Social interactions as an influencer of consumers purchase behavior

(1)

Social interactions as an influencer of consumers purchase

behavior

What makes consumers from collectivist and individualistic country different?

Author: Lenka Halgašová Student number: s1942999

Organization: Rijksuniversiteit Groningen, Faculty of Economics Department: Marketing

Specialization: Marketing Management Qualification: Master Thesis

Supervisors: Dr. Wander Jager

(2)

2 Abstract

This study is intended to determine the effects of social influences in purchase behaviors across cultures. The research was conducted in Slovakia as collectivist culture and in the Netherlands as individualist culture. The results provide insights to what extent both cultures vary in their susceptibility to social influences when purchase decision making. Factor analysis was used to determine the social influence variables and opinion leadership across cultures. The following results of significance testing proved that outgoing normative influence, outgoing informational influence and opinion leadership significantly differ across cultures. Higher engagement with outgoing social influence and the higher number of opinion leaders was found in collectivist culture. This implies that the lower the degree of individualism in culture, the higher the engagement with outgoing normative influence, outgoing informational influence and the higher the number of opinion leaders. While ingoing informational influences play an important role in both cultures, ingoing normative influence seems to lose importance for consumers. These results indicate that informational influence has a stronger impact on consumer purchase behavior in collectivist culture than normative influence. Purchase decisions of collectivist consumers are no more based on what others would approve, as we expected. Instead, desire to make informed choices was indicated as an important factor when purchase decision making.

(3)

3 Acknowledgment

I would like to take this opportunity to sincerely thank Wander Jager for his support and very helpful advices, suggestions, comments, and corrections. Further, I would like to thank Eline de Vries for evaluating my thesis.

(4)

4

1. Introduction

We as consumers tend to be influenced by social interactions with others when we make purchase decisions. Such effect of social interactions varies among consumers across cultures. The introduction of social interactions within two different cultures (individualist and collectivist) and the identification of opinion leaders might reveal different relevance of influence on consumer purchase behaviors. Opinion leaders can directly influence other consumers by giving advice and have stronger effect on consumption than the mass media, but such impact of an opinion leader partially depends on the culture (Cosmas and Sheth, 1980; Hoyer and MacInnis, 2007).

The main purpose of this thesis is to gain understanding of how social interactions can affect consumer purchase behaviors in relation to cultural differences between collectivist and individualist. In order to assess social influence, is necessary to examine similarities and differences in consumer behaviors between two cultures. This cross-cultural comparison will provide detailed information about major drivers of consumer purchase behaviors. Various behavioral factors can determine what knowledge and skills consumers possess, and help to identify opinion leaders. The results of this analysis will provide important information that might be useful when creating effective global marketing activities to target consumers across boundaries.

1.1 Problem analysis

(6)

6

behavior varies among cultures according to consumer preferences, purchase habits and consumption patterns, what has been shown in different purchase behavior of consumers (Cialdini and Goldstein, 2004). The consumption patterns are influenced by several variables (e.g., age, gender, religion, social class, values, etc.) whereas the consumer behavior itself is driven by different sources of social influences and by the form of a social network (Engel, Kegerreis and Blackwell, 1969; Feick and Price, 1987; Katz and Lazarsfeld, 1955). Consumer behavior may have different motives (hedonic or utilitarian) and different needs to satisfy (Jones, Reynolds and Arnold, 2006; Maslow 1954; Max-Neef, 1992).

(7)

7

into consumer behavior and help to explain the values of a particular culture. Therefore the main purpose of this analysis is to determine the impact of social influence on consumers’ purchase intentions regarding individualist and collectivist culture.

1.2 Problem statement

Consumers can learn from and be affected by other consumers’ opinions and others’ actual purchase decisions. The extent to which social interactions of individual culture affect consumers’ decisions depends on susceptibility to ingoing and outgoing normative and informational influence, what led us to the following problem statement:

To what extent can consumers’ purchase behaviour be affected by social interactions in relation to individualist and collectivist culture?

1.3 Structure of the thesis

(8)

8 Psychological factors: • motivation • perception • learning

• beliefs and attitude Cultural factors: • culture • sub-culture • social class Social factors: • reference groups • family

• roles and status Personal factors: • age, life-cycle stage • occupation

• economic situation • life style, personality,

self-concept

2. Theoretical background

2.1 Consumer purchase behavior

Human behavior is not always easy to predict because we cannot look into the consciousness of a person. Nobody can say exactly how consumer´s behavior will look like in the future, but is definitely possible to get an idea about future consumer behavior, given the current consumer behavior and his/her purchase patterns. Consumer behavior can be viewed as search, purchase, use and disposal of a product, what is affected by the mental and social processes that take place before purchasing the product (e.g., an awareness of the needs, product involvement, product selection, risk perception), during the purchase (e.g., the product encounter, consumer buying behavior, the place of purchase) and after purchase (e.g., post-purchase evaluation, the process of consumption) (Hoffman and Turley, 2002; Gardial et al., 1994; Dholakia, 2001). Purchase motives can take place for hedonic or utilitarian reasons. According to Hirschman and Holbrook (1982), hedonic motives are related to the multi sensory (e.g., gratification, fun, enjoyment), fantasy and emotive aspects of one's experience with products. Utilitarian products are less arousing, as they are generally consumed for rational reason of product functionality (Lim and Ang, 2008). For example, a consumer who wants to buy a smart phone may have an interest in both utilitarian (e.g., functional benefits) and hedonic (e.g., design) attributes value (Lee, Kim and Fairhurst, 2009). Consumers frequently consider also the indirect effects of their utilitarian choices, such as social utility. Social utility refers to other people’s reactions and appreciation of a consumer’s private utility (Thompson and Norton, 2011). Consumption of specific products or services can shift from utilitarian to hedonic as an economy of a certain country progresses (Lim and Ang, 2008). According to Kotler (1997), consumer purchase behavior is influenced by four basic factors: cultural, social, personal and psychological (figure 1).

(9)

9

Cultural factors seem to be the most powerful influencers on consumer behavior, because the society and environment where we grew up has deep impact on our behaviors. Social factors, such as consumer’s groups, family, roles and status, affect consumer responses to marketing strategies. Consumers’ purchase choices are also influenced by psychological factors and personal characteristics, such as age, life-cycle stage, a person’s occupation, economic situation, life style, personality and self-concept (Kotler, 1997). All the factors mentioned above are important determinants of identification and understanding consumer purchase behavior.

2.2 Social influence

When buying a certain product consumers are influenced in several ways. One of the most profound determinants affecting consumer behavior is the social influence of those around them such as family, social class and other reference groups. Over time consumers become members of several social institutions and social groups. These social institutions and groups have powerful influence on their behavior. Conforming to such social influence, consumers adjust or reduce their choices of consumption acceptable to the social groups to which they belong, whereas it cannot be forgotten that for different products the social influences differ (Cialdini and Goldstein, 2004). The extent of consumer susceptibility to interpersonal influence depends on cultural and societal values and norms (Mourali, Laroche and Pons, 2005). Consumers with stronger social needs tend to be more open to and derive greater value from social interactions (Hoffmann and Broekhuizen, 2009). Two types of influence, such as normative and informational, affect consumer’s purchase behavior. Normative influences are addressed to compliance to group to feel associated with that group and avoid social expel. Informational influences encompass exchanging information and experiences with people, goods and behaviors (Cialdini and Goldstein, 2004; Hoyer and MacInnis, 2007).

2.2.1 Normative influence

(10)

10

would be expected to conform to the influence of others. The performance of an individual should be visible or known to others. An individual motivated to enhance or support his concept of himself would be expected to accept the influence of a referent by associating himself with positive referents and/or dissociating himself from negative referents. Thus, a person would be identified by taking on the behaviors and opinions that s/he perceives as representative of his/her positive reference groups (Burnkrant and Cousineau, 1975).

Reno, Cialdini and Kallgren (1993) stated that the norms reflect conceptions of what significant others think a person should do and what significant others themselves do. Authors distinguished two types of social norms, descriptive and injunctive. The first, descriptive norms reflect the perception of what most people do in a particular situation, and they inform people about which behaviors are more effective in the sense they lead to the desired outcomes in given circumstances. Descriptive norms provide a standard from which people do not want to deviate. Because people measure the appropriateness of their behavior by how far away they are from the norm, being deviant is being above or below the norm. Specific descriptive normative information can attract an attention of individuals both above and below the norm. The same information may also increase undesirable behavior among individuals (e.g., alcohol consumption) whose behavior is less desirable than the norm (Schultz et al., 2007). Injunctive norms are a source of normative social influence which is based on the fundamental human need to be accepted by others. They specify what people approve and disapprove within the culture. Hence, injunctive norms motivate people to conform because they invoke possible external sanctions in case of counter normative conduct. Consequently, non conformity to injunctive norms is more likely to induce negative emotions than non conformity to descriptive norms (Reno, Cialdini and Kallgren, 1993; Stavrova, Schlösser and Fetchenhauer, 2010). Further social injunctive norms can be internalized by a person and can become its self-expectations and personal obligations or personal norms. This highlights the difference between injunctive norms performing on social and individual or personal level. In contrast to the social injunctive norms, personal injunctive norms are not enforced through sanctions and rewards imposed by other members of the social group, but are specified by the individual him/herself (Stavrova, Schlösser and Fetchenhauer, 2010).

(11)

11

they possess strong or weak personal norms what differs accordingly to their individual characters, hence personal norms increase the amount of behavioral variance. Following a personal norm to express a certain behavior is based on the individual's normative preferences and not on automaticity. The automaticity is more associated with habit (Thøgersen, 2002).

In general, particular norms can direct behavior, but only when they are in focus of a consumer. This means that if only one of the two types of norms, descriptive or injunctive, is prominent in an individual’s consciousness, it will exert the stronger influence on behavior (Gialdini and Goldstein, 2004). Thus, it can happen that descriptive normative information, that describes what others have done, may produce an undesirable effect (e.g., alcohol consumption). By adding an injunctive message indicating the approval of desired behavior, an undesirable effect might be prevented (Schultz et al., 2007). But on the other hand injunctive norms, in comparison with descriptive norms, may be less effective for changing behavior, because they exert strong control effects, limit people’s freedom, and often produce feelings of resentment or psychological reactance (Hoffmann and Broekhuizen, 2009).

(12)

12

makes them excessively fearful of social disapproval, strongly motivated to conform to others’ demands and are more influenceable (Cialdini and Goldstein, 2004; Hoffmann and Broekhuizen, 2009). As a third factor stressing the degree of normative influence is the group characteristics to which a consumer belongs where, for example, a cohesive group communicating on regular basis has a greater opportunity to affect consumer’s behavior (Hoyer and MacInnis, 2007).

2.2.2 Informational influence

When buying a certain product, consumers consult other people (i.e., friends, family or salespersons) to help choose the best available alternative. Informational influences are driven by a desire to form accurate interpretations about reality in order to make more informed decisions and behave correctly (Cialdini and Goldstein, 2004). Informational influences are important to consumers who feel the need to make informed choices, and results from actively requesting information from knowledgeable others (Hoffmann and Broekhuizen, 2009; Hoyer and MacInnis, 2007). The amount of domain knowledge determines product decisions. Consumers who perceive psycho-social risks may actively seek and rely on information from and the opinions of social others to reduce their risk. Hence, consumers who perceive greater psycho-social risks tend to be more susceptible to informational influence, because they are motivated to build their knowledge about socially accepted behaviors and the consequences of their actions for their self-concept (Cialdini and Goldstein, 2004). These consumers ask relevant others, so that they may learn about the correct way to behave and avoid future social embarrassment or psychological discomfort (Hoffmann and Broekhuizen, 2009).

(13)

13

more strongly on this advice compared with those consumers who know more and perceive less risk (Hoffmann and Broekhuizen, 2009; Hoyer and MacInnis, 2007).

2.2.3 Ingoing and outgoing social influence

Consumers can differ in their susceptibility to social influence. The most widely used measure of susceptibility to social influence is that developed by Bearden, Netemeyer and Teel (1989), which distinguishes between susceptibility to informational and normative influences. The scale measures the extent to which consumers choices are influenced by other people. Susceptibility to social influence can be enhanced when individuals are highly sensitive to the judgments others make about them (Netemeyer, Bearden and Teel, 1992). That is, people high in susceptibility to social influence are more likely to purchase products that they perceive will lead others to make favourable attributions about them and less likely to buy products that they perceive will lead others to make negative evaluations of them. Such high susceptibility to social influence is more pronounced for normative than informational influence. Following the construct of Bearden, Netemeyer and Teel (1989), social influences can be classified into four different groups: ingoing normative, outgoing normative, ingoing informational and outgoing informational influence (Jager, 2010). The impact of social influence on consumer purchase behavior was researched many times (Cialdini and Goldstein, 2004; Mourali, Laroche and Pons, 2005; Hoffmann and Broekhuizen, 2009; Broekhuizen, Delre and Torres, 2011), but there is much less known about ingoing and outgoing social influence. I did not come across any research that focused on ingoing and outgoing normative and informational influence in consumer purchase behavior with relation to culture setting. However, one study did find the differences between ingoing and outgoing normative and informational influence regarding gender and age. The observed differences between females and males indicate that males possess more informational outgoing influence than females. Concerning age the results show that older people are less engaged with social influence, except for informational outgoing influence (Jager, 2010).

2.2.4 Sources of social influence

(14)

14

is far more effective than traditional marketing communications, because the word-of-mouth message is perceived as an unbiased third party. The frequency and intensity of word-of-mouth depend on the types of products, the social networks involved, the personality and culture of communicators (Lam, Lee and Mizerski, 2009).

Consumer purchase behavior can be affected by several sources of social influence. Sources of social influence may differ across consumers with regards to the characteristics that influentials possess, their concern and knowledge associated with a specific products or services. Regardless of whether it is due to individuals’ differences or product preferences, the behavioral outcome and consumer’s predispositions toward a certain product are independent of external influences, such as the mass media (Van Eck, Jager and Leeflang, 2011; Watts and Dodds, 2007; Hoyer and MacInnis, 2007).

Social influentials possess several identical features but some features differ from each other, due to which we can see who from the social influentials have the greatest impact on consumer purchase behavior:

 Innovators are more active information seekers of product, are prone to buy new products earlier and enjoy trying out a new innovation. Innovators possess advanced technical knowledge, are able to handle a high degree of uncertainty about an innovation at the time of adoption, usually are users of the product about which they have information and appreciate the benefits of an innovation (Engel, Kegerreis and Blackwell, 1969; Rogers, 2003).

 Early adopters are according to Rogers (2003) referred as a group who adopt a technology after the previous group of innovators. They act as opinion leaders and share similar characteristics with those of innovators but at a slightly lower degree.

 Market mavens have information about many kinds of products and places to shop what includes general marketplace knowledge and expertise. The market maven's influence is based on more general market expertise. Market mavens are not product specific and are not necessarily users of the product about which they have information (Feick and Price, 1987).

(15)

15 2.2.5 Opinion leaders

Opinion leaders are a special and the most relevant source of social influence. According to King and Summers (1970), opinion leaders are defined as a person’s tendency to influence the purchase decisions of other consumers. They are observed by other consumers and are sought out for advice on information about search, purchase, and use of specific products (Flynn, Goldsmith and Eastman, 1996). Such an opinion leader includes the innovators, early adopters and partly the market mavens. Their influences derive from expertise, experience and knowledge, but differ in that the expertise of market mavens is not product specific but rather market based. Market mavens' general marketplace expertise should lead them to earlier awareness of new products and may also lead them to acquire not only general market information, but also in-depth information on selected products (Van Eck, Jager and Leeflang, 2011; Feick and Price, 1987). Opinion leaders play a fundamental role in society when it comes to spread new ideas, values and belief, therefore marketing aims to identify these opinion leaders (Hoyer and MacInnis, 2007). The influence of such a person is direct and derives from its informal status. It is important to note that opinion leaders exert both normative and informational influences, have central network position, significant interpersonal influence, possess more experience or expertise with the product category, have been exposed to or acquired more information about a product, exhibit more exploratory and innovative behavior, and display higher levels of involvement with the product category. Therefore opinion leaders are considered to be the most important source of influences if they are active in a social network (Van Eck, Jager and Leeflang, 2011).

People within and across cultures use some basic dimensions to evaluate their opinion leaders and among these dimensions, different cultures apply different degrees of importance. The more similar cultures are, the more similar are the dimensions used to evaluate opinion leaders, but people of two different groups sharing a common geographical setting might not be always similar in their perceptions of opinion leadership (Cosmas and Sheth, 1980).

2.3 Culture

(16)

16

that characterize a group of people. According to Hoyer and MacInnis (2007), the group of external processes consists of many regional, ethnic or reference groups, social classes and household influences, one´s values, personality and its lifestyle. All of these external processes may directly or indirectly affect consumers’ behavior.

2.3.1 Cultural dimensions

According to Hofstede (2001) cultural dimensions are based on values and are used to describe key differences between various national cultures. Hofstede’s concept distinguishes five basic cultural dimensions:

 Power distance: can be defined as the extent to which the less powerful members of institutions and organisations within a culture expect and accept that power is distributed unequally. Small power distance means minimum inequality among people, minimum hierarchy distance and narrower salary range. At work subordinates expect to be consulted and democratic management style is the most convenient. Large power distance respects inequalities among people, dependence of less powerful people and centralisation of power. Inequality is accompanied by privileges and status. The boss is benevolent autocrat and subordinates are directly told what to do. In large power distance cultures the respect for parents, elders and authorities in general is desired. In small power distance cultures there is more space for own initiative and creativity. Concerning state organisation, for large power distance cultures dictatorship is typical where wealth is unequally distributed. In small power distance cultures human rights are guaranteed equally, countries are more economically developed and there is space for political discussion and change.

(17)

17

with companies owned by other family members and work relationships resemble more family relationship.

 Masculinity/femininity values: the third dimension is connected to gender roles. In masculine society men are supposed to be assertive, aggressive, rational, oriented toward performance and material things. Women are more relationship-oriented and sensitive to others. In masculine societies gender roles are distinct and clearly set – men assertive, women more tender and modest. In feminine societies gender roles overlap and both men and women tend to be more tender and concerned with the quality of life. Masculine societies express sympathy for strong and ambition, where conflicts are solved through fight. Feminine cultures tend to solve problems with compromise. Assertive behaviour is often seen as ridiculous, whereas solidarity and social skills are appreciated. The difference is clear in management and leading roles in companies. In masculine cultures competitiveness and decisive leading style is typical. In feminine cultures there are more women in management, there is strong tendency to cooperate and seek consensus. Masculine countries are performance oriented and they are often active in international conflicts.

 Uncertainty avoidance: characterises the ability and readiness to take risk or avoid it. Cultures with strong uncertainty avoidance tend to plan and set a lot of rules. However strong uncertainty avoidance means natural need for rules, there are negative feelings toward law, rules and politics. In cultures with weak uncertainty avoidance there is a space for new ideas, innovation and different behaviour. People are not afraid of differences, they are more curious and more open to doubt and disagreement. In weak uncertainty avoidance countries rules are not so specified, they do not regulate everything, but they are naturally respected.

 Long-term/short-term orientation: means orientation toward past, present or future. Long term oriented cultures tend to adapt tradition to a modern context. In these cultures social structures and relationships are more stable and longer lasting. Short term oriented countries have respect for traditions, but they are seen as a part of the past. Social structures are not so stable and these cultures are more likely to accept changes.

(18)

18

tend to follow given structure during negotiation. Long-term orientation affects time management during dealing and length of negotiation. Long-term oriented cultures do not mind waiting, long discussions or breaks during negotiation.

2.4 Social influence and culture

(19)

19

individual’s values at the national level: individualism/collectivism, power distance, masculinity/femininity, uncertainty avoidance and long-term/short-term orientation. Consumers of different cultures behave differently. Their relationship orientation to other people in society causes variations in preferences of specific products and has an impact on purchase behaviors. The relationship orientation is associated with the individualism versus collectivism as one of the five Hofstede’s (2001) dimensions of culture. The individualism versus collectivism dimension describes the degree to which individuals look after themselves or remain integrated into groups. This research studies individualism versus collectivism as an important dimension of culture which moderates the relationship between social interactions and consumers purchase behaviors.

(20)

20

H1a: The level of consumer susceptibility to ingoing informational influence is higher for individualist consumers than for collectivist consumers.

H1b: The level of consumer susceptibility to outgoing informational influence is higher for individualist consumers than for collectivist consumers.

People in collectivistic culture can be described as we-conscious. The relationship between attitude and future behaviour is usually not consistent. It may even be a reverse relationship: behaviour (purchase decision making) comes first and defines attitude (De Mooij and Hofstede, 2010). Triandis (1995) defines collectivism as a social pattern that consists of individuals who see themselves as an integral part of one or more collectives, such as family and friends. Even very ambitious consumers with individualist tendencies who grow up in collectivist culture are more likely to incorporate family members' opinions when making a purchase decision (Triandis, 1995). Members of collectivistic culture tend to be susceptible to social influence because they mainly seek social rewards (Broekhuizen, Delre and Torres, 2011). Normative social influence has been found to influence consumers' purchase intentions more strongly in collectivist cultures compared to individualist cultures (Broekhuizen, Delre and Torres, 2011; Lee and Kacen, 2008; Mourali, Laroche and Pons, 2005). In collectivistic and high power distance cultures, people will acquire information more via implicit, interpersonal communication and base their purchase decisions more on feelings and trust (De Mooij and Hofstede, 2010). In general, large power distance cultures tend to be collectivistic. Cultures with large power distance indicate a higher level of inequality of power and wealth within the society (Hofstede, 2001). If this is the case, then the Slovaks should have collectivistic orientation and positive susceptibility to ingoing and outgoing normative influence. This reasoning led to the following hypotheses:

H2a: The level of consumer susceptibility to ingoing normative influence is higher for collectivist consumers than for individualist consumers.

H2b: The level of consumer susceptibility to outgoing normative influence is higher for collectivist consumers than for individualist consumers.

(21)

21

They have higher number of relations, are more innovative, have better product judgment and are less sensitive to normative influence (Van Eck, Jager and Leeflang, 2011). The roles and evaluations of opinion leaders vary across cultures. They can affect consumption of other consumers (Cosmas and Sheth, 1980), and are likely to be favoured in strong uncertainty avoidance cultures (De Mooij, 2011). In line with Hofstede’s (2011) cultural dimensions, the uncertainty avoidance dimension of the Netherlands (53) has very similar ranking as Slovakia (51). If a population has a high uncertainty avoidance value, it will institute strict rules, laws, policies, and regulations. A moderate uncertainty avoidance score, as the Netherlands and Slovakia do, may minimize or reduce the level of uncertainty within the population by enacting rules, laws, policies, and regulations (Hofstede, 2001). Both Slovakia and the Netherlands have a moderate uncertainty avoidance score, but their scores of the individualism/collectivism and the power distance dimension differ significantly (figure 2). Figure 2: Ranking of cultural dimensions (Hofstede, 2011)

Country Individualism Power Distance Index Uncertainty Avoidance Index Netherlands 80 38 53 Slovakia 52 104 51

The cultural dimensions values of Slovakia indicate that the combination of the power distance dimension (high score) and the uncertainty avoidance dimension (moderate score) create societies that are highly rule-oriented. Inequalities of power and wealth have been allowed to grow within the society. This culture is more likely to follow a caste system that does not allow significant upward mobility of its citizens. The combination of the individualism/collectivism dimension and the uncertainty avoidance dimension, both with moderate scores, create societies that tend to maintain harmony, avoid conflicts and follow the rules because collectivist cultures are shame cultures and for breaking rules are seriously judged (Hofstede, 2001). Due to the scores of cultural dimensions discussed above Slovakia seems to be highly rule-oriented culture. Based on lower individualistic orientation of Slovak consumers, it could be said that people from cultures high in individualism would less likely to be opinion leaders. Hence, the following hypothesis is proposed:

(22)

22 2.5 Proposed research framework

The aim of the study is to discuss the extent to which social influences affect consumers when buying a product or service. The objective of the analysis is to propose the research framework related to relationship of social influence (normative/informational), sources of social influence (opinion leader), and culture (individualism/collectivism) to consumer purchase behavior. The research framework (figure 3) was formulated and derived from the hypotheses discussed above.

Figure 3: Proposed research framework

(23)

23

3. Research design

This chapter describes the aims of the survey, the research plan, the questionnaire, the sample size and the analysis. Then the results of the research are presented and discussed.

3.1 Research plan

Based on previous studies, it is believed that social influence has an effect on consumers’ purchase behaviors (Gialdini and Goldstein, 2004; Reno, Cialdini and Kallgren, 1993). The purchase of products, such as smart phones, which are used publicly, is strongly influenced by other consumers (Yang, He and Lee, 2007). The aim of this research is to investigate the strength of social influences and compare consumers’ purchase behaviors across individualist and collectivist cultures. According to Hofstede (1980), countries were selected based on their level of individualism/collectivism. The Netherlands and Slovakia were chosen because they form two culturally distinct groups, shown to differ in their levels of individualism (Figure 4) (Hofstede, 2011). The survey will be conducted both in the Netherlands and Slovakia. Consumers of these two cultures will be invited by email to join the survey and asked to fill in and submit the survey online. The survey is worded in Dutch and in Slovak language.

Figure 4: Ranking of cultural dimensions (Hofstede, 2011) Country Individualism Power Distance Index Uncertainty Avoidance Index Masculinity Long-term Orientation Netherlands 80 38 53 14 44 Slovakia 52 104 51 110 38 3.2 Questionnaire development

(24)

24

proposed by Bearden, Netemeyer and Teel (1989) and recently extended by Jager (2010) has been used. The statements are designed to measure the extent to which consumers desire to identify with the opinion of others through smart phone purchases, are willing to conform to others in their smart phone choices and learn about the smart phone by seeking information from others. 5-point scale, ranged from 5 = strongly agree to 1 = strongly disagree, follows each of the statements. The third section of the questionnaire focuses on the opinion leadership. Opinion leadership is measured with the use of six statements derived from the study of Flynn, Goldsmith and Eastman (1996). The scale measures the extent to which an individual influences the purchasing behavior of others. The scale was adapted to the product domain of smart phones. The six 7-point scale statements, retained in the study of Flynn, Goldsmith, and Eastman’s (1996), were used for the research (ranged from 7 = strongly agree to 1 = strongly disagree).

3.3 The analysis plan

Convenience sampling is used to collect the data for our analysis. The convenience sample technique has been chosen, since respondents are easily accessible and more cooperative (Malhotra, 2004). The analysis is structured as follows: First, demographic variables, namely gender, age, education and employment status, will be described, analyzed and statistically tested. The collection of data is limited to a certain number of respondents who characterize the population of collectivist and individualist culture. Due to this fact statistical significance testing has to be conducted to identify whether our samples are representative for Dutch and Slovak population. The results of a t-test for the age variable, a z-test for the gender variable, and Chi-square test for the education and occupation variable, will determine whether our samples will differ significantly from the real situation in the Slovak and Dutch population. Because our study is using the data of two different countries, we have to examine whether the demographic variables are comparable across cultures. To identify the differences between cultures we will use Chi-square test for categorical variables, such as gender, education and employment status. Since age is a scale variable, a t-test will be used. Finally, the results discussion will be followed by conclusion.

(25)

25

Malhotra, 2004). According the twenty four statements used in our research we will be able to identify and describe the major dimensions. Our research will be approached from an exploratory perspective, since the number of factors is unknown and the structure in the data has to be determined. In this case, our objective is to summarize all of the data in a minimum number of factors, which according to Hair et al. (2010) is in line with Principal component analysis. Thus, Principal component analysis will be conducted, for both cultures separately. After the number of factors will be determined, reliability analysis of the entire scale needs to be conducted. Since a 5-point rating scale is used to answer the questions in our research, the reliability of the scale will be assessed by the calculation of Cronbach’s alpha (Hair et al., 2010). In order to determine the degree of susceptibility to social influence, the scores of summated scales will be used. An analysis of variance (ANOVA) will be employed to find out whether the summated scale scores of social influence variables differ significantly across cultures (Hair et al., 2010). Based on the results of an ANOVA, the discovered differences between cultures will be described.

(26)

26

4. Results

This chapter comprises the results of the entire analysis. Since our survey consists of three parts, the results are presented in three subsections. First, the demographic variables are presented. The first subsection is followed by the entire analysis of social influence and opinion leadership. Consequently the hypotheses investigated in our study will be confirmed or rejected.

4.1 Demographic differences between cultures

In this section the demographic variables will be described, compared and statistically tested. Our purpose is to find out whether or not our samples (cultures) differ significantly from each other. Furthermore, by conducting the tests of significance we are able to determine whether our samples are representative for Dutch and Slovak population with regard to age, gender, education and occupation.

4.1.1 Age

The size of our sample consists of 123 Dutch consumers (individualist culture) and 125 Slovak consumers (collectivist culture).

Dutch consumers in our sample ranged in age from 20 to 61 years, with the mean age of 29,09 years (std.deviation = 7,991). In the comparison we applied the real age of Dutch population in 2010 (table 1). The real age of population averaged higher compared with the age in our sample (Eurostat 1, 2012).

Table 1: The age differences between our sample and the Dutch population

It is necessary to conduct the significance test to compare the average age of our sample with the Dutch population. By performing the t-test for the means we find out that the average age of our sample significantly differs from the Dutch national average age (t-value = -15,976; df = 122; Sig.2-tailed value = 0,000). Due to this fact we consider that the age of our sample is not a good representative of the Dutch population.

(27)

27

Table 2: The age differences between our sample and the Slovak population

To confirm whether our sample significantly differs from Slovak population, in relation to age, the t-test for the means was performed. The t-test for means confirmed that the average age of our sample significantly differs from Slovak national average age (t-value = -6,093; df = 124; Sig.2-tailed value = 0,000). Since significant differences had been found, is has to be determined that our sample is not representative for Slovak population with regard to age.

Next, we would like to examine whether our samples (individualist and collectivist culture) are comparable in relation to age. By performing a t-test we are able to see whether the differences between cultures are significant. The result confirmed significant differences between cultures regarding age (t-value = -2,256; df = 246, Sig.2-tailed value = 0,025). It can be concluded that data we used in our analysis are not equal across cultures due to the significant age differences. Based on Ward (1974) and Jager (2010), consumers' age has a relation with susceptibility to social influences. The age differences may lead to different level of susceptibility to social influence when purchase decision making. Therefore the results of our study might be different if consumers would consist of comparable age groups. 4.1.2 Gender

Our sample of Dutch consumers consists of more males than females (table 3). Concerning the real situation in the Netherlands in 2010, the proportion of men and women slightly differ compare to our sample (Statistics Netherlands 1, 2012).

Table 3: The differences between the proportions of females and males, the Netherlands

(28)

28

In the collectivist culture sample females are slightly predominant than males (table 4). The real situation in Slovakia in 2010 is similar to our sample. The proportion of females is also slightly predominant compare to males (Statistics Slovakia 1, 2012).

Table 4: The differences between the proportions of females and males, Slovakia

The z-test for binominal proportion of females and males confirmed non significant differences between our sample and the real situation in Slovakia (females, Sig.1-tailed = 0,152; males, Sig.1-tailed = 0,344). According to the z-test results, the sample of collectivist culture used in our analysis is considered to be a relevant representative of Slovak population.

In order to find out if our samples (individualist and collectivist culture) are comparable in relation to gender, we have to follow the Chi-square test of significance. According to the result it is concluded that gender is not significantly different for collectivist and individualist culture sample (Chi-square = 1,958; df = 1, Sig.2-sided = 0,162).

4.1.3 Education

The majority of consumers in both cultures attained a master’s degree or a bachelor’s degree (table 5 and 6), what differs from the real situation in the Netherlands and Slovakia (Statistics Netherlands 2, 2012; Statistics Slovakia 2; 2012). This could be possibly caused by conducting the research among consumers who are more likely to have similar characteristics. The Chi-square independence test of categorical variables can confirm whether the data in our samples differs significantly from the real situation in the Netherlands and Slovakia. Due to the data availability, the group of individualist consumers who attained a master’s degree or doctoral degree are grouped together (Statistics Netherlands 2, 2012).

(29)

29

According to the table above the differences between our sample and the real situation in the Netherlands are obvious. Also the Chi-square test confirmed significant differences (Chi-square value = 399,224, df = 3, Sig. value = 0,000).

Table 6: The differences in education levels between our sample and the Slovak population

The Chi-square test confirmed also significant differences between our sample and the real situation in Slovakia (Chi-square value = 299,210; df = 4, Sig. value = 0,000). Based on the results of the Chi-square tests, the samples of both cultures are not appropriate representatives of the Slovak and Dutch population.

The next significance test confirmed that our samples are not equal due to significant education differences between individualist and collectivist consumers (Chi-square = 20,262; df = 4, Sig.2-sided = 0,000).

4.1.4 Occupation

(30)

30

Table 7: The differences in occupation between our sample and the Dutch population

To find out whether our sample differs significantly from the real situation in the Dutch population, the Chi-square test has to be performed. Since the result of Chi-square test is significant (Chi-square value = 423,778, df = 6, Sig. value = 0,000), the sample of individualist culture is not considered to be a representative sample of the Dutch population.

Table 8: The differences in occupations between our sample and the Slovak population

When considering the differences between our sample and the real situation in Slovakia, the result of Chi-square test shows that significant differences were observed (Chi-square value = 39,862, df = 5, Sig. value = 0,000). For this reason our sample cannot be regarded as a representative sample of the Slovak population.

The last significance test, concerning occupation, confirmed that our samples are not equal due to significant differences between individualist and collectivist culture sample (Chi-square = 30,473; df = 7, Sig.2-sided = 0,000).

(31)

31

explain the higher number of students in individualist sample who are still occupied to complete their master studies. Individualist culture is predominantly part-time work oriented. In general, it would be better to gain large sample size in terms of increasing the generalizability of our results. Therefore, larger sample size could increase the heterogeneity within a sample and the homogeneity between the samples, what would make our samples more comparable and representative. As consequences of having larger sample size we could further achieve higher significant differences across cultures regarding ingoing informational, outgoing informational and ingoing normative social influences.

4.2 Analysis of social influence

(32)

32

Table 9: Kaiser-Meyer-Olkin measure and Bartlett's test scores across cultures

In terms of Bartlett’s test of sphericity, the p-values were lower than 0,01 for both culture samples. This means that the variables are uncorrelated and factor analysis will be meaningful (Hair et al., 2008). The next step was to determine the number of factors. When defining how many factors to extract, we took several criteria into consideration. First, only factors for which the Eigenvalues are greater than 1 should be kept. The second and third criterion is the Percentage of variance criterion, which suggests to stop as soon as 60% of the Total variance is explained and when the last factor does not account for more than 5% of the variance. In addition to the three rules, a Scree plot is also a good visual method to apply, when to decide how many factors to extract (Hair et al., 2010). A priori knowledge was also used as an important criterion in this regard (Janssens et al., 2008). According to the results the recommended cut off point was when the Eigenvalues were above 1, which suggests a four factors solution for both cultures. If we apply the rule of 60% of the Total variance explained, three factors can be retained for both cultures. Again, four factors solution occurs, when the last factor does not account for more than 5% of the variance for both cultures. In addition, we also considered the factor solutions provided in the Scree plots (appendices III and IV). The Scree plot outputs, indentifying the elbow shape in the Eigenvalues, indicate that five factors may be an appropriate solution. Before moving on, we are left with three, four or five factors to choose from (Table 10).

Table 10: Results for the number of factors to extract

(33)

33

the origin to redistribute the variance in a more meaningful patter what allows us to better interpret the factors (table 11 and 12).

Table 11: Resulting dimensions for collectivist culture using Varimax orthogonal rotation

(34)

34

Factor loadings above 0,5 were considered to be significant in table 11 and 12 (Hair et al., 2010; Janssens et al., 2008). The rotated component matrices for both cultures show correlations that group nicely into four dimensions. The first dimension shows the statements (variables) associated with outgoing normative influence, while the second groups ingoing normative influence variables. The third group represents ingoing informational influence and the fourth dimension outgoing informational influence. After the factor dimensions were identified, we had to measure the reliability of internal consistency. The reliability analysis assesses the consistency of the entire scale. Since we used a 5-point scale in our questionnaire to evaluate the twenty four statements, we had to determine if the scale is reliable. One way of doing this is by calculating Cronbach’s alpha (Hair et al., 2010). Cronbach’s alpha reliability coefficient ranges between 0 and 1, but the threshold should exceed the lower limit of 0,7, and a 0,6 for exploratory research (Hair et al., 2010). The subscales used for the social influence variables (ingoing informational, outgoing informational, ingoing normative, outgoing normative) were specified and subsequently measured. The Cronbach’s alpha of the four subscales exhibited acceptable levels of reliability (table 13).

Table 13: The results of Cronbach’s alpha calculations

(35)

35

Table 14: The summary of summated scale scores across cultures

In the summary above has been shown that ingoing informational social influence is the most weighted variable for both cultures. Overall higher scores indicate that collectivist consumers suppose to be more socially influenced, which can be confirmed after conducting the significance test. To find out whether the scores of variables significantly differ across cultures an analysis of variance (ANOVA) will be used. Ingoing informational influence, outgoing informational influence, ingoing normative influence and outgoing normative influence are the dependent variables to be measured to determine to what extent has culture, as an independent variable, an impact on susceptibility to social influence. Hence, four ANOVAs are conducted to measure the differences between collectivist and individualist culture regarding social influence variables (appendix V). According to an ANOVA, the results of outgoing informational influence were found marginally significant (F = 3,756; df = 1; p = 0,054). This implies that the results indicate a non significant trend for outgoing informational influence across cultures showing collectivist consumers to be more engaged with social influence (Mean = 2,806; SD = 1,026) than individualist consumers (Mean = 2,555; SD = 1,015). The differences between cultures were statistically significant for outgoing normative influence (F = 8,399; df = 1; p = 0,004). No significant differences have been found between cultures in relation to ingoing informational and ingoing normative influence

(table 15).

Table 15: The results of an ANOVA for difference testing between cultures in relation to social influence

(36)

36

normative influence across cultures. We split out the data in our analysis according to rank value of summated scale of outgoing normative variable. The consumers who ranked higher than value of 3 on the 5-point scale were defined as consumers who are engaged in processes of outgoing normative influence. Next, the gender variable is known as a representative variable of the Slovak and the Dutch population, and the proportions of females and males in our samples do not differ across cultures. For these reasons, only gender as a demographic variable is included in the table below.

Table 16: Culture differences description in relation to outgoing normative influence

The number of consumers who engage in processes of outgoing normative influence is higher in collectivist culture. Their engagement is scored slightly higher compare to individualist consumers. Concerning gender, males are predominantly engaged with outgoing normative influence in both cultures.

4.2.1 Discussion

(37)

37

Table 17: Summary of hypotheses concerning social influence

In spite of the fact that individualist consumers possess lower level of susceptibility to ingoing informational influence, no significant differences between cultures had been found (F = 2,435; df = 1; p = 0,120). Due to these findings, the hypothesis H1a is only partially rejected.

The next result showed that individualist consumers engage less in processes of outgoing informational influence. When we looked at the significance test of differences between cultures, the result is referred to as being marginally significant (F = 3,756; df = 1; p = 0,054). This implies that the hypothesis H1b is rejected.

To determine whether the level of susceptibility to ingoing normative influence is significantly higher for collectivist consumers, we conducted the significance test. The outcome confirmed non significant differences between cultures (F = 2,456; df = 1; p = 0,118). Hence, the hypothesis H2a is only partly confirmed.

The results of the hypothesis H2b indicate that collectivist consumers have more outgoing normative influence than individualist consumers. Supporting the hypothesis H2b, the differences between collectivist and individualist culture had been found significant (F = 8,399; df = 1; p = 0,004).

4.3 Analysis of opinion leadership

(38)

38

Table 18: Kaiser-Meyer-Olkin measure and Bartlett's test scores

To continue with the analysis we had to decide how many factors to extract. Based on the criteria displayed in table 18, we were left with two and three factors solution for the collectivist culture, and one, two and three factors solution for individualist culture (appendices VI and VII). The results of individual criteria with regard to the number of factor to be retained are displayed below in table 19.

Table 19: Results for the number of factors to extract

Based on our results we moved forwards seeking a two factors solution. Next, we rotated the component matrix for collectivist and individualist culture using Varimax orthogonal rotation method. The rotated component matrices generated correlations that group into two dimensions for both cultures (table 20 and 21).

Table 20: Resulting dimensions for collectivist culture using Varimax orthogonal rotation

Table 21: Resulting dimensions for individualist culture using Varimax orthogonal rotation

(39)

39

analysis was conducted by the calculation of Cronbach’s alpha (table 22). The reliability analysis test internal consistency of the entire scale.

Table 22: The output of Cronbach’s alpha calculation

Because the factors in both analyses possess good Cronbach’s alpha, we can continue to compute the summated scale scores and further determine the final number of opinion leaders. The opinion leader statements (variables) are scored on a 7-point scale. Those statements were summed and then combined into a single variable (table 23).

Table 23: The summated scale scores of opinion leader variable across cultures

The summated scale scores of respondents were then used to find out the final number of opinion leaders. The data in our analysis were split out based on the ,,IF condition,, in SPSS software. The ,,IF condition,, selected a group that includes the respondents whose score was higher than value of 4 on the 7-point scale. Higher score makes it more likely that a respondent will be an opinion leader. After the respondents had been selected, we obtain the final number of opinion leaders in both cultures (table 24).

(40)

40

Given the data, it appears that there were 55 (22,18%) respondents in our datasets who were determined as opinion leaders. This means that 34 consumers in collectivist culture and 21 consumers in individualist culture are considered to be opinion leaders. To see whether significant differences occur across cultures, in relation to opinion leadership, an ANOVA had to be conducted (table 25). Opinion leadership is the dependent variable to be measured to find out whether culture, as an independent variable, has an impact on number of opinion leaders across cultures (appendix VIII).

Table 25: The results of an ANOVA for difference testing between cultures

The result of an ANOVA confirmed significant differences between cultures in relation to opinion leadership. Given the significance test results, we can compare opinion leaders across cultures (table 26). Based on comparative and representative nature of gender, this variable is used to define the basic differences of opinion leaders across cultures.

Table 26: Culture differences description in relation to opinion leadership

Opinion leaders comprise predominantly of males in both cultures. The higher gender distinction has been found in individualist culture. Collectivist culture possesses higher number of opinion leaders with slightly higher rank score. The higher the rank score, the higher the tendency to influence the purchase decisions of others as an opinion leader (the higher the probability to be an opinion leader).

4.3.1 Discussion

(41)

41

5. Conclusions

The purpose of our study was to determine the effect of social influence on consumer purchase behavior in two different cultures, the collectivist (Slovak) and the individualist (Dutch) culture. An introduction of our study is given in chapter one. The second chapter gives an overview of the literature and hypotheses. Chapter three gives a summary of the methodology applied in our study. We discussed in chapter four the findings of the data analyses to test our hypotheses. Then, the significance of our results across cultures was tested. The hypotheses developed in the second chapter were then verified with the results of our analyses.

Given the results five variables were examined: ingoing informational influence, outgoing informational influence, ingoing normative influence, outgoing normative influence and opinion leadership. Firstly, we examined an impact of social influence on consumer purchase behavior. The cross cultural comparison showed that collectivist consumers scored higher on all four variables of social influence. However the scores between collectivist and individualist culture varied minimally. This finding turned the attention to determine whether the variables of social influence differ significantly across cultures. An ANOVA confirmed significance of outgoing normative influence and marginal significance of outgoing informational influence cross cultures. Our results in this regard indicate that collectivist consumers are more engaged in processes of outgoing social influence indicating a preference for informational influence over normative influence. This would imply that consumers provide information rather than confirmation to fulfil the norms.

Concerning ingoing informational and ingoing normative influence no significant differences across cultures had been found. Even though the differences across cultures were no significant, the findings showed that ingoing informational influences have the greatest influence on consumers purchase decisions making in both cultures. This indicates that consumers prefer to build knowledge by requesting information from their social environment when making purchase decisions.

The result of ingoing normative influence was consistent with the study of Mourali et al. (2005) and Lee et al. (2008), which confirmed a less important role of normative influence in consumers purchase behaviors when compared with informational influence.

(42)

42

In general, consumers prefer to build knowledge by requesting information from their social environment rather than complying with expectations of others. This means that intention to follow norms to receive a reward or to avoid a punishment is no longer seen as the most relevant influencer when purchase decision making.

5.1 Managerial implications

According to the results of our analysis, social influences play an important role in consumer purchase behavior. Collectivist and individualist consumers are mostly susceptible to ingoing informational influences, when purchase decision making, with no difference between the two cultures. This would imply that collectivist consumers use others as a source of information to comparable extent as individualist consumers do.

Another important finding reveals no difference between cultures in purchase behavior regarding ingoing normative influence. Although consumers do not behave differently across cultures regarding ingoing normative influence, their level of susceptibility is much lower compared to ingoing informational influence. This result indicates that consumers in both cultures prefer to acquire information from others and build knowledge about a product rather than to buy a product to fulfil expectations of others.

Within the scope of outgoing social influence, consumers are more engaged with outgoing informational influence compare to outgoing normative influence, especially in collectivist culture.

When analyzing outgoing social influence, outgoing normative influence plays only a minor role in purchase behaviors. It could likely indicate that the more complex a product is, the less are consumers engaged with normative influence. Furthermore, we examined how differently consumers across cultures are engaged with outgoing social influence. The results proved that collectivist consumers have more outgoing normative influence than individualist consumers. If collectivist consumers possess higher engagement with outgoing informational and outgoing normative influence, they might have been more open to discuss and share information in a specific product category.

(43)

43

outgoing informational and outgoing normative influence. This would imply that the lower the degree of individualism in culture, the higher the number of opinion leaders and the higher the engagement with outgoing informational and outgoing normative influence. When it comes to product diffusion across cultures, the cognition of differences between individualist and collectivist culture is of the essence. It may enable to create appropriate marketing strategies to reach opinion leaders and other consumers (non-opinion leaders).

5.2 Limitations and further research

Although our research provided results that partially confirm the hypotheses, some limitations still exist. First, the limitation is given in the size of our sample. Large sample size can minimize the chances of overfitting the data, maximize the accuracy of data and increase generalizability of results (Hair et al., 2010; Janssens et al., 2008). Although Janssens et al.’s (2008) criterion to minimize the overfitting of data was not satisfied, the criterion of Hair et al. (2010) was met in our analysis. Despite the fact that our sample size was large enough to validate factor analysis, the analysis can suffer from lack of generalizability.

As for the structure of sample, the limitation is set by sending the questionnaires to consumers who are likely to have similar values. The group of predominantly young adults, full-time employed, having a bachelor’s or a master’s degree provides limited ability to accurately determine susceptibility to social influences among consumers across cultures. In other words, the research should cover wider range of different consumers in order to create some more heterogeneity within culture and more homogeneity between the two cultures. Large sample size could make the results of individualist and collectivist culture more comparable and representative.

The survey reveals that lots of consumers do not actually own a smartphone. In collectivist culture 42,4 % and in individualist culture 37,4 % respondents are not smartphone owners. A possible limitation of this research is the biased answers provided by consumers who for any reason do not own a smartphone. Thus, there is a possibility that consumers could not be able to respond like they would in an already experienced purchase situation.

(44)

44

An interesting opportunity for future research would be to investigate what sources of information affect purchase behaviors across cultures. It could efficiently help to reach consumers across cultures, what may enhance our understanding of personal and impersonal influences. To do so, researchers should analyse all possible sources of information in different cultures in an effort to focus on both groups of consumers, opinion leaders and non-opinion leaders.

(45)

45

References

Aggarwal, P., and C. Taihoon, (1997). Surrogate buyers and the new product adoption process: a conceptualization and managerial framework. Journal of Consumer Marketing, Vol. 14, Issue 4/5, pages 391-398.

Arnold, M.J., and K.E. Reynolds, (2003). Hedonic shopping motivations. Journal of Retailing, Vol. 79, Issue 2, pages 77-95.

Arnould, E. J., and C.J. Thompson, (2005). Consumer Culture Theory (CCT): Twenty Years of Research. Journal of Consumer Research, Vol. 31, Issue 4, pages 868-882.

Batra, R., P.M. Homer, and L.R. Kahle, (2001). Values, susceptibility to normative influence, and attribute importance weights: A nomological analysis. Journal of Consumer Psychology, Vol. 11, Issue 2, pages 115–128.

Bearden, W.O., R.G. Netemeyer, and J.E. Teel, (1989). Measurement of Consumer Susceptibility to Interpersonal Influence. Journal of Consumer Research, Vol. 15, Issue 4, pages 473-481.

Broekhuizen, T.J., S.A. Delre, and A. Torres, (2011). Simulating the Cinema Market: How Cross-Cultural Differences in Social Influence Explain Box Office Distributions. Journal of Product Innovation Management, Vol., 28 Issue 2, pages 204-217.

Burnkrant, R.E., and A. Cousineau, (1975). Informational and Normative Social Influence on Buyer Behavior. Journal of Consumer Research, Vol. 2, Issue 3, pages 206-215.

Cialdini, R.B., and N.J. Goldstein, (2004). SOCIAL INFLUENCE: Compliance and Conformity. Annual Review of Psychology, Vol. 55, Issue 1, pages 591-621.

Cosmas, S.S., and J. N. Sheth, (1980). Identification of opinion leaders across cultures: an assessment for use in the diffusion of innovations and ideas. Journal of International Business Studies, Vol. 11, Issue 1, pages 66-73.

Dawar, N., P.M., Parker, and L.J. Price, (1996). A cross-cultural study of interpersonal information exchange. Journal of International Business Studies, Vol. 27, Issue 3, pages 497–516.

De Mooij, M. (2011). Consumer behavior and culture: consequences for global marketing and advertising. Thousand Oaks, CA: Sage Publications, Inc.

De Mooij, M., and G. Hofstede (2010). The Hofstede model. International Journal of Advertising, Vol. 29, Issue 1, pages 85-110.

Social interactions as an influencer of consumers purchase behavior