What is the influence of individual level cultural values on consumer innovativeness, and what is the moderating effect of product involvement? a CVSCALE approach

Master Thesis

Tessa Jonker (11142510)

Under the supervision of

Frauke Mattison Thompson

MSc in Business Administration – Marketing Track Amsterdam Business School

University of Amsterdam June 23rd ₂₀₁₇

– A CVSCALE APPROACH

What is the influence of Individual Level Cultural

Values on Consumer Innovativeness, and what is

the moderating effect of Product Involvement?

1

STATEMENT OF ORIGINALITY

This document is written by Student Tessa Jonker who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

2 TABLE OF CONTENTS Statement of Originality ... 1 List of Tables ... 4 List of Figures ... 4 Abstract ... 5 Introduction ... 6 1. Literature Review ... 8 1.1. Culture ... 8

1.1.1. Hofstede’s dimensions of national culture ... 8

1.1.2. Individual level culture ... 9

1.1.3. CVSCALE ... 9

1.2. Consumer Innovativeness ... 10

1.2.1. The effect of culture on Consumer Innovativeness ... 10

1.2.2. Consumer Innovativeness: EAP scale ... 11

1.3. Product Involvement ... 11

1.3.1. Product Involvement ... 11

1.3.2. Product Involvement and consumers Innovativeness ... 13

2. Research Gap ... 14

2.1. Research question ... 14

2.2. Contribution to the literature ... 14

2.3. Managerial implications ... 15

3. Hypotheses & Conceptual Model ... 16

3.1. Hypotheses ... 16

3.1.1. Power Distance ... 16

3.1.2. Uncertainty Avoidance ... 17

3.1.3. Individualism/Collectivism ... 18

3.1.4. Masculinity/Femininity ... 18

3.1.5. Long-Term Orientation/Short-Term Orientation ... 20

3.2. Conceptual model ... 21

4. Methodology ... 23

4.1. Research Design ... 23

4.2. Sample & Data collection ... 23

4.3. Measures ... 24

3

4.3.2. Independent Variables: Individual Level Cultural Values ... 25

4.3.3. Moderating Variable: Product Involvement ... 25

4.3.4. Control Variables ... 26

5. Results ... 27

5.1. Preliminary analysis ... 27

5.1.1. Normality, Skewness & Kurtosis ... 28

5.1.2. Reliability Analysis ... 29

5.1.3. Factor Analysis ... 31

5.2. Independent T-Test ... 31

5.3. Correlation ... 32

5.4. Hypotheses testing ... 37

5.4.1. Hierarchical multiple regression ... 37

5.4.2. Moderation ... 38

5.5. Summary of the results ... 40

6. Discussion & Conclusion ... 41

6.1. Managerial Implications ... 43

6.2. Limitations & Future Research ... 44

6.3. Summary... 44

References ... 46

Appendix ... 49

Appendix A – Hypotheses prior research ... 49

Appendix B – Consumer Innovativeness: EAP Scale ... 50

Appendix C – Individual Level Cultural Values: CVSCALE... 51

Appendix D – Product Involvement: PII Scale ... 52

Appendix E – Survey ... 53

4

LIST OF TABLES

Table 1 – Correlation Matrix - Smartphone ... 34

Table 2 – Correlation Matrix - Toothpaste ... 35

Table 3 – Correlation Matrix - Total ... 36

Table 4 – Summary Hypotheses ... 40

Table 5 – Hypotheses prior research ... 49

Table 6 – EAP Scale ... 50

Table 7 - CVSCALE ... 51

Table 8 – PII Scale ... 52

Table 9 – Regression Results – Consumer Innovativeness (Smartphone) ... 57

Table 10 – Regression Results – Consumer Innovativeness (Total) ... 57

LIST OF FIGURES Figure 1 – Conceptual Model I ... 21

Figure 2 – Conceptual Model II ... 22

5

ABSTRACT

Researchers have identified that culture can have an effect on different aspects of consumer behaviour. Up until recently, culture has been looked at by comparing different nations and their behaviour. However, due to the increased spreading of cultures and therefore increasing heterogeneous societies, there is the need to measure culture at the individual level. This thesis focuses on measuring individual level cultural values, and determine if similar effects on a particular aspect of consumer behaviour can be found compared to when culture is measured nationally. The aspect of consumer behaviour that is focused upon is consumer innovativeness. For companies, it has become increasingly important to have knowledge about the likelihood of people to try out their products, as they rely increasingly on the success of new product innovations. Another concept that is also examined is whether or not the product involvement level of a person moderates the effect of individual level cultural values on consumer innovativeness. A survey was conducted, leading to a total of 399 respondents. Analysing the data showed that not a similar effect was found when measuring culture at the individual level compared to nationally. Only the Uncertainty Avoidance dimension of culture showed to have an effect on the innovativeness level of an individual. This finding contradicts earlier findings of researchers who took national culture indices as the predictors of their study. This is an important finding, as it shows that managers can’t simply rely on the national culture indices for predicting a person’s tendency to try out new product innovations.

Keywords: National Culture, Individual Level Cultural Values, CVSCALE, Power Distance, Uncertainty Avoidance, Individualism/Collectivism, Masculinity/Femininity, Long-Term Orientation/Short-Term Orientation, Product Involvement, Consumer Innovativeness

6

INTRODUCTION

There has been a lot of research into the effect that culture has on different aspects of consumer behaviour. For example, there have been studies looking into the effect of culture on brand perception (Foscht, Maloles III, Swoboda, Morschett, & Sinha, 2008), impulsive buying behaviour (Kacen & Lee, 2002) and the adoption of innovations (Steenkamp, Hostede, & Wedel, 1999; Van Everdingen & Waarts, 2003). However, up until recently, most research has focused on the effect of national culture. National culture has been studied extensively by Hofstede (1980, 2001) and he divided national culture into five different dimensions. The dimensions are used in a way in which nations receive a score based on if they score high or low on a certain dimension. For example, Western countries score relatively high in individuality, whereas Eastern cultures are more collectivistic.

Since firms need to deal with different cultural environments, as they often operate across nations, it is important for them to understand what specific influence culture can have on the behaviour of their consumers in order for firms to anticipate on these varying cultures and play into the demands of their consumers. Therefore, Hofstede’s dimensions (Hofstede, 2001) have been used to look at the effect of all cultural dimensions on various consumer behaviour aspects.

However, according to Yoo, Donthu, & Lenartowicz (2011), it might not always be the best approach to look at national culture and its effect on consumer behaviour. They argue that Hofstede’s method has been used in such a way that individual people are equally assigned Hofstede’s national culture indices just because of what nation they are from (Yoo et al., 2011). This leads to for example all Chinese being treated as collectivists and all Americans as individualists, while it would be perfectly possible that, when looked at individually, a Chinese person is more of an individualist and an American person has a collectivist cultural orientation. Therefore, to avoid this generalization by nation, Yoo et al. (2011) developed the Individual Cultural Value Scale (CVSCALE).

The CVSCALE is a measure which gives individuals a score on all of Hofstede’s dimensions (Yoo et al., 2011) and can be used to see what effect individual level cultural orientation can have on consumer behaviour as well as what this influence looks like (what dimensions lead to what behaviour).

One aspect of consumer behaviour that has been examined when trying to depict the effect of national culture, is consumer innovativeness (Steenkamp et al., 1999; Van Everdingen & Waarts, 2003). Consumer innovativeness refers to the tendency of people to try out new products or product innovations. According to these studies regarding the effect of national culture on consumer innovativeness, the fact that a nation scores high or low on certain dimensions has a significant impact

7 on consumers’ likelihood to adopt new product innovations (Steenkamp et al., 1999; Van Everdingen & Waarts, 2003). This is very important for firms to know, as they need to determine where to introduce their new innovations, and would thus like to know where it will most likely be adopted by their consumers.

However, as explained it might not be truthful to generalize an entire nation to a certain score on Hofstede’s dimensions, as many inhabitants might score entirely different when looked at individually. In the case of product adoptions, this could lead to firms targeting a nation because its national culture dimensions indicate that the inhabitants are likely to try out new product innovations. This, while in fact many people living within that country could be, when looked at individually, hesitant to try out new products. Therefore, it is a lot more relevant to look at the effect of individual level cultural values on consumer innovativeness.

Therefore, this research will use the CVSCALE to measure people’s individual level cultural values and determine what the effect is on consumer innovativeness. This research can contribute to existing literature by determining if similar findings can be made when culture is measured at the individual level instead of by national culture indices. It will provide insights or contribute to existing literature both if individual level culture has a similar effect on the innovativeness of consumers as well as when it doesn’t, as it will then show how individual level cultural values affects consumers’ adoption of innovations and how this differs from the effect of national culture. It will also contribute to existing literature on individual level cultural values, as it will add construct to the CVSCALE, and at the same time add to knowledge regarding consumers’ innovativeness as it will show what effect the culture of individuals can have, which as mentioned is very important for companies to have insights into. It is expected that the dimensions of culture have a similar effect on consumer innovativeness when looking at individual level cultural values as opposed to national culture.

This thesis further contains a literature review, discussing existing research in this area and all the key aspects of this research, the research gap and proposed research question, hypotheses and conceptual framework, methodology, results and finally a discussion and conclusion.

8

1. LITERATURE REVIEW

Due to the world becoming ever more globalized, with firms often operating across several countries, there is the need to research the effect that culture has on consumer behaviour. Particularly because firms that operate across several countries, need to deal with different cultural environments, and these varying cultures can lead to differences in consumer behaviour. These differences are important to study for marketers to anticipate on these various cultures and the differences in consumer behaviour because of it.

1.1. CULTURE

Culture is defined as ‘the collective programming of the mind that distinguishes the members of one

group or category of people from others’ (Hofstede, 2011, p. 3) and the term is most commonly used

for ethnic groups, nations and for organizations (Hofstede, 2011). Hofstede (2001) developed a five-dimensional measure of culture that is used to compare the impact of culture across nations. This measure has been widely accepted and used, and has contributed a lot to cross-cultural research.

1.1.1. HOFSTEDE’S DIMENSIONS OF NATIONAL CULTURE

Hofstede (2001) identified and labelled the following five dimensions of national culture: Power Distance, Individualism/Collectivism, Masculinity/Femininity, Uncertainty Avoidance and Long-/Short-Term Orientation.

Hofstede (2011) explained his dimensions in a nutshell, which includes the following definitions: Power Distance refers to the extent to which the less powerful individuals in an organization or institution accept and expect an unequal distribution of power. Uncertainty Avoidance refers to a society’s tolerance for uncertainty or ambiguity. Individualism/Collectivism is the degree to which people in a society are integrated into groups (expected to look after himself/herself vs. share close ties). Masculinity/Femininity refers to the distribution of values between the genders. Long-/Short-Term Orientation is the degree to which societies have a long-term focus (e.g. building relationships) or focus more on short-term results.

The dimensions have been used to explain variance in consumer behaviour, such as brand perception Foscht et al. (2008), online behaviour (Chau, Cole, Massey, Montoya-Weiss, & O’Keefe, 2002), impulsive buying behaviour (Kacen & Lee, 2002) and consumer innovativeness (Steenkamp et al., 1999; Van Everdingen & Waarts, 2003). These studies show that national culture can have a significant impact on consumer behaviour.

9 However, marketers in this day and age are confronted with increasingly multicultural marketplaces, due to for example globalization, increased mobility and the development of technological advancements, such as the Internet. Increased mobility, of for example labour, could lead to people from a certain nation to work in another, and coming into contact with different cultures. This could in turn lead to the diffusion of cultures (mixing of cultural values). The Internet also is a way in which different cultures come into contact with each other, and are transferred. All in all, therefore, the world is becoming increasingly cross-cultural. According to Yoo et al., (2011), it may not work to look blindly at national culture anymore and using that to target individual consumers, because as they say ‘national boundaries do not always characterize the cultural values of people’ (Yoo & Donthu, 2002, p. 94).

1.1.2. INDIVIDUAL LEVEL CULTURE

Yoo et al. (2011) suggest that in managerial situations, the reflection of culture at the individual level is more important and relevant. Hofstede’s metric has been used in such a way that individuals are assigned Hofstede’s national culture indices by their national identity (Yoo et al., 2011), so for example treating all members of one country as collectivists. This method is acceptable when the unit of analysis is a country, and has generated a lot of insights into what effect culture can have on various consumer behaviour aspects, but it is not appropriate when a study looks into the effect of an individual’s cultural values (Yoo et al., 2011). Using Hofstede’s metrics and assigning individuals a score based on what nation they are from, would be generalizing that all individuals from a certain country have the same individual cultural orientation, whereas it is perfectly possible that a person from a mostly collectivist nation, when looked at individually, is more of an individualist.

1.1.3. CVSCALE

Especially in the increasingly heterogeneous societies, this generalization by nation might not be appropriate anymore. Therefore, these national indices can’t be generalized to all inhabitants in the same nation, thus it is beneficial to have an alternative measure of culture at the individual level. Therefore, Yoo et al. (2011) developed the CVSCALE: Individual Cultural Value Scale. This scale assesses Hofstede’s cultural dimensions at the individual level.

According to Yoo et al. (2011) managers, especially those dealing with global products and brands, will benefit from the scale because they can find equivalent market segments across countries based on consumer segments showing similar cultural values, and then they can apply a similar strategy to those consumer segments in different countries. This means that they can specifically target for example individualistic and masculine-oriented consumer segments across different countries instead of just

10 identifying and targeting a couple of countries that meet this profile specifically. Thus, they can identify a consumer segment matching their product based on individual level cultural values, and develop a strategy to target this segment worldwide if they wanted. In a world with cultures spreading out and mixing (otherwise known as cultural diffusion), due to globalization and technological advancements, leading to more heterogeneous markets, this individual targeting would make more sense than solely targeting countries that fit the profile culturally the best.

Yoo et al. (2011) suggested that future research should keep validating their measure, for example using larger samples or demographically diverse samples, but more importantly, meaningful research would be to test prominent theories of the influence of national culture at the individual level instead (Earley, 2006; as cited in Yoo et al., 2011). As they argue, ‘a convenient and productive start will be to

replicate existing studies to see if similar findings can be made when culture is measured at the individual level’ (Yoo 2009; as cited in Yoo et al., 2001, p. 206).

There have been several researchers who have also noticed the importance and adopted the individual perspective rather than the national culture approach. Cao & Phan (2015) looked into the effect of individual level culture on service quality expectations, Patterson, Cowley, & Prasongsukarn (2006) examined the effect of individual culture on perceptions of justice (in service failure recovery) and Thompson, Newman, & Liu (2014) researched the effect of individual level collectivist values on brand loyalty. However, there has been no research into the effect of individual level cultural values on consumers’ adoption of innovations yet.

1.2. CONSUMER INNOVATIVENESS

Consumer innovativeness is defined by Steenkamp et al. (1999) as ‘the predisposition to buy new and

different products and brands rather than re- main with previous choices and consumption patterns’ (p. 56).

Companies rely increasingly on the success of new product introductions, making this a very relevant area of consumer behaviour to focus upon (Steenkamp et al., 1999; Van Everdingen & Waarts, 2003).

1.2.1. THE EFFECT OF CULTURE ON CONSUMER INNOVATIVENESS

Steenkamp et al. (1999) studied the effect of three dimensions of national culture on consumers, which were Individualism/Collectivism, Uncertainty Avoidance and Masculinity/Femininity, on the innovativeness of consumers. Hofstede’s dimension indices were used for the countries studied in this research, although the authors used an updated version of these national culture values which are provided by Hoppe (1990; as cited in Steenkamp et al., 1999).

11 Steenkamp et al. (1999) hypothesised that the higher the Individualism or the higher the Masculinity in a country, the higher the level of innovativeness, but a negative relationship between Uncertainty Avoidance and innovativeness (thus the higher the Uncertainty Avoidance, the lower the innovativeness). They found support for these three hypotheses.

Besides the study by Steenkamp et al. (1999), also De Mooij & Hofstede (2011) reported in their research the link between national culture and the adoption of innovations, and that they correlate negatively with Uncertainty Avoidance, and positively with Individualism.

Van Everdingen & Waarts (2003) on the other hand noticed that in prior studies not all dimensions of Hofstede were examined, so they looked at the influence of all Hofstede’s dimensions on innovation (adoption of a new IT system).

These studies show that national culture indeed has an effect on consumers’ adoption of product innovation, but there has been no research into the effect of individual level cultural values on consumers’ adoption of innovations yet.

1.2.2. CONSUMER INNOVATIVENESS: EAP SCALE

Consumer innovativeness in the above described research was determined by using the consumer-specific Exploratory Acquisition of Products (EAP) scale (Baumgartner & Steenkamp, 1996; as cited in Steenkamp et al., 1999). This is a 10-item scale that ask questions to individuals to determine a person’s willingness to try new/unfamiliar product innovations. However, in this study a shortened version with the highest loading questions were used, leading to a 5-scale item (Steenkamp et al., 1999).

1.3. PRODUCT INVOLVEMENT

1.3.1. PRODUCT INVOLVEMENT

Besides cultural influences, there are other factors that might determine an individual’s likeliness to adopt or try out new product innovations. One factor that is important to consider is an individual’s product involvement.

Product involvement relates to the importance that consumers attach to the product and their level of interest in it. With some product types, consumers might be highly involved, for example if it is a product that is very relevant to them and something they would often use. Also, it could be when it concerns something that is expensive so they want to search a lot of information about it before making an actual purchase decision, because they do not want to make a bad choice. Other product

12 types however, consumers might be less involved in, possibly because it is not a relevant product for them, or it is a product that requires little information search before buying.

Laurent & Kapferer (1985) have developed a measure for consumer product involvement. As can be seen from this measure, there are several aspects that contribute to an individual’s level of involvement. Laurent & Kapferer (1985) indicate five different aspects, which include a person’s interest in a product (its personal meaning or importance), the hedonic value of the product (its ability to provide pleasure), the sign value of the product (the degree to which it expresses a person’s self) or the perceived risk related to a product purchase (importance of potential negative consequences associated with a poor choice as well as the probability of making such a poor choice). Other researchers agree with this perspective and also describe product involvement as a multidimensional and product-specific construct (Dholakia, 1997).

So, according to Laurent & Kapferer (1985), one aspect that is closely related to product involvement is the risk construct, where they distinguish between two aspects of risk in their measure, namely risk importance and risk probability. People that are highly involved are in part that involved due to the perceived risk that they have or experience when buying that particular product, and the importance of the consequences of making a wrong purchase. Because the risks of a bad purchase are high when the price of the product is high, consumers are likely to be involved, so price can serve as an indicator of involvement (Rothshild, 1979; as cited in Laurent & Kapferer, 1985).

Dholakia (1997) has looked into the relationship between risk and product involvement, and mentions many similarities between the two constructs, but also some differences. Both constructs incorporate the notion of ‘importance’ of a product class to a consumer, and both high levels of risk and involvement are known to result in more extensive information search by the consumer (Celsi & Olsen, 1988; as cited in Dholakia, 1997). On the other hand, risk looks more at the negative consequences arising from a bad purchase decision while involvement is also affected by positive consequences (such as the hedonic value mentioned earlier). Related to this, is the discussion whether perceived risk should be treated as an antecedent of involvement (Bloch, 1981), as one of its dimensions or building blocks (Laurent & Kapferer, 1985), or as its consequence. What is clear however, is that the two constructs are closely related, and that a high perceived risk often corresponds with a high level of involvement (even though other factors also influence the level of involvement).

Besides Laurent & Kapferer (1985), also Zaichkowsky (1985, 1994) developed a scale to measure the involvement construct, which was named the Personal Involvement Inventory (PII). The scale was developed since there was little agreement as to how the construct should be measured, even though researchers did agree that studies of low versus high involvement were interesting and important.

13 Zaichkowsky (1985, 1994) therefore, developed a semantic differential scale, first consisting of 20 items but after being revised a 10-item scale, which can be used to measure a consumers’ product involvement.

1.3.2. PRODUCT INVOLVEMENT AND CONSUMERS INNOVATIVENESS

According to both Bauer, Sauer, & Becker (2006) and Lin & Chen (2006), product involvement has an effect on consumer decision making or consumer purchase decisions. However, what does this mean for consumer innovativeness?

As mentioned, consumer innovativeness has to do with an individual’s likelihood or predisposition to buy new and different products rather than remain with previous choices. However, choosing to buy a different type of product is also a form of a consumer purchase decision. This means that consumer innovativeness might also be influenced by the level of involvement of a person in the new/different product that they are buying. Therefore, product involvement is an important factor to consider when looking at consumer innovativeness.

14

2. RESEARCH GAP

The prior described study by Steenkamp et al. (1999), as well as others on the same topic (De Mooij & Hofstede, 2011; Van Everdingen & Waarts, 2003), clearly shows that national culture has an effect on the adoption of innovations. Meaning that the different cultural dimensions have either a positive or negative effect on the adoption of innovations. However, as previously indicated, it might be more relevant and important to look at individual level cultural values instead of national culture, due to globalization and the spreading of cultures, and thus increasingly heterogeneous societies in terms of culture. This (consumer innovativeness) is an area of consumer behaviour of which the effect of individual level cultural values on it has not been studied yet, but as explained before, is becoming increasingly important for companies to have knowledge about as they rely increasingly on the success of new product innovations (Steenkamp et al., 1999; Van Everdingen & Waarts, 2003).

Furthermore, there might be another factor that also affects consumer innovativeness, and that is consumers’ product involvement, so it is important to consider this factor as well.

2.1. RESEARCH QUESTION

Thus, this research will focus on the effect of individual level cultural values on consumer innovativeness, by using the CVSCALE proposed by Yoo et al. (2011), and will examine the possible moderating effect of product involvement. Therefore, the research question of this thesis will be as follows:

What is the influence of Individual Level Cultural Values on Consumer Innovativeness, and what is the moderating effect of Product Involvement?

2.2. CONTRIBUTION TO THE LITERATURE

This research contributes to existing literature by determining if similar findings can be made when culture is measured at the individual level. If so, it is determined that individual level culture has a similar effect on the innovativeness of consumers as national culture, and firms can act upon this knowledge, for example by finding segments with individuals globally that meet the criteria in terms of individual level cultural values for adopting innovative products, and base their strategy on this. If not a similar effect is found, it can provide important insights in how individual level cultural orientation effects consumer innovativeness, and how this differs from the effect of national culture found in previous research.

15 It contributes both to research on individual level cultural values, by adding construct to the CVSCALE, and to research into consumer innovativeness, by showing the effect of individual level culture on it. Furthermore, it adds to existing research regarding product involvement, as the possible influence of product involvement on consumer innovativeness is examined.

2.3. MANAGERIAL IMPLICATIONS

Besides contributing to literature about individual level cultural values and consumer innovativeness, the results of this research have important managerial implications. As mentioned, firms are increasingly relying on the success of product innovations, and thus the adoption of those (Steenkamp et al., 1999; Van Everdingen & Waarts, 2003). Sometimes, companies even have it written in their policies that a certain percentage of their sales must come from new products introduced in the last couple of years. Therefore, managers will benefit from gaining knowledge regarding what aspects influence the adoption of their innovations.

As explained before, managers might be able to find equivalent market segments across countries based on consumer segments showing similar cultural values, and then they can apply a similar strategy to those consumer segments in different countries. This means that they can specifically target for example individualistic and masculine-oriented consumer segments, if those cultural values for example suggest high levels of consumer innovativeness, across different countries instead of just identifying and targeting a couple of countries that meet this profile specifically.

16

3. HYPOTHESES & CONCEPTUAL MODEL

In the following chapter, I build upon the literature to formulate the hypotheses of this research. The hypothesized relationships are then illustrated in a conceptual model (Figure 1 & Figure 2).

3.1. HYPOTHESES

The hypotheses of this research were formed partially based on several prior research studies (Appendix A/Table 5) regarding both the effect of (national) culture on consumer innovativeness and research on the influence of product involvement, and will be explained one by one in the next section.

3.1.1. POWER DISTANCE

First of all, is the Power Distance dimension, which constitutes one of the Independent Variables in this study.

The Power Distance dimension refers to the extent to which a member accepts and or expects there to be an unequal distribution of power. So, when Power Distance is high, there is an expectation and acceptance of unequal distribution of power. Power Distance is related to conservatism and maintaining a certain status quo (Steenkamp, 2001; as cited in Yeniyurt & Townsend, 2003) and people are therefore possibly less open to new ideas and products, because they accept the way things are and that there is a certain level of inequality. According to Van Everdingen & Waarts (2003) high levels of centralization/formalization (so high Power Distance and hierarchical structures) are associated with low levels of innovation, so agree with Yeniyurt & Townsend (2003) on the influence of Power Distance on innovation.

Hence, I suggest the following hypothesis:

H1a. There is a negative relationship between Power Distance and Consumer Innovativeness. Thus, the higher the Power Distance, the lower the Consumer’s Innovativeness.

Besides the direct effect of individual level cultural values on consumer innovativeness, also the moderating effect of product involvement will be examined.

As importance and involvement are strongly related or even build up by risk perceptions, as explained before, higher product involvement often is associated with bigger risks.

Thus, the effect that Power Distance has on consumer innovativeness (negative relationship) is expected to be magnified by product involvement, so that an even lower level of consumer innovativeness is shown with higher levels of product involvement. This since with higher levels of

17 product involvement, there is often a higher level of perceived risk involved, and the individuals that are already aversive of trying out new product innovations, thus individuals with high Power Distance levels, will be even less likely to try out new products.

Hence, leading to the following hypothesis:

H1b. The relationship between Power Distance and Consumer Innovativeness is moderated by Product Involvement, so that this relationship is stronger for those consumers high in product involvement compared to those consumers low in product involvement.

3.1.2. UNCERTAINTY AVOIDANCE

Secondly, the next Independent Variable includes the Uncertainty Avoidance culture dimension. If people score low on Uncertainty Avoidance, this means that they accept that there is uncertainty in life, and they are comfortable with a little ambiguity so they do not strongly avoid it. Whereas if a person scores high on Uncertainty Avoidance, they are intolerant for uncertainty, and often show intolerance for people or ideas that are unlike what is normal to them (Hofstede, 2011). This leads to the reasoning that people who score high in Uncertainty Avoidance are hesitant to try out new products, as they are intolerant for deviant products or ideas, so they rather avoid it.

Thus, this leads to the hypothesis:

H2a. There is a negative relationship between Uncertainty Avoidance and Consumer Innovativeness. Thus, the higher the Uncertainty Avoidance, the lower the Consumer’s Innovativeness.

Product involvement could magnify the relationship between Uncertainty Avoidance and consumer innovativeness, so that an even lower level of consumer innovativeness is shown for people with higher values of product involvement. This again since the higher the involvement, the more risk is (perceived to be) involved with a particular product, and risk aversive people will be even less likely to try out new products if there is a higher perceived risk. So, people who are higher in Uncertainty Avoidance will even be less likely to try out new products if the product is perceived to be involving and thus risky.

This leads to the following hypothesis:

H2b. The relationship between Uncertainty Avoidance and Consumer Innovativeness is moderated by Product Involvement, so that this relationship is stronger for those consumers high in product involvement compared to those consumers low in product involvement.

18

3.1.3. INDIVIDUALISM/COLLECTIVISM

The third Independent Variable in this research is Individualism/Collectivism. This cultural dimension looks at the extent to which people are integrated into groups; so are either focused on the individual, or on the collective. Collectivistic people are characterized by collective decisions, so conform the norms of the group. On the other hand, individualistic people feel like they can make their own decisions. Hence, people who score high on Individualism, would be more prone to try out an innovation, as they can decide more easily for themselves if they want to or not, whereas collectivist people might stick more to what the rest of the group does and will stay within the confinements of their group.

Many researchers agree on this positive effect, like Van Everdingen & Waarts (2003), who state that for example individuals in organizations in individualistic countries have more freedom to develop or try new products than in collectivist countries, or Yeniyurt & Townsend (2003), who state that previous research results indicate that highly individualistic countries have higher coefficients of innovations (Yaveroglu & Donthu, 2002; as cited in Yeniyurt & Townsend, 2003).

Therefore, in contrast with Power Distance and Uncertainty Avoidance, Individualism is expected to have a positive effect on consumer innovativeness:

H3a. There is a positive relationship between Individualism and Consumer Innovativeness. Thus, the higher the Individualism, the higher a Consumer’s Innovativeness.

So, as mentioned, I theorize that the higher the Individualism, the higher the consumer innovativeness, or in other words, the more likely they are to try out new products. Product involvement can have a moderating effect on this relationship, in a way that that individualistic people will be less likely to be innovative when they perceive the risk too high for this.

This leads to the following hypothesis regarding the moderating effect of product involvement: H3b. The relationship between Individualism and Consumer Innovativeness is moderated by Product Involvement, so that this relationship is weaker for those consumers high in product involvement compared to those consumers low in product involvement.

3.1.4. MASCULINITY/FEMININITY

Fourthly, there is the Masculinity/Femininity dimension which will be another Independent Variable in this research. The effect of Masculinity/Femininity is mostly argued by the characteristics that are derived from being either masculine or feminine. According to Hofstede (2011), masculine values

19 include being assertive and ambitious, while feminine values stands for being caring and modest. Reasoning from this point of view, results in the view that with high scores of Masculinity, people are more ambitious and therefore probably more prone to trying out a new product, whereas high scores of Femininity stand for modesty and more reserved characteristics, and would make people more hesitant to try out new things. So, high scores of Masculinity would result in higher levels of consumer innovativeness.

Researchers that have looked at the effect of this dimension, but then at the national level, have all had the same hypothesis regarding this dimension, and expected an effect that is in line with the above described reasoning. They thus all expected a positive relationship between Masculinity and innovativeness.

Even though they all shared the same direction in their hypothesis (positive effect), not all shared the same results. Singh (2006) hypothesized that in more masculine cultures there will be a higher propensity to innovate, based on the character traits that belong to Masculinity, and found support for his hypothesis. Steenkamp et al. (1999) as well hypothesized the relationship to be positive, and also found support for this. However, Yeniyurt & Townsend (2003) also thought there would be a positive effect, but could not confirm this in their research. Whereas on the other hand Van Everdingen & Waarts (2003) argued that there would be a positive relationship, but actually found a negative influence to be true.

However, following the above explained reasoning, and in line with the hypothesised effect of all researchers, still a positive relationship is expected. It could contribute to existing literature if clear support can be found for this.

Hence, this leads to the following hypothesis.

H4a. There is a positive relationship between Masculinity and Consumer Innovativeness. Thus, the more masculine, the higher consumers’ adoption of innovations.

Also, this direct effect could possibly be moderated by the level of product involvement of consumers. Masculine people might be more ambitious to try out new products than feminine consumers, but this effect could be affected by the level of product involvement, as higher involvement levels are related to higher levels of perceived risk. So even though masculine consumers might be more prone to try out new product innovations, this effect could be expected to be weaker for high levels of product involvement.

20 H4b. The relationship between Masculinity and Consumer Innovativeness is moderated by Product Involvement, so that this relationship is weaker for those consumers high in product involvement compared to those consumers low in product involvement.

3.1.5. LONG-TERM ORIENTATION/SHORT-TERM ORIENTATION

One of the dimensions of which the effect of it on consumer innovativeness has not been studied often, is Long-/Short-Term Orientation. This is also the dimension that was most recently defined, compared to the first four, by Hofstede (2001).

However, Van Everdingen & Waarts (2003) saw that many studies left out one or several dimensions, and saw this as a flaw, so they did look into the effect of Long-Term Orientation on innovation. According to them, cultures with a long-term orientation are characterized by values like persistence, adaptations of traditions to new circumstances, and personal adaptability (Van Everdingen & Waarts, 2003). All these values are asked for when having a long-term focus in life. These values are in line with being adaptive, and this can lead to innovative behaviour. These cultures are open to adapting to new circumstances, and possibly therefore also new ideas and products.

On the other hand, Short-Term Orientation is characterized by values like respecting tradition, and the focus is more on what happened in the past and what happens on the short-term. Therefore, such cultures are expected to be less innovative.

This leads to the fifth hypothesis:

H5a. There is a positive relationship between Long-Term Orientation and Consumer Innovativeness. Thus, the more someone has a Long-Term Orientation, the higher the Consumer’s Innovativeness

However, the product involvement variable could have an effect on the above described relationship. Long-Term Oriented people might be more adaptive to new circumstances, and therefore be more innovative than Short-Term Oriented people who are more adherent to their traditions, but they might also be more considerate of the possible risks involved in the long run. Therefore, with high levels of product involvement, and subsequently often high levels of perceived risks, this direct effect between Long-Term Orientation and consumer innovativeness might be less strong.

21 H5b. The relationship between Long-Term Orientation and Consumer Innovativeness is moderated by Product Involvement, so that this relationship is weaker for those consumers high in product involvement compared to those consumers low in product involvement. 3.2. CONCEPTUAL MODEL

As mentioned, the conceptual model in figure 1 and 2 represents the concepts investigated in this research and shows the hypothesized relationships between the Independent Variables (which constitute the cultural dimensions) and consumer innovativeness, and the moderating effect of product involvement.

The Independent Variables in this research represent the cultural dimensions proposed by Hofstede (2001), but then measured at the individual level, instead of nationally, by means of the CVSCALE (Yoo et al., 2011). These dimensions include the following: Power Distance, Uncertainty Avoidance, Individualism/Collectivism, Masculinity/Femininity, Uncertainty Avoidance and Long-/Short-Term Orientation.

As this research will look into the effect of those dimensions on consumer innovativeness, that will be the Dependent Variable of this research. The figure below (Figure 1) shows these variables and their relationships, along with the hypothesized effects.

22 However, there is another factor that is important to consider with an individual’s likeliness to adopt or try out a new product innovation, and that is product involvement.

Product involvement relates to the importance that consumers attach to the product and their level of interest in it, and can range from low to high involvement. As mentioned, according to (Bauer, Sauer, & Becker, 2006; Lin & Chen, 2006) product involvement has a significant effect on consumer purchase decisions, thus this could indicate that it might also have an influence on consumer innovativeness, as choosing to try out a new product could also be seen as a purchase decision.

Therefore, in this research, the possible moderating effect of product involvement will be examined. This is visualised in Figure 2. As shown in the figure, the Independent Variables are all the individual level cultural value dimensions and the Dependent Variable is consumer innovativeness, but the moderating effect of product involvement is now included in the conceptual model.

23

4. METHODOLOGY

This chapter will address the research design of this study. This includes an explanation of what kind of research approach was used, what kind of variables were being measured and the sample and data collection.

4.1. RESEARCH DESIGN

This study has a deductive research approach, thus it will test theoretical propositions (the hypotheses specified in chapter 3) by using a research strategy that is specifically designed for the purpose of testing this, and will be explanatory in nature as it will depict the relationships between variables. It was executed in a cross-sectional collection method through a self-administered online survey. For this research, a quantitative approach, by means of surveys, has been used to test all the variables of this research (the cultural dimensions, product involvement and consumer innovativeness). Surveys were best suited to test the variables that were intended to be measured, which will be explained within this chapter.

4.2. SAMPLE & DATA COLLECTION

The sampling technique that was used is non-probability sampling, specifically convenience sampling. There are not many criteria for respondents for this research, mostly because the cultural values and its effect on innovativeness is actually intended to be measured individually, so therefore convenience sampling is a sufficient method.

Since the variables are intended to be measured individually, there is not a specific population aimed for, as theoretically all people could qualify to be included in this research. However, in order to investigate if culture influences consumer innovativeness, there has to be enough variance in the cultural dimensions as well as the levels of consumer innovativeness. Therefore, the survey has been spread foremost amongst students living and studying in Amsterdam. Students in Amsterdam, for example at the University of Amsterdam, come from various cultural backgrounds and various countries, but have at the same time been in contact with the Dutch culture because they have been living and studying here, so cultural diffusion might have taken place (in various degrees). This should result in a heterogenous sample in terms of cultural values, and the effect of these different cultural values on consumers’ innovativeness can be studied.

The survey has been distributed in the Netherlands, via university email and social media. As the survey will foremost be distributed amongst Dutch and International students, the survey will be in English.

24 Most Dutch and International students in the Netherlands have an adequate level of English (as they often even study in English), and this will avoid translation bias of the survey.

The survey has been pre-tested to ensure that the respondents clearly understand the questions, and that there are no problems with filling in the survey. As there were no alterations to be made, the survey was then distributed after the pre-test.

After deleting respondents who did not fill in the survey completely (due to several partially complete surveys that were recorded in Qualtrics after a certain time limit of the survey being open to be filled in), a total number of 399 respondents remained.

4.3. MEASURES

4.3.1. DEPENDENT VARIABLE: CONSUMER INNOVATIVENESS (EAP SCALE)

The Dependent Variable is consumer innovativeness, so the consumers’ level of innovativeness had to be determined. In order to do that, the EAP (exploratory product acquisition) scale of Baumgartner & Steenkamp (1996) was used. This EAP scale is a 10-item scale determining how high people score on enjoying taking chances in buying products they are unfamiliar to (hence the name exploratory product acquisition, EAP).

Roehrich (2004) looked into consumer innovativeness and the different concepts and measurements for it. According to Roehrich (2004), there are many consumer innovativeness scales, but the EAP scale is one of the most representative scales. Also, the measure provides average predictive validity for innovative behaviour.

The EAP scale shows good reliability with alpha coefficients ranging from .75 to .84 across different samples (Bearden & Netemeyer, 2011).

The respondents however, will be asked questions in two different contexts. Half of the respondents will answer the questions about smartphones and the other half about toothpaste. This is to ensure that there will be varying levels of involvement in the responses. This will be explained in detail in chapter 4.3.3. However, since the respondents are asked questions in two different contexts, toothpaste and smartphone, some questions of the EAP scale do not work, as they are regarding another purchase context (restaurants). Therefore, these questions are omitted and this resulted in a total of 7 items asked to determine consumer innovativeness.

However, Steenkamp et al. (1999) also successfully used a shortened version (5-items instead of 10-items) with the highest loading questions from the EAP scale in their study. So, not all items of the EAP

25 scale had to be used to get good results in their study. This indicates that there should be no problem with only using 7 out of the 10 items of the EAP scale. Thus, items with the highest loading factors without being about a specific purchase context were used. The total EAP scale with indication of which items were omitted can be found in the appendix (Appendix B/Table 6). When testing the data from the survey, the reliability analysis shows the reliability of the scale items used, which is discussed in chapter 5.

4.3.2. INDEPENDENT VARIABLES: INDIVIDUAL LEVEL CULTURAL VALUES

Furthermore, the Independent Variables of this study had to be measured. As explained, the Independent Variables consist of the cultural dimensions of Hofstede (2011), but then measured at the individual level by means of the CVSCALE, which was developed by Yoo et al. (2011). This scale is an individual representation of Hofstede’s dimensions, and gives individuals a score for each of those dimensions based on certain questions. These questions have been determined and tested by Yoo et al. (2011). The questions were asked on a 5 point Likert scale, and were thus easily represented in a survey data collection method. The items (questions of the CVSCALE) that have been asked can be found in the appendix (Appendix C/Table 7), sorted by each dimension separately.

The reliability of the CVSCALE scale has been tested in various countries by the developers of the scale themselves by computing the Cronbach’s Alpha (Yoo et al., 2011). However, also other researchers have looked at the reliability of the scale, like Prasongsukarn (2009). Prasongsukarn (2009) aimed with his research to validate the CVSCALE in Thailand, and scale reliability was assessed by using the Cronbach’s Alpha coefficient. It was computed to test the reliability of each dimension. The measure showed to achieve satisfactory levels of reliability as all yielded a value of more than 0.6: the Cronbach Alpha was .63 for Power Distance, .81 for Uncertainty Avoidance, .81 for Individualism/Collectivism, .61 for Masculinity/Femininity and .85 for Long-Term Orientation/Short-Term Orientation (Prasongsukarn, 2009).

As there are five culture dimensions, there are five Independent Variables in this research (of which the effect will be tested on the level of consumer innovativeness), as was shown in the conceptual model in chapter 3 (Figure 1 & Figure 2).

4.3.3. MODERATING VARIABLE: PRODUCT INVOLVEMENT

However, as pointed out in the literature review, it might be possible that consumers show different levels of consumer innovativeness, depending on the level of involvement in the product. Therefore, product involvement has been measured.

26 Product involvement was measured by means of the Product Involvement Inventory scale (PII), developed by Zaichkowsky (1985) as a 20-item scale, but later on revised to a 10-item scale (Zaichkowsky, 1994). This revised scale was used to determine the level of product involvement of the respondents of the survey, and can be found in the appendix (Appendix D/Table 8).

As I hypothesize different effects of the Independent Variables on the Dependent variable, depending on either a high or a low level of product involvement, respondents of this survey answered questions about two different products. Half of the respondents answered the questions about product involvement and consumer innovativeness about toothpaste, and the other half about smartphones. This since based on prior research, toothpaste generally leads to lower levels of involvement than smartphones, providing data of both high involved and low involved individuals, and so a comparison of the effects of product involvement can be made (Zaichkowsky, 1994).

4.3.4. CONTROL VARIABLES

The following variables are used as control variables. Control variables include elements which are kept constant, in order for them to not interfere with the effect that is being tested, as they too might have an impact.

Age

One variable that might have an influence on consumer innovativeness is age. However, this is not intended to be measured and should be avoided to interfere with the results of the research. Therefore, it will be a control variable. Age will be measured in the survey by having people fill in their age in years (as opposed to selecting an age category, because this will lead to categorical values instead of ratio values).

Gender

Another variable that might interfere with the concepts measured in this study is gender. This since one of the independent variables measures the culture in terms of masculinity/femininity, and gender could influence this dimension. To avoid interference, gender will be determined and used as a control variable.

Income

Finally, Income will also be used as a control variable to avoid interference with the results. Income might influence the fact if people actually have the money to often buy new products or not, so their tendency to be innovative, so also differences in income will be controlled for.

27

5. RESULTS

The following chapter describes the results of the analysis of the data from the survey that was conducted.

5.1. PRELIMINARY ANALYSIS

After deleting only partially completed surveys that were recorded, a total of n=399 surveys remained. As mentioned before, all respondents answered the questions about the individual cultural values, and then (approximately) half answered the questions determining scores for product involvement and consumer innovativeness about toothpaste and the other half about a smartphone. Thus, the data with which the analysis is conducted consists of a sample of n=399, of which n=2oo for toothpaste and n=199 for smartphones.

Reverse scored items were recoded and frequency checks were done to detect missing or abnormal items to be dealt with, which only showed one value of age to be text instead of numerical. Subsequently, mean scale items were computed constituting the following variables:

- Power Distance - Uncertainty Avoidance - Individualism/Collectivism - Masculinity/Femininity

- Long-Term Orientation/Short-Term Orientation - Product Involvement – Smartphones

- Product Involvement – Toothpaste - Consumer Innovativeness – Smartphones - Consumer Innovativeness – Toothpaste.

All variables are a mean average of the various Likert scale items constituting that particular variable, except the product involvement variables. As explained, this scale includes 10-items on a 7-point Likert scale, and are summed up to determine a total product involvement score. Thus, this item can have values ranging from 10 to 70. However, the mean average for Individualism was reversed, as the questions were asked in a way in which higher scores mean a higher level of Collectivism (thus a lower level of Individualism) while the hypothesis regarding this variable is formulated in a way in which a higher level of Individualism shows a higher level of consumer innovativeness. Thus, to avoid confusing with the direction of correlations or Beta scores in the analysis, the mean score was reversed to represent levels of Individualism.

28 Furthermore, also a total distribution of product involvement and consumer innovativeness was computed, including both scores for toothpaste and smartphones simultaneously, showing the total distributions of these variables among all respondents (n=399):

- Product Involvement – Total - Consumer Innovativeness – Total

For the Independent Variables with a continuum (e.g. Individualism/Collectivism), the primary mentioned aspect is the direction in which is being measured, meaning that the higher individualism, the higher the score, and the lower individualism (so more collectivistic) the lower the score.

5.1.1. NORMALITY, SKEWNESS & KURTOSIS

Normality checks showed that the Independent Variables showed bell-shaped distributions, and Skewness and Kurtosis levels were within the accepted values of -1 and 1 (except for Uncertainty Avoidance and Long-Term Orientation/Short-Term Orientation, but they were only marginally higher than |1|; Kurtosis = 1.096 and Kurtosis = 1.022 respectively). Shapiro-Wilk however, showed significant p-values (<0.001), so indicate non-normal distributions, but this test has been known to be very powerful and very often reject the null hypothesis of a normal distribution (Field, 2013).

Normality checks of the product involvement and consumer innovativeness variables showed that the variables were normally distributed for toothpaste (bell shaped distribution, Skewness and Kurtosis levels between -1 and 1, and not significant Shapiro Wilk p-values p=0.291 and p=0.142). Product involvement for smartphones does show a bell shape distribution, but had too high Kurtosis levels (1,739) and a significant p-value for the Shapiro-Wilk test (p<0.001) so a non-normal distribution. Consumer innovativeness for smartphones showed Skewness and Kurtosis levels within the acceptable range, but the Shapiro-Wilk p-value was significant here as well (p<0.001). Again, this is due to the known powerfulness of the test which often rejects the null hypothesis of a normal distribution. However, the variable does show a bell-shaped distribution.

However, when looking at the total distribution of product involvement (including both scores for toothpaste and smartphones simultaneously), Skewness and Kurtosis were well within bounds. Shapiro-Wilk showed a p-value of p=0.038, so would still reject H0, but is close to the significant level. The distribution of product involvement in total is therefore considered to be normally distributed. This variable was used in the independent sample t-test, to determine if product involvement statistically differed between the smartphone and toothpaste condition, which will be explained later on in this chapter.

29 The total distribution of consumer innovativeness (including both scores for toothpaste and smartphones simultaneously) was also looked at. It shows a bell-shaped distribution, and Skewness and Kurtosis were well within the accepted levels of -1 to 1, indicating a normal distribution. Shapiro-Wilk again however showed a significant p-value (p=0.002), but this is again due to the known power of this test.

5.1.2. RELIABILITY ANALYSIS

Dependent Variable

Firstly, the reliability scores of the Dependent Variable, consumer innovativeness, was looked at. As mentioned before, the EAP-scale itself, developed by (Baumgartner & Steenkamp, 1996), consisted of more items. However, as some items were about specific products already, these were eliminated. Therefore, it is very important to see if the scales still lead to reliable scores. The results were as follows.

The consumer innovativeness scale (for toothpaste) has high reliability, with Cronbach’s Alpha =.83. The corrected item-total correlations indicate that all the items have a good correlation with the total score of the scale (all above .40). Also, none of the items would substantially affect reliability if they were deleted.

The consumer innovativeness scale (for smartphones) has high reliability, with Cronbach’s Alpha =.85. The corrected item-total correlations indicate that all the items have a good correlation with the total score of the scale (all above .40). Also, none of the items would substantially affect reliability if they were deleted.

So, consumer innovativeness scored high in reliability, so the measures were good and the results can be used to draw conclusions about the hypotheses.

Independent Variables: Individual Level Cultural Values (CVSCALE)

As mentioned in Chapter 3, when the CVSCALE was tested in earlier research, all cultural dimensions yielded a satisfactory level of reliability with a Cronbach Alpha value of more than 0.6. The testing of the data from the survey that was conducted in this research lead to similar results for all but one dimension. The Cronbach Alpha coefficients were as follows.

The Power Distance scale has a good reliability, with Cronbach’s Alpha =.63 The corrected item-total correlations indicate that all the items have a good correlation with the total score of the scale (all above .30). Also, none of the items would substantially affect reliability if they were deleted.

30 The Uncertainty Avoidance scale also has good reliability, with Cronbach’s Alpha =.67 The corrected item-total correlations indicate that all the items have a good correlation with the total score of the scale (all above .30). Also, none of the items would substantially affect the reliability coefficient if they were to be deleted.

The Individualism/Collectivism scale has high reliability, with Cronbach’s Alpha =.79 The corrected item-total correlations indicate that all the items have a good correlation with the total score of the scale (all above .40). Also, none of the items would substantially affect reliability if they were deleted. The Masculinity/Femininity scale has good reliability, with Cronbach’s Alpha =.66 The corrected item-total correlations indicate that all the items have a good correlation with the item-total score of the scale (all above .30). Also, none of the items would substantially affect reliability if they were deleted. However, the Long-Term Orientation/Short-Term Orientation scale showed less convincing results. It shows a Cronbach’s Alpha score of =.53. The corrected item-total correlations indicate that not all of the items have a good correlation (above .30) with the total score of the scale. However, none of the items would substantially affect reliability if they were deleted.

So, Power Distance, Uncertainty Avoidance, Individualism/Collectivism and Masculinity/Femininity all scored good reliability scores, but the Long-Term Orientation/Short-Term Orientation was not as convincing. This needs to be taken into consideration for further results.

Moderator Variable

Besides the Dependent Variable and Independent Variables, also the Moderator variable was tested for reliability. This was divided into the scores of involvement for the respondents who received the questions about toothpaste and about smartphones.

The product involvement scale (for toothpaste) has high reliability, with Cronbach’s Alpha =.71. The corrected item-total correlations indicate that most of the items have a good correlation with the total score of the scale (above .30). However, there were a few items with a lower correlation score (.127, .221, .282). This could be since the involvement scale is quite broad, asking for how much a product is needed, relevant, fascinating. In the case of toothpaste, many people find it not the most involving product, but they still indicate they need it. This leads to differences in correlation levels. However, none of the items would substantially affect the reliability score if they were to be deleted.

The product involvement scale (for smartphones) also has high reliability, with Cronbach’s Alpha =.92. The corrected item-total correlations indicate that all the items have a good correlation with the total

31 score of the scale (all above .50). Also, none of the items would substantially affect reliability if they were to be deleted.

5.1.3. FACTOR ANALYSIS

For the CVSCALE, besides the reliability, also the validity of the measure was tested using a principal axis factoring analysis (PAF).

The Kaiser-Meyer-Olkin measure verified the adequacy of the sampling for the analysis, with KMO = .747. Bartlett’s test of sphericity indicated a significant value of p <0.001, indicating that correlations between items were sufficient for PAF.

As expected, the first five factors reported eigenvalues above Kaiser’s criterion of 1, and cumulatively explained 44,69% of the variance. However, there were two more factors with eigenvalues above 1, leading to a cumulative 53,55% of the variance explained. However, the scree plot provided more insight. In agreement with Kaiser’s criterion, the scree plot revealed a point of inflection after five factors. So, it was levelling off after the fifth factor. After the seventh factor, there was another point of inflection, but it was way less apparent than the first, indicating that five factors should be retained. Thus, five factors were retained and rotated with an Obliminin rotation. The items that cluster on the same factors suggest that factor 1 represents Power Distance, factor 2 represents Individualism/Collectivism, factor 3 represents Uncertainty Avoidance, factor 4 represents Long-Term Orientation/Short-Term Orientation and factor 5 represents Masculinity/Femininity.

For the first four dimensions, all the items loaded highly on the appropriate factors and no item loaded on more than one factor, supporting the independence of the constructs and providing evidence of their validity. However, same as with the reliability analysis, the fifth dimension (Long-term Orientation/Short-Term Orientation) did not lead to convincing scores. Several items loaded onto several factors, and they did not all have high loading scores either. This, together with the lower Cronbach Alpha for this variable, indicates that conclusions about this dimension/variable need to be drawn with caution.

5.2. INDEPENDENT T-TEST

As explained earlier, product involvement and consumer innovativeness were measured asking about two different products, toothpaste and smartphones. This was done because prior research indicated that they on average score differently on product involvement (toothpaste scores on average lower than smartphones) and to test the moderating effect of this variable, there needed to be respondents in various degrees of involvement.

32 However, to test if respondents in fact did score on average lower in involvement (significantly) in case of toothpaste as opposed to smartphones, an independent sample t-test was used.

The product involvement variable used was the total distribution of involvement levels of respondents, including both the ones asked questions about toothpaste and about smartphones (Product Involvement – Total). As mentioned earlier, this variable has a normal distribution. The two samples are also independent, as respondents were either asked about smartphones or about toothpaste, and not both.

As grouping variable, respondents were given a score to indicate what product they were asked about (0=Smartphone, 1=Toothpaste).

An independent sample t-test was then conducted to compare product involvement in the smartphone and toothpaste conditions. Levene’s test indicates to reject equality of variances between the two means (p=0.013, so p<0.1). Looking then at the scores of the t-test, there is a significant difference in the scores for smartphone (M = 51.81, SD = 9.87) and toothpaste (M = 43.41, SD = 7.69) conditions; t(373,70) = 9.48, p<0.001. These results suggest that either a smartphone or toothpaste lead to significant different levels of product involvement of people.

5.3. CORRELATION

Subsequently, a correlation matrix was compiled, to examine the linear relationships between the main variables of interest within this study.

The following tables (Table 1, Table 2 & Table 3) provide an overview of the correlations between the variables. The first table looks at correlations between the Independent Variables (individual level cultural values) and correlations with product involvement and consumer innovativeness within the smartphone condition, and the other within the toothpaste condition.

As can be seen from Table 1, there are three Independent Variables that show a significant correlation with the Dependent Variable consumer innovativeness in the smartphone condition. These Independent Variables include Uncertainty Avoidance (r=-.21), Individualism/Collectivism (r=.14) and Long-Term Orientation/Short-Term Orientation (r=-.15). The directions of the correlations are in line with the hypothesis for Uncertainty Avoidance and Individualism/Collectivism (H2a, H3a), but not for Long-Term Orientation/Short-Term Orientation (H5a).

In the toothpaste condition (Table 2), there are no significant correlations between one of the Independent Variables and consumer innovativeness. Product involvement does show a significant correlation with consumer innovativeness (r=-.15). This indicates that people with a higher level of