Effects of Consumer Dispositional Attitude on Purchase Intention in an Emerging Market

Consumer ethnocentrism

The term consumer ethnocentrism describes those consumers who believe in their own culture, see others only from their own set of values’ perspective and decline people who do not belong to their group or one who does not believe in their set of values and traditions. People with ethnocentric beliefs have an in-group and they tend to favour them and discard others who do not believe in their culture and traditions and are held in contempt (Shimp and Sharma, 1987).

Shimp and Sharma (1987) conceptualized “consumer ethnocentrism as the beliefs held by consumers about the appropriateness, indeed the morality, of purchasing locally-made products instead of foreign-made products”. For an ethnocentric consumer, their own country’s products are superior to a foreign country’s products, even though the quality of foreign products is superior, but still, they prefer their nation’s products (Raju, 1995; Riefler and Diamantopoulos, 2007). Ethnocentric consumers do not believe in buying foreign products as it is against their moral values, they believe it would hurt the domestic economy, thereby resulting in job losses for their fellow countrymen, not only that, but foreign brands also even pose a threat to indigenous companies (Cleveland et al., 2009).

Consumer cosmopolitanism

Cosmopolitanism dates to the eighteenth century, when people assumed themselves to be citizen of the world and their desire to adopt different cultures made them stand out from others. Cosmopolitans are those who have an intention to be recognized beyond their nation and culture and those who spend their lives seeking experiences both within and outside their nation (Thompson and Tambyah, 1999; Cannon and Yaprak, 2002; Saran and Kalliny, 2012). In the early 1950s, many sociologists viewed cosmopolitans as outgoing people, beyond their community, and influenced by world culture more than their local traditions. Consumer cosmopolitanism is a construct that can be measured in two ways. One is that cosmopolitan consumers have cultural openness, where they are willing to adopt other cultures and are open to other cultural experiences, having a positive sense of competence towards foreign cultures.

Cosmopolitan consumers are outgoing, open to exploring other cultures and have world-mindedness. Even though they accept their local cultures, due to their yearning to be associated with a global identity they have a positive attitude towards global brands (Cleveland et al., 2011). The other view is that consumer cosmopolitanism shows that a consumer embraces foreign cultures through open-mindedness and the diversity it brings due to the products available from different nations as well as cultural origins and they are positive towards consumption of foreign-made products. In this context, cosmopolitan consumers have an unprejudiced opinion towards foreign cultures and in experiencing their authenticity (Riefler et al., 2012).

Attitude towards brand

“Attitude towards brand is defined as an individual’s internal evaluation of the brand” (Mitchell and Olson, 1981). A claim by Lutz (1991) says that for a consumer, attitudes function as a strainer for how they perceive an object. Past studies have confirmed the noteworthy effect of attitude towards purchase intention (Zhang et al., 2005). Jin and Kang (2011) have established that for Chinese consumers, attitude plays an important factor that predicts purchase intention of global apparel brands.

Purchase intention

Consumers evaluate products and services available to them before any purchase decision, so this evaluation finally results in a transaction which is termed as purchase intention (Zeithaml, 1988). Purchase intention is referred to as the probability that consumers are going to buy the given product (Schlosser et al., 2006). Alden et al. (1999) have argued in the literature that there is a belief that marketers are employing local and global consumption positioning for their products. Previous studies aimed to study the impact of consumers’ evaluation of global and local brands in both emerging and developed economies and their influence on purchase intention (Alden et al., 2006).

Researchers in the past have previously studied consumer dispositions such as ethnocentrism, cosmopolitanism, etc. in various research contexts and extensive research has been conducted and has had a diverse view on its impact on attitude and purchase intention. The Figure 1 below shows the conceptual framework of this study.

Figure 1. Conceptual framework - a schematic representation.

Theoretical background and hypotheses development

Consumer ethnocentrism is a psycho-social construct that is relevant to an individual-level personality trait, which initially evolved from being a sociological concept (Levine and Campbell, 1972). Consumers associated with certain beliefs, attitudes, behaviours, etc. have adopted certain social identities, which is also known as in-group, due to which the intention to buy local brand increases, hence promoting ethnocentrism and believing that they are superior to others (Turner et al., 1987).

Shimp and Sharma (1987) measured consumer ethnocentrism and concluded by saying that consumers have high ethnocentric tendencies and often purchase domestic products rather than global products, and also consider domestic products to be superior, as later supported by Netemeyer et al.(1991) and Chryssochoidis et al. (2007). Consumers, when deciding to buy between local and global brands, will favour products from countries that possess a similar culture (Sharma et al., 1995; Lantz and Loeb, 1996; Watson and Wright, 2000). Consumers consider the products made in their country as group related products and products of foreign countries are taken as unrelated group products, so naturally, they tend to favour products related to their group (Verlegh, 2007).

Consumers in emerging economies are inclined to believe that local products are not as good as foreign products and favour global products more (Balabanis et al., 2001; John and Brady, 2011). The consumer’s love towards their nation results in high consumer ethnocentrism (Sharma, Shimp, and Shin, 1994). Due to this, the consumer’s evaluation of product influences in such a way that, they overestimate the domestic product’s quality and underestimate the foreign product’s quality. This would result in a positive attitude towards local products and a negative attitude towards global products (Sharma et al., 1995; Watson & Wright 2000).

Hence the following hypotheses

H1a: There is a negative relationship between consumer ethnocentrism and attitude towards global brands.

H1b: There is a positive relationship between consumer ethnocentrism and attitude towards local brands.

Cosmopolitan consumers are not affected by their home culture and are open to different cultures, they transcend any boundaries or setting to experience foreign cultures and, in a way, they are motivated to achieve social status by favouring global products (Thompson and Tambyah, 1999; Cannon and Yaprak, 2002). Cosmopolitan consumers are the ones who are outgoing, open-minded, and willing to explore other cultures with a motivation of understanding and learning from other cultures’ experiences (Weij et al., 2015). Cosmopolitans from developed as well as emerging economies generally prefer buying foreign products (Caldwell et al., 2006). But Riefler et al., (2012) argue that although cosmopolitan consumers may not prefer products from their nation, they are well-versed in their local traditions and culture.

Therefore, we hypothesise that

H2a: There is a positive relationship between consumer cosmopolitanism and attitude towards global brands.

H2b: There is a negative relationship between consumer cosmopolitanism and attitude towards local brands.

Consumers have a predisposition towards certain brands, be it global or local (Verlegh and Steenkamp, 1999; Magnusson, Westjohn, and Zdravkovic, 2011). This may be either positive or negative, and either way, it influences their purchase decision. Consumers’ attitude towards brands originates from their past experiences, and it affects their purchase intention (Costa et al., 2016; Herz and Diamantopoulos, 2017).

Hence the following hypotheses

H3a: There is a significant relationship between attitude towards global brands and purchase intention.

H3b: There is a significant relationship between attitude towards local brands and purchase intention.

Measures

The questionnaire consisted of scales concerning constructs of consumer ethnocentrism with 5 measurement items, consumer cosmopolitanism with 12 measurement items, attitude towards brands with 5 measurement items, and purchase intention with 5 measurement items. The study used a 5-point Likert scale labelled from 1 (strongly disagree) to 5 (strongly agree). For the measurement of the independent variables and dependent variables. Table 1 has a list of the measuring items.

Sampling design and data collection

Sampling is done in two stages. In the first stage, the sampling method is employed to draw the products for the consumer survey. Therefore, the judgment sampling technique is employed under the nonprobability sampling method to draw products and brands from the FMCG sector. The FMCG sector is identified for the study using these three criteria: (i) availability of global and local brands in the same product category, (ii) low price point of the products, and (iii) accessibility to a broad range of consumers (Narang, 2016).

The product category with higher penetration and higher market share identified for this research is biscuits from food and beverages. Since India is the highest consumer of biscuits in the world and these products have a low price point, easily accessible to consumers, it has consumers from all the strata (TRA Research, 2022).

In the second stage of sampling, data were collected using the convenience sampling method. Convenience sampling was particularly relevant in this research because researchers needed to choose respondents who are already familiar with the chosen brands (Narang, 2016). In addition to this, researchers chose respondents who were reasonably knowledgeable about global and local brands. A pilot study was done with a sample size of 84 to ensure that there are no major flaws in the study instrument and to check the reliability and validity of the scale used in this research.

Data analysis

The demographic description of the data was conducted using IBM-SPSS software (RRID:SCR_019096), and the evaluation of the measurement model was done using Smart PLS 4 (RRID:SCR_022040). Investigating the convergent validity and discriminant validity of the measurement model allowed for an evaluation of its quality. Outer loadings were checked to negate the measurement items with poor loadings, and such items were deleted from the analysis (Gefen and Straub, 2005).

Ethical considerations

Before distributing the survey questionnaire, all survey participants provided their written informed consent. Before asking for the respondents' signed informed consent, an information sheet outlining the research's goals, findings, and consequences was included in the form's content. It contains statements stating that any personal information will be kept confidential and that the data obtained will be used only for research and publication purposes (see Extended data, Shanbhogue & Ranjith, 2023). In addition, the Manipal Institute of Management ethics committee and doctoral advisory committee panel gave their approval before conducting this research (date of approval 24.02.2022).

Reliability assessment

Hair et al. (2012) describe that reliability and validity need to be assessed for the measurement model. The specific measures for internal consistency and reliability are composite reliability, convergent validity, and discriminant validity. As per Hair et al. (2012), composite reliability is used to evaluate internal consistency, and average variance extracted (AVE) is used to determine convergent validity. The researchers adopt the Fornell-Larcker criterion and the heterotrait-monotrait ratio (HTMT) to assess the discriminant validity Hair et al. (2012).

The first criterion of composite reliability has been assessed for internal consistency. Hair et al. (2012) recommend composite reliability for measuring the internal consistent reliability of the PLS-SEM approach rather than Cronbach’s alpha value. The threshold value for composite reliability is above 0.60. It is evident from Table 3, that internal consistent reliability among all constructs of the measurement model is acceptable as the values are above the threshold values. Since Cronbach alpha value and composite reliability value for all constructs are above 0.70, it is declared that the constructs used for the study have internal consistent reliability Hair et al. (2012). Therefore, all the values of composite reliability are greater than the average variance extracted for all the constructs. The average variance extracted (AVE) for each of the constructs exceeds the threshold value of 0.50, thus confirming the convergent validity of the scale (Nunnally, 1994).

Discriminant validity

Discriminant validity is about measuring the individuality of a construct among other constructs of the measurement model. It measures the extent of the distinctness of one construct from the other constructs. The researchers use the Fornell-Larcker criteria, and the HTMT to assess the discriminant validity.

Fornell-Larcker criteria

The results of the Fornell-Larcker criteria for the discriminant validity in Table 4 demonstrate the square root of every construct exceeding its correlations with the other construct. Therefore, it is evident from the result that discriminant validity exists among the constructs used for the study.

HTMT

Henseler et al. (2015) proved through their simulation models that the HTMT ratio is better than the Fornell-Larcker criterion for discriminant validity. The heterotrait-hetro method correlations geometric mean is divided by monotrait-hetro method correlations average to derive the HTMT ratio.

As per Henseler et al.’s (2015) suggestion, to understand the discriminant validity that has been established among the constructs, the HTMT ratio value should be below 0.90. Table 5 indicates a good discriminant validity among the constructs since the HTMT ratio of all the constructs among other constructs is less than 0.80. After assessing the reliability, construct validity, and discriminant validity researchers find that there is good reliability and validity among the indicators. Therefore, structural model assessment and hypotheses testing are done in the next section.

Structural model and hypotheses testing

The researchers validate the model with the PLS-SEM approach and examine the conceptual framework developed based on the literature support. The researchers adopt the PLS-SEM algorithm to compute the path coefficients of the structural model. The Figure 2 shows the structural model assessment of this study.

Figure 2. Structural model assessment.

A strong relationship between the constructs exists if the path coefficient values are not near zero. If the values are near zero, it indicates a weak relationship between the constructs Hair et al. (2012). The bootstrapping technique evaluates the significance level of the relationship between the constructs. The evaluation of the structural model will show the significance level, and the t-value above 1.96 declares the relationship between the constructs is significant.

Since the coefficient determination is the criterion by which the structural model is measured for its predictive accuracy (Henseler et al., 2012; Hair et al., 2012), the coefficient of determination, i.e., the R² value is considered to validate the model. The model stands weak, with a value of R², the coefficient of determination being 0.013 for attitude towards global brands, weak with a value of 0.311 for attitude towards local brands, reasonably strong with a value of 0.644 for purchase intention of global brands, and 0.614 for purchase intention of local brands as shown in Table 6.

Model fit

For model fit, the value of Standardized Root Mean Square Residual (SRMR) should be less than the threshold value of 0.08 or 0.10 according to Henseler et al. (2014). Since the value of SRMR is less than the threshold value as shown in Table 7, it represents a good model fit. Another indicator of a good model fit is Normed Fit Index (NFI). According to Lohmöller (1989), the value of NFI lies between 0 and 1, the closer the NFI is to 1, the better the fit. Since the value of NFI is closer to 1 as shown in Table 7, it is considered a good model fit.

The structural model assessment and bootstrapping results in Table 8 summarize the path coefficient values, t-values, p-values, significance levels, and bootstrap confidence intervals for the model. The results of hypothesis testing for the model are illustrated in Table 9.

Limitations and scope for further research

Firstly, the sample, given its homogeneity, does not perfectly represent the entire spectrum of consumers, as a high number of respondents in this study are college/university students. The difference among educational qualifications still exists, which might lead to differences in their behavioural aspects of attitude formation or attitudinal dispositions. Secondly, this study discusses a specific brand or product category, hence this limitation is also one of the scopes for future research to include more product categories or a set of specific brands for the consumers to evaluate and compare before making a purchase decision. Finally, this study has not considered any moderators in the research model. Top researchers in this discipline have advocated that consumer disposition constitutes differently depending upon the contextual factors in terms of consumption, therefore future research should replicate this study by accounting for such aforementioned factors influencing consumers’ attitude and purchase intention.