The Effect of Consumer Affinity and Country Image Toward Willingness to Buy

Abstract

Purpose - This research aims to determine whether the consumer affinity and ethnocentrism as well as the image of a foreign country (Japanese product as the most popular product in Indonesia) are able to influence behavior related to the perceived risk and willingness to buy foreign products from the affinity country. Research design, data, and methodology - Using survey techniques with 164 respondents, the study uses structural equation model with confirmatory factor analysis (CFA). To ensure the research objective and appropriate respondent, then we select an individual who have interest on Japanese culture & language. The primary and secondary data used in this study. Primary data refers to information collected directly from respondent by questionnaires dissemination while secondary data is provided from well-established literatures. Results - The results show us that the ethnocentrism has dominant affection role compared to affinity in order to influence consumer behavior meanwhile, the product country image has cognition role to evoke consumer desire to consume foreign products. Conclusions - From a theoretical perspective, the study contributes to international marketing literature by refining the conceptualization of the consumer affinity construct and highlighting its multidimensional nature. The consumer affinity research need to enrich in term of the context and the different culture and situation.

1. Introduction

A changing trend and lifestyle is a cycle of each generation to another. These changes mostly triggered by individual interest about an object, person, or even stuff. A theory that capable to explain about this mechanism is the emotional attachment theory (Bowlby, 1980). Based on this theory, an interaction of a person to an object is strongly influenced by how strong the emotions possessed by individual towards reference. Solomon (2013), introduced the concept of referent power which consumers voluntarily adapt the attitude and purchase products that have relevance to the referent in order to make it look identical. This concept strengthened the basis of individual motives purchasing foreign products. Those phenomena above can be explained from various perspectives. In some cases, an individual admiration of other countries is caused by the similarity of culture, language and economic situation of the host country which in turn creates a bond (Josiassen, 2011). In other cases, the passion to interact with the host country’s culture, people and tradition in a personal experience (by travelling, live or work abroad) might cause the admiration as well (Riefler et al., 2012).

Attitudes and behavior of Indonesian consumers tend to view local products as inferior products and prefer to use foreign products. Jaffe and Nebenzahl (2006) stated that the consumer in-group have ethnocentric tendencies as a general attitude that allows individuals to have a sense of affinity towards certain foreign countries to foster a positive attitude. This type of attitude might be serious threat for domestic product because one side, consumer has ethnocentric tendency as general attitude but in other side, they have an affinity for a specific foreign products as simultaneously (Oberecker et al., 2008). Therefore, this segment of consumer needs to investigate by more research to be particularly attractive targeting purposes.

The following subject of interest to do this research are firstly, to determine the level of consistency that varies from previous studies on the effect of each concept related and also the outcomes in influencing people. Secondly, to test the effect of the model in the context of one of the developing country consumers in ASEAN, namely Indonesia. This is based on the statement Erdogan and Uzkurt (2010), that the results of a study examined the relationship between consumer ethnocentrism, product attitude and other demographic variables were varies and had different strength from one country to another. Finally, the subject matter of this research was to evaluate the relationship between the concept of consumer affinity, consumer ethnocentrism and country image in the simultaneous effect on perceived risk and willingness to buy foreign product. The recent study will enrich the previous studies not only in term of context but also the foundation of the behavioral differences response among different culture (country).

2. Theoretical Foundation

onsumer affinity (CA) defined as a “feeling of love, sympathy, and a certain attachment to foreign countries from personal experience or normative exposure which influences consumer when making decisions related to the products and services derived from the affinity country” (Oberecker et al., 2008). Consumer affinity understood as a positive feelings directed to a reference (in the context of this study the referent is an affinity country) and produce an action of consumption behavior. Moreover in specific, an affinity describes positive emotions towards a country rather than a general mood or affection in common (Oberecker & Diamantopoulos, 2011).

Consumer ethnocentrism (CET) defined as beliefs held by consumers about the appropriateness and morality in buying foreign products (Josiassen, 2011). The essence of this concept is about countrymen supposed to buy and use products made in their own country for the purpose of prosperity and escalate the development of the nation itself. In general, the expectation of this construct is to be able to describe the level of "discrimination against foreign products" (Baughn & Yaprak, 1996), of an overall consumer response (Josiassen et al., 2011). CET is influenced by two factors; the first is the affection of love for his own country, and secondly is normative pressure that consumers feel when they do not buy domestic product (Olsen et al., 1993).

Country image defined as a set of COO (country-of-origin) associations arranged into groups in such a way that has meaning (Keller, 1993; Pappu, Quester, & Cooksey, 2007). The construct provides insight into what the underlying consumer attitudes towards products manufactured in certain countries and thus, it has a vital role to strengthen the consumer's perception of a product. The way country image works is similar to halo effect which introduced by Throndike (1920), cited on Robbins and Judge (2010), that is describe about the influence of cognitive biases that occur when an overall impression of someone or something is inferred from the results of generalization of one dominant characteristic. This effect is interpreted by the belief that the perception of a product is in the minds of consumers built from the cognitive attributes of product and it is largely determined by the consumer perspective to portray a country in general.

Product country image defined as total beliefs held by individuals to a product of a country (Nagashima, 1970). Roth and Romeo (1992) conducted a study of the concept of country of origin as the root of the product of country image and formulate that there are four main dimensions of this construct, specifically; innovativeness, design, prestige, workmanship. The essence that can be drawn from this construct is that consumers are cognitively using all known knowledge (or which they consider to know) about the products of a country as an indicator of beliefs (Bloemer, Brijs, & Kasper, 2009).

In general, perceived risk is defined by Yuksel and Yuksel (2007) as a construct with multiple aspects potential that include financial issue (loss of income or waste of money), performance issue (inappropriate requirements or below the standard), physical issue (personal illness, injury or health risk), psychological issue (emotional distress) and social issue (seen as outdated or have a lower status). In the context of the study congruity, the closest definition of perceived risk is the consumer's perception of the negative consequences to be borne associated with consuming certain product categories (Dowling & Staelin, 1994).

Willingness to buy defined as a consumer’s desire to buy foreign products due to the fulfillment of an expectation of a product (Kumara & Canhua, 2010). The consumers found the values of foreign products that are not available on local products so they decided to use the foreign product. Moreover, Breidert (2006) explains that there are two distinct constructs in determining how much consumers are willing to pay for a product or service. These constructs include, the maximum price and the reservation price. However, the difference between the two constructs are often difficult to explain. Therefore, the researchers and practitioners are prefer to use willingness to buy as terminology to describe the situation that explained above.

3. Research Hypotheses

This study is a replication refers to research model Oberecker and Diamantopoulos (2011) with several adjustments. In previous studies, the potential role of affect is highlighted by empirical evidence indicating that consumer affinity construct has multidimensional nature. The results provides support for Jaffe and Nebenzahl’s (2006) consumer segmentation model in which negative sentiments toward foreign products in general (as captured by consumer ethnocentrism) and positive sentiments toward a specific foreign country (as captured by consumer affinity) can coexist. However on the other side broke the statement Jaffe and Nebenzahl (2006), who state that “given their conflicting emotions, the behavior of these consumers cannot be predicted”. The study results clearly indicate that affinity feelings outweigh ethnocentric tendencies in affecting perceived risk and willingness to buy products from the affinity country.

[Figure 1] Previous Result found by Oberecker & Diamantapoulos (2011).

Consumer Affinity, Perceived Risk and Willingness to Buy

In the context of affinity country (with the assumption that consumers have sufficient level of familiarity about the country and the product which to be selected and prioritized), the initial assumption is that the consumers will directly affect their desire to buy the product created by affinity country which then minimize the influence of perceived risk. Based on statement Oberecker, Riefler and Diamantapoulos (2008), consumer will choose a product from affinity country when confronted with several products due “heightened trust” (Mitchell, 1999). Thus:

[H 1a] Consumer affinity positively affects consumers’ perception of risk associated with the products derived from affinity country.

The subject of consumer behavior about experiential hierarchy of effects (Solomon, 2013) explains us that "consumers act on the basis of their emotional reaction". Those affection describes how we “feel” about an object. Hazan and Shaver (1994) argued consumers with strong attachment to an object will attempt to maintain the intimacy with the selected object. In relation to consumer mechanism against foreign products, Verlegh (2007) consumers who have admiration for a country will deliberately buy products from foreign countries for the purpose of establishing a bond. Wongtada et al. (2012) reinforces the assumption that the construct of affinity is able to affect the way consumers evaluate and willingness to buy foreign products. Thus;

[H 1b] Consumer affinity positively affects consumers’ willingness to buy products derived from affinity country.

Ethnocentrism, Product Country Image, Perceived Risk and Willingness to buy

Similar to the discussion of affinity, ethnocentrism has a similar impact on the group. Lang et. al. (2012) said that ethnocentric consumers are believed to have a negative initial perception towards foreign products. It encourages consumers to gone further. An interesting fact revealed consumer ethnocentric has mindset and belief that by purchasing foreign products is a mistake since it is not only reflects the unpatriotic attitude of a citizen, but also harm the economy and cause many citizens lost their jobs (Kaynak & Kara, 2011).

[H 2a] Consumer ethnocentrism negatively affects product country image affinity country.

Furthermore, the impact of a consumer confidence in the origin of a product is encouraged, then the level of perceived risk will increase without regard to the "perceived quality" (Keller, 2008) from products including the value inside (e.g., performance, features, conformance quality, reliability, durability, serviceability, style and design). Ethnocentric consumers will only have one view that buying foreign products have a greater risk against country whether it underway currently or it happens in the future. Thus, the consumers constrained by this thought and reluctant to buy foreign products. The consequence is ethnocentric consumers will have a negative evaluation toward foreign products (Sharma, Shimp, & Shin, 1995), perceived risk on foreign products is about increasing (Jung & Khau, 2006) and the interest to buy foreign products is on the wane (Durvasula, Andrews, & Netemeyer, 1997).

[H 2b] Consumer ethnocentrism positively affects perceived risk associated with products derived from affinity country.

[H 2c] Consumer ethnocentrism negatively affects towards willingness to buy products derived from affinity country.

Country Image and Product Country Image

The study of relationships among macro image of country (Martin & Eroglu, 1993) and micro (Roth & Romeo, 1992), have been found in many literatures and other scientific studies. Basically, general beliefs of a country in terms of economic conditions, political structures, or cultural aspect are able to affect consumer perceptions of products created by the country concerned (Heslop & Papadopoulos, 1993; Pappu, Quester, & Cooksey, 2007). In fact, in the context of the evaluation towards foreign brands which COO is the only reference when consumers were clueless (Johansson, Douglas, & Nonaka, 1985), it is significantly affects consumer evaluation against attributes of the product concerned (Erickson, Johansson, & Chao, 1984; Han, 1989; Johansson, Douglas, & Nonaka, 1985).

[H 3] Country image positively affects Product Country Image.

Product Country Image, Perceived Risk and Willingness to buy

The evaluation of the product-country image shows that the product of a country is able to directly affect consumer behavioral intentions (Ekinci & Hosany, 2006; Heslop et al., 1998). It is described by Agarwal and Teas (2001) that "extrinsic product cues" give effect to the "perceived risk". The meaning of the statement above is that consumers are strongly rely on extrinsic cues to evaluate or calculate the losses that might appear. Other research proclaimed “country-of-origin image” is simulteanously increase perceived quality while reduce performance risk of products (Thorelli et al., 1989). Performance risk is understood as the negative consequences to be borne when a product is used, while performance itself defined as "dimension of perceived quality” (Keller, 2008). In other words, when the risk performance decreases consumers will lose doubts towards products while increasing the desire to own foreign products of host country because of the quality issue.

[H 4a] Product country image negatively affects perceived risk associated with products derived from affinity country.

[H 4b] Product country image positively affects willingness to buy products of affinity country.

Perceived Risk and Willingness to buy

In some situation, when the perceived risk of consumers towards foreign products reaches a certain level, then their desire to purchase those products will decline. The more level perceived generated increase, the more desire of consumption diminished.

Semakin meningkat level perceived yang dihasilkan, semakin menurun pula keinginan untuk mengkonsumsi produk. This is because consumers think of the negative consequences that come out of the evaluation process in their minds.

[H 5] Perceived risk negatively affects Willingness to buy product derived from affinity country.

4. Method

We using structural equation model (SEM) and maximum likelihood estimation (MLE) to test all hypotheses. To test the validity of all measurement, confirmatory factor analysis (CFA) was used. Meanwhile, to ensure the reliability of measurement item, we using composite reliability (CR) and variance extracted (VE). The data used of the study sourced by primary and secondary data. Primary data refers to information collected directly from respondent by questionnaires dissemination (Sekaran & Bougie, 2011), while secondary data is provided from well-established sources (e.g., literatures, articles, scientific papers and websites).

The study using the Japaness product as the most popular product in Indonesia such as automotive, electronic, food and bevergare, etc. Before performing the main test, the authors conducted pilot tests twice. First test is to determine the affinity country as the object of the research. Second test aims to test the validity and reliability of the measurement instrument model before conducting the main test. First test is conducted by distributing questionnaires to 41 undergraduate students of Faculty of Economics and Business, Universitas Indonesia. The result of this test reveals Japan as the country of affinity. Next, the second test is conducted to 30 undergraduate students of Faculty of Linguistics, Universitas Indonesia. The result of this test reveals that the measurements of the study have sufficient validity and reliability for further assessment.

The next step is a preparation of main study. The samples of the main study are totaled 164 respondents. The respondents are selected by absolute prerequisite with the possession of interest about language and culture of host country. In this term, we select an individual who have interest on Japanese culture & language. This is based on pilot study which conducted in early time to determine the host country that appeal to the population of samples. According to Crookes and Schimdt (1991), an individual who learn foreign language as its second languages will have positive attitude to the target language group and desire to connect and being close with the community. The desire has in common with the meaning of affinity therefore, the authors tries to integrate the two concepts together. Furthermore, to ensure the respondents’ characteristics are relatively homogenous, then we select respondents who domiciled on the Greater Jakarta area. In this manner, the nuisance variables can be controlled (Sekaran & Bougie, 2011). After the data collection process is complete, the next stage is data processing. Software used to assist in processing the data of this research is LISREL 8.7. Confirmatory data analysis study has three stages in the data processing, firstly a test of overall compatibility model. Second is an analysis of model validity and third is an analysis of model reliability. First stage is to determine whether there are offending estimates such as negative error variance and standardized loading factor below the standard. Second stage is to analyze the validity of model by referring the t-values and standardized loading factor scores of observed variables of the model. The last stage is to analyze the reliability of model by referring the composite reliability and variance extracted scores from the standards.

5. Results

The explanation of the main study results begin with descriptive data. It presented by several categories, such as type of questionnaires distribution, gender, age and the domicile place of respondents. Begin with the type of questionnaires distribution, a total of 48.78% of questionnaires distributed physically by visiting campus and central library of Universitas Indonesia, whilst 51.22% is distributed by online via direct links. Direct link has valid period and feature that restrict access to the link from the same URL only after they already submit the answers. Once the collection period ends, then we disable the direct link to prevent further access. The other classification is the gender consisted of 40.24% male and 59.76% female with the average of age is 22.2 years with the majority of respondents (36%) live in Jakarta.

Consumer affinity (CA) was measured using seven items developed by Netemeyer, Bearden and Sharma (2003). It is measured on five-point-scale from 1 (slightly), 3 (moderate) and 5 (extremely). The CA measurement item focusing on positive emotion and eliminate the negative one to ensure the optimum responses. We measured Consumer Ethnocentrism (CET) with five items using 5-point Likert scale. The CET items was originally developed by Shimp & Sharma (1987). To ensure the comprehensive measurement, the country image was operationalized as the combined of macro and micro country image. We measured macro country using five semantic differential scale (Martin & Eroglu, 1993). Product country image was used as micro country image specifically using four semantic differential scale (Nagashima, 1970). For measured perceived risk (PR) and willingness to buy (WTB), we used three item with 5-point Likert scale. The following [Table 1] below is a result of the validity and reliability test of all measurements.

[Table 1] > Measurements

Source: Authors

From the [Table 1] we can figure out there were several indicators of measurement that not qualify the standards. According to Malhotra (2010), standardized loading score must above 0.5 at minimum in order to be statistically significant, although the ideal number is 0.7. We then eliminate the indicators which are not valid for further analysis. We also indicate all t-values are higher than the critical point (>1.96). The constructs also must possessed a qualification for CR (>.60, Bagozzi & Yi 1988) and AVE (>.50, Fornell & Lacker 1981) as well. Next stage, we investigated the overall goodness of fit of model. Overall fit was satisfactory (=208.69, d.f.=192; RMSEA=.023; SRMR=.069; NNFI=.98; GFI=.90; and CFI=.98), and all relevant loadings were substantial and fairly significant (see [Table 1]). We assessed discriminant validity by comparing the AVEs with the squared correlation between each pair of constructs and by constructing 95% confidence intervals around the interconstruct correlations. In conclusion, Firstly PCI can be explained by the construct of CET and CI by only 11%. Thus, there are unobserved variables which can explain the residual of 89% correlation. Secondly, PR can be explained by the construct of PCI and CET by only 29%. Thus, there are unobserved variables which can be explain the residual of 71% correlation. Lastly, WTB can be explained by the construct of PCI and CET by only 36%. Thus, there are unobserved variables which can be explain the residual of 64% correlation. [Table 2] shows the intercorrelations among the study constructs.

[Table 2] The Structural Model

Source: Authors

Final stage is the analysis of t-value model coefficients. Analysis of the t-value is stated by t-value of relationships between constructs > 1.96 considered statistically significant ([H1]accepted), while the relationship with the t-value which does not meet the requirement is considered uncorrelated ([H0] accepted). Based on the results, there were 2 of 9 coefficients in the structural model which have not eligible. In conclusion, from 9 hypotheses constructed of the study, there were two hypotheses rejected because the data did not support the hypotheses. [Figure 2] provide the summary of overall analysis of the study.

[Figure 2] The Structural Model

6. Discussion

In this study, we conceptually and empirically advance Oberecker, Riefler, and Diamantapoulos’s (2008) proposed consumer affinity construct. On the basis of this conceptualization, we developed and rigorously tested a consumer affinity scale and provided evidence for the latter’s validity and reliability. Furthermore, we incorporated consumer affinity, country image, and consumer ethnocentrism in an overall conceptual model and examined the impact on perceived risk and willingness to buy products from the affinity country. From a theoretical perspective, our study contributes to international marketing literature by refining the conceptualization of the consumer affinity construct and highlighting its multidimensional nature. According to previous study, the results indicate that affinity feelings outweigh ethnocentric tendencies in affecting perceived risk and willingness to buy product from the affinity country. Our results from this study are less consistent with the previous study. A significant negative influence were affected the relationship of consumer affinity toward perceived risk (t=-3.34, p<.01). This outcome indicated that intense feeling of affinity is able to reduce the risk perception towards foreign products (Oberecker & Diamantapoulos, 2011). However, the absence of a positive effect on the relationship of consumer affinity toward willingness to buy indicated the result was inconsistent and did not fit the initial expectations. Previous study had proved that consumer affinity significantly affect people willingness to buy foreign products (t=2.71, p<.01). The cause of insignificant relationship between two constructs is presumably because the majority of respondents experienced an error in filling out the questionnaire due to imperfect translation or the respondents itself whose were not able to understand correctly. It is proven by poor score of reliability of the questionnaire. Vida and Reardon (2008) speculate that consumer affinity for a country translates directly into shopping preferences. However, our findings empirically did not support this proposition and show that emotions of affinity might not role in decision-making situations in which rational criteria are expected to be predominant.

Another theoretical insight provided by our study on the interaction of consumer ethnocentrism towards product country image, perceived risk and willingness to buy. Firstly, there is a positive effect from consumer ethnocentrism towards product country image which in previous study the relationship between two constructs were not proven empirically (t=.48, p-value: n.s.). Secondly, consumer ethnocentrism was able to affect consumers in term of increasing the level of risk perception (Jung & Khau, 2006), and makes consumers only have one point of view that buying foreign products are risky and cause consumers reluctant to buy foreign products. In previous study, Oberecker and Diamantapoulos (2011) failed to proof this relationship. Finally, consumer ethnocentricsm (CET) negatively influenced the customer willingness to buy. This result inline with the work of Durvasula, Andrews dan Netemeyer (1997).