The Consumer Ethnocentrism And Country Of Origin Effect

Published Date: 02 Nov 2017

Andreea Iacob [1] and Lartey G. Lawson [2] 

Consumer ethnocentrism and Country of origin effect:

The brand perception of Romanian consumers concerning Danish beer products

Purpose of the article: The present study using the Consumer Ethnocentric Scale (CETSCALE) seeks to evaluate the level of consumer ethnocentrism (CE) and its implication on the consumers’ evaluation of a foreign brand. Furthermore it seeks to test whether country of origin matters for ethnocentric and non-ethnocentric consumers. It builds up on a previous study, undertaken by the authors, concerning the impact of consumer ethnocentrism on the general brand perception of the Romanian consumers, with the aim to test the ethnocentric level and country of origin effect on a specific foreign beer brand.

Design/methodology/approach: Data were collected using a questionnaire that was administered to 325 respondents using the mall intercept approach to select them (with 317 usable questionnaires after data editing). We used exploratory and confirmatory factor analysis, which justifies the uni-dimensionality of the CETSCALE. Cluster analysis was performed in order to divide the sample into ethnocentric and non-ethnocentric consumers. Principal Component Analysis to estimate component factor scores, and used hierarchical regression analyses to test our hypotheses.

Findings: The results showed that the sample of Romanian consumers can be characterized as marginally ethnocentric.

Originality/value: Although the research on consumer ethnocentrism and country of origin effect is quite substantial in the area of consumer research, yet the literature has focused almost exclusively on high involvement products. The current study..

Keywords: Consumer ethnocentrism, country of origin, Romania, brand perception.

Paper type: Research paper

Introduction

International marketing researchers have long been concerned with determining whether consumers are predisposed towards a preference for domestic products, as opposed to foreign products (Evanschitzky at al., 2007). Altintaş and Tokol (2007) present that one factor which may influence a consumer’s decision to buy a domestically-produced rather than a foreign-made product is the concept of consumer ethnocentrism (CE). Alongside CE, empirical studies on the effects of country of origin (COO) on brand perception shows that usually consumers perceive products made in developed countries to be of higher quality compared with products made in emerging markets (Pappu, Quester, and Cooksey, 2007; Josiassen and Harzing, 2008; Usunier and Cestre, 2008). In a study conducted by Hamin and Elliott (2006) in countries such as Mexico, the Philippines, Jordan and Nigeria, the consumers seem to rate products from more developed countries more highly than domestic products. These authors also note that alongside the generalized preference for products originating from developed countries, there is also evidence to suggest that some consumers will always prefer to buy products manufactured in their home country (Beverland, 2001; Pecotich and Rosenthal, 2001; Dmitrovic, Vida and Reardon, 2009). When a consumer makes a reference to the country of origin of a product the perception is automatically drawn by the quality of the product, thus it can be called country-specific (Balabanis and Diamantopoulos, 2004; Lantz and Loeb, 1996), while the extent to which the effect depends on the product category, is called product-specific (Balabanis and Diamantopoulos, 2004; Watson and Wright, 2000; Sharma et al., 1995, in Piron, 2000; Smith, 1993; Cordell, 1992).

In spite of the extensive research about COO effects and CE on consumers in developed markets to a greater extent than in emerging markets, some inconclusive findings arise: some studies show a negative effect of consumer ethnocentrism (CET) on the evaluation of foreign products by consumers in emerging markets (Klein et al., 2006), whereas others show either no significant effect (Huddleston et al., 2001), or different effects based on product type (Hamin, 2006), perceptions about domestic products and conspicuous consumption tendencies (Wang & Chen, 2004), and attitudes towards brands and advertising (Reardon et al., 2005). Thus there is no consensus on the role of CE in COO effects for consumers in emerging markets.

When it comes to COO effects, there are two issues that need to be addressed. Firstly, most studies have focused on high involvement products where consumers look for other cues such as price or design when making their purchase decision. To date, there have been few studies on the impact of consumers’ COO perceptions on low-involvement products. For example, one study conducted by Chryssochoidis et al. (2006) on Greek consumers, showed that ethnocentrism affects not only consumer beliefs, but also how the quality of domestic and foreign products are evaluated, thus making the appearance of the COO effect more obvious at the level of the product type (product-specific). Thus, it is not clear what role COO plays in shaping consumers’ brand perception of low involvement foreign goods or whether its effect is the same for low-involvement products as for high-involvement products (Ahmed et al. (2002).

Second, the majority of COO studies have been conducted in developed countries. Some studies which were conducted in developing countries are: Zhang (1996)Piron (2000), Kaynak and Hyder (2000), Kaynak and Kara (2002), Wang and Chen (2004), Ahmed and d’Astous (2007); Pecotich and Ward (2007). COO research in developing or non-western countries, especially Eastern European markets remains less explored.

Taking the above into consideration, this study examines the level of consumer ethnocentrism (CE) and its implication on the consumers’ evaluation of a foreign brand. Furthermore it seeks to test whether country of origin matters for in connection to the consumers’ brand perception. It builds up on a previous study, undertaken by the authors, concerning the impact of consumer ethnocentrism on the general brand perception of the Romanian consumers, with the aim to test the ethnocentric level and country of origin effect on a specific foreign beer brand.

The rest of the paper is structured as follows. In section 2, we shortly reviewed the extant literature on consumer ethnocentrism, country of origin and brand perception and formulate hypotheses for the empirical investigation in section 3, again drawing on knowledge from the existing literature. This is followed by section 4, where a presentation of the methodological approaches used for the empirical investigation is presented and the results of the empirical analyses are reported. Section 5 discusses the findings, noting their implications for marketing strategy formulation.

Literature review and hypothesis development

The Country-of-origin effect on brand perception

H1: A developed country of origin has a positive impact on brand perception.

Consumer ethnocentrism and brand perception

H2: The more ethnocentric Romanian consumers are the more likely it is that they will have a negative perception of foreign brands.

The Romanian beer market

Methodology

This paper is the third in a series of three papers. The first paper is titled, Brand Perception, Ideology and Ethnocentrism: How strong are the links in Post-Communist Romania. The paper focused on identifying and quantifying the relationships between Ideological preferences and ethnocentrism as well as between Ideological preferences with ethnocentrism and perception of branding. The second paper investigated the impact of demographic factors on the perception of branded products after accounting for Consumer Ethnocentric Tendency and ideological preferences.

While the first two papers focused on the general perception of the consumers towards brands, this third paper is more specific in choosing a particular Western brand and the aim is to examine the effect a developed country of origin has on a low involvement product, the Tuborg beer brand, and how ethnocentric tendencies affect the consumers’ perception of the brand.

Data Collection Process

The questionnaire was first written in English and then translated into Romanian by one of the authors (who is a Romania by birth) and then back-translated by another Romanian fluent in both English in Romanian to make sure that all idioms and Romanian expressions were correct. It was then pretested on Romanian students studying at Aalborg University. The questionnaire was then administered to the respondents in Bucharest, Cluj-Napoca and Sibiu – three major Romanian cities. A mall intercept method was used to sample 325 respondents for the study. The mall intercept approach has been used in a number of previous consumer ethnocentric studies (Saffu and Walker, 2006). The data quality from these previous studies has been found to be comparable to telephone interviews (Bush and Hair, 1985). Respondents filled in nearly 90 per cent of the questionnaires without assistance from the researchers. The data were screened for univariate outliers and data editing was performed in order to identify omissions and errors in responses. Eight questionnaires were deleted in the process, leaving 317 questionnaires for the final analysis.

Survey Instrument

The questionnaire was developed containing three major sections. The first section covers the demographic data, specifically age, gender, education and household income.

The second section of the questionnaire contains questions relating to the beer brand Tuborg. Each of the items was evaluated using a 5 item Likert-type scale. Responses to the Likert scale ranged from strongly disagree (coded 1) to strongly agree (coded 5).

The third section of the questionnaire consists of questions related to consumer ethnocentrism. CETSCALE (Consumer Ethnocentric Tendencies Scale), a 17-item questionnaire, developed by Shimp and Sharma (1987) is being used to measure consumer ethnocentrism which is based on the same 5 ranged Likert-type scale as described earlier. This instrument has been used previously in various cultural and cross-cultural settings such as: the USA, France, Germany, Korea, and Japan, but also CEE countries (Huddleston et al., 2001, Lindquist et al., 2001). High CETSCALE scores indicate a high level of ethnocentrism, while low CETSCALE scores indicate a low level of ethnocentrism. Justification for using the scale was based on confirmation of its reliability and validity by a number of researchers (Netemeyer et al., 1991; Saffu and Walker 2006, Saffu and Scott, 2009).

Table 1. The percentage distribution of the socio-demographic profile of the 317 respondents

Demographic characteristics

%

Gender

[Female]

57.4

[Male]

42.6

Age in years

< 26

33.1

26 - 45

42.9

> 45

24.0

Education

[Primary]

6.3

[Secondary]

28.1

[University]

46.4

[Post University]

19.2

Occupation

[Students]

26.2

[Unemployed]

16.1

(Employee]

11.0

[Employed with Higher Education]

34.1

[Employer]

12.6

Monthly income a

Zero

20.2

< 1501

38.8

1501- 2500

23.3

> 2500

17.7

Note:a in RON; 1 RON = 4.4 EUR

Data Analysis

The data analysis was in four steps. First, we did a descriptive analysis to get an overview of the respondents’ replies to the items in the questionnaire. Second, we used the Cronbach’s alpha to analyse the internal consistency for all items. Finally, we conducted hierarchical multiple regression analyses to investigate our hypotheses. All analyses were done using the SPSS version 19.

Brand perception

Three characteristics for the perception of the Tuborg brand have been evaluated by the Romanian consumers (see Table 2). A total of 51.7% of the respondents agree with the fact that they trust the quality of the brand, while 48.9% believe that the TUborg brand is attentive with their needs. It further shows that 33.4% of the respondents would choose Tuborg instead of other beer brands, compared to 18.8 who disagree with this statement.

Table 2. Percentage distribution of respondents scores and mean scores for brands perception (n = 317)

Instrument items a

Totally

disagree

Rather

disagree

Neither agree nor disagree

Rather

agree

Totally

agree

Mean

Std

1. I trust the quality of the products of this brand

.9

4.1

43.2

27.4

24.3

3.70

.915

2. I think that this brand is always looking to improve its products to better satisfy consumers’ needs.

1.3

2.2

47.6

30.9

18.0

3.62

.847

3. I prefer this brand over other brands in the same product category.

7.9

10.7

47.9

20.2

13.2

3.20

1.057

Note: a The response format is a five-point Likert-type scale (1= strongly disagree, 5= strongly agree.)

In order to test the hypothesis of whether a developed country of origin, in this case Denmark has a positive outcome on how Romanian consumers perceive the brand a cross tabulation between these items is needed. The pearson chi square test revealed a significance level of .000 for all three items which denoted the brand perception of Tuborg, thus the analysis can be performed. When looking at the table below, the results of over 62% show an obvious remark that consumers, who have a positive perception towards the brand Tuborg, also associate the Western country of origin, Denmark, as having a positive effect on their overall brand perception, thus hypothesis 1 is approved.

Table 3. The percentage distribution between Country of origin and brand perception items

Tuborg a Danish beer brand

"Trust the quality of the brand"

"The Tuborg brand is attentive with consumers’ needs"

Prefer Tuborg over other brands

Disagree

Neutral

Agree

Disagree

Neutral

Agree

Disagree

Neutral

Agree

No effect

1.2

96.5

2.4

1.2

92.9

5.9

3.5

94.1

2.4

Positive effect

0

17.5

82.5

1.8

24.6

73.7

12.3

25.4

62.3

Negative effect

12.7

29.7

57.6

6.8

37.3

55.9

35.6

36.4

28

Consumer ethnocentrism and brand perception

Results of the Hierarchical Regression Analyses

As mentioned earlier, we used hierarchical regression models to examine the links between country of origin, consumer ethnocentrism, and brand perception. According to Kahane (2001) hierarchical regression models allow researchers to control for the possible effect of a set of variables and then investigate if other variables can predict a significant amount of variance in a dependent variable. For example, in this study, we need to control for "anti-monopoly" and "free market economy" factors while we investigate the link between "communist ideology" and consumer ethnocentrism in Romania. This accounts for our choice of the hierarchical regression model. The estimated functional form for our first model is as follows:

Where

i represents an individual respondent; Y is a vector of components factor scores of interest. β0 is the grand mean, Zk are first level independent variables of interest, Xj are the second level variables of interest. The β0, βj and γk are the parameter coefficients to be estimated and εijk the error term assumed to be normally distributed N(0,σ2). This general model is specified for ethnocentric constructs - "taxing imports", "buying Romanian products", and "supporting the Romanian economy" and for brand perception constructs - "buying branded products", "quality of branded products ", and "Romanian brands".

The results for the ethnocentric constructs are presented in Table 9. They show that Romanian consumers with communist ideological preference turn to support "taxing imports", "buying Romanian products", and "supporting the Romanian economy". The sizes of the unique R2 for these relationships are 7.9 %, 7.8 % and 15.4 % for "taxing imports", "buy Romanian products", and "support Romanian economy" respectively. That is, when we disregard consumer "anti-monopoly" and "free market" ideological preferences, consumers’ communist ideological preferences account for about 8 % of the variations in consumer ethnocentric tendencies reflected in consumers’ attitude to "taxing imports" and "buy Romanian products" as well as 15 % of the variation in "support Romanian economy".

Furthermore, we found that consumers with anti-monopoly ideological orientation are likely to exhibit ethnocentric tendencies expressed in terms of "buy Romanian products" and "support Romanian economy". The unique R2 of these relationships are 4.6 % and 1 % respectively. Finally, the relationships between consumers’ "free market" ideological disposition and ethnocentric tendencies expressed in "taxing imports" and "support Romanian economy" are negative. Their unique R2 are 1.2 % and 2.7 % respectively. These findings therefore support hypothesis 1 – i.e. communist ideology has direct impact on Romanian consumers’ ethnocentric dispositions.

The results of the impact of ideology and ethnocentrism on brand perception are presented in Tables 10 and 11. They show that without accounting for consumer ethnocentrism (Table 11, Model 1), there is a positive and statistically significant relationship between consumers’ communist ideological dispositions and their willingness to "buy Romanian brands" (p<0.001). Those with and "anti-monopoly" orientation also show a moderately significant link with the "buy Romanian brand" (p<0.06).

Table 4. Distribution of respondents and mean estimates of Brand perceptions by COO and CE

 

Respon-dents

 

"Trust the quality of the brand"

"The Tuborg brand is attentive with consumers’ needs"

Prefer Tuborg over other brands

%

Mean

SE

Mean

SE

Mean

SE

Total

317

100

0

0,051

0

0,047

0

0,059

COO effect  

No effect

85

26,8

3.035

0,084

3.047

0,081

2.965

0.105

Positive effect

114

36,0

4.289

0,072

3.951

0.070

3.763

0.091

Negative effect

118

37,2

3.610

0,071

3.450

0.068

0.332

0.089

Ethnocentric Tendencies (CE)

Tax on Imports, component 1

Low-CE

263

83

3.700

0.057

3.608

0.052

3.183

0.065

High-CE

54

17

3.704

0.125

3.685

0.115

3.296

0.144

Buy Romanian Products, component 2

Low-CE

265

83.6

3.657

0.056

3.596

0.052

3.140

0.064

High-CE

52

16.4

3.923

0.126

3.750

0.117

3.519

0.146

Support Romanian Products, component 3

Low-CE

257

81.1

3.673

0.057

3.599

0.053

3.144

0.066

High-CE

60

18.9

3.817

0.118

3.717

0.109

3.450

0.136

"Trust the quality of the brand"

Tuborg a Danish beer brand

No effect

Positive effect

Negative effect

Disagree

6.3

0

93.8

Neutral

59.9

14.6

25.5

Agree

1.2

57.3

41.5

"The Tuborg brand is attentive with consumers’ needs"

Tuborg a Danish beer brand

No effect

Positive effect

Negative effect

Disagree

9.1

18.2

72.7

Neutral

52.3

18.5

29.1

Agree

3.2

54.2

42.6

"The Tuborg brand is attentive with consumers’ needs"

Tuborg a Danish beer brand

No effect

Positive effect

Negative effect

Disagree

9.1

18.2

72.7

Neutral

52.3

18.5

29.1

Agree

3.2

54.2

42.6

The principal component analysis of the data was performed to test whether all three items load on one component, and the results produced one component factor which explains 74.4% of the total variance. The Kaiser-Meyer-Olkin measure of sampling adequacy was 0.699, and Bartlett’s sphericity test was significant (2 = 369.361, p < 0.000).

Table 3. Initial factor loading for consumer perception of brands (n = 317)

Instrument items

Initial Factor loading a

I trust the quality of the products of this brand

.895

I think that this brand is always looking to improve its products to better satisfy consumers’ needs.

.874

I prefer this brand over other brands in the same product category.

.817

Notes: a Only loadings with a value of greater than 0.5 are of relevance.

Analysis and results

The Analysis of the data was done using the statistical analysis from the SPSS version 19. The significance value was at a two-tail with an alpha level of 0.05. Cronbach’s alpha was used to analyse the internal consistency for all instruments.

The Cronbach Alpha for all variables is presented in table 2. If the result is greater than 0.8 then the answers are considered reliable. If the score is below 0.2 then the answers are not reliable at all. When looking at the result of the study one can see that the answers are reliable, with a fairly low reliability of the Ideology constructs.

Table 2. Reliability Statistics

Variables

Items

Cronbach’s Alpha

Brand perception

9

.728

Ideology

5

.249

Consumer Ethnocentrism

17

.947

Preliminary results

The exploratory factor analysis was used to examine the factorial validity of the constructs. The 31 items were estimated using an exploratory factor analysis with principle component analysis and Varimax rotation. So that factor analysis can be used, the sample must be 100 or greater (Hair et al., 1998, Tabachnick & Fidell, 2007), which in this case is of 317. Firstly, it was observed that the factor loadings were at least .3, suggesting reasonable factorability. Secondly, the Kaiser-Meyer-Olkin measure of sampling adequacy was .864, above the commonly recommended value of .6, and Bartlett’s test of sphericity was significant (2 = 4335.8, p < 0.000). The diagonals of the anti-image correlation matrix demonstrated that the measures of sampling adequacy were over .5. Finally, the communalities were above .3, further confirming that each item shared some common variance with other items. Therefore, it was considered appropriate to factor analyse the data.

Results

Degree of Ethnocentrism among Romanian Consumers