The current literature suggests that the concept of consumer innovativeness is universally applicable. Innovators are believed to be novelty seekers and risk takers independent of their national identity, and therefore to be attracted to similar characteristics of an innovation across most countries. However, research in intercultural marketing has shown that cultural norms and values have varying influences on the adoption of innovation, a finding that seems to contradict the assumption that the relationship between consumer innovativeness and adoption of innovation is universally uniform. This research investigates the effects of consumer innovativeness on attitude toward a service-based innovation across three European countries. The results of a multi-group structural equation modeling show that the relationship between consumer innovativeness and attitude toward innovation varies across the three dimensions of perceived novelty, perceived value, and perceived risk.
The process of globalization has led to the increasing belief among marketing managers that consumer behaviors in different nations are converging into general common patterns that may serve as a common ground for predicting the adoption of an innovation (Yalcinkaya, 2008). The success of some recent new products tends to confirm this belief, as millions of Apple’s iPhones, Nintendo’s Wiis, and Asus computers have been sold in many countries with different value systems. However, innovations do not all succeed in the same manner, and in fact, most new products and services tend to fail in the market (Gourville, 2006), often because they have not been adapted to local markets (Meng et al., 2009 and Yalcinkaya, 2008).
Since national values and norms strongly determine motivation and behavior (Markus and Kitayama, 1991, Park et al., 2012 and Yeniyurt and Townsend, 2003), consumers in different countries have differing perceptions, preferences, and values that influence their intention to adopt an innovation (Dwyer et al., 2005, Suh and Kwon, 2002 and Yang and Jolly, 2009). One evidence of these differences is the varying technology adoption rates of countries that share similar economic situations (Erumbana and de Jong, 2006 and van Everdingen and Waarts, 2003). For example, one of the main explanations of the varying adoption rates of broadband within the European Union lies in cultural and social factors (Fife and Pereira, 2002).
However, even if the literature shows evidence of the influence of cultural values and norms on adoption behavior (Hofstede, 2001 and Steenkamp et al., 1999), few studies have investigated the validity of adoption models across nations (Yalcinkaya, 2008). Instead, research using popular adoption models such as the technology acceptance model (Davis et al., 1989) and the unified theory of acceptance and use of technology (Venkatesh et al., 2003) attempts to generalize consumer behavior at an international level. One important assumption of these models is that the relationships between the predicting and predicted variables are universal.
Consumer innovativeness determines one’s tendency toward novelty-seeking and risk-taking behavior (Hirschman, 1980). The concept often implies that innovators hold the same profile independently of their national culture. For example, innovators are universally believed to be novelty-seekers, risk-takers and independent judgment-makers (Midgley and Dowling, 1978 and Rogers, 2003). However, this assumption overlooks the influence of cultural values and norms on consumers' behavior toward innovation—an influence that is especially critical to the concept of consumer innovativeness since innovators play a central role by adopting the innovation earlier and diffusing it to subsequent segments of the population.
Because of the centrality of innovators to the adoption process, investigation of how consumer innovativeness may differ across cultures in its influence on perceptions, attitude, and adoption of innovation is particularly important to understanding how new products should be adapted to local differences (Fowler and Bridges, 2010). This importance underlies the increased attention cross-national study of adoption behavior has been receiving from both academics and practitioners (Yalcinkaya, 2008). Thus, although studies have provided evidence of variation in consumer innovativeness across countries (Tellis et al., 2009) and of differences in how national culture affects innovativeness (Steenkamp et al., 1999), little if any research has validated the influence of consumer innovativeness on perceptions and adoption behavior. This disregard is particularly apparent in the context of services, as most research so far has focused on products.
Many researchers seem to have assumed that the pattern by which consumer innovativeness influences perceptions and attitudes is similar across countries. Therefore, the objective of this research is to investigate the role of culture in moderating the relationships between consumer innovativeness, perceptions, and attitudes toward an incremental product innovation in a service context. Given the novelty of the product under study, we used attitude instead of adoption in order to better capture consumer initial evaluation of the new product. A structural model was tested using data collected in the UK, France, and Germany. Results show that consumer innovativeness influences perceptions differently across countries and that perceptions produce dissimilar effects on attitude across countries. For marketing managers, this finding illustrates the importance of taking into account the influence of culture on the attitude toward service-based technological innovations, and the corresponding need to adapt retailing strategies, to increase the chances of success of a new product or service in the local market.
4.1. CFA and model fit
The Kaiser–Meyer–Olkin test (>0.80) and Bartlett’s test of sphericity (p<0.01) confirmed that the data were appropriate for factor analysis (Schumacker and Lomax, 2004). Using principal component analysis and varimax rotation, we found that the 18 items loaded correctly with their corresponding factor, with all loadings above 0.70 except one item for perceived value. We therefore decided to remove this item, after which the outcome satisfied the requirement for sufficient homogeneity and validity in structural equation modeling that at least three items measure each factor (Byrne, 2001). Reliability was also satisfactory for all factors as all Cronbach’s alphas were above 0.70. Table 2 shows the items and their corresponding factors as provided by SPSS 18. Following Hu and Bentler (1999) and Byrne (2001), we used a combination of relative indices (IFI and TLI), CFI, a noncentrality-based index (RMSEA), and an absolute fit index (SRMR) to assess the fit of the structural model. This combination provided a solid ground for assessing model fit using some of the most reliable and sample size-independent indices. All indices met the recommended threshold: IFI (0.93), TLI (0.90) and CFI (0.93) were above 0.90, RMSEA (0.05) was below 0.60, and SRMR (0.07) was below 0.80 (Byrne, 2001). As a consequence, the model presented good fit and thus acceptable configural invariance. To test metric invariance, we observed the chi-square difference between a constrained versus an unconstrained model in AMOS, and found that the difference was non-significant. Therefore, we concluded that the measurement was invariant across the three groups.4.2. Hypothesis testing
Assessment of the relationships between the three models was performed by a multigroup analysis procedure in AMOS 16. The results of the analysis appear in Table 3. All regression coefficients were significant at the 0.05 level except the effect of perceived risk on attitude for the French sample. We used the critical ratios computed by AMOS to compare the regression coefficients between the three models (two-tailed z-test). As a critical ratio above 1.96 implies that the difference between two regression coefficients is significant at 0.05 level (Byrne, 2004), we used the critical ratios to test our six hypotheses. On the relationship between consumer innovativeness and perceived novelty, the difference between the three regression coefficients was significant. Consumer innovativeness had a greater effect on perceived novelty in the UK and France than in Germany, a result supporting the hypothesis that consumer innovativeness influences perceived novelty differently across countries (H1).We found similar results for the relationship between consumer innovativeness and perceived value. Perceived value was more salient in Germany than in the UK and France. Thus H2, which hypothesized that consumer innovativeness influences perceived value differently across countries, is also supported. With respect to risk, consumer innovativeness had a negative effect on perceived risk in the UK and Germany, but the relationship was found to be insignificant in France. Thus H3, which posits that consumer innovativeness influences perceived risk differently across countries, is partially supported.
Perceived novelty had an important effect on attitude toward innovation in all three countries and no significant differences were found between countries, implying that novelty is a salient determinant of attitude in all three countries. Thus, H4, which hypothesized that perceived novelty influences attitude differently across countries, is not supported. However, the effect of perceived value as a determinant of attitude was significantly stronger in France than in the UK and Germany. Therefore, H5, which hypothesized that perceived value influences attitude differently across countries, is supported.
Finally, differences in the influence of perceived risk on attitude were significant in all three countries. Perceived risk was a stronger determinant of attitude in Germany than in the UK and France. In France, perceived risk did not seem to affect respondents' attitude toward the new service. Therefore H6, which hypothesized that perceived risk influences attitude differently across countries, is supported. Table 4 summarizes the hypotheses, results and findings.
