How Third-Party Information Influences Perceived Quality

(1)

How Third-Party Information

Influences Perceived Quality

The Moderating roles of Brand Reputation and

Susceptibility to Interpersonal Influence

(2)

How Recommendations Influence Perceived Quality

The Moderating roles of Brand Equity and Susceptibility to Interpersonal

Influence

Anamaria Cristina Marinescu

Faculty of Economics and Business, Department of Marketing

Master Thesis

26 of June 2013

Aweg 2a 9718 CS Groningen T: +31(0)681229453 E: ana.marinecu@gmail.com Student number: s1946226 Supervisor: Dr. Jia Liu

(3)

Management summary

(4)

Preface

It has been clear before, but even clearer now that academic research needs strength of mind and meticulous thinking. I have learned much throughout the process, and I can clearly say that my analytical capabilities have increased. The process which has been employed in writing this thesis helped a lot, pushing me to manage my deadlines, and giving me insight in problems that we were all confronting with. I also believe that working in a group creates the peer pressure, making each and every one of us eager to do good so that we do well in our group. An extra bonus is of course also the fact that when one has a small problem, you can always ask your group members and see if they had encountered it as well and can give some advice. I am certain that I have learned a lot from this process, and have gained countless respect for all the researchers and authors that do this for a living; for their tenacity and patience as well as the countless insight that they provide through their papers.

(5)

1. I

NTRODUCTION

According to USA Today, when Oprah Winfrey first recommended a book in 1996, people began to instantly read it. To this day, her influence with people's choices has made bestsellers from no-names (Minzesheimer, B. 2011). Another example is Top Gear's Jeremy Clarckson whom many people consider a car expert and listen to his opinion about which car is good or not (Shunk, C. 2009). It can thus be said that when consumers evaluate products, their perceptions of that product is often relied not only on marketing cues, but on the opinion of third-party's reviews or ratings, assumption which is supported by the findings of Archibald, et al. (1983), as well as Lyons and Henderson (2005).

Furthermore, a survey conducted in 2008 has shown that out of 600 respondents, 90% stated that they trusted the reviews read online and 86% of the respondents were checking product reviews before purchase (Miller, 2008). In the paper of Moe and Trusov (2011), the authors conclude that product ratings are of importance and have an impact on product sales. Furthermore, research shows that information given either by users or by experts has an influence on consumer decision making (Ardnt, 1967, Duhan et al., 1997; Gilly et al. 1998; Olshavsky and Granbois, 1979).

Research has shown that in situations where the quality of a product is not easily measured, consumers seek insights from trustworthy experts and third party ratings (Akdeniz, et al., 2013; Jiang, et al. 2008). Furthermore, the nature of the recommendations is of great importance. Regarding this, researchers have found that positive or negative recommendations can influence product trial adoption (Arndt, 1967) the behavior of switching to other products (Wagenheim and Bayon, 2004), and consumer attitudes toward the product (Bone, 1995). A further study that shows the difference in reactions when exposed to positive or negative recommendations has been performed by Burzynski and Bayer (1977), in which individuals were exposed to positive or negative commentaries before going to see a movie. People who were exposed to positive commentaries evaluated the movie better than people exposed to negative commentaries.

(8)

influenced into looking for and accepting said recommendations. While previous research into this topic has taken into account brand equity (Akdeniz et al., 2013), characteristics of opinion leaders (Lyons and Henderson, 2005), or trust cues (Benedicktus et al., 2010), the research is lacking when looking at the characteristics of the consumer. More precisely, there is little to no literature dealing with the influence that a consumer's susceptibility to influence has on the strength of the relationship between recommendations and perceived product quality. It is important to study the characteristics of the consumer in this context, especially how easily influenced one is, because this particular characteristic in a consumer could determine whether the recommendations are effective or not.

Looking into previous research concerning the topic of susceptibility, McGuire (1968) terms the phenomenon of people listening to another's opinion as susceptibility to interpersonal influence. The author concludes in his paper that susceptibility to interpersonal influence is a general personality trait that varies across persons. A more recent study of Bearden, Netemeyer and Teel, (1989) build upon the conclusions of McGuire (1968), with the authors developing a scale to measure consumer susceptibility to interpersonal influence. In the article the authors make a distinction between general susceptibility to interpersonal influence, which had been previously researched, and consumer susceptibility to interpersonal influence. Thus, the concept is defined by the authors as the tendency to learn about products and services by either observing or seeking information from others, the need to identify oneself with a product or to enhance one's personal image through the procurement of a product, or willingness to conform to others expectations when it comes to buying products (Bearden et al., 1989).

(9)

1.1 Problem Statement

Though there have been papers which researched the influence of third-party information (Akdeniz et al., 2013; Lyons and Henderson, 2005; Benedicktus et al., 2010), there has been no research into how consumer susceptibility would affect this interaction. The aim of the present study is to research in the influence that third-party information has on perceived product quality, taking into account the moderating roles of brand equity and consumer susceptibility to interpersonal influence. Thus the problem statement can be translated in the following way:

How does consumer susceptibility to interpersonal influence and brand equity influence the relationship between recommendations (positive vs. negative) and perceived quality?

The problem statement will be answered with the help of statistical analysis. First, a survey will be conducted in which questions from the scale provided by Bearden et al. (1989) will be used to measure consumer susceptibility to interpersonal influence, combined with questions designed to test the two levels of recommendations, namely consumer generated information and opinion leader generated information. Furthermore, questions regarding the influence of brand equity will be used. The responses of the questionnaire will make the base for the data set which will be subsequently statistically analyzed. A Linear regression model will be used to analyze the main interaction effect between third-party information and perceived product quality as well as the moderating effect of the two variables, brand equity and consumer susceptibility to interpersonal influence.

1.2 Research Questions

The Problem Statement can be translated into the following research questions:

1. How do positive and negative recommendations affect perceived product quality?

(10)

3. How does consumer susceptibility to interpersonal influence moderate the effect of positive and negative recommendations on perceived product quality?

1.3 Theoretical and Managerial Relevance

The present paper contributes to the academic community by filling a gap in literature concerning the moderating effect that consumer personality characteristics such as susceptibility to interpersonal influence, has on the influence of recommendations on perceived product quality. Thus it adds another dimension which has not been researched before to the topic of social influence in marketing. Furthermore, Brand Equity will be investigated as a moderator in the relationship between recommendations and perceived quality, thus adding another dimension to the literature whole.

As presented in the section above, user as well as expert generated ratings can have an influence on sales, consumer purchase behavior or even subsequent ratings. While consumers might sometimes tend to trust expert opinions since they have a better knowledge of the product or situation, there is also the case that consumers look more for reviews and ratings generated by other users, hoping that they have encountered the same situations. Building upon the paper of Bearden, et al., (1989) about the 'Measurement of Consumer Susceptibility to Interpersonal Influence', in which the authors develop a scale to measure this concept, this paper will use that scale in order to uncover the role that a person's susceptibility to social influence plays in the relationship between third party ratings and perceived product quality. Furthermore, the present study extends the work of Senecal and Nantel (2004) by introducing the moderators of Brand Equity and Susceptibility to interpersonal influence. The presence of the two moderators together has not been studied yet, and thus the current paper is presented with the opportunity to shed light on that issue.

(11)

for managers to look more closely at the personality traits of their consumers. With the help of internet, sites like, for example amazon, zalando, or bol can see if a user is buying more frequently products recommended by others, and could thus adjust the targeting strategy for those strategies.

1.4 Structure

(12)

2. L

ITERATURE

R

EVIEW

The following section of the paper will analyze the variables to be used in detail, using existing literature. Definitions will be provided for every variable, and their interactions will be analyzed as well, thus creating the necessary hypotheses which will be later tested in the analysis. The section starts with the independent variable – recommendations–, then the independent variable – perceived product quality – will be defined, as well as the relation between the two will be discussed. Lastly, the two moderators – brand equity and susceptibility to interpersonal influence – will be introduced with their respective effects.

2.1 Recommendations

In the book of Hoyer et al.(2013), the authors talk about social influence, which the act of informing and either implicitly or explicitly receiving pressure from either other individuals, groups or the mass media. In line with this view, the paper of Senecal and Nantel (2004) also treats recommendations as a type of social influence. The current paper will adopt this view. Thus, the focus of the current paper will be on non-marketing sources of influence, which are defined by Hoyer et.al.(2013) as sources of information that have no ties to the marketing organization such as friends, blogs, reviews, news or virtual communities. According to Nelson (1974), due to an asymmetry in the information between sellers and buyers, product ratings, and reviews have emerged. Thus, in order to solve the problem on information asymmetry, product information is being delivered to consumers by experts or other consumers, as shown in the paper of Faulhaber and Yao (1989). In the paper of Palmatier, Dant, Grewal & Evans, (2006), the authors define recommendations as the process of mentioning positively a company or a product to another potential customer. However, as Mahajan, Muller and Kerin (1984) argue, the assumption that only positive information is being transferred to other consumers is weak, because consumers may also transfer negative information to others. Furthermore, as mentioned by Senecal and Nantel (2004), the research made on recommendations has mainly been done from the point of Word-of-mouth and personal influence.

(13)

at research regarding recommendations, reviews and word of mouth, one can see that much research has focused on the positive side of this topic (Fornell et al. 2006; Luo and Homburg 2007). On the other side, Mahajan, Muller and Kerin, (1984) conclude that negative recommendations have the tendency to influence consumers more than positive information would. As an example, a study by Chevalier and Mayzlin (2006) in which the authors investigate the effect that online consumer reviews have on book sales, shows that negative reviews lead to a greater decrease in sales than positive reviews do to an increase in sales. As these studies point out, it is important to look at positive as well as negative recommendations, since the effect of the two differ.

It is thus important to mention existing definitions of the two concepts of positive and negative recommendations. Concerning negative recommendations, Luo (2007) defines this concept as the display of negative opinions about product/service consumption, or a proof of customer complaint. Other papers regard the concept of negative recommendations as the reaction that customers have when a service fails (Godes and Mayzlin, 2004; Mahajan, Muller and Kerin, 1984), or the communication of negative experiences with products (Anderson, 1998; Wangenheim 2005). At the opposite end of the spectrum lie positive recommendations. In the article of Anderson (1998), positive word of mouth is seen as the act of spreading word about pleasant or novel experiences the customer has had with the product or service, and recommending it to others.

2.2 Perceived Product Quality

(14)

extrinsic quality cues as well as a distinction between experience and credence quality attributes. The model, as well as the paper brings a deeper understanding of the process of quality perception.

Perceived product quality has been treated in many papers as an important topic, thus generating a multitude of definitions regarding the concept. In the paper of Herbig and O'Hara (1994), perceived product quality is viewed as the judgment that a consumer has about the way that product conforms to specifications. In the paper of Parasuraman, Zeithaml (1988), the concept is seen as the consumer's evaluation concerning the product's superior value added capability. Another definition is presented by Zeithaml (1988), in which perceived quality is seen as a global assessment which is characterized by a high abstraction level and concerns the judgments made by consumers concerning a product's (or service's) overall excellence or superiority, definition which is also used by Rowley (1998). Another definition used by Aaker (1991) looks at perceived quality as being the perception of overall quality or superiority of a product or service, relative to other products or services which have the same purpose. As it can be seen throughout the years there have been many definitions on perceived quality; however it can be seen that most definitions consider perceived quality as a judgment or assessment which cones from the consumer regarding the quality of a product or service. However these perceptions can vary per consumer, an assumption which is supported by Ophuis and van Trijp (1995). A highly comprehensive and extensive definition of perceived product quality is being offered by Steenkamp (1990) which states that the concept is an idiosyncratic value judgment which deals with the fitness for consumption This is based upon the conscious or unconscious processing of quality cues relating to quality attributes in the context of personal and situational variables Below, an overview of the definitions can be seen.

Definitions Authors

Perceived product quality is viewed as the judgment that a consumer has about the way that product conforms to specifications

Herbig and O'Hara (1994)

Perceived product quality is the consumer's evaluation concerning the product's superior value added capability

Parasuraman, Zeithaml (1988)

(15)

characterized by a high abstraction level and concerns the judgments made by consumers concerning a product's (or service's) overall excellence or superiority

Perceived quality is the perception of overall quality or superiority of a product or service, relative to other products or services which have the same purpose.

Aaker (1991)

Perceived product quality is an idiosyncratic value judgment which deals with the fitness for consumption This is based upon the conscious or unconscious processing of quality cues relating to quality attributes in the context of personal and situational variables

Steenkamp (1990)

Table 1: Overview of Perceived Quality Definitions

While looking at the process of perceived product quality, it is important to mention that perceived quality differs from objective quality. In the paper of Zeithaml (1988), perceived quality is seen as an abstract concept which refers to a specific consumption setting. On the other hand, objective quality represents the actual technical excellence of a product and can be easily measured, as described by Monroe and Krishman (1985).

2.3. Relationship between the two variables

(16)

By looking further into literature, it can be seen that there exists the belief that recommendations or reviews can significantly influence the decisions of consumers (Ardnt, 1967, Duhan et al., 1997; Gilly et al. 1998; Olshavsky and Granbois, 1979). There have been many studies that have examined the effect that recommendations can have on the way consumers perceive a product or service. For example, an early study of Mahajan et.al.(1984) concluded that word of mouth is a significant predictor of movie attendance. A later study of Reddy et al.(1998) found that critics presented in newspapers have a significant impact on the success of Broadway shows. However, on the other side, in their study of the movie industry, Eliashberg and Shugan (1997) concluded that the critics posted are not an influencer, but rather predictors of future consumer behavior. Furthermore, authors such as Senecal and Nantel (2004) reporting that recommended products were chosen twice as often by participants that consulted product recommendations. Another noteworthy example is the article of Chevalier and Mayzlin (2006), in which the authors had found that in the book industry, consumer purchase behavior is influenced by online user book ratings. Thus, based on previous research it can be said that recommendations has a general influence on the behavior of consumers. Furthermore, by looking at the literature presented, it can be seen that recommendations have an influence on the way people perceive quality (Mahajan et al., 1984; Reddy et al., 1997). It can further be inferred that if a consumer is influenced to purchase a product by recommendations (Mayzlin, 2006; Senecal and Nantel, 2004), then it is to be assumed that the consumers had viewed that product as having a better quality then the rest.

(17)

that those movie-goers spread the word about a movie which they have heard positive things about due to the fact that their perceived quality of that movie has increased as a result of the recommendations.

This can be translated into the following hypothesis:

H1: Positive recommendations have a positive influence on the perceived quality of the product, while negative recommendations have a negative effect.

2.3 Moderation Effects

In addition to the main effect which is covered by the independent variable recommendations, the present study will address two moderating effect. The moderating variables are Brand Equity and Susceptibility to Interpersonal Influence, and will be discussed in the following sections.

2.3.1 Brand Equity

There have been many ways in which Brand Equity has been seen and tried to be measured. These vary from perspectives such as financial (Simon and Sullivan, 1993), the value of the brand to an acquiring firm (Mahajan, Rao and Srivastava, 1991), or the importance of the brand name (Sharkey, 1989; MacLachlan and Mulhern, 1990). However, the widely used definition of Brand Equity, which came to be known as customer-based Brand Equity was developed by Keller (1993). In his paper, Keller (1993) defines Brand Equity as the differential effect of brand knowledge on consumer response to the marketing of the brand (Keller, 1993, p.2). Furthermore, the author mentions that in order for customer-based Brand Equity to occur, the person must know the brand and must hold in their memory strong, positive associations to the brand.

2.3.1.1. Components and consequences of Brand Equity

(18)

has high brand knowledge, they will react more positively to the marketing of the brand than to the marketing of an identical non-branded product. Furthermore, brand knowledge is divided in brand awareness which relates to the strength of the brand in the memory of the consumer, and brand image, which refers to the perceptions about the brand (Keller, 1993).

The paper of Aaker and Biel (1993) concludes that one of the main ways in which favorable associations are concerning the brand are formed in consumer memory is through advertising. Studies have shown that advertising of a brand can increase the perceived quality of experience goods (Nelson, 1974), or how the perceived cost of the marketing campaign influences the product quality expectations (Wright, 1989). Furthermore, the paper of Keller (1993) states that a brand with high equity would have a positive effect on the sales of that brand as well as in making consumers less sensitive to price and thus be willing to pay a higher premium.

2.3.1.2 Moderating Effect

(19)

products based on the brand equity of the store in which these products are sold. This view is supported by Dodds, Monroe and Grewal, (1991).

A further article which touches upon the connection of brand equity and recommendations is by authors Herr, Kardes and Kim (1991). In this article the authors found that positive recommendations lead to favorable brand attitudes, and that these are formed regardless of other diagnostic information which might be available. However, the same article found in one of their experiments that the effect of recommendations is weakened when the impression on the product is well rooted in the mind of the consumer and if it is negative. In light of the current paper, it can be said that once a product has a favorable brand attitude, negative recommendations will not have such a great impact, compared to when the brand is not as strong. This leads to the following hypothesis:

H2: Brand Equity (BE) moderates the effect of recommendations on perceived quality, such that high BE will strengthen the effect of recommendations on perceived quality, compared to low BE.

2.4 Consumer Susceptibility to Interpersonal Influence

According to the book of Hoyer et al. (2013), the sources of influence which are marketing or non-marketing, and among which are opinion leaders and other consumers, can exert two types of influence, namely normative or informational. This is in concordance to the paper of Deutsch and Gerard (1955) in which the authors state that normative and informational influences are the two ways in which interpersonal influence is manifested. Thus normative influence refers to the social pressure experienced by people which urges them to conform to the expectations of others (Burnkrant and Cousineau, 1975). According to Wooten (2006), the consumer can be sanctioned by its peers if he/she does not give into norms, or can be rewarded if the norms are respected.

(20)

opinion, which has been termed by McGuire (1968) as susceptibility to interpersonal influence. McGuire (1968) concludes in his paper that susceptibility to interpersonal influence is a general personality trait that varies across persons. The concept developed by McGuire is taken one step further in the paper of Bearden et.al. (1989) calls this interpersonal influence as consumer susceptibility to interpersonal influence, and defines it as the tendency to learn about products and services by either observing or seeking information from others, the need to identify oneself with a product or to enhance one's personal image through the procurement of a product, or willingness to conform to others expectations when it comes to buying products.

Due to the fact that the aim of this paper is to find the effect of information on consumers, it is only the first part of the definition presented by the authors that is relevant. Thus, for clarification purposes consumer susceptibility to interpersonal influence will be seen as the tendency to learn about products and services by either observing or seeking information from others.

2.4.1 Consequences of being susceptible

In their paper, Bearden et al. (1989) have developed a scale to measure this multidimensional construct, which is split into normative and in informational items. The normative items deal with how easily influenced a person is by others in their behavior, while the informational items deal with the tendency to ask for the opinion of others. The authors concluded that both normative and informational influences were positively related to one's motivation to comply with the expectations of others (Bearden et al., 1989). Furthermore, additional tests on the scale revealed that people who scored high in the normative part of the scale were more prone to conformity and more concerned with the reaction of others, and that people who scored high on either of the dimensions were low on self-esteem.

(21)

The paper of Burkrand and Cousineau (1975) studies the informational and normative influences in the case of consumer buying behavior. The paper concludes that people base their evaluations of products on other people's reviews or evaluations. Furthermore, their perceived quality of a product is better in the presence of evaluations compared to when the evaluations are missing.

It can thus be inferred that people who have a tendency to conform to normative and/informational influence, or as termed by Bearden et.al. (1989) susceptibility to interpersonal influence, are more likely to accept product recommendations and to have a better opinion of the products if others have the same opinion. Thus, the following hypothesis emerges:

H3: Consumer Susceptibility to Interpersonal Influence (SII) moderates the effect of recommendations on perceived quality, such that, under high SII, recommendations have a stronger impact on the

perceived quality, compared to low SII.

2.5. Conceptual Model

(22)

(23)

3. M

ETHODOLOGY

3.1 Introduction

The following section, the methodology will be presented. The research procedure used in order to investigate the research question and subsequent hypotheses will be described as well as the underlying manipulations and measures of the different variables.

3.2 Research Design

The current paper will employ conclusive casual research, which according to the book of Malhotra, Birks and Wills (2012) it is used in order to measure concepts, examine causal relationships between variables, and test specific hypotheses. Furthermore, causal research is used to determine cause an effect relationships of, in the current case, the independent variable recommendations, on the dependent (perceived quality), as well as the moderating effects of susceptibility to interpersonal influence and brand equity. Furthermore, this will be put into application through an experimental design. This design has been chosen due to the fact that the author wants to uncover the cause as to why perceived quality changes when associated with recommendations; if it is only due to recommendations or not, and what role do the two moderators play.

3.3 Participants and sample

(24)

3.4 Variables

The following section will describe the different variables, as well as the way in which the independent variable and two moderators are to be manipulated or measured. Furthermore, the experimental design will be described and shown, and the expected relations between the variables used in the current research.

3.4.1 Description and manipulation of the variables

The following variables will be used in the research: Independent variable: Recommendations

Dependent variable: Perceived Quality Moderator 1 variable: Brand Equity

Moderator 2 variable: Susceptibility to Interpersonal Influence

The above mentioned variables will be further described with regard to their measurements and manipulations for the current research.

3.4.1.1 Independent variable: Recommendations

As presented before, the independent variable for the current study is recommendations. The recommendations will be presented to the respondents as a written scenario. The variable will be manipulated by presenting different respondents conditioned scenarios depicting either positive or negative recommendations. The approach of the current paper is similar to the approach of Senecal and Nantel (2004), in their paper. In their paper, the authors investigate the influence of online recommendations on product choice. In their experiment, the authors have also used scenarios in depicting various situations and various types of recommendations. The authors have used four treatment levels for the different levels of the variable recommendation that was present in the study. In the present case, there will be two treatment levels for the variable recommendations.

(25)

talks about tablet pc’s. The reason for choosing such a product is due to the fact that it is a product which has a relatively high involvement. Thus it is more likely that a person would turn to listen to a second opinion.

The positive scenario will contain sentences like: ‘it has greater processing memory than other tablet models’, ‘the screen resolution is good and the camera takes great pictures’, ‘I am very pleased with this product, and recommend it to you’, or ‘it has a good android system which makes it so fast you can load an internet page in seconds’, and ‘I strongly think that it is a very good products, and at a very reasonable price as well.’. In the case of the negatively recommended scenarios, these will contain sentences like ‘It has less processing memory than other tablets, making it slow to switch between tabs’, ‘the screen resolution is low, and the camera has only 3 mega pixels, making it hard to take a proper picture.’, as well as ‘the battery life is very low’, and ‘the design is terrible, not stylish at all. It’s not worth the price’. The different product specifications, whether positive or negative have been gathered from the internet by the researcher. The author looked at different websites on which experts post reviews, and by using a slight amount of exaggeration, created the scenario.

In order to check the manipulation for the scenarios, the questionnaire will also contain 3 questions on a 7 point Liker scale. The questions will ask the participant to rate what sort of attributes were depicted in the scenario (1: very positive to 7: very negative), how the tone of the recommender was (1: very positive to 7: very negative), and how satisfied the recommender was with the product (1: very satisfied to 7: very dissatisfied). A detailed scale can be found in appendix 1.1

3.4.1.2 Dependent variable: Perceived Quality

(26)

the paper of Dodds, Monroe and Grewal (1991) on a 7 point Likert scale and the present paper will adopt this scale as well.

The questionnaire contains statements like ‘The likelihood that this product is dependable is’, ‘The workmanship of this product would be’, and ‘The likelihood that the product would be reliable is’, and ‘The product would seem to be durable’, which are measured with 1: Strongly agree to 7: Strongly disagree. The statement ‘The product presented should be of’ is measured from 1: Very good quality to 7: Very poor quality. For the detailed scale, see appendix 1.2.

3.4.1.3 Moderator 1: Brand Equity

The first moderating variable in this research is Brand Equity. The levels of this variable are high BE, and low BE. In the scenario part of the questionnaire, there will be versions in which the product presented in the scenario will have a high BE, meaning that it will be a known, famous brand, and in another scenario it will be an unknown brand. In order to measure if the manipulation of brand equity was successful, a scale will be used. The scale used to measure Brand Equity will be taken from the paper of Yoo andDonthu (2001), in which the authors develop a scale to measure consumer-based brand equity. The scale developed by the authors has 6 items for perceived quality, 3 for brand loyalty, 2 for brand awareness, and 3 for brand associations, measured on a 5 point Liker scale. From this scale, the present paper will use the three items for brand loyalty, and five items for brand awareness/associations. An overview of the statements from the scale can be found in appendix 1.3

(27)

3.4.1.4 Moderator 2: Susceptibility to Interpersonal Influence

The second moderator to be used in the present study is Susceptibility to Interpersonal Influence (SII). This concept will be measured according to the measurement scale developed by Bearden et al. (1989). In their paper, Bearden et al. (1989) had originally started with a pool of 166 items which were later reduced to 15. Out of these 15 items, the authors extracted a 12 factor item-scale which contains four informational items, and eight normative. The current paper will use the scale developed by Bearden et al. (1989), which contains statements such as ‘I rarely purchase the latest fashion styles until I am sure my friends approve of them.’, ‘It is important that others like the products and brands I buy’, ‘To make sure I buy the right product or brand, I often observe what others are buying and using.’, or ‘I often consult other people to help choose the best alternative available from a product class.’, which are measured from 1: Strongly disagree to 7: Strongly agree. The measurements of the scale are taken from the study of Bearden et al (1989). The table with the 12 statements is presented in detail in appendix 1.4.

The variable Susceptibility to Interpersonal Influence (SII) will be measured at the beginning of the questionnaire, and will serve as measurement of the personality of the respondent.

3.4.2 Study Design

The design of the study is a 2 (Recommendations: positive vs. negative) x 2 (BE: high vs. low) x 2 (SII: high vs. low). Due to the fact that SII will be measured, and only Recommendations and BE are being manipulated, there will be four conditions, which are shown in the figure below in Figure 1. The participants to the study will be randomly assigned to one of these four conditions, with a minimum of 40 respondents per conditional group.

Brand Equity

Recommendations

Positive/ High Positive/ Low Negative/ High Negative/ Low

(28)

3.5 Questionnaire and Procedure

Participants in the current research will be given a hyperlink which randomizes the four different questionnaires which were made for each of the four conditions. This randomization procedure assures that every respondent has equal chance to be assigned to one of the four conditions. The questionnaire will start with a neutral opening statement which will give a brief introduction to the study, however without disclosing any vital information. The next part of the questionnaire will contain the measurement questions regarding a person's susceptibility to interpersonal influence, followed by a set of demographic questions. The core of the questionnaire will contain a scenario in which a product of either high or low Brand Equity is depicted, followed by either positive or negative recommendations. A detailed description of the questionnaire can be found in appendix 2.1, followed by the 4 different scenarios in appendix 2.2.

(29)

4. R

ESULTS

4.1 Introduction and plan of analysis

The following section will include the results of the experiment in which the respondents have been assigned to one of the four conditions described in the methodology part. These conditions will be assessed and analyzed in the current section.

The current chapter will first look at the sample population at large by looking at the frequencies and see whether there are differences in the random assignment of the respondents over the four conditions of the experiment. Looking into more detail, the manipulations will be checked with according tests, as well as the reliability of the scales used.

Once an assessment of the scales has been made, and the general characteristics of the sample have been described, the key part of the paper will follow, which is represented by the actual results of the analysis, described per hypothesis.

In order to test the hypotheses which were built in the literature review, several statistical testes will be performed on the data set. The statistical instrument to be used will be linear regression, which is appropriate due to the fact that there are continuous variables in the data set. Through this method, the effect of the independent variable Recommendations can be assessed, together with the two moderating effects of Brand Equity and Susceptibility to Interpersonal Influence respectively.

4.2 Sample characteristics

The following section will give a general description of the sample based on demographics. The description will be accompanied by graphs for better visualization. The frequency tables which accompany the graphs can be found in Appendix 2.1.

4.2.1 Gender

(30)

there are differences in the way the gender is distributed across the sample, these differences are not large, and it can thus be said that they are almost equally divided. The figure can be found in appendix 3.1.

4.2.2 Age

As it can be seen in the graph below, the sample of 160 participants mainly consists of the age group between 18 and 25 years old, which had a frequency of 117 (73.1%). The second largest age group, 26 to 30 years old, and falls far behind the first one, with a frequency of 31 (19.4%). The other age groups are very small, with frequencies of maximum 4 (over 45). The figure can be found in appendix 3.1.

4.2.3 Occupation

The majority of the sample sizes of participants are made of students, with a frequency of 104 (65%). The second somewhat large segment is employed people, with a frequency of 34 (21.3%), followed by working students, with a frequency of 11 (6.9%). The smallest segment is those unable to work, with a frequency of 1. The figure can be found in appendix 3.1.

4.2.4 Brand bought

It is interesting to see across the population how many participants have bought the brand with which they were presented. Out of 160 participants, 80 were presented in the scenario a High equity brand (Samsung), and 80 were presented a Low equity brand (Medion). In the graph it can be seen that even though there are still many respondents that did not buy the high equity brand Samsung (48 out of 80), there is a considerable higher number of respondents that did not buy the low equity brand Medion (68 out of 80). The figure can be found in appendix 3.1.

4.2.5 Product used

(31)

recommendations and reviews. The graph below describes this distribution. The figure can be found in appendix 3.1.

As it can be seen from the graph, the majority of the sample (frequency of 140), is not a user of pc tablets. Thus it can be inferred that the majority of the participants are not very knowledgeable regarding pc tablets, and would thus be more prone to listen to recommendations.

4.3 Random assignment

The following section will look at the previously presented demographic variables and see whether they have been randomly assigned for every condition of the study. This will be done using a Chi square test, which shows the frequency with which the data is spread over the conditions.

4.3.1 Gender

Regarding the gender demographic variable, the Chi square test shows that the male and female are almost equally distributed among the 4 conditions of the experiment, with a Chi (df.= 3) value of 0.508 and a p value of 0.917. This means that the null hypothesis that the proportion in each response group is equal is not rejected. Even though there are more females in total in the sample population, the number of females exceeds the males only by 20. Thus, it can be said that according to gender, the sample is randomly distributed. The full table can be found in appendix 3.2.1.

4.3.2 Age

(32)

4.3.2 Occupation

From the occupation point of view, the majority of the sample size is represented by students, which are randomly distributed over the four conditions. Employed is the next level of occupation which is most prominent, and this shows some problems, with counts of 11, 9, 4 and 10 over the four conditions. The Chi (df.= 15) value is of 21.391 with a p value of 0.125, indicating that the null hypothesis that the proportion in each response group is equal is not rejected, and that the sample is randomly distributed. The full table can be found in appendix 3.2.3

4.3.3 Brand bought

Looking from the perspective of whether the participants have bought or not the brand over the four conditions, the test shows a Chi (df= 3) test value of 12.539 with a p value of 0.006, which indicated that the null hypothesis is rejected. This means that the sample is not randomly distributed, thus posing problems for future analysis. This will be taken into account in subsequent tests. The full table can be found in appendix 3.2.4

4.3.4 Product used

As is the case with the previous variable, based on whether the participants have used the type of product presented in the scenario or not, the random distribution has been tested. The test shows Chi (df.=3) with a value of 2.743 and a p value of 0.433. This means that the null hypothesis that the proportion in each response group is equal is not rejected, which indicates that the sample size is randomly distributed in this case. The full table can be found in appendix 3.2.5

4.4 Manipulation checks

(33)

4.4.1 Manipulation check Recommendations

After the scenario has been shown in the questionnaire, a series of three questions were asked to the participants in order to discern if the manipulations were done well or not. Every participant was asked to rate how positive the scenario was on a 7 point Liker scale. The questions have been checked for internal consistency using a Cronbach’s alpha test, and have shown a value of 0.976, indicating a high internal consistency.

Looking at the group statistics table it can be seen that positive recommendations have M= 5.99 and SD= 0.450, while the negative recommendations have M = 1.69 and SD= 0.393. Furthermore, the test shows a value of t (df.= 158) = 64.411 with a p value of 0.000 < 0.05. As it can be seen, the respondents that were shown positive scenarios, perceived the scenarios as positive, while the respondents which were shown negative scenarios, perceived them as negative. This means that the manipulation was successful. The table can be found in appendix 3.4.1.

4.4.2 Manipulation check Brand Equity

The manipulation check with regard to Brand Equity (high or low) has been checked in the experiment by asking respondents to rate the brand based on the scale developed by Yoo and Donthu (2001) which checked the loyalty one had with the brand, the awareness for the brand, and the brand associations.

By first taking a look at the group statistics, it can be seen that the high equity group ha M= 3.43 and SD= 0.819, while the low equity group has M= 2.54 and SD= 0.592. These smaller values can be due to the fact that the scale to measure BE has less items. By looking at the Levene’s test for equality of variances it can be seen that the assumption that the variances are equal of broken, since the test has F= 11.077 with a p value of 0.001 < 0.05. This means that the variability of the two conditions is different, and thus the researcher will look at the second row of the table to see the t test. Here the value of the test is t (df.= 143.862) = 7.871 with a p value of 0.000 < 0.05, making the manipulation a success. Table can be found in appendix 3.4.2.

4.5 Scale Reliability

(34)

scenario. All of the scales will be assessed on their internal consistency and reliability by using Cronbach’s Alpha measure of reliability. According to Malhorta (2007), in order to continue with the scale, a minimal alpha value of 0.6 must be achieved. Below, a table overview with the Cronbach’s Alpha values of the different scales is given.

Scale Number of Items Cronbach’s Alpha

Recommendations 3 0.958

Susceptibility 12 0.904

Brand Equity 7 0.841

Perceived Quality 5 0.936

Table 4: Cronbach’s Alpha scores for internal consistency and reliability

It is important to mention that initially the Cronbach’s Alpha for the Brand Equity scale was 0.658, and the one for susceptibility was initially 0.899.Even though the alpha value was over the required 0.6, by looking at the table containing the alpha if item deleted, it can be seen that if question BE_5 is deleted from the scale, the alpha would rise to 0.841. For susceptibility, the item SII10 was deleted, thus reaching an alpha of 0.904 Thus this item has been deleted. All other scales have an alpha score above 0.8, which indicated a very good reliability and internal consistency. A detailed table of the Cronbach’s Alpha for every scale can be found in Appendix 3.5.

4.6 Testing the hypotheses

The following section will test the hypotheses presented in the Literature Review. A regression analysis will be performed in order to see the effects of the independent variable Recommendations and of the two moderators, SII and BE. The following section will present the statistical analysis per hypothesis, thus dealing with every variable. The statistical output for the regression analysis can be found in appendix 2.4.

(35)

Before looking at every effect in part, it is important to write the regression equation. First the general regression for the particular model presented in the paper will be written, followed by a version in which the coefficients are presented.

Yi = α0 +β1x1i +β2x2i + β3 (x1i*x2i) + β4x4i + β5 (x1i*x4i) +

ε

i

Yi: Perceived quality

x1i: Recommendations

x2i: Susceptibility to interpersonal influence

(x1i*x2i): Interaction term between recommendations and susceptibility to interpersonal influence

x4i: Brand equity

(x1i*x4i): Interaction term between recommendations and brand equity

εi:

Error term

i: Respondent i

Perceived_Q= 2.731+ 1.771* Recommend +0.241* SII- 0.285 (Recommend*SII) + 0.361* BE + 0.224(Recommend * BE) +

ε

By having a general look at the regression output, it can be seen that the model has an adjusted R square of 0.443, and the F test has a value of 26,290, with a p value of 0.000. The F-test, F= 26.290 with a p value of 0.000, indicates that the model has a rather good prediction value. Below a table of the coefficients will be shown, which will be further discussed in the following sections.

(36)

BE ,361 ,123 1,455 ,148

BE_moderator ,224 ,066 ,639 ,523

Table 5: Regression coefficients

It can thus be seen from the model that there are two predicting variables which show a statistically significant influence on the dependent variable Perceived Quality. These variables are Recommendations and Susceptibility to interpersonal influence (SII).

4.6.1 Hypothesis 1: Recommendations

The independent variable for the present study is Recommendations, which are measured as positive or negative. The dependent variable is Perceived quality. For the sake of a better overview, the hypothesis regarding recommendations is presented in the Literature Review will be repeated below.

H1: Positive recommendations have a positive influence on the perceived quality of the product, while negative recommendations have a negative effect.

It serves to be reminded that the variable Recommendations has been manipulated by the researcher, and is measured in the analysis as a dummy variable of 0 (negative) and 1 (positive).

(37)

same magnitude, but of different sign effect on perceived quality. More concrete, and in concordance with the hypothesis, negative recommendations have a negative effect on the way consumers perceive the quality of a product.

4.6.2 Hypothesis 2: Brand Equity

Regarding the third and final hypothesis, there is no statistically significant effect of the interaction term BE_mod (t= .639, p>.05 (Sig.=.523). Looking at the direct effect of the variable BE, it can be seen that this effect is not statistically significant either (Sig.= .148). To make sure that there are no other factors intervening with the result, it is important to check if the variable has been randomly assigned to the 4 conditions. The manipulation checks and checks for random assignment have been done before proceeding with the regression. Thus as mentioned before in the random assignment section, regarding ‘brand bought’, the data set is not randomly distributed, and thus this could have posed problems in the analysis. This could have been the cause of the insignificant statistical result. Concluding, it is clear that hypothesis 2 (Brand Equity (BE) moderates the effect of recommendations on perceived quality,

such that high BE will strengthen the effect of recommendations on perceived quality, compared to low BE.) is not supported.

4.6.3 Hypothesis 3: Susceptibility to interpersonal influence

Looking further in the coefficients table, it can be seen that the interaction effect between SII and Recomend is not statistically significant (Sig.= .110). This leads to the conclusion that there is no moderating effect of SII, and that hypothesis 3 (Consumer Susceptibility to Interpersonal Influence

(SII) moderates the effect of recommendations on perceived quality, such that, under high SII, recommendations have a stronger impact on the perceived quality, compared to low SII) is not

confirmed.

(38)

controversial, since susceptibility is a personality measure, and it is strange that it has an effect on perceived quality on its own. It can thus be inferred that there exists a future possibility for a moderating effect to be proven, however not in the span of the current paper.

4.7. Discussion of results

The following section will discuss in detail the results of the analysis presented above. In the case of supported hypotheses, a short discussion about the results will be given. In the case in which the hypotheses are not supported, possible reasons as to why they were not supported are given.

As it was expected from the literature review, when recommendations positive recommendations are presented to the consumer, their perception of product quality increases. Looking back at the statement of Steekamp (1990), we can see that the author believes that when the true state of the quality attribute is unknown, extrinsic quality cues are used to influence the perceived state of the quality attribute. As mentioned before in the current paper, extrinsic quality cues are treated as recommendations, and the analysis proves the fact that consumers make inferences about product quality from these recommendations. In the questionnaire presented to participants the product on which recommendations were made was specially chosen as a product on which quality inferences could not be easily made. Thus the product category pc tablets have been used. The results show that the first

hypothesis is fully supported. Thus, positive recommendations influence the perceived quality in a

positive manner, and negative recommendations influence it in a negative way. The findings regarding negative recommendations is in concordance with the assumptions of authors Herr, Kardes and Kim (1991), that when negative recommendations, such as word of mouth were presented to the respondents, negative quality perceptions were made. Looking at the results regarding positive recommendations, this is in line with the findings of Duan, Gu, and Whinston (2008), that a positive valence of recommendations leads to positive customer opinions, and as stated in the literature review, it can be assumed to be similar to perceived quality.

(39)

section the researcher will address the possible causes found for this insignificant effect. Firstly, it is important to take a look at the brands used in the questionnaire. It is possible that due to the fact that the majority of the respondents were students between the age of 18 and 25 that this type of consumer was biased regarding the brand and thus did not truly regard the brand with high equity (Samsung) as having true high equity. However, the most possible cause for the lack of statistical influence is the fact that the manipulation of the variable BE was not entirely successful. This can be due to a failure in measurement, a misunderstanding of the scale questions by the participant, or even the fact that the high equity brand has not been perceived as high equity by the participants. Either or all of these reasons can be the cause of the fail in manipulation, and thus leading to an insignificant statistical result.

(40)

5. CONCLUSION

The current section will present a conclusion the paper, as well as some limitations. Furthermore, opportunities for future research will be discussed; opportunities which can either validate the current paper, or look at different perspectives.

5.1 Conclusion

The paper has the aim to look into the manner in which the perception of product quality can be influenced by factors other than the physical qualities of a product. For the scope of the research, the author has chosen recommendations (positive or negative) as the main factor which could influence perceived quality. In turn, two other factors are presumed to affect the relationship between recommendations and perceived quality. These two other moderating factors, susceptibility to interpersonal influence and brand equity are thought to strengthen the relationship when they are at a high level, compared with their low level.

The method of analysis for these assumptions has been an online survey in which the recommendations and brand equity of the product presented were manipulated. Scales have been used to measure perceived quality and susceptibility to interpersonal influence, as well as to check for the correct manipulations of recommendations and brand equity. The subsequent analysis of the data has shown that indeed recommendations influence perceived quality, thus validating the first assumption made by the researcher. The second and third assumptions, regarding brand equity and susceptibility to interpersonal influence were however not valid. However, the results indicate that susceptibility to interpersonal influence has a direct influence, independent of the interaction, on perceived quality, which would suggest that susceptible people see products better than insusceptible ones. The results of this paper have both academic and managerial implications, which will be discussed at length in the following section.

5.2 Academic and managerial implications

(41)

negative recommendations. The current paper has brought these two concepts together in one study, testing the effects of these recommendations on perceived quality, and thus further validating the works of previous authors in this area. Furthermore, the introduction of the personality measure susceptibility to interpersonal influence has produced interesting results. Although the author strongly believes that given a larger sample size, susceptibility could act as a moderator, it must be mentioned that in the current paper it did not act as a moderator, but it had an unexpected direct effect. For the academic world, this could mean that it is worthy to pay more attention to such personality measures in contexts of advertising and perception of goods.

Looking strictly at the outcomes of this study, the most critical recommendation for managers would be to have a close look at the ratings and reviews posted by users. Whether it is other users or experts, consumers take into account recommendations regarding products, and the valence of these recommendations helps them in making the difference between the ways they perceive the quality of a product. Due to the fact that word of mouth recommendations are hard to track, and due to the fact that consumers are digitalizing more and more, checking online recommendations makes most sense for managers. Thus, looking at the valence of recommendations marketers can anticipate a change in consumer opinion of their product, or even a change in the sales of the product. The presence of negative recommendations and opinions can be countered by the marketer through a campaign to counter the negative opinions and prevent a decrease in the perceived quality of the product.

5.3 Limitations and future research

Every paper is presented with limitations, which give way for future research, and the current paper is no different. Thus, the current section will present the various limitations which in the mind of the author are present for this paper, as well as the opportunities for future research on the current subject.

(42)

of a large number of respondents that can be considered as a limitation. It must be mentioned that the population has not been diverse enough, which is shown clearly by the fact that the sample is not entirely randomly distributed. This could have been the cause for some errors in the results, and thus it is an important limitation. Further to be taken into account is the participants' understanding of the questionnaire. While all participants had a grasp of the English language and could communicate in it, some of the participants did not have such a vast vocabulary and there have been several participants that have admitted that some questions were formulated in a rather difficult manner. It is thus the author's opinion that a lack of understanding of the questions could have led to faulty answers from the participants and thus creating some problems with the analysis.

However, the fact that the paper has limitations paves the way for future research. Looking at opportunities for further research the author can see many directions in which the research undergone in this paper can look toward. Firstly, a replication of the study with a larger sample size can be done, as well as a different manipulation of the variable brand equity in order to prove that the assumptions made by the author are true. Furthermore, it can be interesting to include more variables in the model. For example the source of the recommendations can have an effect, as done in the paper of Senecal and Nantel (2004). Furthermore, testing if product involvement or the nature of the product (hedonic vs utilitarian) have any influence on the way recommendations influence perceived quality could be an interesting subject for future research. Another possibility for future research can use either purchase intention, or a measure of company sales as the dependent variable, in order to see the effects on these variables.

(43)

6. R

EFERENCES

Aaker, D. A. (1991), Managing Brand Equity. New York: the free press.

Aaker, D. A. and Biel, A. L. (1993), Brand Equity & Advertising: Advertising's Role in Building Strong

Brands. Psychology Press.

Ailawadi, K. L., Lehmann, D. R., and Neslin, S. A. (2003), "Revenue Premium as an Outcome Measure of Brand Equity," Journal of Marketing, 67 (Oct.), 1-17.

Akdeniz, B., R. J. Calantone, and C. M. Vorhees (2013), "Effectiveness of Marketing Cues on Consumer Perceptions of Quality: The Moderating Roles of Brand Reputation and Third-Party Information," Psychology and Marketing, 30 76-89.

Alba, J. et al. (1997), "Interactive Home Shopping: Consumer, Retailer, and Manufacturer Incentives to Participate in Electronic Marketplaces," Journal of Marketing, 61 (Jul.), 38-53.

Anderson, E. W. (1998), "Customer Satisfaction and Word of Mouth," Journal of Service Research, 1 (August), 5-17.

Archibald, R. B., Haulman, C. A., and Moody Jr, C. E. (1983), "Quality, Price, Advertising, and Published Quality Ratings," Journal of Consumer Research, 9 (Mar.), 347-56.

Arndt, J. (1967), "Role of Product-Related Conversations in the Diffusion of a New Product," Journal

of Marketing Research, 4 (Aug.), 291-5.

Bearden, W. O., Netemeyer, R. G., and Teel, J. E. (1989), "Measurement of Consumer Susceptibility to Interpersonal Influence," Journal of Consumer Research, 15 (Mar.), 473-81.

Benedicktus, R. L. et al. (2010), "Conveying Trustworthiness to Online Consumers: Reactions to Consensus, Physical Store Presence, Brand Familiarity, and Generalized Suspicion," Journal of

(44)

Bone, P. F. (1995), "Word-of-Mouth Effects on Short-Term and Long-Term Product Judgments," Journal of Business Research, 32 (3), 213-23.

Burnkrant, R. E. and Cousineau, A. (1975), "Informational and Normative Social Influence in Buyer Behavior," Journal of Consumer Research, 2 (Dec.), 206-15.

Burzynski, M. H. and Bayer, D. J. (1977), "The Effect of Positive and Negative Prior Information on Motion Picture Appreciation," The Journal of Social Psychology, 101 (04/01; 2013/04), 215-8.

Chevalier, J. A. and Mayzlin, D. (2006), "The Effect of Word of Mouth on Sales: Online Book Reviews." Journal of Marketing Research, 43 345-354.

Darley, J. M. and Gross, P. H. (1983), "A Hypothesis- Confirming Bias in Labelling Effect," Journal of

Personality and Social Psychology, 44 20-33.

Dawar, N. and Pillutla, M. M. (2000), "Impact of Product-Harm Crises on Brand Equity: The Moderating Role of Consumer Expectations," Journal of Marketing Research, 37 (May), 215-26.

Deutsch, M. and Gerard, H. B. (1955), "A Study of Normative and Informational Influence upon Individual Judgment," Journal of Abnormal and Social Psychology, 51 629-636.

Dodds, W. B., Monroe, K. B., and Grewal, D. (1991), "Effects of Price, Brand, and Store Information on Buyers' Product Evaluations," Journal of Marketing Research, 28 (Aug.), 307-19.

Dudycha, L. W. and Naylor, J. C. (1966), "Characteristics of the Human Inference Process in Complex Choice Behavior Situations," Organizational Behavior and Human Performance, 1 (9), 110-28.

Duhan, D. F. et al. (1997), "Influences on Consumer use of Word-of-Mouth Recommendation Sources," Journal of the Academy of Marketing Science, 25 283-295.

Eliashberg, J. and Shugan, S. M. (1997), "Film Critics: Influencers Or Predictors?" Journal of

(45)

Faulhaber, G. R. and Yao, D. A. (1989), ""Fly-by-Night" Firms and the Market for Product Reviews," The Journal of Industrial Economics, 38 (Sep.), 65-77.

Fornell, C. et al. (2006), "Customer Satisfaction and Stock Prices: High Returns, Low Risk," Journal of

Marketing, 70 (Jan.), 3-14.

Gilly, M. C. et al. (1998), "A Dyadic Study of Interpersonal Information Search," Journal of the

Academy of Marketing Science, 26 (April 01), 83-100.

Godes, D. and Mayzlin, D. (2004), "Using Online Conversations to Study Word-of-Mouth Communication," Marketing Science, 23 (September 21), 545-60.

Herbig, P. A. and O'Hara, B. S. (1994), "Quality is in the Eye of the Beholder," Journal of Professional

Services Marketing, 10 (10/07; 2013/04), 19-33.

Herr, P. M., Kardes, F. R., and Kim, J. (1991), "Effects of Word-of-Mouth and Product-Attribute Information on Persuasion: An Accessibility-Diagnosticity Perspective," Journal of Consumer

Research, 17 (Mar.), 454-62.

Hoyer, W. D., D. J. Machlnnis, and R. Pieters (2013), Consumer Behaviour. South-Western Cengage Learning.

Jiang, P., Jones, D. B., and Javie, S. (2008), "How Third-Party Certification Programs Relate to Consumer Trust in Online Transactions: An Exploratory Study," Psychology and Marketing, 25 839-58.

Luo, X. and Homburg, C. (2007), "Neglected Outcomes of Customer Satisfaction," Journal of

Marketing, 71 (Apr.), 133-49.

(46)

Mahajan, V., Rao, V. R., and Srivastava, R. K. (1990), "Development, Testing, and Validation of Brand Equity Under Conditions of Acquisition and Divestment," 14-5.

Mahajan, V., Muller, E., and Kerin, R. A. (1984), "Introduction Strategy for New Products with Positive and Negative Word-of-Mouth," Management Science, 30 (December 01), 1389-404. Malhotra, N. K.; Birks, D. F. and Wills, P. (2012), “Marketing research : an Applied Orientation” Englewood Cliffs: Prentice Hall International

McGuire, W. J. (1968), "Personality and Susceptibility to Social Influence," Handbook of Personality

Theory and Research, 2 1130-87.

Michelle L. R. and Brady M. K. (2007), "Consumer Responses to Performance Failures by High‐ Equity Brands," Journal of Consumer Research, 34 (December), 537-45.

Miller, S. "Kudzu.Com Survey Reveals Online Reviews significantly Impact Consumer Spending," (accessed march 2013), [available at http://www.reuters.com/article/2008/12/09/idUS159105+09-Dec-2008+PRN20081209].

Minzesheimer, B. (2011), "How the 'Oprah Effect' Changed Publishing," (accessed 10 March), [available at http://usatoday30.usatoday.com/life/books/news/2011-05-22-Oprah-Winfrey-Book-Club_n.htm].

Mizerski, R. W., Golden, L. L., and Kernan, J. B. (1979), "The Attribution Process in Consumer Decision Making," Journal of Consumer Research, 6 (Sep.), 123-40.

Moe, W. W. and Trusov, M. (2011), "The Value of Social Dynamics in Online Product Ratings Forums," Journal of Marketing Research, 48 444-456.

How Third-Party Information Influences Perceived Quality