of independent platforms. The research on third party platforms was mainly on the consensus of online consumer reviews on a single particular independent platform.
This leads to the following hypothesis:
H2. The impact of online consumer reviews on the consumer’s attitude toward a product is larger when there is inter consensus between the online consumer reviews on both independent platforms.
2.3 Elaboration Likelihood Model
2.3.1 Consumer’s product involvement
20 Researches widely agree that there are high-involvement consumer and low-involvement consumers (Petty & Cacippo, 1984; Laurent & Kapferer, 1985; Zaichkowsky, 1985; Mittal & Myung-Soo, 1988; Park,Lee & Han, 2007). Whereas highly involved consumers have a high motivation and high interest for the search of product information and purchase of a particular product, low involved consumers have low motivation and a low interest for the search of product information and purchase for that particular product. So, product involvement is related with the motivation to process information in ELM. The ELM provides a perspective on how consumers process information of online consumer reviews. Low involved consumers are not motivated to process the online consumer reviews and are persuaded by the number of online consumer reviews. Highly involved consumers will seek a lot of information, because they are motivated to process the information provided by online consumer reviews (Park,Lee & Han, 2007). The degree of involvement has also a strong influence on the degree of change in the consumer’s attitude toward a product. Consumers with a high degree of involvement are influenced more by OCR’s compared to consumers with a low level of involvement regarding the product (Lee, et al., 2008). Park, Lee & Han (2007) state that highly involved
consumers are affected by the review quantity, they show as well that review quantity is significant even when the arguments of the online consumer review are weak.
Thus, highly involved consumers have a high motivation and high interest for the search of product information. Also consumers with a high degree of involvement are influenced more by OCR’s compared to consumers with a low level of involvement. High-involvement consumers will be more likely to look at the quality of reviews compared to low-involvement consumers, but are mostly influenced by the quantity. Consumer with low involvement are also more likely to only look at the quantity, they may consider many reviews as an indicator that the product is very popular. The expectation is that high-involvement consumers are influenced most by the online consumer reviews of the two independent platforms. This leads to the following hypothesis:
H3a. The higher a consumer’s product involvement, the larger the impact of seeing more instead of one online consumer review on the consumer’s attitude toward a product.
21 because these consumers are less involved with the topic, therefore, they would rely more on online consumer reviews which are consistent with the average rating (Sun-Jae & Jang-Sun, 2009, Man Yee, et al., 2009). The above leads to the following hypothesis:
H3b. The higher a consumer’s product involvement, the lower the impact on the consumer’s attitude toward the product when there is inter consensus among the independent platforms.
Highly involved consumers have a high motivation and high interest for the search of product information like OCR’s, and are influenced more by these OCR’s compared to low involved consumers. Low involved consumers lack motivation and have no interest for searching product information and are less influenced by OCR’s. Therefore, it is likely that the degree of product involvement has a direct effect on the consumer’s attitude toward a product. The direct effect is expected to be stronger for highly involved consumers.
The above leads to the following hypothesis:
H3c. The higher a consumer’s product involvement, the larger the impact of the direct effect of product involvement on the consumer’s attitude toward a product.
2.3.2 Consumer’s product knowledge
In this study, product knowledge is assumed to moderate the effect of OCR’s on attitude. The effect of product involvement can be explained by using the Elaboration Likelihood Model. But first, product knowledge consists of two components, familiarity and expertise (Alba and Hutchington, 1987). Familiarity is defined as the number of product-related experiences acquired by a consumer. Expertise is the capability of the consumer to perform product-related tasks successfully. Product knowledge about the product has a strong influence on WOM effects (Richards & Root-Shaffer (1988).
22 effortful processing of the online consumer reviews, however this relates with the motivation (involvement) of the consumer (Park,Lee & Han, 2007). When a certain product is familiar and liked, consumers are willingly to defend their liking of that product, by analyzing mainly the positive online consumer reviews about the product (Ahluwalia, 2002).
Accordingly, consumers would consider the recommendation rating more if they are familiar with the discussion forum or platform. An explanation could be that familiar consumers have a better understanding of the rating system mechanism, and therefore, use the rating system more as an indicator of the product’s popularity. In the field of OCR’s and its effect, prior knowledge or familiarity has been used as a moderator before, prior knowledge moderates the impact of online consumer reviews on eWOM credibility (Man Yee et al., 2009). Sun-Jae & Jang-Sun (2009) use the the degree of involvement and prior knowledge as well as moderators. The results show that prior knowledge only partially moderate the relationship between the ratio of positive online consumer reviews and the degree of change in the consumer’s attitude toward the product. Consumers with prior knowledge turned out to be more sensitive to negative online consumer reviews compared to consumer with no prior knowledge. However, the focus in the current research is on the effect of online consumer reviews of an additional independent platform compared to the effect of online consumer reviews of only one independent platform on the consumer’s attitude toward a product.
23 The above leads to the following hypothesis:
H4a. The higher a consumer’s product knowledge, the larger the impact of seeing more instead of one online consumer review on the consumer’s attitude toward a product.
Cialdini (1993) proposed that consumers use consistent information as an evidence of the truth. When the consumer is not familiar to the product, the consumer is more likely to adopt consistent information. Consumers without prior knowledge are expected to blindly follow the consistent information given. However, still the influence is expected to be stronger for consumers with high product knowledge given the findings of Sridhar & Srinivasan (2012).
The above leads to the following hypothesis:
H4b. The higher a consumer’s product knowledge, the larger the impact on the consumer’s attitude toward the product when there is inter consensus among the independent platforms.
The ability of consumers to understand the rating systems of OCR’s on platforms related to the product depends on the degree of product knowledge. Consumers with high product knowledge are familiar with the product and related platforms, while consumers with low product knowledge are not or less familiar with the product and related platforms. Sridhar & Srinivasan (2012) show that consumers with high product knowledge are expected to be influenced more by OCR’s compared to consumers with low product knowledge. Therefore, it is proposed that the degree of product knowledge has a direct effect on the consumer’s attitude toward the product.
The above leads to the following hypothesis:
24
2.4 Conceptual model
Figure 2. Conceptual model
25
3. Methodology
The first section of chapter 3 is the research design, this section discusses the participants and the experimental design. The procedure follows where is explained what participants needed to do for the experiment. In section 3 are the measurements of the concepts discussed, a quick overview in the form of a table is given. The last section of this chapter is the plan of analysis. But first let us start with the type of product & platform chosen.
Type of product chosen:
A product can be categorized into two types of products namely, utilitarian and hedonic products. The goal of utilitarian products is maximizing the utility for the consumer, with the product’s tangible attributes. Purchasing utilitarian products is goal-oriented consumption, and is motivated primarily by the desire to fill a basic need or to complete a functional or practical task.
The goal of hedonic products is to maximize the satisfaction of emotional wants. Purchasing hedonic production is pleasure oriented consumption, and is motivated generally by the desire for sensual pleasure, an experience concerning feelings, intuition, fantasy and fun (Hirschman & Holbrook 1982; Sen & Lerman 2007; Strahilevitz & Myers 1998). Consumers generally find negative reviews most useful for utilitarian products, because consumer are likely to find that reviewer’s opinions are external motivations, which are product related. Contrary, for hedonic products consumer are likely to ignore negative reviews instead of value the negative reviews. The reason behind this behavior is that for hedonic products, consumers are more likely to relate the negative opinions of the reviewer to internal reasons of the reviewer, which are not related to the product. In this research the type of product used will be a hedonic product, more specifically a movie.
The internet movie database (IMDb) categorizes movies into 22 genres:
1. Action 7. Horror 13. Biography 18. Music
2. Animation 8. Musical 14. Crime 19. Mystery
3. Comedy 9. Romance 15. Drama 20. Sci-Fi
4. Documentary 10. Sport 16. Fantasy 21. Thriller
5. Family 11. War 17. History 22. Western
26 According to Nielsen (2012) the most popular movie genre in the US was ‘action/adventure’. 61% of the participants indicated that action/adventure movies are the movies they would like to see the most in theaters. So, the movie chosen to build the OCR upon for this experiment is Thor: The Dark World. This movie is categorized in the most popular genre, ‘action/adventure’.
Type of platforms chosen:
In this research there are two independent platforms used, on which people can give their opinions in the form of online consumer reviews. Platforms have different sets of goals and can be categorized by 11 different goals (Schaupp, 2006). However, according to the literature it is complex to label a platform into one type of category, because a platform can have multiple goals (see appendix A for for the different goals and definitions). The chosen independent platforms are:
- YouTube its primary goal: Entertainment
- IMDb its primary goal: Informed decision - unbiased
Trustpilot.com shows that their users (objective consumers) rate YouTube 7,9 and IMDb 9,1, out of 10. Both score high and are therefore, perceived as trustworthy platforms.
The online consumer review systems are different for both independent platforms as they both have different primary goals. YouTube shows how often the (short) video or trailer is played by viewers, the rating by the number of thumbs up & the number of thumbs down, while also OCR’s with content can be posted by consumers. IMDb shows the average star rating information, and the total number of reviews.
3.1 Research Design
First, the participants are described, followed by the experimental design.
3.1.1 Participants
27
3.1.2 Experimental design
The quantitative research will be executed in the form of a between participants design, specifically a 2 (consensus platform 1: + vs. -) x2 (consensus platform 2: + vs. -) design with two hanging control conditions (see figure 3). Participants assigned to one of the two hanging control conditions will only view positive OCR’s of one platform. This experimental design will be used because this research will consist of two analyses. The hanging control conditions are added in order to test the possible difference in the effect of OCR’s on 2 independent platform compared to the effect of OCR’s on one independent platform on the consumer’s attitude toward a product (analysis 1). Therefore, hanging control condition 1 shows only positive OCR’s on YouTube, and hanging control condition 2 only positive OCR’s on IMDb. The effect of these conditions will be compared with the effect of condition 6, where participants view positive OCR’s on both YouTube and IMDb. Conditions 3,4,5 and 6 will be used to test the effect of inter consensus on the consumer’s attitude toward a product (analysis 2). In order to collect data for the qualitative research, questionnaires will be used. The questionnaires will be distributed online. The participants are randomly assigned to six conditions.
Figure 3. Experimental design
Analysis 1:
28
Figure 4. Experimental design: analysis 1
Condition 1: Hanging control condition (1) Positive OCR’s on YouTube Condition 2: Hanging control condition (2) Positive OCR’s on IMDb Condition 6: Positive OCR’s on Youtube & Positive OCR’s on IMDb
Analysis 2:
The second analysis tests the effect of inter consensus among OCR’s placed on the independent platforms on the consumer’s attitude toward the product. The two hanging control conditions are not used in this analysis because in these conditions the participants only viewed one platform. The other four conditions are the possible outcomes in viewing two platforms (see figure 5).
Figure 5. Experimental design: analysis 2
Condition 3: Negative OCR’s on YouTube & Positive OCR’s on IMDb (no consensus)
Condition 4: Negative OCR’s on Youtube & Negative OCR’s on IMDb (negative inter consensus) Condition 5: Positive OCR’s on YouTube & Negative OCR’s on IMDb (no consensus)
29
3.2 Procedure
The participants were approached online via the social media channel Facebook. A link was published on a personal Facebook page and participants were approached via personal messages to take part in a study concerned the effect of online consumer reviews on the consumer’s attitude toward a product. In addition, the participants were told that they had a chance of winning 25 Euros. When the participants agreed to take part in the study, they were primed that they were in a certain situation. More specifically that they decided to go the movie theatre to watch Thor: The Dark World, however before going to the movie theatre the participants would get online consumer reviews from two independent platforms about Thor: The Dark World. The participants were asked to watch the online consumer reviews thoroughly. The participants who were assigned to condition 1 & 2 received only positive reviews from one independent platform (YouTube and IMDb) (both hanging conditions in order to test the addition of a second independent platform to just one). The participants of group 3,4,5 and 6 got two see reviews from the independent platforms YouTube and IMDb (inter consensus condition). After having seen the online consumer reviews the participants completed measures of attitude, product involvement and product knowledge. The last part of the questionnaire were some demographic questions.
3.3 Measurement of concepts
The measurements of the concepts will be discussed in this section. The first two subsections discuss the manipulations of the independent variables. The measurement of the dependent variable attitude follows in subsection 3.3.3. In the fourth and fifth subsections are the measurements of product involvement and product knowledge explained. In the last subsection shows an overview of the concepts in the form of a table.
3.3.1 Manipulation of seeing OCR’s on one or two independent platforms
30 overall rating of thumbs up and thumbs down, and the positivity in textual reviews. For the positive manipulation the number of thumbs up was 18.815 and the number of thumbs down 525. Since the main goals of YouTube are to entertain users and to give the users the opportunity to socialize by giving opinions, a positively written online consumer review was added for the manipulation of YouTube. The overall rating of IMDb is indicated by a star rating of 1 till 10 and the number of users who rated the movie. The rating of the movie in the manipulation was an average star rating of 7,6. Lovett et al. (2013) found that the order of importance of the three most important drivers in eWOM is social, functional, and emotional. Therefore, the participants who were assigned to condition 6 of getting to see both online consumer reviews of YouTube and IMDb, got to see the online consumer reviews on YouTube first, followed by the average online consumer rating on IMDb.
31
Figure 6 Positive manipulation of YouTube
32
3.3.2 Manipulation of inter consensus among independent platforms
The same product Thor: The Dark World and the same independent platforms were used in order to measure the effect of inter consensus among the independent platforms on the consumer’s attitude toward the product. The level of positivity and negativity of the online consumer reviews on YouTube is indicated by the overall rating of thumbs up and thumbs down, and the positivity/negativity in textual reviews. For the positive manipulation the number of thumbs up was 18.815 and the number of thumbs down 525 (positive intra-consensus on the platform). For the negative manipulation the numbers were reversed, 525 thumbs up and 18.815 thumbs down ( negative intra-consensus on the platform). A positively written online consumer review was added for the positive manipulation of YouTube and a negatively written online consumer review was added for the negative manipulation of YouTube. The overall rating of IMDb is indicated by a star rating of 1 till 10 and the number of users who rated the movie. The rating of the movie in the positive manipulation (positive intra-consensus on the platform) is an average star rating of 7,6 . The rating of the negative manipulation (negative intra-consensus on the platform) is an average star rating of 4,6, both manipulations had 79.402 reviewers.
There were four possible conditions for the respondent:
Condition 3: The participant got to see negative OCR’s on YouTube (-), and positive OCR’s on IMDb (+). No consensus (-+) among the independent platforms.
Condition 4: The participant got to see both negative OCR’s on YouTube (-), and negative OCR’s on IMDb (-). Therefore, there was negative inter consensus (--) among the independent platforms. Condition 5: The participant got to see positive OCR’s on YouTube(+), and negative OCR’s on IMDb (-). No consensus (+-) among the independent platforms.
Condition 6: The participant got to see both positive OCR’s on YouTube (+), and positive OCR’s on IMDb (+). Therefore, there was positive inter consensus (++) among the independent platforms.
33 different for condition 5 and 6. While in condition 5 there was no consensus. So, respondents in condition 5 may not have consciously viewed that there was no consensus between the reviews.
Below are the manipulations of negative OCR’s on YouTube and negative OCR’s on IMDb presented. It is actually what the respondents of condition 4 saw, inter consensus among the negative OCR’s on the two independent platforms. The respondents of condition 3 & 5 saw one of the above presented manipulation of positive OCR’s on one independent platform (figure 6 & 7), and these respondents saw also a manipulation of negative OCR’s on the other independent platform (figure 8 & 9).
Figure 8 Negative manipulation of YouTube
34 After having viewed the manipulations of the platform(s) the respondent needed to answer some questions of the questionnaire. The buildup of the questionnaire is presented below in table 1(see the full questionnaire in appendix B). The questions related to each variable and the Cronbach’s Alpha’s are discussed in the following sections.
Table. 1 Buildup questionnaire
3.3.3 Measuring the dependent variable attitude
In order to measure the dependent variable attitude a 7-point Likert scale of Osgood et al. (1957) with three items was used. Therefore, a semantic differential scale is used. The following three bipolar items should be rated on a 7-point Likert scale: 1. Bad/Good; 2. Unfavorable/Favorable & 3. Negative/Positive. The reliability analysis of these items showed a Cronbach’s Alpha of 0.96, which implies that the items are reliable. The three items are combined in a composite measure, attitude.
3.3.4 Measuring product involvement
Zaichkowsky (1985) researched product involvement and came up with a measurement scale of product involvement. The measurement scale consist of a 7-point Likert scale with 20 items.
Laurent & Kapferer (1985) created a different measurement method of product involvement. A 4 item scale was developed, and every item consists of a minimum of 3 items. With their research they proved that the method was valid. Mittal et al. (1988) used the 4 item scale of Laurent & Kapferer Buildup of questionnaire
1: A short introduction
2: Manipulation of OCR’s of the platform(s)
3: Questions concerning the respondent’s attitude toward the product and intention of going to the theatre to watch the movie.
4: Questions concerning the respondent’s product involvement
5: Questions concerning the consumer’s product knowledge
6: Manipulation check: Do you feel there is consensus between the reviews of the two platforms?
35 but their remark was that there was not a clear distinction between brand-choice involvement and product involvement. Mittal et al. (1988) used the 4 items separately for brand-choice involvement and product involvement and added measures for all 4 facets. The items developed had good internal reliabilities for the perceived importance of brand choice, sign value of the product, sign value of the brand, and the hedonic value from the product. The other three facets, perceived brand risk, perceived product risk and hedonic value at brand level, had modest reliabilities. In this research the measurement scale of Zaichkowsky (1985) was used, because Zaichkowsky’s research method relates to the concept of situational involvement. Participants have to score the 20 bipolar items at fairly high speed without worrying or puzzling too much.
Thus, the items used in this measurement scale of product involvement are measured on a 7-point Likert scale. 10 of the 20 items were reversed when used in the questionnaire, and therefore, these 10 items are re-coded into different variables. The reliability analysis of these items showed a Cronbach’s Alpha of 0.95, which implies that the items are reliable. The 20 items are combined in a composite measure, product involvement. For the full set of items see table 2, but three example items are:
- Unimportant/Important - Not needed/Needed -Unappealing/Appealing
3.3.5 Measuring product knowledge
36
3.3.6 Overview of the measurements of concepts
Variable Construct Literature Scale Items
IV Type of
independent platform
IV Consensus Manipulation check:
7 point Likert scale, 1 item.
Scale 7-point Likert scale: positive vs negative.
Question: Do you feel there is consensus between the reviews of the two platforms? DV Attitude toward the product Osgood, et al. (1957). Muehling, Laczniak, and Stoltman (1991)
7 point Likert scale, 3 items, Cronbach’s alpha: 0,96.
1.Good vs Bad.
2. Favorable vs Unfavorable 3. Positive vs Negative
Control variable for DV
Intention to go to the particular movie
7 point Likert scale, 1 item.
Scale 7-point Likert scale: positive vs negative
Question: Your intention was going to the theatre to watch Thor 2: The Dark World. How would this be changed after reading the reviews?
Moderator Product
involvement
Zaichkowsky (1985)
7 point Likert scale, 20 items, Cronbach’s alpha: 0,95.
1. important/ unimportant [R] 2. of no concern/ of no concern to me. 3.irrelevant/ relevant 4. means a lot to me/ means nothing to me [R] 5. useless/ useful
6. valuable/ worthless [R] 7. trivial/ fundamental
8. beneficial/ not beneficial [R] 9. matters to me/ doesn’t matter [R] 10. uninterested/ interested
11. significant/ insignificant [R] 12. vital/ superfluous [R] 13. boring/ interesting 14. unexciting/ exciting
15. appealing/ unappealing [R] 16. mundane/ fascinating 17. essential/ nonessential [R]
18. undesirable/ desirable 19. wanted/ unwanted [R] 20. not needed/ needed
Moderator Product
knowledge
Flynn, Goldsmith, and Eastman (1996)
7 point Likert scale, 6 items, Cronbach’s alpha: 0,97.
1. I feel quite knowledgeable about ….
2. Among my circle of friends, I’m one of the ‘experts’ on … 3. I rarely come across a …. that I haven’t heard of. 4. I know pretty much about….
5. I do not feel very knowledgeable about ….. [R]
6. Compared to most other people, I know less about …. [R] 7. When it comes to …., I really don’t know a lot. [R] 8. I have heard of most of the new … that are around. Table 2. Overview of measurements of concepts
37
3.4 Descriptives
3.4.1 Basic descriptives
Descriptive analysis is used to provide insight into the demographic characteristics of the respondents by giving overviews of frequencies. The number of respondents participated in this research is 197. 59% of the respondents are male (117) and 41% of the respondents are female (80), see figure 10. While the ratio of the Dutch population is 49,5% male and 50,5% female (CBS, 2013), so the sample does not corresponds with the actual Dutch population on gender.
Figure 10. Gender.
38
Condition: Number of
respondents
1: Only positive OCR’s YouTube 26
2: Only positive OCR’s IMDb 32
3: Negative OCR’s YouTube & positive OCR’s IMDb 30
4: Both negative OCR’s on both Youtube & IMDb 38
5: Positive OCR’s on YouTube & Negative OCR’s on IMDb 37
6: Both positive OCR’s on both Youtube & IMDb 34
Total number of respondents 197
Table 3. The number of respondents per condition
3.4.2 Means of attitude toward the product and intention
The means of the six conditions regarding the attitude toward the product per condition will give some basic understanding of the data. The intention of going to the movie after having seen the OCR’s was measured as a control variable of attitude.
OCR’s on 2vs1 independent platform
39
Figure 11. Mean scores conditions of OCR’s on 2vs1 independent platform
Inter consensus
The outcomes of the post-hoc test with respect to the mean attitude toward the product, show that each condition significantly differs from another (see figure 12). So, the mean attitude toward the product is significantly different between the condition with positive inter consensus (condition 6) and the condition with negative inter consensus (condition 4). Surprisingly, also between condition 3 (positive YouTube - negative IMDb) and condition 5 (negative YouTube - positive IMDb) there is a significant difference in the respondent’s attitude toward the product.
40
Figure 12. Mean scores conditions of inter consensus
3.5 Homogeneity of slopes
3.5.1 OCR’s on 2vs1 independent platform (analysis 1)
41
3.5.2 Inter consensus (analysis 2)
The first custom ANCOVA analysis contained the independent variable inter consensus, dependent variable attitude toward the product and the covariates product involvement and product knowledge. The results indicate that product involvement has a significant influence (0,031) on the effect of inter consensus on the consumer’s attitude toward the product, thus homogeneity of slopes. Therefore, a fixed factor of product involvement was created by using the median of product involvement. Product involvement was recoded into a different variable and created two groups: Group 1 low conforming (below median) and group 2 highly conforming (above median). Then another custom ANCOVA analysis was performed including the fixed factor of product involvement, thus the only covariate left is product knowledge. The results of the second custom ANCOVA analysis showed that product knowledge has a significant influence (0,04) on the effect of inter consensus * product involvement on the consumer’s attitude toward the product. Before using product knowledge as a fixed factor in this analysis, the bivariate correlation coefficient was used to find out if product involvement and product knowledge correlate. The bivariate correlation coefficient of 0,24 indicate that there is low correlation between the two covariates. Taking into account that the sample will be very small when using also the fixed factor of product knowledge, the full factorial ANCOVA analysis contained both fixed factors of product involvement and product knowledge (see chapter 4 Results).
3.6 Plan of analysis
The plan of analysis with respect to the first question (analysis 1) is as follows:
Analysis of demographic data to provide insights in the group of respondents has already been done. Aggregate the variables by using a reliability analysis (Cronbach’s Alpha) is also completed. Thus: 1. Find the effect of the first independent variable which is proposed to influence attitude by means of an ANCOVA analysis; and 2. Find moderating effects of product involvement and product knowledge of the relation between the independent variables and attitude.
The following methods will answer the second research question (analysis 2):
42
4. Results
In this chapter are the results of the full factorial ANCOVA discussed for both analyses.
Figure 13 Overview analysis 1 and 2
4.1 OCR’s on independent platforms (2 vs 1)
The between subject effects on the dependent variable, attitude toward the product.
Source Significance level
Corrected model ,000
Intercept ,002
2vs1 independent platform (IV) ,365
Product Involvement (Covariate) ,000
Fixed factor product knowledge ,976
2vs1 independent platform * fixed factor product knowledge ,133
Table 4. The full factorial ANCOVA of analysis 1
43
4.2 Inter Consensus
The between subject effects on the dependent variable, attitude toward the product. The full factorial ANCOVA of this analysis contains no covariates. It contains the independent variable inter consensus, the fixed factors of product involvement and product knowledge and the dependent variable attitude toward the product.
Source Significance level
Corrected model ,000
Intercept ,000
Inter consensus (IV) ,000
Fixed factor product knowledge ,145
Fixed factor product involvement ,019
Inter consensus * fixed factor product knowledge ,058
Inter consensus * fixed factor product involvement ,016
Fixed factor product knowledge * fixed factor product involvement ,396 Inter consensus * fixed factor product knowledge * fixed factor product involvement
,039
Table 5. The full factorial ANCOVA of analysis 2
44 interaction effect of inter consensus with both fixed factors is also significant. The fixed factors do not have an interaction effect on the consumer’s attitude toward the product.
The first plot (figure 14) shows the same results as figure 12 the mean score of the attitude toward the product differs significantly between the each of the four conditions.
The largest difference is between negative inter consensus condition (4) and the positive inter consensus condition (6).
Figure 14. Consensus and mean attitude
The following plot (figure 15) shows the direct effect of the covariate product involvement. Low involved consumers are number 1 and the highly involved respondents are number 2 on the horizontal axe. The difference in the respondents attitude toward the product is large between the low- and highly involved consumers. The low involved respondents have more favorable attitudes toward the movie compared to the high involved respondents.
45 Surprisingly the effect of inter consensus is larger for high involved respondents compared to low involved respondents. The results indicate that high involved respondents tend to value the negative OCR’s more compared to the low involved respondents. However, when there is positive inter consensus among the OCR’s on both platforms has a stronger influence on the attitudes of the highly involved respondents. OCR’s on IMDb have a stronger influence compared to OCR’s on YouTube on the respondent’s attitude toward the product. When there is inter consensus between the OCR’s on YouTube and IMDb the effect of the OCR’s on the respondent’s attitude toward the product is the greatest.
46
Figure 17. The interaction effect of consensus and product knowledge
47
Figure 18. The interaction effect of consensus, product involvement and (low) product knowledge
When respondents are low involved and have also low product knowledge than they tend to value the positive OCR’s more than the negative OCR’s. When OCR’s on both independent platforms are negative (negative inter consensus), than the attitude toward the product becomes unfavorable. The effect of OCR’s on independent platforms is stronger on the attitude of highly involved respondents with low product knowledge compared to low involved consumers with low product knowledge. Highly involved respondents with low knowledge form stronger unfavorable attitudes toward the product when OCR’s on YouTube are positive and the OCR’s on IMDb are negative compared with OCR’s with negative inter consensus among the independent platforms.
48
Figure 19. The interaction effect of consensus, product involvement and (high) product knowledge
The above results lead to the acceptation or rejection of hypotheses:
49
Hypothesis Accepted
/ Rejected
H1. The impact of online consumer reviews on the consumer’s attitude toward a products is larger when reviews are shown on two independent platforms than on a single independent platform.
H2. The impact of online consumer reviews on the consumer’s attitude toward a product is larger when there is inter consensus between the online consumer reviews on both independent platforms.
H3a. The higher a consumer’s product involvement, the larger the impact of seeing more instead of one online consumer review on the consumer’s attitude toward a product.
H3b. The higher a consumer’s product involvement, the lower the impact on the consumer’s attitude toward a product when there is inter consensus among the independent platforms.
H3c. The higher a consumer’s product involvement, the larger the impact of the direct effect of product involvement on the consumer’s attitude toward a product.
H4a. The higher a consumer’s product knowledge, the larger the impact of seeing more instead of one online consumer review on the consumer’s attitude toward a product. H4b. The higher a consumer’s product knowledge, the larger the impact on the consumer’s attitude toward a product when there is inter consensus among the independent platforms. H4c. The higher a consumer’s product knowledge, the larger the impact of the direct effect of product knowledge on the consumer’s attitude toward a product.
50
5. Discussion
In this chapter the research questions are being answered based on the data presented in the previous chapter. Recommendations are made concerning managerial implications. The third section discusses the limitations concerning the current research and suggestions for future research are offered.
5.1 Conclusions
5.1.1 Research question one
The purpose of the first research question was to investigate whether the effect of OCR’s on the consumer’s attitude toward a product is stronger when a consumer views OCR’s on two independent platforms than when a consumer views OCR’s on a single independent platform. Secondly, if product involvement and product knowledge have an influence on the effect of OCR’s on the consumer’s attitude toward the product.
51 are (established known) independent platforms, so consumers trust and believe that the information provided by consumers is credible and trustworthy. According to Duan et al. (2008) are consumers capable of judging the quality of a product from online consumer reviews, without being influenced by the volume of OCR’s. Therefore, the consumer does not need to search for more OCR’s on different independent platforms, and already form favorable attitudes toward the product by viewing positive OCR’s on a single independent platform. Thus, the consumer’s attitude may be influenced by OCR’s on one independent platform.
5.1.2 Research question two
The purpose of the second research question was to investigate whether the effect of OCR’s on the consumer’s attitude toward a product is stronger when there is inter consensus among the OCR’s on the two independent platforms. Secondly, if product involvement and product knowledge have an influence on the effect of inter consensus among OCR’s on the consumer’s attitude toward a product.
Inter consensus
52 An interesting finding is the significant difference in the mean score of the respondent’s attitude toward the product between condition 3 and 5 (negative YouTube & positive IMDb versus positive YouTube & negative IMDb) A possible explanation is the finding that OCR’s on IMDb have a stronger impact on the consumer’s attitude toward a product compared to OCR’s on YouTube. Consumers are more likely to view OCR’s on IMDb compared to YouTube, and a lot of consumers check IMDb before watching the chosen movie.
Interaction effects
The theory of ELM indicates that low involved consumers are less motivated to process the online consumer reviews and are persuaded by the number of online consumer reviews. Highly involved consumers will seek a lot of information, because they are motivated to process the information provided by online consumer reviews (Park,Lee & Han, 2007). When the consumer is able to understand the online consumer review system of the platform, they are more likely to engage in deliberate and effortful processing of the online consumer reviews, however this relates with the motivation (involvement) of the consumer (Ahluwalia, 2002; Park,Lee & Han, 2007). Results show interaction effect between the direct effect of product involvement and product knowledge. However , the results of the current study indicate that the effect of inter consensus on the consumer’s attitude toward the product is influenced by product involvement and product knowledge. There is an interaction effect of inter consensus, product involvement and consumers with high knowledge. According to Alba & Hutchinson (1987) have consumers with high product knowledge the ability to evaluate the product by their previous experiences with the product and are less influenced by OCR’s. While the results of the current research indicate that highly involved respondents with high product knowledge are influenced most by OCR’s. In line with the findings of Zhang and Watts (2003) & Man Yee et al. (2009), OCR’s will deliver a positive influence on the information adoption of the consumer when the consumer has product knowledge and when this is consistent the influence will be even greater.
