Cross-Cultural Differences in Electronic Word-of-Mouth Influence on Purchase Intention

Table of Contents

Acknowledgement

Abstract

List of Figures

List of Tables

List of Abbreviations

Chapter 1: Introduction
1.1 Research Objective and Questions
1.2 Dissertation Structure

Chapter 2: Theoretical Background
2.1 Definition of Culture
2.2 Justification of Adopted Conceptualisation

Chapter 3: Literature Review
3.1 From WoM to eWoM
3.2 Purchase Intention
3.3 Culture
3.3.1 Cultural Convergence
3.3.2 Espoused National Cultural Values
3.3.3 Power Distance (PD)
3.3.4 Uncertainty Avoidance (UA)
3.3.5 Individualism/Collectivism (IND/COL)
3.3.6 Masculinity/Femininity (MAS/FEM)

Chapter 4: Research Methodology
4.1 Research Philosophy
4.2 Approach to Theory Development
4.3 Research Design and Purpose
4.4 Research Strategy
4.5 Experimental Parameters
4.6 Pretest
4.7 Measurement
4.7.1 Measurement of Culture
4.7.2 Measurement of Purchase Intention
4.8 Sampling
4.8.1 Volunteer Sampling
4.8.2 Convenience Sampling
4.8.3 Sample Size and Response Rate
4.9 Research Instrument
4.10 Pilot Test
4.11 Experimental Procedure
4.12 Ethical Principles
4.13 Methodological Limitations

Chapter 5: Research Analysis and Findings
5.1 Demographic Profile of the Sample
5.2 Scale Reliability
5.3 Descriptive Statistics
5.4 Paired Samples T-Test
5.4.1 Justification
5.4.2 Check for Outliers
5.4.3 Assumption Checks
5.4.4 Results
5.5 Multiple Regression Analysis
5.5.1 Justification
5.5.2 Check for Outliers
5.5.3 Assumption Checks
5.5.4 Results

Chapter 6: Discussion
6.1 Findings on the Effect of eWoM on Purchase Intention (RQ1)
6.2 Findings on the Influence of Espoused National Cultural Values on the Effect of eWoM on Purchase Intention (RQ 2 - 5)

Chapter 7: Conclusions
7.1 Theoretical Contributions
7.2 Managerial Recommendations
7.2.1 Standardised e-Marketing Program
7.2.2 Incorporate Social Media Plug-in on the Website
7.2.3 e-Marketing Program to Fertilise eWoM
7.2.4 Internationalisation through e-Commerce
7.3 Limitations and Future Research
7.4 Conclusion

Bibliography

Appendices
Appendix 1: Boolean Search Strings
Appendix 2: Pretest Web Questionnaire
Appendix 3: Web Questionnaire of the main Research
Appendix 4: Consideration of Ethical Principles
Appendix 5: Demographic Sample Information
Appendix 6: Histogram and P-P Plot for Purchase Intention Difference
Appendix 7: Tests of Normality of Purchase Intention Difference
Appendix 8: Correlation Matrix
Appendix 9: Scatterplot of Standardised Residuals
Appendix 10: Histogram and P-P Plot of Standardised Residuals
Appendix 11: Tests of Normality of Standardised Residuals

Abstract

International marketers are trying to encourage positive electronic word-of-mouth (eWoM) about their products and services. The question of whether consumers in different cultures respond differently to eWoM is critical if marketers are to leverage the potentially global power of eWoM. The central objective of this study is to critically evaluate whether espoused national cultural values at the individual level influence the effect of eWoM on consumers’ purchase intention

Prior cross-cultural eWoM research mostly studied culture at the country level. This paper draws on perspectives in cultural psychology and cross-cultural research that argue that individuals espouse national cultural values to different degrees. Therefore, predicting the influence of culture on individuals’ behaviour necessitates to assess cultural values by personality tests at the individual level of analysis. Yet, no research can sufficiently answer the question of how individual level culture may influence the effect of eWoM on purchase intention. The present research addresses this gap by measuring espoused national cultural values of power distance, uncertainty avoidance, individualism/collectivism, and masculinity/ femininity at the individual level of analysis and investigating their influence on the effect of eWoM on purchase intention

An experiment, using a repeated measures design, was conducted with 100 subjects from 18 countries. The results reveal that, as expected, consumers’ purchase intentions are significantly higher after reading eWoM than after reading factual information on a company website. Further, the results show that, contrary to expectations, this effect is not significantly influenced by the national cultural values an individual espouses. This implies that traditional market segmentations based on culture are of limited relevance in the online market place. Theoretical and managerial implications are discussed

Keywords: electronic word-of-mouth; eWoM; culture; espoused culture; espoused cultural values; purchase intention

List of Figures

Figure 1: Structure of the dissertation

Figure 2: Academic and technological development from WoM to eWoM

Figure 3: Research model

Figure 4: Scatterplot of purchase intention factual against purchase intention eWoM

Figure 5: Purchase intention means and 95% confidence interval associated with factual information and eWoM information

List of Tables

Table 1: Definitions of culture

Table 2: Cultural dimensions adopted in prior research

Table 3: Distinction between individualism and collectivism in terms of individuals’ self-concept

Table 4: Adopted research strategies and data collection instruments in similar studies

Table 5: Critical evaluation of culture scales

Table 6: Measures taken to increase response rate

Table 7: Data requirements for this research 48-49

Table 8: Measures taken to improve validity

Table 9: Scale reliability

Table 10: Means and standard deviations of key constructs

Table 11: Assumption checks for paired samples t -test

Table 12: Paired sample t -test for purchase intention factual & purchase intention eWoM

Table 13: Assumption checks for multiple regression analysis

Table 14: Results multiple regression analysis

List of Abbreviations

illustration not visible in this excerpt

Chapter 1: Introduction

In the era of Web 2.0 and social networking sites (hereafter SNS), consumers are able to share opinions about products and services with other consumers across the globe (Kim et al., 2016). This phenomenon is commonly called electronic word-of-mouth (hereafter eWoM) and has recently gained renewed interest from academics and marketers due to its powerful impact on one of the most fundamental aspects of consumer behaviour, namely purchase intention (Kotler and Armstrong, 2016). SNS, such as Facebook and WeChat, are a major catalyst of eWoM (Kotler and Keller, 2016) and have become worldwide the predominant social communication channel among young adults (Balaji et al., 2016). More than 2 billion people use SNS, which is about 29% of the world’s population (Statista, 2016). Thus, eWoM in SNS surmounts national and cultural borders (Kasabov, 2016) and provides international marketing managers a powerful marketing tool to reach and influence consumers around the world (Balaji et al., 2016; Martensen and Mourisen, 2016). Therefore, this study focuses specifically on eWoM on SNS.

Recognising this potential, most marketers embrace e-marketing through SNS to engage with their customers (Malgorzata, 2016) and to encourage positive eWoM about their products and services (Kotler and Keller, 2016; Solomon et al., 2013). These opportunities, however, come not without challenges. If international marketing managers are to leverage the potentially global power of eWoM, they must understand potential cultural differences between consumers, since culture is recognised as a major influence on their purchase decisions and their responses to marketing programs (Kotler and Armstrong, 2016). Studying the effect of eWoM on consumers’ purchase intentions across cultures will help international marketing managers to understand how consumers in different markets may respond to eWoM, and provide valuable guidelines to encourage eWoM across cultures.

Most previous research investigates consumer behaviour from the perspective of national culture. Differences in research results are ascribed to national-level differences on Hofstede’s (1980) cultural dimensions: Power Distance (PD), Uncertainty Avoidance (UA), Individualism/Collectivism (IND/COL), and Masculinity/Femininity (MAS/FEM).

However, the impact of national culture on consumer behaviour has been challenged by numerous scholars. The fast advancement in technology is making the world smaller and consumer needs around the world are becoming more similar, leading to the emergence of a common global consumer culture (e.g., Levitt, 1983; Alden et al., 1999; Triandis, 1982). The Internet has become like an own virtual “country” with a common culture which makes differences in national-level culture less relevant in an online environment (Johnston and Johal, 1999; Burgmann et al., 2006).

Along the line, culture scholars (e.g., Srite and Karahanna, 2006; Straub et al., 2002) have argued that using national-level differences on Hofstede’s (1980) cultural dimensions to predict individual behaviour means to substitute stereotypes for individual traits. This is because cultural borders between countries are blurring, and individuals vary greatly in the degree in which they espouse their national culture.

Rather, as Srite & Karahanna (2006) propose, cultural traits should be assessed through personality tests at the individual level of analysis (Tyler et al., 2000). Thus, espoused national cultural values should be treated as an individual difference variable (Srite and Karahanna, 2006). Supporting this notion, research (e.g. Yoon, 2009; Srite and Karahanna, 2006; Shin et al., 2007) has shown that culture at the individual level can considerably impact consumer behaviour and is more meaningful in today’s globalised world.

This contrast in scholarly perspectives leads this paper to an intriguing question: does the effect of eWoM on purchase intention differ across individuals who espouse different cultural values? Purchase intention is the focal point of the marketer’s effort (Kotler and Armstrong, 2016). A systematic literature review has shown, however, that research investigating the influence of espoused national cultural values on the effect of eWoM on purchase intention is scarce (see appendix 1). This is considered a substantial gap in the literature. Therefore, the main contribution of the present study is to tackle this gap. The paper investigates whether the effect of eWoM on the receiver’s purchase intention differs depending on the receiver’s espoused national cultural values. These values are defined as the degree to which an individual embraces the values of his or her national culture (Srite and Karahanna, 2006)

Following the above presented critique of using national-level differences on Hofstede’s (1980) cultural dimensions to predict individual behaviour, this research operationalises culture as espoused national cultural values by applying Hofstede’s (1980) four cultural dimensions on the individual level. This approach differs considerably from previous research on cultural differences in behavioural intentions following received eWoM (Christodoulides et al., 2012; Goodrich and de Mooij, 2014).

The two approaches, however, are not mutually exclusive but provide complementary perspectives on the influence of culture on behavioural intentions (Ford et al., 2003). That is, whereas previous WoM and eWoM studies have investigated how culture influences behavioural intentions on an aggregated level, the method taken in this paper sheds light onto how culture can manifest at the individual level and influence each consumer’s purchase intention (Srite and Karahanna, 2006).

This study also furthers our understanding of individual characteristics of eWoM receivers. This area is recognised as vitally important since the same eWoM message may provoke very different responses in different receivers (Chaiken, 1980). Prior research has investigated individual characteristics as moderators of behavioural intentions following exposure to eWoM (e.g., Doh and Hwang, 2009; Park et al., 2007). However, very few studies have considered the receiver’s espoused national cultural values as individual difference variables. In response to a call for research on how further receiver characteristics affect purchase intention (Cheung et al., 2012), the present paper incorporates an additional set of receiver characteristics — espoused national cultural values — into existing eWoM behaviour models.

1.1 Research Objective and Questions

Based on the discussed rationale, the central objective of this study is to critically evaluate whether espoused national cultural values at the individual level influence the effect of eWoM on consumers’ purchase intention.

To do so, this paper puts forward five research questions:

1. How does eWoM influence purchase intention?
2. How does the espoused national cultural value of power distance influence the effect of eWoM on purchase intention?
3. How does the espoused national cultural value of uncertainty avoidance influence the effect of eWoM on purchase intention?
4. How does the espoused national cultural value of collectivism influence the effect of eWoM on purchase intention?
5. How does the espoused national cultural value of masculinity influence the effect of eWoM on purchase intention?

1.2 Dissertation Structure

The paper consists of seven main chapters (see figure 1).

Chapter two provides the theoretical background of culture. Different definitions of culture are discussed and a justification for the adopted conceptualisation of culture in this paper is presented.

Chapter three reviews relevant literature on WoM, eWoM, consumer behaviour, and culture. Drawing on these theoretical underpinnings, the paper develops expectations regarding the relationship between eWoM and purchase intention, and the moderating role of culture.

Chapter four discusses the research methodology. An experimental research strategy has been employed throughout the research process.

Chapter five presents a justification of the chosen statistical analysis (i.e., paired samples t- test and multiple regression analysis), and the results.

Chapter six discusses the results in relation to the research questions and links them to previous research.

Lastly, chapter seven presents theoretical implications and managerial recommendations. Limitations of the study are acknowledged, suggestions for future research are given, and a general conclusion is provided.

Figure 1: Structure of the dissertation Source: Author, 2016

illustration not visible in this excerpt

Chapter 2: Theoretical Background

2.1 Definition of Culture

Culture is a complex concept with a plethora of wide-ranging and contradictory definitions (Straub et al., 2002). Straub et al. (2002) classify these into three main schools of thought. Based on this classification, table 1 provides an overview of the main authors and their definitions of culture within each school of thought.

The first school of thought defines culture as consisting of patterned ways of thinking that are based on values and shared across people in a society (Straub et al., 2002). These values act on people’s cognitions, attitudes, and behaviours (ibid). The second school of thought defines culture not in terms of its composition but with regard to learned ways of problem solving (Ford, 1942). The third group of scholars sees culture as more abstract, and in some cases esoteric or spiritual (Straub et al., 2002).

Definitions based on shared values represent the most common view on culture (Straub et al., 2002). Notably, the definition put forward by Hofstede (1980), which is one of the most widely accepted and cited theory of culture, falls into this school of thought (Straub et al., 2002). Hofstede defines culture as “the collective programming of the mind which distinguishes the members of one human group from another” (1980, p.260). He spells out four cultural dimensions: PD, UA, IND/COL, and MAS/FEM. Later, Hofstede and Bond (1988) added a fifth dimension the original four dimensions, namely long-term versus short-term orientation (LTO).

Table 1: Definitions of culture Source: Author, 2016

illustration not visible in this excerpt

2.2 Justification of Adopted Conceptualisation

Fernandez et al. (1997, pp.43-4) claim Hofstede’s framework to be "a watershed conceptual foundation for many subsequent cross-national research endeavors”. Moreover, a major analysis of journals and authors on international business identified Hofstede’s work as one of the most impactful studies in the field (Chandy and Williams, 1994). In fact, Hofstede’s conceptualisation of culture has been adopted by numerous researchers to investigate a large array of marketing issues (Steenkamp, 2001). For example, Moon et al. (2008) successfully adopted Hofstede’s conceptualisation to investigate the impact of culture on consumers’ purchase intentions. Although the study context was different from the present research, this indicates that Hofstede’s framework suits research on cultural differences in purchase intentions.

Moreover, Hofstede’s framework is by far the predominantly adopted conceptualisation of culture in prior cross-cultural eWoM research. Table 2 provides an overview of relevant prior consumer behaviour research that adopted Hofstede’s conceptualisation. As illustrated, only a fraction of papers assessed the fifth dimension of LTO.

Seen from a practical perspective, international marketers often use Hofstede’s framework for cultural and market segmentation (Liu et al., 2001). Kale (1991) argues that Hofstede's (1980) framework is the most appropriate one for practical application in cross-cultural marketing, like comparing the receptivity to marketing communication across cultures. This is because most other conceptualisations of culture lack comprehensiveness for marketers, are not empirically supported, and allow only the direct comparison of countries, but no indirect comparison to a scale in order to facilitate measurement (ibid).

Drawing on the presented rationale, this study adopts Hofstede’s (1980) original four dimensions to operationalise culture as espoused national cultural values on an individual level. However, following majority of prior research, this study will not assess the fifth dimension of LTO. As argued by Lam et al. (2009), this dimension does not lend itself to clear hypotheses development regarding eWoM.

Table 2: Cultural dimensions adopted in prior research Source: Author, 2016

illustration not visible in this excerpt

Chapter 3: Literature Review

3.1 From WoM to eWoM

Interpersonal influence has always exerted a significant impact on consumer behaviour (e.g., Asch, 1952; Bearden and Etzel, 1982). For hundreds of years, consumers are sharing opinions and experiences regarding products and services through informal conversations with their family, friends, and acquaintances "over the clothesline" and "across backyard fences” (Burton and Khammash, 2010; Whyte, 1954). This phenomenon is known as word- of-mouth (hereafter WoM). After WoM has been discovered as a powerful “interpersonal consumer network” which significantly affects consumers’ purchase decisions (Ryan and Gross, 1943), interest into the phenomenon from scholars and practitioners took off.

Fast forward to 2016: the digital revolution has changed everyone’s life and has become the most significant influence on consumer behaviour (Solomon et al., 2013). Today, the Internet is the backbone of the society (ibid). Wide spread adoption of computers, smartphones, tablets, social media, and web cams are enabling practically every consumer in virtually any place of the world to create and share content with an infinite numbers of consumers by one click on a keypad (ibid). The Internet has become the primary channel for information search and it has become part of consumers’ pre-purchase routine to seek advice from others online (Chaffey et al., 2009). As illustrated in figure 2, this development has transformed WoM into an even more forceful, digital form of eWoM. Hennig-Thurau et al. (2004, p.39) proposed one of the first and most widely cited definition of eWoM:

“ any positive or negative statement made by potential, actual, or former customers about a product or company, which is made available to a multitude of people and institutions via the Internet. ”

Figure 2: Academic and technological development from WoM to eWoM Source: Author, 2016

illustration not visible in this excerpt

3.2 Purchase Intention

A consumer’s purchase decision is the most important aspect of consumer behaviour (Kotler and Keller, 2016) and the focal point of the marketer’s effort (Kotler and Armstrong, 2016). Purchase intention is defined as the likelihood that an consumer will purchase a particular product or service based on the interaction between needs, attitude and perception towards the product or service (O’Cass and Lim, 2002). Purchase intention is a widely accepted predictor of actual purchase decisions (East et al., 2016; Grewal et al., 1998).

Generally, consumers’ ability to determine product quality and seller reliability prior to the purchase decision is limited (Blackwell et al., 2006). This results in an uncertainty about the product or service choice which consumers seek to minimise through acquisition of information in the pre-purchase phase (Wang and Wang, 2010; Bone, 1995). However, information search typically involves high opportunity costs regarding the searcher’s time (Stigler, 1961). Thus, consumers have to be selective regarding the information sources that they consult (ibid). They especially search for credible information from sources they consider as reliable (Wangenheim and Bayon, 2004).

It is generally agreed upon that consumers consider WoM and eWoM information as more trustworthy and credible than marketer generated information (Solomon et al., 2013; Blackwell et al., 2006; Day, 1971). This is because information from peer consumers is typically independent of marketers’ selling intents (Dichter, 1966). Moreover, consumers generally perceive the source of eWoM information as similar to themselves. That is, although the senders of eWoM may or may not be similar to the reader regarding demographics and lifestyles, they are similar to readers in that they are fellow consumers (Bickart and Schindler, 2001). Thus, the information they provide are expected to reflect actual product performance, which makes the information more relevant to the reader than most marketer generated information (ibid). Information originating from sources that are recognised as more trustworthy and relevant tend to be more persuasive (Hovland and Weiss, 1951; Wilson and Sherrell, 1993; Hass, 1981; McGuire, 1969; Price et al., 1989).

This makes eWoM an important source of information and a major influence on consumer purchasing behaviour (Kotler and Armstrong, 2016). The literature is saturated with studies that confirm that eWoM directly causes purchase intention (e.g., Beneke et al., 2016; Erkan and Evans, 2016; Plotkina and Munzel, 2016; Chen et al., 2016). The direction of the effect is intuitive: positive eWoM increases purchase intention, while negative eWoM decreases purchase intention (Plotkina and Munzel, 2016).

East et al. (2016) find that positive eWoM has a greater effect on purchase intention than negative eWoM. Moreover, Martensen and Mouritsen (2016) suggest that positive eWoM is the main mediator of the relationship between a company’s marketing program and consumer responses, whereas negative WoM only slightly mediates product quality. Based on this, positive eWoM is more impactful than negative eWoM and more interesting from a theoretical as well as a practical perspective. Therefore, the present study focuses on positive eWoM.

Drawing on the strong empirical evidence of prior research, it is expected that received positive eWoM will lead to a significant increase in consumers’ purchase intention.

3.3 Culture

Cultural factors are the most basic influence on consumers’ wants and behaviour (Kotler and Armstrong, 2016). This is because human behaviour is to a large extend determined by the values, perception, and wants that are learned while growing up in a society (ibid). Children growing up in different parts of the world are likely to be exposed to different values. Therefore, consumer purchasing behaviour may differ greatly between countries (Kotler and Keller, 2016; Keller, 2013).

However, most of the above discussed eWoM studies have investigated purchase intention in a single Western country, while the impact of culture has rarely been considered (Kasabov, 2016; Christodoulides et al., 2012). The few papers that investigate cross-cultural differences in behavioural intentions following eWoM typically do not measure the respondents’ culture in situ. Rather, researchers attribute differences in observed behaviour between respondents from different countries to differences in Hofstede’s (1980) country-level sores.

However, the suitability of national culture to predict an individual’s behaviour has been challenged by numerous scholars from two major literature streams, namely cultural convergence theory, and espoused national culture. The arguments put forward by these streams are discussed in the following.

3.3.1 Cultural Convergence

Scholars of the cultural convergence stream argue that the vast penetration of the Internet, cable and satellite TV, and telecommunication networks that link people across the globe have led to a convergence of lifestyles and the emergence of a common global consumer culture (Kotler and Keller, 2016). Cultures are becoming more similar with needs, wants, behaviours, consumption patterns, and ideologies that are increasingly shared across cultures (Kotler and Armstrong, 2016; Ritzer, 2011; Clarke, 2004; Coughlin, 2000).

This phenomenon becomes especially apparent in an online environment. Social culture is constructed through communication and socialisation which includes virtual processes on the Internet as well as those in the physical world (Johnston and Johal, 1999). As the Internet possesses its own mores and etiquette (Scheuermann and Taylor, 1997), it has become like an own virtual “country” with its own social society and a common global culture which transcends the geographic-nationhood idea of culture (Johnston and Johal, 1999; Burgmann et al., 2006). When individuals are online, they are primarily citizens of this virtual culture rather than citizens of their national country (Johnston and Johal, 1999). The common society of the internet establishes a common context between national cultures that is different yet affiliated with each other. In this vein, the Internet society facilitates the transference of meaning between nationalities (Nicovich and Cornwell, 1998). Thus, the impact of national culture on an individual consumer’s behaviour becomes less salient in an online environment (Nordstrom, 1991; O’Reilly, 1993; Levitt, 1983; Ohmae, 1985; Roberts and Boyacigiller, 1984; Triandis, 1982). This notion is empirically supported by Lichy (2012), who finds a significant convergence in Internet user behaviour in different national cultures.

3.3.2 Espoused National Cultural Values

Literature that advocates the concept of espoused national cultural values is another literature stream that questions the suitability of national culture as a predictor of individual consumer behaviour in nowadays’ globalised world. Donthu and Yoo (1998) argue that due to considerable within-country heterogeneity in many countries, culture is not synonymous with the term country. For example, within the USA, subcultures such as Caucasian Americans, Hispanic Americans, African Americans, and Asian Americans are very different from one another, making it improper to use one uniform national culture to describe individuals (ibid). Supporting this notion, Srite and Karahanna (2006) posit that at the individual level, national culture is manifested through an individual's espoused national cultural values which refer to the degree to which an individual embraces the values of his or her national culture.

Straub et al. (2002) argue that the values an individual espouses are not only defined by the national culture, but also influenced and modified by an individual’s membership in professional, organisational, ethnic, religious, and various other social groups. Each of these groups has its own specialised culture and value set. Therefore, individuals vary greatly in the degree in which they espouse their national culture (ibid). As the effect of national culture on an individual’s behaviour depends on the degree to which an individual adopts different national cultural values, it is inappropriate to use aggregated country scores on cultural dimensions to predict individual behaviour (Ford et al. 2003; McCoy et al. 2005; Srite and Karahanna, 2006). Doing so means to commit “ecological fallacy” (Robinson, 1950): assuming that one can validly use ecological correlations to substitute for individual correlations (Srite and Karahanna, 2006) and substituting stereotypes for individualistic traits (Straub et al., 2002).

Notably, Hofstede (1994) himself argues that his country-level scores cannot predict individual behaviour. Rather, if one is to investigate the impact of national culture on individual behaviour while avoiding ecological fallacy, each individual’s culture must be measured in situ (Straub et al., 2002) , and be treated as an individual difference variable (Srite and Karahanna, 2006). This approach is conform with perspectives in several culture related fields (ibid). First, it is in line with a psychological anthropology perspective which stresses the fact that an individual’s culture and personality cannot be seen in isolation as they interact and therefore, culture should be measured through personality tests at the individual level (Hofstede, 1984). Second, the perspective is supported by social identity theory (Tajfel, 1981), which argues that individuals see themselves as belonging to various kinds of culture such as professional, organisational, ethnic, and national culture, and even though they all assist in defining the individual, certain cultures are more important to the individual in certain situations (Ford et al., 2003).

Based on these perspectives, the present paper draws attention to the absence of research investigating whether differences in espoused national cultural values influence the effect of eWoM on an individual’s purchase intention. Notably, on an individual level of analysis, Schumann et al. (2010) find that the relationship between WoM and perceived service quality is significantly moderated by espoused UA. Further, Lam et al. (2009) find that espoused PD, espoused COL, and espoused MAS have a significant positive effect on an individual’s information acquisition and information sharing behaviour towards personal contacts. Motivated by the findings of these pioneering studies, this paper expects that espoused national cultural values will also have a significant impact on the effect of eWoM on purchase intention. To tackle the aforementioned void, this study assesses individual subjects’ espoused national cultural values. These values are identified in terms of Hofstede’s (1980) four national cultural dimensions, however, they are measured at an individual level (Donthu and Yoo, 1998).

Although most findings of prior cross-cultural studies related to eWoM may not suit to predict individual consumers’ behaviour, they still provide valuable insights for this paper. These findings are discussed in the following sections for each of Hofstede’s cultural dimensions. The conceptual model of the present research is depicted in figure 3.

Figure 3: Research model

Source: Author, 2016

illustration not visible in this excerpt

3.3.3 Power Distance (PD)

Hofstede (2001, p.83) defines PD as follows: “the power distance between a boss B and a subordinate S in a hierarchy is the differences between the extent to which B can determine the behaviour of S and the extent to which S can determine the behavior of B”.

In high PD cultures, individuals in powerful positions determine what is right and valuable (Hofstede, 1991). Therefore, individuals tend to base their decisions on the judgment of their superiors (Hofstede, 1998) and are more likely to comply with their superiors' opinions since they fear to disagree with them (Hofstede, 1984). As consumers are aware that power is often coercive rather than legitimate, they tend to exhibit a general distrust of others (Hofstede, 1980). However, this distrust needs not to be true for an individual’s personal network (Dawar et al., 1996) whose opinions may be consulted before making purchase decisions (Pornpitakpan, 2004).

Therefore, as found by Dawar et al. (1996) and Pornpitakpan (2004), consumers in high PD countries rely more on personal sources of information such as friends, family members and acquaintances, than consumers in low PD cultures do. Accordingly, their studies further found that consumers in high PD countries seek less opinions from impersonal sources such as magazines, TV and salespeople, than consumers in low PD cultures do. This is in line with Goodrich and De Mooij (2014), who find that individuals in high PD cultures favour C2C communication over marketer generated information and that trust in online forums and usage of eWoM in social media for purchase decisions is positively correlated with PD.

By contrast, people in low PD cultures believe that prestige, power, and wealth should be distributed more equally (Hofstede, 2001). Hierarchical structures are flatter, decision-making is more decentralised, individual thinking is encouraged, and supervisor- subordinate communication and cooperation are generally smoother (Hofstede, 1991). These culturally learned behaviours influence their decision making. That is, consumers in low PD countries depend less on others and base decisions more on own evaluations of facts and data (Goodrich and De Mooij, 2014).

In summary, prior research on the national-level of analysis has established that consumers in high PD countries rely more on personal sources of information in their decision making process than individuals in low PD countries. Since contacts within SNS are generally members of focal users’ personal networks rather than strangers (Erkan and Evans, 2016a; Chu and Choi, 2011), eWoM within SNS can be classified as a trustworthy and personal information source. Based on the findings of prior research, it is expected that the more an individual espouses the national cultural value of PD, the stronger will be the effect of eWoM on consumers’ purchase intention.

3.3.4 Uncertainty Avoidance (UA)

Hofstede (2001) argues that people in different cultures deal in different ways with the given unpredictability of the future. UA is “the extent to which the members of a culture feel threatened by uncertain or unknown situations” (Hofstede 2001, p. 161). People in high UA cultures exhibit a low tolerance for ambiguity which results in higher levels of anxiety (Hofstede, 2001). Therefore, high UA cultures strive to reduce ambiguity and risk (Kale and Barns, 1992), which is manifested in a high use of rules and rituals that help to provide the desired predictability to their lives and avoid encountering unfamiliar situations (Hofstede, 2001). By contrast, consumers in low UA cultures accept uncertainty as inevitable and believe that they can influence their own lives (Hofstede 2001).

A way to reduce ambiguity and uncertainty of a purchase decision is to seek advice or assurance from trusted others (Schumann et al., 2010). Confirming this notion, Dawar et al. (1996) find that individuals from cultures high in UA exhibit stronger opinion seeking from personal sources than low UA cultures do. Accordingly, cultures high in UA exhibit lower use of impersonal information sources, than those low in UA. Furthermore, Money et al. (1998) support this notion in a B2B context, finding that Japanese business customers, who are high in UA (Hofstede, 1980), consult considerably more personal referral sources than Americans, who score much lower in UA (Hofstede, 1980).

In sum, it has been shown on a national-level of analysis that consumers in high UA countries are seeking more opinions from personal information sources in their decision making process than consumers in low UA countries. Therefore, as eWoM in SNS is a personal information source (Erkan and Evans, 2016a), it is expected that the more an individual espouses the national cultural value of UA, the stronger will be the effect of eWoM on consumers’ purchase intention.

3.3.5 Individualism/Collectivism (IND/COL)

According to Hofstede (2001), IND/COL refers majorly to two aspects: first, the extent to which individuals prioritise their personal goals as opposed to those of the group, and second, the degree to which they define their self-image in terms of “I” as opposed to “we”. Drawing on two influential papers by Markus and Kitayama (1991) and Triandis (1989), IND and COL can be further distinguished in terms of two dimensions of individuals’ self-concept: the construal of the self (independent versus interdependent) and aspect of the self (private self versus collective self). These dimensions are considered as major determinants of cultural differences in behaviour (Heine et al., 1999). Table 3 contrasts IND and COL in terms of individuals’ self-concept.

As becomes apparent from the table, people in collectivistic countries assign priority to group goals (Gudykunst, 1997), exhibit a greater concern about how others may regard or be affected by their actions (Srite and Karahanna, 2006), and emphasise harmony in person-to- group relations (Bond and Smith, 1996). This group-oriented nature of individuals in collectivistic countries affects consumers’ information acquisition and decision making behaviour. That is, consumers’ opinion tends to be predetermined by their in-group (Hofstede, 2001). For example, the Chinese, which is a collectivistic culture, modify their opinion more easily in terms of their social environment than Western individualists do (ibid). In line with this, researchers find that consumers from collectivistic countries tend to make group-based decisions rather than individual decisions and are more susceptible to opinions of in-groups and reference-groups than those in individualistic countries (Cialdini et al., 1999; Srite and Karahanna, 2006; Bond and Smith, 1996; Doran, 2002). Moreover, individuals in collectivistic countries seek more information from personal sources (Goodrich and de Mooij, 2014; Chu and Choi, 2011; Fong and Burton, 2008; Money et al., 1998), whereas individuals in individualistic countries tend to rely more on mass media (De Mooij, 1998; 2001; as cited in Hofstede, 2001).

Table 3: Distinction between individualism and collectivism in terms of individuals’ self-concept Source: Author, 2016 based on theories of Markus and Kitayama (1991) and Triandis (1989)

illustration not visible in this excerpt

[...]