A Model of Factors Affecting Foreign Brand Trust

This study aims to investigate the effect of perceived quality on foreign brand trust and the moderating role of consumer ethnocentrism and negative subjective norms on foreign brand trust. A self-report survey of 170 Iranian home appliances shoppers was administrated. The results showed that perceived quality has a positive direct effect on foreign brand trust (β=0.623, t= 9.8) but consumer ethnocentrism and negative subjective norms have no significant effect on perceived quality. The analysis also indicated that consumer ethnocentrism and negative subjective norms moderate the relationship between perceived quality and foreign brand trust negatively, respectively (β=-0.229, t= -2.39; β=-0.131, t= -2.04).


IntroductIon
Today, globalization phenomenon has created numerous opportunities for companies to go beyond their national borders.Iran is one of the attractive markets for Korean home appliances due to the population of approximately 70 million.Although there are some Iranian producers of household appliances including: Pars Electric, RTC, Pars Khazar, OSUN, Sinjer, Yakhsaran, NiKala and so on, but foreign household appliances brands are dominant throughout Iran's market.According to chief of Iranian home appliance association, about 95% of Iranian market has been captured by foreign brands (Hamshahrionline.ir, 2010).Among foreign brands, South Korean brands such as: LG and Samsung are leaders in Iran household appliances market.Iranian home appliances showed a low quality compared to its high price (Hamshahrionline.ir, 2010).In recent years, South Korea is among the top six exporter countries to Iran.This research investigates the relationship between perceived quality with brand trust and moderating role of Iranian consumer ethnocentrism and negative subjective norms.According to Delgado-Balester ( 2004) "trust has seldom been distinctly examined in end-consumer studies, especially those concerning consumer-brand domains" hence, the first objective of this study is to examine the direct effect of perceived quality on brand trust.The second goal is researching about moderating effects of consumer ethnocentrism and negative subjective norms on the relationship between perceived quality and brand trust.

Perceived quality (PQ) and brand trust (BT)
In brand literature, researchers claim that brand is sign of quality.For example, Kotler and Armstrong (2004) contended that brands are means and signs of quality.Based on previous studies, brand name is antecedent of perceived quality but this claim is not true because brand is based on continuous quality.Improving quality continually results in brand name.Therefore, before a brand becomes famous, quality could not infer from brand name.However, after establishing a brand name that is outcome of ongoing high quality, brand appears to be a sign of quality.Quality is a subjective phenomenon.Consumer assessment of product quality determines its quality.In marketing literature, the construct of perceived quality has been widely acknowledged as the primary driver of purchase intention ( Jacoby & Olson, 1985).In general, perceived quality is defined as a consumer's evaluation of a brand's overall excellence based on intrinsic (e.g., performance, durability) and extrinsic cues (e.g., brand name, warranty) (Kirmani & Baumgartner, 2000).There are many conceptual and experimental studies about investigating perceived quality as an independent variable.Iglesias and Guillen (2004) found positive impact for perceived quality on satisfaction.Several studies have argued that a product's quality is a crucial determinant of future purchase and word-of-mouth intentions (Baker & Crompton, 2000;Cronin & Taylor, 1994;Parasuraman et al., 1988).Unfortunately, there are no studies about effect of perceived quality on brand trust explicitly.Delgado-Balester and Aleman (2005) said "trust a brand implicitly means that there is a high probability or expectancy that the brand will result in positive outcomes for the consumer".Reliability is as focal component of brand trust (Delgado-Balester & Aleman, 2005).Usually, high reputation stems from high quality hence high quality can result in foreign brand trust.If a brand has a high quality, reliability will be guaranteed.in other words brand quality is a sign of brand reliability.If consumer trusts the brand, it can be concluded that consumer believed that the brand has high quality.Thus, one can say that: H1: Korean brand perceived quality (PQ) has a positive impact on the Korean brand trust (BT) among Iranian consumers.

Consumer ethnocentrism (CE) and brand trust (BT)
The "consumer ethnocentrism" concept was developed first time by Shimp and Sharma (1987) as a domain-specific sub-set of ethnocentrism.Shimp and Sharma (1987) viewed consumer ethnocentrism as "a domain specific concept for the study of consumer behavior with marketing implications".Consumer ethnocentrism was regarded as a "unique economic form of ethnocentrism that captures the beliefs held by consumers about the appropriateness and indeed morality of purchasing foreign-made products" (Shimp and Sharma, 1987).Consumers have different levels of ethnocentrism which impacts attitudes and intentions towards buying foreign goods (Klein, 2002;Ulrich & Zuzana, 2003).In foreign brand purchasing, one of the most studied variables is consumer ethnocentrism.As Supphellen and Gronhaug (2003) found "When foreign brands are commonly regarded as superior to domestic brands, consumer ethnocentrism has positive impact on attitudes toward domestic brands but no effect on attitudes toward foreign brands".Previous studies (Shimp & Sharma, 1987;Klein et al., 1998) have found negative relationship between consumer ethnocentrism and purchasing foreign products.Other researchers also found the negative relationship between ethnocentrism with purchase foreign products and tendencies to evaluate them.Therefore, it is proposed that: H2: Iranian consumer's ethnocentrism (CE) has a negative effect on the Korean brand trust (BT)

Negative subjective norm (NSN) and brand trust (BT)
Social environment surrounds all life.Society accepted opinions and norms have pivotal role in decision-making and beliefs.Subjective norms reveal important referents beliefs such as: family, friends or peers that their norms impact judgments.Subjective norms can be positive or negative.Subjective norm is an integral component of theories of reasoned action (Ajzen & Fishbein, 1973), as well as planned behavior (Ajzen, 1991).According to Fishbein & Ajzen (1975), subjective norms refer to "perceived pressures on a person to perform a given behavior and the person's motivation to comply with those pressures."In both of them, subjective norm has immediate effect on behavioral intention.Negative subjective norms can result in people's fear of making purchases and make their judgments negative.Brand trust is a consumer judgment about reliability of a specific brand.When reference groups have negative norms and beliefs about foreign brand it can reduce brand trust.In high context cultures such as IRANIAN culture, subjective norms and social environment have more power on consumer behavior and decisions, especially on brand trust.Although there is no study about impact of subjective norm on brand trust, but the effects of subjective norms on other variables examined in many previous researches.Many researchers have called for greater inclusion of social factors, especially social norms, in the study of trust (Williams, 2001;Lewicki et al., 1998).Jeffries and Becker (2008) found that subjective cooperative norms have positive relationship with attitudinal trust.Kim et al. (2009) indicated that subjective norms have positive effect on attitude toward the use and intention to reuse b2c ecommerce website.Thus, it is hypothesized that: H3: Negative subjective norms (NSN) have a negative effect on the Korean brand trust (BT) among Iranian consumers.

Moderating role of consumer ethnocentrism
The relationship between perceived quality and brand trust moderates negatively by consumer ethnocentrism.In a study on the US and Korean consumers, however, Steenkamp et al. (2003) found that the level of consumer ethnocentrism moderated the effect of country of origin on perceived product quality.When consumer ethnocentrism is low their beliefs and attitudes are more positive.in other words, relationship between perceived quality and brand trust is more positive in consumers with low degree of ethnocentrism in comparison to consumers with high degree of ethnocentrism.Thus, it is hypothesized that: H4: The relationship between Korean brand perceived quality (PQ) and Korean brand trust (BT) among the Iranian consumers who have a low degree of ethnocentrism is greater than the consumers that have a high degree of ethnocentrism.

Moderating role of negative subjective norms
People live in a social context and affected by the social norms surrounding them.Social context has a very strong impact on individual attitudes and beliefs and Iranians are highly contextually affected.Subjective norms have been studied in theory of reasoned action and planned behavior extensively in many researches.Subjective norm is the antecedent of behavioral intention in Ajzen and Fishbein's theory of reasoned action and theory of planned behavior.Wong and Zhou (2005) found that social compliance has a positive effect on the relationship between product quality and purchase intention among Chinese consumers.H5: The relationship between Korean brand perceived quality (PQ) and Korean brand trust (BT) among the Iranian consumers who have a low degree of negative subjective norms is greater than the consumers that have a high degree of negative subjective norms.

Data Source and Sample
200 questionnaires were distributed among home appliances shoppers in AMINHOZOR Bazaar (Iran's most famous market for home appliances) in Tehran from Jun 15 to Jun 25 2013.180 questionnaires were backed (response rate equals 0, 90%).Ten questionnaires have missing data and excluded from analysis.Self-report method used for completing questionnaire.The respondents were aged 23 to 68 years.57.3 % of sample was males and 42.7% were females.12.7% of respondents were under 30 years old, 84.7% were between 30-50 years old and 2.5% were between 50-70 years old.

Measures
Consumer ethnocentrism scale contains 6 items adopted from Shimp and Sharma (1987).4 items for measuring negative subjective norms were designed based on Hansen et al. (2004) and Wilson et al. (1975).7 items for measuring perceived quality of Korean home appliances were designed based on Snoj et al. (2004).Korean brand trust assessed by 8-items scale developed by Delgado-Balester (2004).Appendix A shows the questionnaire items.The two famous Korean brands in Iran home appliance market: LG and Samsung, hence the perceived quality and brand trust were specific to these brands.All items scored on a five-point scale ranging from completely disagree (1) to completely agree (5).

Means Standard deviation and correlations
Means standard deviation and correlation among research variable reported in Table1.The perceived quality has the highest mean (3.96) and consumer ethnocentrism has the lowest mean (2.3).The means score showed this important reality that ethnocentrism of the sample was under the mean and indicated the weak ethnocentrism.It may be because of diversity in Iranian sub cultures such as Kurdish, Turkish, Fars, Baloch and other minor cultures.As we anticipated, brand trust has positive correlation with perceived quality(r = 0.70, p<0.01).Correlation results show that there is negative relationship between brand trust and consumer ethnocentrism(r = -0.22,p<0.05).Negative subject norms toward Korean household appliance has positive correlation with consumer ethnocentrism (r = 0.19, p<0.01).
Tab At the next step, the conceptual model and hypotheses needs to be tested.According to Sharma et al. (1981), there are two principle methods for detecting moderating effects: sub-group and moderated regression.Most of past studies use moderated regression by centering the latent variables and deploy the ordinary least squares (OLS) procedure.One of the biggest weaknesses of this analyzing strategy is ignoring factor structure.When a latent variable is used as a moderator for another latent variable, the interaction terms should be computed for the array of manifest variables of two latent variables (independent and moderator).In this situation, moderated structural equations models (MSEM) need to be deployed.MSEM cannot be done in traditional analyzing methods and software, hence PLS Graph version 3.0 (developed by Wynne W. Chin in 2001) applied that is capable of removing this problem using Kenny and Judd (1984)

Discriminant validity
To study discriminant validity, the average variance extracted (AVE) in exploratory factor analysis in pairs of all the constructs should be estimated.Discriminant validity exists when AVE is greater than the squared correlation between pairs of factors (Fornell and Larcker, 1981).Results of the tests of AVE found satisfactory.The upper diagonal of Table 2 shows the results of the tests and the results of the intercorrelations squared between the constructs shown in the lower diagonal. Tab

Uni-dimensionality check of the constructs
Uni-dimensionality means there is only one construct underlying the data, not several.Uni-dimensionality check is necessary when the manifest variables connected to their latent variables in a reflective way (Tenenhaus et al., 2005).There are three tools available for the uni-dimensionality check of a construct: principal component analysis of the construct Cronbach's α and Dillon-Goldstein's ρ.A construct is essentially uni-dimensional, if the first eigenvalue of the correlation matrix of the construct manifest variables is larger than 1 and the second one smaller than 1, or at least very far from the first one.A construct is also considered as uni-dimensional when Cronbach's α and Dillon-Goldstein's ρ values are larger than 0.7 (Tenenhaus et al., 2005).2. R square of the tested model equals 0.571.In other words, 57 percent of brand trust variance resulted from the research variables and 43 percent attributed to unknown variables.Perceived quality has a strong positive effect on brand trust (β = 0.62, t = 9.75), hence, hypothesis H1 was confirmed.Based on the results, consumer ethnocentrism does not have main effect on brand trust (β = -0.035,t = -0.81),therefore, the hypothesis H2 was rejected.Negative subjective norm has no main effect on brand trust (β = -0.05,t = -1.04),and then H3 was not confirmed.Two interaction terms made for detection of moderating effects of consumer ethnocentrism and negative subjective norm on relationship between perceived quality and brand trust.Results shows that consumer ethnocentrism has negatively moderated the relationship between perceived quality and brand trust (β = -0.22,t = -2.40,p<0.05).Perceived quality has stronger impact on brand trust in consumers who have low score in ethnocentrism than consumers who have higher score in ethnocentrism.PLS output depicts that negative subjective norms negatively moderates the NSN relationship between perceived quality and brand trust among Iranian consumers (β = -0.13,t = -2.32,p<0.05).Perceived quality has stronger impact on the brand trust in the consumers who have a low score in the negative subjective norms than the consumers who have a higher score in the negative subjective norms.Sharma et al. (1981) framework for identifying the type of moderator variable applied.First model extended to include main effects of consumer ethnocentrism and negative subjective norm on brand trust (Figure 3).In this model, R square for perceived quality as a dependent variable of negative subjective norm and consumer ethnocentrism is only 7 percent.This means that only 7 percent of perceived quality can attribute to independent variables including: negative subjective norm and consumer ethnocentrism and 93 percent is the function of unknown variables.In the extended model, consumer ethnocentrism has a negative main effect on the perceived quality (β = -0.24,t = -3.84,p<0.05) and consumer ethnocentrism moderates the relationship between perceived quality and brand trust negatively (β = -0.22,t = -2.40,p<0.05).Therefore, consumer ethnocentrism falls in the quasi moderator category.Negative subjective norms falls in the pure moderator category because it moderates the relationship between perceived quality and brand trust negatively (β = -0.13,t = -2.32)and does not have an impact on the perceived quality (β = -0.05,t = -0.78).In the extended model, perceived quality has a strong positive effect on the brand trust (β = 0.62, t = 9.28), consumer ethnocentrism has not main effect on brand trust (β = -0.03,t = -0.82),and negative subjective norm has not main effect on brand trust (β = -0.05,t = -1.02).The extended model shows that consumer ethnocentrism moderates the relationship between perceived quality and brand trust negatively (β = -0.22,t = -2.48,p<0.05).In the extended model, negative subjective norms negatively moderates the relationship between perceived quality and brand trust (β = -0.132,t = -2.78,p<0.05). -

dIscussIon and conclusIon
This research showed negative impact of consumer ethnocentrism on the evaluation of the quality of foreign brand (β = -0.29,t = -4.35,p<0.05).This result is consistent with previous researches such as Balabanis and Diamantopoulos (2004), Klein et al., (1998), Shimp and Sharma (1987) and tong and li (2013).The more ethnocentrism the lower perceived quality.There is no study about moderating effect of consumer ethnocentrism on the relationship between perceived quality and brand trust but the negative effect of Iranian consumer ethnocentrism is similar to Liu et.al (2005) that found Chinese ethnocentrism has negative moderating effect on attitudes towards signs in both English and Chinese bread shop.Iran has a high context culture.In this type of culture, subjective norms and social environment has a pivotal role for foreign brand success and failure.Further research can be conduct for identifying other consequences of foreign brand trust such as: perceived value, foreign brand advertising, post-purchase services and guarantee (especially for durable goods), prior experience about foreign brand.Studies in the area of foreign brand trust consequences including intention, satisfaction, actual purchasing, foreign brand loyalty and word of mouth advised.Studying foreign brand trust or purchasing about the specific foreign brand such as: U.S and England are of the interest because of Iranian negative perceptions about the political and economic actions of these governments.One of the limitations of this study is about external validity because the sample is limited to AMINHO-ZOR Bazzar customers.Another limitation of this study is its relatively small sample.
procedure for MSEM.Another problem in investigation of moderating effect is pointing out Chow test.Chow test is a statistical test developed byChow (1960).Chow test reveals if there is any significant difference in regression coefficients of two or more groups.If the answer is positive then moderated regression or subgroup analysis can be used and it should try to interpret interaction terms by beta coefficients.Two Chow tests conducted to see if the set of coefficients in the regression for high ethnocentrism and negative subjective norm was significantly different from the set of coefficients in the regression for low ethnocentrism and negative subjective norms.
. 2 -Discriminant validity test using AVE and correlation methods.Source: own.
were not significant and can be omitted from the model.So we can draw the final model (fig-ure4).In the final model consumer ethnocentrism has a negative main effect on the perceived quality (β = -0.29,t = -4.35,p<0.05).Consumer ethnocentrism also moderates the relationship between perceived quality and brand trust negatively (β = -0.22,t = -2.52,p<0.05). in this model perceived quality has a strong positive effect on the brand trust (β = 0.63, t = 9.83).Negative subjective norms moderates the relationship between perceived quality and brand trust negatively (β = -0.15,t = -2.39,p<0.05).