Ethnocentric beliefs and country-of-origin (COO) effect

38 downloads 0 Views 266KB Size Report
encompasses Greece, Italy and Holland and yellow cheese, ham and beer. ... the influence of COO-effect depends not only on the country of origin or the ...
The current issue and full text archive of this journal is available at www.emeraldinsight.com/0309-0566.htm

EJM 41,11/12

1518 Received December 2005 Revised June 2006

Ethnocentric beliefs and country-of-origin (COO) effect Impact of country, product and product attributes on Greek consumers’ evaluation of food products George Chryssochoidis Laboratory of Agribusiness Management, Agricultural University of Athens, Athens, Greece

Athanassios Krystallis National Agricultural Research Foundation (NAGREF), Athens, Greece, and

Panagiotis Perreas Laboratory of Agribusiness Management, Agricultural University of Athens, Athens, Greece Abstract

European Journal of Marketing Vol. 41 No. 11/12, 2007 pp. 1518-1544 q Emerald Group Publishing Limited 0309-0566 DOI 10.1108/03090560710821288

Purpose – The present study using the Consumer Ethnocentric Tendencies Scale (CET-SCALE) aims to evaluate the level of consumer ethnocentrism (CE) and its implications on their evaluation of food products. Furthermore, it seeks to examine the level at which country of origin (COO) effect is activated (country, product or attribute) per consumer cluster of different level of CE in a food evaluation context. Design/methodology/approach – For attaining the above aims, a questionnaire was developed and completed by 274 respondents. The set of countries of origin and products under consideration encompasses Greece, Italy and Holland and yellow cheese, ham and beer. Findings – The use of the CET-SCALE pinpointed that the sample can be characterised as marginally ethnocentric. Exploratory and confirmatory factor analyses justified the uni-dimensionality of CE. Cluster analysis allocated the sample into two clusters, the ethnocentric and the non-ethnocentric. The results showed that ethnocentrism affects not only consumer beliefs, but also the way perceived quality of domestic and foreign products are evaluated, culminating in the appearance of COO-effect. In ethnocentric consumers, the COO effect is activated at the initiatory level of the country a food product originates in (country-specific), except when the foreign country of origin is given, where the COO effect is activated at the level of the product type (product-specific). In the non-ethnocentric cluster, COO does not lead to an overall acceptance or rejection, but instead it affects the evaluation of specific product attributes (attribute-specific). Research limitations/implications – The survey suffers the limitation of focusing on the influence of ethnocentric beliefs in food products evaluation and not on their real impact on final purchasing behaviour. Consumer ethnocentrism and COO effect are linked together, but the stimulus that activates their link differs according to the strength of ethnocentric beliefs held by consumers; that given, different marketing strategies should be applied depending on the level of CE of the target-group selected. Originality/value – Internationally, the issue of COO-effect is comprehensively examined, yet the literature has focused almost explicitly on hi-tech or fashion products and services. This fact attaches particular importance to the present study, which is concerned exclusively with food products. Keywords Ethnocentrism, Consumer behaviour, Country of origin, Food products Paper type Research paper

Introduction International empirical studies concerning the influence of country-of-origin (COO) effect in the evaluation of various products by consumers reveal mixed and sometimes contradictory results, possibly due to different combinations of products, samples and countries where the studies were conducted (Kaynak and Kara, 2002). The majority of surveys underline the multidimensionality of COO-effect. Reference to the country of origin of a product made on its label influences consumers’ perceptions regarding its quality (country-specific) (Balabanis and Diamantopoulos, 2004; Lantz and Loeb, 1996), yet the magnitude of the effect depends on the product category (product-specific) (Balabanis and Diamantopoulos, 2004; Watson and Wright, 2000; Sharma et al., 1995, in Piron, 2000; Smith, 1993; Cordell, 1992). Moreover, several other studies suggest that the influence of COO-effect depends not only on the country of origin or the product category, but on specific product attributes as well (attribute-specific) (Supphellen and Rittenburgh, 2001; Juric and Worsley, 1998; Johanson et al., 1985, in Showers and Showers, 1993). The present study attempts to assess the impact of ethnocentrism in the evaluation of specific food products by Greek consumers. Internationally, the issue of COO effect is comprehensively examined, yet the literature has focused almost explicitly on hi-tech or fashion products (automobiles, home appliances, computers, apparel etc) and services (e.g. air transportation), living the important area of food products virtually unexplored. Few recent exceptions constitute the surveys of Juric and Worsley (1998) (food products in general) and Orth and Firbasova (2003) (yogurt). This particular fact attaches importance to the present study, which is concerned exclusively with food products. Given the fact that older consumers or people with low educational level are expected to exhibit highly ethnocentric attitude in their evaluations of Greek versus foreign food products, the survey focuses on relatively young and well-educated consumers. This was necessary since the study involves foreign products, so a fair degree of familiarity of consumers with the products and countries in question was a prerequisite. The selection of three different countries (Greece, Holland and Italy) and three kinds of products (beer, ham products and cheese) intends to pinpoint the level at which COO effect emerges (country or product-specific). Also, the use of the list of food attributes introduced by Steptoe et al. (1995) is essential for determining whether the COO effect emerges at the food attributes level (attribute-specific). Consumer ethnocentrism and COO effect Consumer ethnocentrism According to various studies, the cause of the appearance of COO effect can be found in consumer ethnocentrism (CE) (Lantz and Loeb, 1996; Lee and Ganesh, 1999; Stoltman et al., 1991). The origins of the CE construct can be traced back to the work by Sumner (1906, in Balabanis and Diamantopoulos, 2004), who introduced ethnocentrism as a general construct reflecting the view of things in which one’s own group is the centre of everything, and all others are scaled and rated with reference to it. Sumner’s conception of ethnocentrism is based on the formation of “we-group” feelings, whereby the in-group is the focal point and all out-groups are judged in relation to it. Sumner emphasised a dichotomous structure of ethnocentrism: an unfavourable attitude toward out-groups; and a favourable attitude toward the in-groups. The in-groups determine the standard of judging other groups and the willingness to associate with them.

Ethnocentric beliefs and COO effect 1519

EJM 41,11/12

1520

CE is the application of the ethnocentric construct in the economic environment and has inherited the main premises and properties of ethnocentrism. It is defined as a “trait-like property of an individual’s personality” and encompasses “the beliefs held by the consumers about appropriateness, indeed morality, of purchasing foreign-made products” (Shimp and Sharma, 1987). Sharma et al. (1995, in Piron, 2000) and Rawwas and Rajendran (1996) demonstrated that CE may lead to overestimation of specific attributes and overall quality of domestic products and to an underestimation of that of foreign products. However, according to Watson and Wright (2000), these attitudes may connote consumer behaviours but they are not equivalent to them, since the latter are product-specific. Shimp and Sharma (1987) postulated that CE can provide an explanation as to why consumers prefer domestic over foreign products even when there is no obvious reason for such a preference (e.g. when the foreign products are of better quality or cheaper). There are many factors that affect CE, with the type of product per se being one of them. It has been found that the lever of CE varies among product categories. Sharma et al. (1995, in Piron, 2000), indicated that the less important a product category the greater the ethnocentric tendencies and behaviour exhibited by consumers. Also, Javalgi et al. (2005) found that the impact of CE in purchasing intention of a particular product is moderate when this product is perceived as absolutely necessary. Moreover, the impact of CE depends on the level of development of consumers’ home country. According to Wang and Chen (2004), consumers from a developed country tend to appreciate more favourably domestic products over imported ones, hence resulting to an increase of the impact of CE on buying the former and rejecting the latter. The reverse has been observed in developing countries, where consumers perceive foreign products (especially those originating to prestigious countries) as superior compared to their domestic counterparts. Furthermore, Balabanis and Diamantopoulos (2004) argued that similarity between countries of origin (in terms of culture and level of economic competitiveness) is unrelated to preference or rejection of foreign products. Additionally, they found that CE is sometimes negatively related to preferences for foreign products, yet it is mostly unrelated, leading to the conclusion that, overall, CE is a more consistent predictor of preferences for domestic products rather than for foreign products. In other words, CE leads to consumers preferring domestic products but not necessarily rejecting foreign ones. Finally, they postulated that the CE impact varies significantly among different product categories and countries of origin. For that reason, they suggest that companies should not depend solely on CE levels of target markets when foreseeing potential success or failure of their products. CE and COO effect: review of recent literature In the international literature one can find many diverse definitions of the COO effect (Khachaturian and Morganosky, 1990; Okechuku, 1994; Schaefer, 1995; Ahmed and d’Astus, 1995, 1996; Kucukemiroglou, 1997; Knight and Calantone, 2000). According to Wang and Lamb (1983, in Al-Sulaiti and Baker, 1998), the COO effect is an obscure, intangible obstacle that a product (or service) confronts when entering a new market. This obstacle is manifested with the form of negative disposition, on behalf of consumers, towards the newly imported product or service. The entire literature concerning the COO effect suggests that consumers’ perceptions about imported

products and their attributes are affected by the country of origin (Papadopoulos, 1993) and related to CE. The country of origin, as an information cue, activates various ethnocentric or not beliefs and the antecedent knowledge of consumers, which subsequently affect the interpretation and evaluation of product attributes. The phenomenon of evaluating products based on judging the country of origin is called COO effect. Since the beginning of the last decade, many changes have occurred in the international economic environment, including the globalisation of markets and the fast growth of Asian markets, which overall offered new potential for deeper investigation of the subject. As a result, research that took place from 1990 onwards tackled more thoroughly the COO effect. Following, the most representative studies of the past decade are being presented. Smith (1993) studied the beliefs of American consumers regarding imported products whose labels pinpointed not their specific country of origin but their continent of origin. Findings suggested that the use of such labels might moderate the negative bias that some consumers held against products originating from developing countries. For instance, results revealed that consumers evaluated products from South America as equal, in terms of quality, with the products originating from Western Europe. Maheswaran (1994) found that, in case of existence of unambiguous information for a specific product (in his case: computers), experienced consumers base their choices upon the evaluation of specific product attributes, whereas less experienced consumers rely on the country of origin. On the other hand, when information lacks or it is ambiguous, both types of consumers rely on country of origin so as to infer the quality of the product. Moreover, Gurhan-Canli and Maheswaran (2000) argued that the extent to which a product’s country of origin cue is utilised depends on the level of consumer involvement in its purchasing. The fewer motives consumers have to engage actively in the buying process, the greater the probability to utilise the country of origin to infer quality, whereas highly motivated consumers are expected to focus on specific product’s attributes. In the latter case, country of origin is perceived as a mere informational cue and not as a measurement of overall quality. Yarpak and Baughan (1991) and Han (1989) found that CE influences significantly the preferences of consumers, not only indirectly – through the evaluation processes of each product attribute, but also directly, by affecting the formation of positive or negative purchasing intention (COO effect). On the other hand, according to Wall et al. (1991), the evaluation of the quality of a product and its country of origin might be linked together, but the latter was found of minor importance when purchasing intention has being evaluated. Similar results were reported by Rahman (2000), who found that the COO effect is prominent in product evaluation, but does not solely affect the final purchasing behaviour, which in turn is affected by other factors, such as price. Roth and Romeo (1992) formulated a theoretical framework for the relationship between consumer preferences for a country’s products, and perceptions of a country’s culture, economy and politics. COO effect is created when the skills of a COO do not correspond with its product’s attributes that are considered by consumers as important. They argued that consumers’ evaluations of a specific product from country X are based on the match between product and country. According to these authors, consumers prefer country X as an origin for specific

Ethnocentric beliefs and COO effect 1521

EJM 41,11/12

1522

products when they believe that there is a match between the perceived “strengths” of country X and the skills that are needed for manufacturing the product under consideration. A preference for Swiss watches or German cars, for example, might be explained by the perception of the workmanship of Swiss or German engineers respectively. Same findings were observed by Moon and Jain (2002), while investigating the influence of CE in the formation of positive or negative disposition towards foreign advertisements. Juric and Worsley (1998) argued that the COO effect is attribute-specific. A product originating from a particular country may be evaluated favourably on one attribute (e.g. taste of French wines) but unfavourably on another (e.g. safety of French food products). The magnitude of the COO effect on consumer’s choices was also explored by Watson and Wright (2000) in their study of New Zealand consumers. In case where imported products did not have domestic substitutes (competitive of foreign ones), then similarity in terms of culture and politics between two countries was found to be a major factor influencing the evaluation of products. The authors also found that highly ethnocentric consumers, under the afore-mentioned circumstances, tend to prefer products from “similar” countries. It is notable that, when a domestic substitute product exists, New Zealanders preferred the domestic over the imported, even when the foreign was perceived of better quality or cheaper. On the contrary, Supphellen and Rittenburgh (2001) in their survey of Polish consumers found that, when foreign products are significantly better compared to domestic ones, ethnocentric consumers were “forced” to conform to the overall public opinion, which preferred the imported products. Kaynak et al. (1994) examined the impact the profile of a country has in the area of services. They studied the beliefs of American consumers regarding 24 airline services and discovered major differences between the users of local airlines and the consumers that used both domestic and foreign airlines. Brunning (1997) examined the case of Canadian airlines and found that COO was affecting consumers’ attitudes in such extent that it was ranked as the second most important factor after price. Additionally, Al-Sulaiti and Baker (1998) in their survey in airlines of Qatar found that foreign airlines were more attractive than the domestic, but contrariwise, consumers preferred to travel using domestic airlines or those originating in the broader region of Persian Gulf. Finally, similarly to the case of tangible products, de Ruyter et al. (1998) observed that the less important or useful a service the greater the impact of COO in consumer perceptions. Methodology Aims of the study – selection of products and countries Using the findings of the previous studies as a point of departure, the present study aims not only to evaluate the level of ethnocentrism of Greek consumers, but also to investigate the CE-COO effect relationship and examine the implications of the fore mentioned notions on consumers’ perceptions regarding imported food products. Analytically, the study has the following major aims: (1) The assessment of the level of CE is of the utmost importance, since CE constitutes a powerful motive that activates the COO effect (Lantz and Loeb, 1996; Lee and Ganesh, 1999; Stoltman et al. 1991). Also, it has been suggested that CE explains a greater proportion of variance in purchasing behaviour as

compared to elements of the marketing mix (Herche, 1994), thus constituting an important strategic component that should be taken into account by marketing practitioners. Hence, the present study aims at the assessment of the level of ethnocentrism of Greek consumer in a food evaluation context. (2) In the international literature there are several scales that have been used for the assessment of CE. However, the most commonly used measurement instrument is the Consumer Ethnocentric Tendencies Scale (CET-SCALE), introduced by Shimp and Sharma (1987). However, even though the validity of CET-SCALE has been repeatedly proven, its use was limited mostly in measuring CE in developed countries (USA, Canada, Japan, etc.). In validity tests of a shorter version of CET-SCALE in less developed countries (e.g. Hungary, Czech Republic, Poland) differences were revealed underlying the necessity for testing the scale in more culturally variable countries (Javalgi et al. 2005; Kaynak and Kara, 2000). Consequently, the second aim of the study is to examine the applicability of CET-SCALE in the case of Greek consumers, as well as the conceptual examination of the notion of CE, which is measured through CET-SCALE. (3) According to some studies (Watson and Wright, 2000; Herche, 1994) CE is a concept closely linked with purchasing behaviour. Given that the analysis of the latter is a crucial element for the implementation of any marketing strategy, it would be useful to examine if the sample of Greek respondents can be divided into groups of consumers with different level of CE, as well as identifying their socio-demographic profile. The literature review demonstrated a divergence of views regarding consumers’ demographic characteristics that are related to CE. In some studies, for instance, educational level has been found to correlate negatively with CE (Orth and Firbasova, 2003; Sharma et al. 1995, in Piron, 2000), while in others it was found independent from CE (Javalgi et al., 2005). Similarly, income was found important in the studies of Sharma et al. (1995, in Piron, 2000) and Bailey and Pineres(1997), while it was found of questionable importance in the studies conducted by Han (1990), McLain and Sternquist (1991) and Javalgi et al. (2005). Consequently, the third aim of the study is to identify the existence of different consumer clusters with varying level of CE and their distinctive demographic characteristics. (4) It has been mentioned that COO effect is multidimensional and affects consumers’ perceptions at the level of country (country-specific), it can become more concrete, concerning only specific products or it can appear at an even more analytic level, concerning specific attributes of two substitute products originating in two different countries. The identification and verification of the level at which COO effect is activated is important for marketing practitioners, since it is obvious that different levels of reference require implementation of different strategies. Thus, the fourth aim of the study is the identification of the level at which COO effect is activated per cluster of different level of CE in a food evaluation context. For attaining the above aims, a questionnaire was developed and completed though personal interviews by 274 respondents. The set of countries of origin under consideration in the food evaluation context of the present survey encompasses Greece,

Ethnocentric beliefs and COO effect 1523

EJM 41,11/12

1524

Italy and Holland. The latter two of these where included because both countries export a large amount of products to the Greek food market. Avoidance of more “exotic” countries was necessary so as to ensure an acceptable degree of familiarity of consumers with food products originating from the set of countries examined. Also, less developed countries were excluded from the survey in order to prevent consumer bias, since, as we have seen, products originating from such countries are often perceived as low quality products. The same philosophy was applied while forming the set of products under consideration. Beer, ham products and yellow cheese are considered as representative products of the afore-mentioned countries, ensuring familiarity of consumers as well as satisfying a substantial degree of “compatibility” between the products and the countries’ profile. Finally, the selection of products was carried out keeping in mind that all three products had to have at least one domestic (Greek) competitive product and employ ingredients of different origin (namely: ham is meat based; cheese is milk based whereas beer is crop based), an important issue for food products. Construction of questionnaire, data collection and sample description The questionnaire consists of three parts: in the first part, respondents were asked to evaluate several product attributes using seven-point Likert-type agreement questions with end-points 1 ¼ “totally agree” to 7 ¼ “totally disagree”. The evaluation was carried out using the following combinations of products and countries: . Greek beer vs Dutch beer; . Greek ham products vs Italian ham products; and . Greek yellow cheese vs Dutch yellow cheese. The evaluation criteria were derived from the set of 36 questions formulated by Steptoe et al. (1995), which concerned the overall evaluation criteria that consumers bear in mind when purchasing food products. The number of questions was modified per product, since not all questions included in the list by Steptoe et al. (1995) were applicable. Thus, the evaluation of beer was based on 28 characteristics, whereas ham products and yellow cheese were evaluated using 33 and 32 characteristics respectively. The second part of the questionnaire comprises the CET-SCALE (Shimp and Sharma, 1987), which consists of 17 Likert-type questions with end-points 1 ¼ “totally agree” to 7 ¼ “totally disagree”. The scale was carefully translated and tailored to fit the Greek food market realities. This scale has been extensively used and validated in many studies (most recent examples: Balabanis and Diamantopoulos, 2004; Orth and Firbasova, 2003; Kaynak and Kara, 2002; Steenkamp and Baumgartner, 1998), but only one concerned the relationship between CE and COO effect in a case of food products (Orth and Firbasova, 2003). The last section of the questionnaire consists of questions regarding purchasing habits for Italian and Dutch food products in general, as well as demographic characteristics. Data collection lasted 1.5 months (from January 15 to February 28 2005). After appointments made to visit each respondent at his/her place, the questionnaires were self-completed with the proper instructions and clarifications. Thus, no missing values

were recorded. The time of completion varied between 15 and 30 minutes. It is worth mentioning that, according to instructions given, the evaluation for each product’s characteristic was carried out in pairs, meaning that every product characteristic derived from Steptoe’s list was evaluated simultaneously for both of the countries of origin per food product under examination. The socio-demographic profile of respondents is depicted in Table I. The majority of respondents are relatively young (average age: 37.6 years) and well educated, conforming to the study’s aims as described earlier. Meta-analytic research demonstrated that using young respondents (most of the past studies used convenience samples comprising students) did not lead to a systematic overestimation of the COO effect (Verlegh and Steenkamp, 1999; Peterson and Jolibert, 1995; Liefeld, 1993, in Balabanis and Diamantopoulos, 2004).

Ethnocentric beliefs and COO effect 1525

% 1. Age ,25 26-35 36-45 46-55 55 or .

13.9 36.9 24.5 16.1 8.8

2. Educational level Non formal education Elementary or nine-year circle High-school and/or two-year technical University graduate Post-graduate or PhD holder

0.4 5.2 43.1 39.8 11.7

3. Number of persons in the household 1-2 3-4 5 or .

34.3 54.0 11.7

4. Personal monthly income (e, pre-tax) ,9,000 9,000-17,999 18,000-24,999 25,000-34,999 35,000 or .

11.7 45.3 19.0 11.3 12.8

5. Presence of under-age children in the household Yes No

28.1 71.9

6. Marital status Single Married Divorced/widowed Unmarried couple

38.3 56.6 4.0 1.1

7. Gender Male Female

46.7 53.3

Table I. Sample’s socio-demographic profile (n ¼ 274)

EJM 41,11/12

1526

Analysis and results Preliminary results Purchasing habits in relation to the products under consideration. More than one in four respondents (27 per cent) buys Italian products more than once per month, whereas 10.9 per cent neither has bought nor has ever considered buying them. The corresponding percentages for Dutch products are 31.4 per cent and 10.2 per cent. It is notable that in both cases (Italian and Dutch products in general) purchasing frequencies are relatively high. As for the purchasing frequency of the products under evaluation irrespectively of their country of origin, 20.8 per cent of the sample purchases beer at least once a week, whereas non-beer drinkers are limited to 4.7 per cent. In the case of ham and yellow cheese, the corresponding percentages are 34.7 per cent and 34.7 per cent (users at least once a week) and 2.2 per cent and 2.6 per cent (non-users). Preferences per product type. A total of 28 characteristics of beer have been evaluated first by consumers in paired comparisons per criterion for Greek as opposed to Dutch beer (see Table II).

Evaluation characteristics: beer . . .

Table II. Consumer evaluation of Greek and Dutch beer (n ¼ 274)

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28.

Is easily consumable/drinkable Contains no additives Tastes good Contains natural ingredients Is not expensive Is familiar Is easily available in shops and supermarkets Is good value for money Cheers me up Smells nice Can be consumed/drunk very simply Helps me cope with stress Has a pleasant texture Is packaged in an environmentally friendly way Comes from country I approve of politically Is like the beer I drunk when I was younger Contains no artificial ingredients Keeps me awake/alert Packaging looks nice Helps me relax Takes no time to consume/drink Keeps me healthy Makes me feel good Has the country of origin clearly marked Is what I usually drink Helps me to cope with life Can be bought in shop close to where I live/work Is cheap

Agree (totally/rather) Greek Dutch (%) (%) 88.7 36.1 81.0 52.9 70.8 95.3 98.2 64.6 60.9 70.1 87.2 27.0 65.0 54.4 – 42.3 23.4 15.7 56.6 56.9 76.3 9.9 56.9 75.2 71.2 8.4 93.4 54.0

63.9 23.7 65.3 36.5 36.1 68.2 63.1 41.2 45.6 54.0 72.3 21.9 52.6 50.0 30.4 19.3 14.2 10.9 58.8 48.2 65.7 7.3 47.8 68.2 27.4 4.7 54.7 29.9

Disagree (totally/rather) Greek Dutch (%) (%) 3.0 31 7.3 18.2 15.0 2.8 0.7 11.3 17.2 11.3 4.4 46.0 13.9 23.4 – 24.8 40.5 59.5 11.7 24.1 10.2 66.1 21.9 8.8 16.8 71.1 1.5 15.3

12.0 39.4 12.8 23.0 33.6 19.3 18.2 18.2 11.7 15.7 7.7 49.6 17.2 23.0 28.2 39.1 42.0 60.2 9.5 25.9 11.7 66.8 26.3 10.9 48.5 71.5 24.1 29.9

Greek beer was evaluated more favourably than Dutch beer in all characteristics except for its package attractiveness/appearance (question 19). However, it is worth mentioning that 19.3 per cent of the sample is totally unfamiliar with Dutch beer (q.6) and only 27.4 per cent consumes Dutch beer on a regular basis (q.25). Regarding availability, Greek beer is rated more favourably by over 90 per cent of respondents (q.7, 27). The characteristics which exhibited the most prominent differences (always in favour of Greek beer) were: ease of use (q.1), price and value for money (q.5, 8), taste, aroma, feel in mouth (q.3, 10, 13), ingredients/additives (q.2, 4), as well as familiarity with the product / traditional image (q.6, 16). Greek ham products were evaluated more favourably, as compared to the Italian ones, in 31 out of the 32 characteristics under consideration (see Table III).

Evaluation characteristics: ham . . . 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33.

Is easily consumable/edible Contains no additives Is low in calories Tastes good Contains natural ingredients Is not expensive Is low in fat Is familiar Is nutritious Is easily available in shops and supermarkets Is good value for money Cheers me up Smells nice Can be consumed/eaten very simply Helps me cope with stress Helps me control my weight Has a pleasant texture Is packaged in an environmentally friendly way Comes from country I approve of politically Is like the delicatessen I ate when I was a child Contains no artificial ingredients Keeps me awake/alert Packaging looks nice Helps me relax Is high in protein Takes no time to consume/eat Keeps me healthy Makes me feel good Has the country of origin clearly marked Is what I usually eat Helps me to cope with life Can be bought in shop close to where I live/work Is cheap

Agree (totally/rather) Greek Italian (%) (%) 93.4 9.1 17.5 91.6 26.3 36.5 13.1 94.2 41.2 97.1 45.3 20.4 67.2 93.4 7.3 9.1 62.4 35.0 – 43.8 10.9 5.5 50.0 8.4 57.7 87.6 10.9 25.9 73.4 76.6 8.4 93.1 33.9

67.2 6.2 10.9 75.9 16.8 12.8 8.4 66.1 36.1 74.5 27.0 15.3 56.6 85.8 4.7 5.5 51.1 29.2 36.3 19.3 4.7 4.0 48.5 4.7 49.3 78.8 5.8 19.0 63.5 26.6 3.3 59.9 12.4

Ethnocentric beliefs and COO effect 1527

Disagree (totally/rather) Greek Italian (%) (%) 2.4 70.1 65.7 2.6 37.2 35.0, 63.5 2.2 32.8 1.1 22.6 47.8 13.5 2.6 74.5 75.2 13.5 30.0 – 30.3 63.5 75.9 18.2 75.9 17.9 3.3 63.5 43.4 10.2 13.1 72.6 4.0 27.0

10.9 65.3 67.9 6.2 39.8 51.1 65.3 19.7 35.4 13.1 28.8 50.4 16.8 5.8 75.2 76.3 15.3 32.1 26.5 40.6 65 77.4 16.5 77.7 20.7 6.6 67.6 48.5 15.7 51.5 75.9 21.5 47.1

Table III. Consumer evaluation of Greek and Italian ham products (n ¼ 274)

EJM 41,11/12

1528

Almost 20 per cent of the sample is unfamiliar with Italian ham (q.8), whereas 26.6 per cent consume it regularly (q.30). Similarly to beer, more than 90 per cent of the sample rated more favourably the Greek ham as much more available (q.10, 32) than the Italian. Great differences were also observed in the ease of use (q.1), taste and aroma (q.4, 13) and familiarity with the product/traditional image (q.8, 20). One third of the sample (33.9 per cent) considers Greek delicatessen as cheap (q.33), whereas the corresponding percentage for the Italian products is limited to 12.4 per cent. Both products from Greece and Italy were unfavourably evaluated in terms of health and safety issues (q.21, 5, 2). Greek yellow cheese was evaluated more favourably than the Dutch cheese in 29 out of 32 characteristics (see Table IV). Unlike the previous two product types, only 8.1 per cent of the sample is unfamiliar with yellow cheese of Dutch origin (q.8). The most favourable characteristic of Greek

Evaluation characteristics: yellow cheese . . .

Table IV. Consumer evaluation of Greek and Dutch yellow cheese (n ¼ 274)

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32.

Is easily consumable/edible Contains no additives Is low in calories Tastes good Contains natural ingredients Is not expensive Is low in fat Is familiar Is nutritious Is easily available in shops and supermarkets Is good value for money Cheers me up Smells nice Can be consumed/eaten very simply Helps me cope with stress Helps me control my weight Has a pleasant texture Is packaged in an environmentally friendly way Comes from country I approve of politically Is like the cheese I ate when I was a child Contains no artificial ingredients Keeps me awake/alert Packaging looks nice Helps me relax Takes no time to consume/eat Keeps me healthy Makes me feel good Has the country of origin clearly marked Is what I usually eat Helps me to cope with life Can be bought in shop close to where I live/work Is cheap

Agree (totally/rather) Greek Dutch (%) (%) 93.1 36.5 19.0 90.5 41.6 34.3 17.5 94.2 75.2 97.8 45.6 24.5 71.9 92.3 7.7 14.2 71.5 42.7 48.9 24.1 7.3 48.2 7.3 87.6 35.4 33.9 72.6 80.7 11.7 96.0 35.8

83.6 17.9 19.3 75.9 21.9 22.6 18.2 81.0 60.6 87.2 32.5 16.1 57.7 84.3 5.8 13.5 63.5 37.2 31.1 29.2 7.7 4.0 48.5 6.9 78.5 24.1 27.7 66.4 48.5 7.3 81.4 23.0

Disagree (totally/rather) Greek Dutch (%) (%) 2.2 34.3 52.2 3.3 21.9 37.2 51.8 2.5 8.0 0.4 24.1 44.5 10.2 1.5 65.7 66.5 9.5 18.6 29.9 42.0 67.5 15.3 65.7 3.6 31.4 35.1 7.3 10.2 61.3 1.1 31.4

6.2 46.0 54.4 8.4 33.6 45.3 52.2 8.1 12.8 5.9 28.4 50.0 13.1 3.7 68.7 67.2 12.0 21.6 26.4 37.6 57.6 70.1 13.2 66.5 6.9 37.6 40.2 10.2 32.9 64.2 8.0 35.4

cheese which the overwhelming majority of respondents favours over the Dutch ones, concern its better availability (q.10, 31). Greek yellow cheese is regularly consumed by 80.7 per cent of respondents (q.29), whereas the corresponding percentage for the Dutch cheese is 48.5 per cent, significantly higher compared to Dutch beer and Italian ham. Substantial differences in favour of the Greek cheese are also observed in other characteristics, such as taste (q. 4) and price (q.6). Another important finding regards consumer perception that Greek cheese is a safer or healthier to consume cheese (q.2, 21), but Dutch cheese is better in terms of its fat/calories content (q.3, 7). Similarly to beer, Dutch cheese is evaluated as packed in a more attractive way (q.23). Finally, it is notable that the majority of respondents argued that all three types of Greek products bear on their labelling a clear indication of their origin (q.28). Results of CET-SCALE analysis. Data collected for the entire sample lead to the estimation of mean CE at 3.85 in the 1-7 scale (SD ¼ 1:05), revealing a marginally ethnocentric sample of consumers (see Table V). CET-SCALE questions with the most salient results are summarised as follows: 89.8 per cent of the sample argues that buying Greek products helps creating jobs in Greece (q.3). However, the opposite does not appears to apply, since 61.3 per cent of respondents believe that buying foreign products does not augment the unemployment rate in Greece (q.11). On the other hand, 64.6 per cent of the sample disagrees with the statement that “buying foreign products constitutes an CET-SCALE 1. 2. 3. 4. 5. 6.

Greek people should always buy Greek-made products instead of imports Only those products that are unavailable in Greece should be imported Buy Greek-made products. Keep Greece working Greek products, first, last and foremost Purchasing foreign-made products is anti-Greek It is not right to purchase foreign-made products, because it puts Greeks out of jobs 7. A real Greek should always buy Greek-made products 8. We should purchase products manufactured is Greece instead of letting other countries get rich out of us 9. It is always better to purchase Greek products 10. There should be very little trading or purchasing of goods from other countries unless out of necessity 11. Greeks should not buy foreign products, because this hurts Greek business and cause unemployment 12. Barriers should be put on all imports 13. It may cost me in the long-run but I prefer to support Greek products 14. Foreigners should not be allowed to put their products in our markets 15. Foreign products should be taxed heavily to reduce their entry into Greece 16. We should buy from foreign countries only those products that we cannot obtain within our own country 17. Greek consumers who purchase products made in other countries are responsible for putting their fellow Greeks out of work Total mean value a

Meana

St. dev.

3.07 3.20 1.99 3.74 5.03

0.10 0.12 0.07 0.11 0.10

4.07 4.67

0.11 0.11

4.03 3.45

0.12 0.11

3.33

0.11

3.81 4.22 3.37 4.93 4.15

0.11 0.10 0.10 0.10 0.11

3.63

0.11

4.80 3.85

0.11 1.05

Note: 1 ¼ strongly agree, 2 ¼ agree, 3 ¼ rather agree, 4 ¼ neither . . . nor . . .; 5 ¼ rather disagree, 6 ¼ disagree, 7 ¼ strongly disagree

Ethnocentric beliefs and COO effect 1529

Table V. Mean value and standard deviation, CET-SCALE (n ¼ 274)

EJM 41,11/12

1530

anti-Greek act” (q.5) and, similarly, 56.6 per cent dissent from the statement that “a real Greek should buy only Greek products” (q. 7). A notable finding is also observed in relation to the statement that foreigners should not be allowed to sell their products in Greece (q. 14), where 64.2 per cent of respondents disagree. Finally, 55 per cent of the sample believes that it is always better to buy domestic products (q.9) and 58.8 per cent prefers to buy Greek products, even if they are more expensive than the foreign ones (q.13). Principal components and factor analyses of CET-SCALE The conceptual examination of ethnocentrism aims to test the applicability of the CET-SCALE on the present sample of young and well-educated Greek consumers, thus fulfilling the second aim of the study. The notion of CE is conceptually explored using principal components and factor analyses methodology (SPSS v. 12). Principal components analysis revealed one factor (eigenvalue 10.13) explaining 59.6 per cent of the total variance (see Table VI). The loadings of all 17 variables of the principal component are higher than 0.6, meaning that all variables measure the same concept of consumer ethnocentrism, which in this case is found to be uni-dimensional, as described in the relevant literature (Orth and Firbasova, 2003; Kaynak and Kara, 2002). However, the rotated solution in factor analysis (varimax with Kaiser normalisation) revealed that the 17 variables can be grouped in two almost equally strong factors with exceptionally high reliability (Cronbach’s alpha values higher than 0.90), while the two factors together explain 66.03 per cent of the total variance. The first factor conceptually contains the expression of a combined strong-national-geared behaviour (var.5, 7, 8, 17) and an almost hostile attitude towards foreign products, whose purchase is regarded as detrimental for the society (var.6, 11). Moreover, factor 1 encompasses an utterly drastic attitude against imported products and foreign countries: “diminution of imports”, “imposition of heavy taxes on imported goods”, “foreigners should not be allowed to sell their products in Greece” (var.12, 14, 15). The afore-mentioned variables represent the most radical statements of CET-SCALE and reflect the social reaction against the “menace” foreign products constitute. Consequently, factor 1 is termed as “hard” ethnocentrism. On the other hand, factor 2 encompasses statements with relatively more “temperate” meaning. For instance, the following variables “Only those products that are unavailable in Greece should be imported” (var.2), “There should be very little trading or purchasing of goods from other countries unless out of necessity” (var.10) and “It is always better to purchase Greek products” (var.9) are statements that do not prompt for the overall rejection of foreign products, but simply emphasise a preference towards domestic products and urge for their protection. Factor 2 can thus be termed “soft” ethnocentrism. Confirmatory factor analysis was then performed for the factor pattern suggested by the exploratory analysis. Variables 5, 7, 8, 17, 6, 11, 12, 14 and 15 were specified as indicators of the “hard” ethnocentrism factor and the remaining eight CET-SCALE variables were specified as indicators of the “soft” factor. Two models were estimated by means of maximum likelihood (using LISREL 8.54): an independence model, assuming that the interrelationships between all variables were zero, and the hypothesised model outlined above. The independence model was clearly rejected

Principal components loadingsa 1. Greek people should always buy Greek-made products instead of imports 2. Only those products that are unavailable in Greece should be imported 3. Buy Greek-made products. Keep Greece working 4. Greek products, first, last and foremost 5. Purchasing foreign-made products is anti-Greek 6. It is not right to purchase foreign-made products, because it puts Greeks out of jobs 7. A real Greek should always buy Greek-made products 8. We should purchase products manufactured is Greece instead of letting other countries get rich out of us 9. It is always better to purchase Greek products 10. There should be very little trading or purchasing of goods from other countries unless out of necessity 11. Greeks should not buy foreign products, because this hurts Greek business and cause unemployment 12. Barriers should be put on all imports 13. It may cost me in the long-run but I prefer to support Greek products 14. Foreigners should not be allowed to put their products in our markets 15. Foreign products should be taxed heavily to reduce their entry into Greece 16. We should buy from foreign countries only those products that we cannot obtain within our own country 17. Greek consumers who purchase products made in other countries are responsible for putting their fellow Greeks out of work Eigenvalue Explained variance (%) Cronbach alpha

Factor analysis loadings Factor Factor 1 2

0.80

0.73

0.72 0.61 0.86 0.80

0.51 0.80

0.78 0.75 0.71

0.86 0.84

0.69 0.76

0.83 0.77

0.71

0.51

0.70 0.71 0.55

0.69

0.52 0.64

0.70

0.79

0.67

0.70

0.72 0.78 10.13 59.62

1531

0.62

0.79 0.87 0.73

Ethnocentric beliefs and COO effect

0.58 0.70 5.90 34.72 0.9388

5.32 31.30 0.9130

Note: a The table displays only loadings greater than 0.500

(x 2 ½136 ¼ 14; 617:96). A chi-square difference test indicated a non significant improvement in fit from the independence model to the hypothesised model: x 2 ½118 ¼ 719:51 (satisfactory fit when chi square value is less than three times the degrees of freedom) and root mean squared error of approximation RMSEA ¼ 0:137. The final estimates are shown in Figure 1 (standardised solution). Identification of different consumers’ clusters (cluster analysis) The 17 variables of the CET-SCALE were used in the subsequent level of analysis in order to examine the possible existence of consumer clusters with different CE level. After the initial implementation of hierarchical cluster analysis, the procedure of quick clustering (k-means cluster analysis) followed, with the option to identify two to six

Table VI. Principal component and factor analyses, CET-SCALE (n ¼ 274)

EJM 41,11/12

1532

Figure 1. Confirmatory factor analysis model of consumer ethnocentrism (standardised estimates, two-factorial hypothesis)

clusters, according to the size of the sample. Information about the distribution of consumers in the clusters with the aid of a nominal variable is created (cluster membership variable) for both the hierarchical and the k-means solutions. The option of two clusters is finally selected (bivariate correlation between the hierarchical cluster membership variable and the k-means cluster membership variable ¼ 0:841, p , 0:01): cluster 1 comprises 164 consumers (59.8 per cent of the sample) and cluster 2 comprises 110 consumers (40.2 per cent of the sample). The analytical description of the clusters in relation to the 17 CET-SCALE variables is as follows (see Table VII):

CET-SCALEa 1. Greek people should always buy Greek-made products instead of imports 2. Only those products that are unavailable in Greece should be imported 3. Buy Greek-made products. Keep Greece working 4. Greek products, first, last and foremost 5. Purchasing foreign-made products is un-Greek 6. It is not right to purchase foreign-made products, because it puts Greeks out of jobs 7. A real Greek should always buy Greek-made products 8. We should purchase products manufactured is Greece instead of letting other countries get rich out of us 9. It is always better to purchase Greek products 10. There should be very little trading or purchasing of goods from other countries unless out of necessity 11. Greeks should not buy foreign products, because this hurts Greek business and cause unemployment 12. Barriers should be put on all imports 13. It may cost me in the long run but I prefer to support Greek products 14. Foreigners should not be allowed to put their products in our markets 15. Foreign products should be taxed heavily to reduce their entry into Greece 16. We should buy from foreign countries only those products that we cannot obtain within our own country 17. Greek consumers who purchase products made in other countries are responsible for putting their fellow Greeks out of work Mean ethnocentrism per cluster

Cluster 1 (n1 ¼ 59:8%) Std Mean dev. 2.10

1.092

Cluster 2 (n2 ¼ 40:2)) Std Mean dev. 4.53

1.379

t

t-test Sig.

16.2

*

2.02 1.49 2.62 4.15

1.315 0.779 1.307 1.641

4.95 2.73 5.42 6.33

1.319 1.401 1.176 0.756

18.0 9.37 18.0 12.9

* * * *

3.01

1.414

5.65

0.982

16.9

*

3.70

1.659

6.12

0.865

14.0

*

3.02 2.62

1.568 1.407

5.55 4.68

1.254 1.514

14.1 11.5

* *

2.35

1.134

4.78

1.455

15.4

*

2.82 3.46

1.388 1.520

5.28 5.35

1.142 1.253

15.4 10.8

* *

2.76

1.253

4.27

1.596

8.7

*

4.26

1.570

5.93

1.098

9.6

*

3.52

1.746

5.10

1.354

8.0

*

2.71

1.383

5.00

1.465

13.1

*

3.91 2.97

1.756 1.407

6.11 5.16

0.817 12.2 1.225

*

Notes: a 1 ¼ strongly agree, 2 ¼ agree, 3 ¼ rather agree, 4 ¼ neither . . . nor . . ., 5 ¼ rather disagree, 6 ¼ disagree, 7 ¼ strongly disagree; * Statistically significant for p , 0:01

.

.

Cluster 1 (n1 ¼ 164): ethnocentric consumers. Mean agreement with all 17 statements is relatively low (2.97 in the 1-7 scale). Given that agreement score value lower than 4 denotes ethnocentric respondents, cluster 1 encompasses individuals with relatively pronounced ethnocentric beliefs. Cluster 2: non-ethnocentric consumers. Mean agreement with all 17 statements of CET-SCALE is 5.16, indicating a cluster that comprises individuals with relatively weak ethnocentric beliefs.

Ethnocentric beliefs and COO effect 1533

Table VII. Means, standard deviations and statistically significant differences of clusters, CET-SCALE (n ¼ 274)

EJM 41,11/12

1534

Table VIII. Demographic profile of the two clusters

As for demographic variables, age, educational level, marital status and family size exhibit statistically significant differences between the two clusters (see Table VIII). Age has been repeatedly mentioned in the literature as a variable with high discriminating power among consumers in terms of their ethnocentrism (Orth and Firbasova, 2003; Sharma et al. 1995, in Piron, 2000). Marital status and family size turned out to be statistically significant presumably due to their relation with age. Ethnocentric consumers (cluster 1) are mainly older (especially in the categories over 46 years old), as opposed to non-ethnocentric consumers (cluster 2) whose overwhelming majority is younger than 35 years. As for education, it is impressive that the large majority of non-ethnocentric are highly educated (tertiary level) as compared to ethnocentric that are mostly limited to secondary education. Finally, it is worth noting that none of the variables concerning purchasing frequency for the products/countries under consideration was found statistically different between the two clusters (see Table IX).

Characteristics

Cluster 1 (59.8%): ethnocentric (%) (%)

Sex Male Female

43.9 56.1

50.9 49.1

Age , 35 years old . 35 years old

41.0 59.0

65.5 34.5

Marital status Married Unmarried Divorced/widower Unmarried couple

65.9 30.6 2.4 1.1

42.7 50.9 6.4 0.0

Educational level University graduate or higher

42.1

65.4

Personal monthly income (e, pre-tax) , 9,000 9,000-18,000 18,000-25,000 25,000-35,000 . 35,000

9.8 48.8 18.9 10.4 12.2

14.5 40.0 19.1 12.7 13.6

Family size 1-2 3-4 5 or more

31.7 55.5 12.8

38.2 51.8 10.0

Under-age children Yes No

27.4 72.6

29.1 70.9

Notes: * Statistically significant for p , 0:05; n.s. non statistically significant

Cluster 2 (40.2%): non-ethnocentric t-test Sig. 1.299

n.s.

16.790

*

18.362

*

31.915

*

2.848

n.s.

15.791

*

0.921

n.s.

Cluster 1 (59.8%) Ethnocentric (%)

Cluster 2 (40.2%) Non-ethnocentric (%)

Italian products purchasing frequency I have never purchased, but I intend to do so I have never purchased nor thought of I have purchased but I won’t again I purchase once or twice a year I purchase few times a year I purchase few times a month I purchase in a weekly basis

12.2 4.9 6.7 16.5 32.3 24.4 3.0

9.1 5.5 0.9 10.0 35.5 30.9 8.2

Dutch products purchasing frequency I have never purchased, but I intend to do so I have never purchased nor thought of I have purchased but I won’t again I purchase once or twice a year I purchase few times a year I purchase few times a month I purchase in a weekly basis

7.3 3.7 6.1 7.9 38.4 28.0 8.5

7.3 1.8 3.6 9.1 34.5 36.4 7.3

Beer purchasing frequency At least once a week Few times a month Once a month Few times a year Never

20.7 27.4 11.6 34.8 5.5

20.9 28.2 15.5 31.8 3.6

Delicatessen purchasing frequency At least once a week Few times a month Once a month Few times a year Never

35.4 37.2 9.8 14.6 3.0

33.6 37.3 16.4 11.8 0.9

Yellow cheese purchasing frequency At least once a week Few times a month Once a month Few times a year Never

46.3 32.3 9.1 9.8 2.4

47.3 39.1 4.5 6.4 2.7

Characteristics

t-test

Sig.

12.284

n.s.

Ethnocentric beliefs and COO effect 1535

3.499

n.s.

1.409

n.s.

4.137

n.s.

3.708

n.s.

Note: n.s. non statistically significant

Assessment of product evaluation criteria per cluster The process of the evaluation of products was repeated for each cluster (see Tables X and XI). Comparing the attributes of consumers in the two clusters towards Greek and Dutch beer, it is found that ethnocentric consumers evaluate the two products differently (statistically significant differences) in 26 out of 27 attributes (96.3 per cent). On the contrary, only in 12 attributes (44.4 per cent) the two products are perceived differently by the non-ethnocentric. Regarding ham products from Greece and Italy, ethnocentric

Table IX. Purchasing behaviour of the 2 clusters in terms of the products under evaluation, per cent

Is easily consumable/edible Contains no additives Is low in calories Tastes good Contains natural ingredients Is not expensive Is low in fat Is familiar Is nutritious Is easily available in shops and supermarkets Is good value for money Cheers me up Smells nice Can be consumed/eaten very simply Helps me cope with stress Helps me control my weight Has a pleasant texture Is packaged in an environmentally friendly way Are like the products I ate when I was a child Contains no artificial ingredients Keeps me awake/alert Packaging looks nice Helps me relax Is high in protein Takes no time to consume/eat Keeps me healthy Makes me feel good Has the country of origin clearly marked Is what I usually eat Helps me to cope with life Can be bought in shop close to where I live/work Is cheap

0.003 * 0.610 (n.s.) 0.741 (n.s.) 0.834 (n.s.) 0.092 (n.s.) 0.000 * 0.320 (n.s.) 0.000 * 0.052 (n.s.) 0.000 * 0.002 * 0.253 (n.s.) 1.000 (n.s.) 0.038 * * 0.707 (n.s.) 0.530 (n.s.) 0.294 (n.s.) 0.200 (n.s.) 0.000 * 0.031 * * 0.469 (n.s.) 0.011 * * 0.052 (n.s.) 0.264 (n.s.) 0.117 (n.s.) 0.158 (n.s.) 0.196 (n.s.) 0.221 (n.s.) 0.000 * 0.274 (n.s.) 0.000 * 0.000 *

0.000 * 0.857 (n.s.) 0.000 * 0.000 * 0.000 * 0.000 * 0.001 * 0.000 * 0.001 * 0.000 * 0.000 * 0.001 * 0.000 * 0.000 * 0.137 (n.s.) 0.082 (n.s.) 0.001 * 0.002 * 0.000 * 0.007 * 0.145 (n.s.) 0.526 (n.s.) 0.001 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.001 * 0.000 * 0.000 *

0.0 * 0.030 * * – 0.595 (n.s.) 0.006 * 0.000 * – 0.000 * – 0.000 * 0.000 * 0.735 (n.s.) 0.360 (n.s.) 0.089 (n.s.) 0.083 (n.s.) – 0.283 (n.s.) 0.139 (n.s.) 0.002 * 0.071 (n.s.) 0.090 (n.s.) 0.000 * 0.540 (n.s.) – 0.400 (n.s.) 0.530 (n.s.) 0.625 (n.s.) 0.302 (n.s.) 0.000 * 0.863 (n.s.) 0.000 * 0.000 *

0.000 * 0.000 * – 0.000 * 0.000 * 0.000 * – 0.000 * – 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.001 * – 0.000 * 0.000 * 0.000 * 0.000 * 0.003 * 0.414 (n.s.) 0.000 * – 0.000 * 0.041 * * 0.000 * 0.000 * 0.000 * 0.019 * * 0.000 * 0.000 *

0.000 * 0.000 * 0.009 * 0.000 * 0.000 * 0.001 * 0.105 (n.s.) 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.014 * * 0.054 0.000 * 0.001 * 0.000 * 0.000 * 0.008 * 0.127 (n.s.) 0.010 * – 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 * 0.000 *

0.330 (n.s.) 0.000 * 0.164 (n.s.) 0.029 * * 0.000 * 0.027 * * 0.141 (n.s.) 0.017 * * 0.000 * 0.001 * 0.348 (n.s.) 0.510 (n.s.) 0.390 (n.s.) 0.348 (n.s.) 0.025 * * 0.339 (n.s.) 0.777 (n.s.) 0.804 (n.s.) 0.000 * 0.000 * 0.158 (n.s.) 0.004 * 0.259 (n.s.) – 0.852 (n.s.) 0.003 * 0.334 (n.s.) 0.927 (n.s.) 0.000 * 0.235 (n.s.) 0.000 * 0.035 * *

Yellow cheese Greek and Dutch Ethnocentric Non-ethnocentric

Note: Question 19 “Comes from country I approve of politically” was left out since there is no corresponding reply for the equivalent Greek product; * statistically significant differences for p , 0:01; * * statistically significant differences for p , 0:05; n.s.: non statistically significant

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33.

Characteristics of evaluation

Table X. Differences per cluster between Greek and Dutch beer, Greek and Italian delicatessen and Greek and Dutch cheese (paired t-tests) Delicatessen Greek and Italian Ethnocentric Non-ethnocentric

1536

Beer Greek and Dutch Ethnocentric Non-ethnocentric

EJM 41,11/12

Characteristics of evaluation 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28.

Is easily consumable/drinkable Contains no additives Tastes good Contains natural ingredients Is not expensive Is familiar Is easily available in shops and supermarkets Is good value for money Cheers me up Smells nice Can be consumed/drunk very simply Helps me cope with stress Has a pleasant texture Is packaged in an environmentally friendly way Is like the beer I drunk when I was a child Contains no artificial ingredients Keeps me awake/alert Packaging looks nice Helps me relax Takes no time to consume/drink Keeps me healthy Makes me feel good Has the country of origin clearly marked Is what I usually drink Helps me to cope with life Can be bought in shop close to where I live/work Is cheap

Significance Ethnocentric Non-ethnocentric 0.000 * 0.164 (n.s.) 0.000 * 0.000 * 0.000 * 0.008 * 0.000 * 0.004 * 0.000 * 0.122 (n.s.) 0.002 * 0.000 * 0.000 * 0.046 * * 0.316 (n.s.) 0.000 * 0.027 * * 0.264 (n.s.) 0.000 * 0.007 * 0.000 * 0.000 * 0.258 (n.s.) 0.000 * 0.033 * * 0.000 * 0.038 * *

0.002 * 0.058 (n.s.) 0.650 (n.s.) 0.000 * 0.073 (n.s.) 0.200 (n.s.) 0.000 * 0.285 (n.s.) 0.000 * 0.250 (n.s.) 0.000 * 0.000 * 0.283 (n.s.) 0.961 (n.s.) 0.105 (n.s.) 0.029 * * 0.005 * 0.001 * 0.000 * 0.000 * 0.000 * 0.000 * 0.257 (n.s.) 0.072 (n.s.) 0.105 (n.s.) 0.000 * 0.530 (n.s.)

Notes: Question 15 “Comes from country I approve of politically” was left out since it was common for the two products; * statistically significant differences for p , 0:01; * * statistically significant differences for p , 0:05; n.s.: non statistically significant

consumers evaluate them differently in 27 out of 33 attributes (81.8 per cent) whereas the non-ethnocentric in only 13 (37.5 per cent). Correspondingly, the ethnocentric consumers perceive Greek yellow cheese as differing from the Dutch one in 29 out of 32 attributes (93.5 per cent) and the non-ethnocentric consumers in only 15 out of 32 attributes (48.3 per cent). The same pattern can be observed when comparing two products from the same country (beer and yellow cheese from Holland). Ethnocentric respondents perceive the two products as different in 22 out of their 27 (common) attributes (81.4 per cent), whereas non-ethnocentric consumers evaluate them differently only in 14 attributes (51.8 per cent). Interestingly, for the overwhelming majority of attributes, both ethnocentric and non-ethnocentric consumers evaluate more favourably the Greek products over their foreign origin counterparts. More specifically, ethnocentric consumers evaluate Greek products as superior to foreign alternatives in all attributes apart from the additional chemical substances content in ham products. The non-ethnocentric consumers also evaluated Greek products more favourably, albeit the differences between products of

Ethnocentric beliefs and COO effect 1537

Table XI. Differences per cluster between Dutch beer and yellow cheese, per cluster (paired t-tests)

EJM 41,11/12

1538

Greek and non-Greek origin are not as prominent as in the case of the ethnocentric. Striking differences were observed in the price, value for money, ease of consumption and availability/familiarity in beer and ham. Regarding Greek yellow cheese, price along with attributes related to health and safety are the most favourably evaluated by the non-ethnocentric consumers. A notable fact is that both ethnocentric and non-ethnocentric perceive the three Greek product types as traditional (“Are like the products I ate when I was a child”), which adds greater emotional value. Discussion The evaluation of CE level of (relatively younger) Greeks consumers is the first aim of the study. Use of the CET-SCALE (Shimp and Sharma, 1987) showed an average value of 3.85 in the 1 to 7 scale. This result pinpoints that the sample of mainly younger and well-educated respondents used in the present survey can not be characterised as particularly ethnocentric. Principal component and confirmatory factor analyses were used to fulfil the second aim. While the former extracted two (explaining 59.62 per cent of total variance), the latter confirmed only one component justifying the uni-dimensionality of CE postulated in the literature. As for the third aim, cluster analysis statistically imposed the allocation of the sample in two clusters. The first cluster (59.8 per cent of the sample) was characterised as ethnocentric (average value of 2.97) and encompassed the older and less educated individuals of the sample. The second cluster (40.2 per cent) was characterised as non-ethnocentric (average value of 5.16) and comprises mostly young and well-educated respondents. The fore mentioned findings confirm most of the international literature, where it has been agreed that consumer ethnocentrism is directly related to age (the older one person, the more likely to exhibit ethnocentric beliefs), while there is a reverse relation of ethnocentrism with education (the higher the education, the lesser the level of ethnocentrism). Moreover, the incongruity observed in age increments underlines the existence of a sudden change regarding ethnocentric beliefs: the highest percentages in every age increment can be observed in the ethnocentric cluster, excluding the respondents up to 35 years old whose majority belonged to the non-ethnocentric cluster. Consequently, it seems that from the last generation onwards (namely people below the age of 35), consumers are less ethnocentric as compared to previous generations. This phenomenon can be attributed to the fact that younger individuals are more acquainted with foreign countries and more receptive to the products they produce, given Greece’s accession in the European Union (EU) during the early 1980s. It is thus possible that modern consumers (between 26-35 years old) are less prejudiced when evaluating foreign products and adopt a more sceptical stance towards traditional stereotypes such as “Greek origin equals good quality”, etc. Finally, concerning the fourth objective, cluster analysis results showed that ethnocentrism affects not only consumer beliefs but also the way the perceived quality of domestic and foreign products is evaluated, culminating in the appearance of COO effect. Examination of preferences of the ethnocentric consumers highlights the existence of statistically significant differences in the way they evaluate Greek and foreign products in favour of the former. These differences were observed for the vast majority

of product attributes (26 out of 27 beer attributes, 27 out of 32 ham attributes and 28 out of yellow cheese attributes). Moreover, it is worth noting that in all three product categories, the attribute of “clear indication of the country of origin” was evaluated statistically differently meaning that ethnocentric consumers do pay attention in the indication of Greece as the country of origin in food products. It can be thus concluded that, in the case of ethnocentric consumers, the COO effect is activated at the initiatory level, namely that of the country a food product originates in (country-specific), and not at the level of the product type (product-specific) or product attributes (attribute-specific), since in every domestic vs. foreign comparison the Greek product was evaluated distinctively and overwhelmingly higher, with no variations from product to product or from attribute to attribute (see Figure 2). Especially regarding the comparison between Dutch beer and Dutch yellow cheese, ethnocentric consumers evaluated differently 22 out of their 27 common attributes (81.4 per cent), connoting that they perceive these two products as two different “entities”, regardless their country of origin. That leads to the conclusion that when the foreign

Ethnocentric beliefs and COO effect 1539

Figure 2. Level of activation of the link between consumer ethnocentrism and COO effect per product origin during product evaluation

EJM 41,11/12

1540

country of origin is given, then the COO effect is activated at the level of the product type (product-specific). Concerning the non-ethnocentric consumers, there were also statistically significant differences observed in the way they evaluated Greek and foreign products, albeit in a lesser degree. The non-ethnocentric generally evaluated more favourably the domestic products however, unlike the ethnocentric, foreign products were not overall rejected: many of their attributes were evaluated as superior comparing to the corresponding attributes of their Greek alternatives. In every comparison between the country product and the foreign alternative non-ethnocentric consumers perceived them as having significant differences in no more than the 50 per cent of their attributes. It is worth noting that the attribute of “clear indication of the country of origin” was not found statistically different, meaning that non-ethnocentric consumers do not pay attention necessarily in the indication of Greece as the country of origin in food. Given that statistical differences were not observed for the majority of attributes, it cannot be claimed that country of origin has a distinctive effect over non-ethnocentric consumers’ evaluations. On the contrary, the partially only differentiation in some attributes denotes the existence of COO effect activated at the level of product attributes (attribute-specific). The country of origin does not lead to an overall acceptance or rejection, but instead it affects the evaluation of specific product attributes. Similar findings were extracted during the Dutch beer vs Dutch yellow cheese comparison. From the body of 27 common attributes, only 15 (55.6 per cent) were found statistically significant. This relatively modest percentage leads to the conclusion that non-ethnocentric consumers do not perceive two products of foreign origin as two entirely different “entities”, at least not as entirely as in the case of the ethnocentric consumers. Since the differences are located in specific attributes, we presume that the COO effect is salient not at the level of product but at the level of attribute (attribute-specific). Non-ethnocentric respondents evaluated several foreign products’ attributes positively, especially the attributes concerning their perceived quality. More specifically, they appeared to appreciate “taste” and the “feeling in the mouth” of the Dutch beer and Italian ham. As for the Dutch yellow cheese, it was favourably evaluated in terms of “taste”, “feeling in the mouth”, “nutritional value”, “low fat/calorie content” and “relaxation”. Finally, all three foreign product types where evaluated positively in terms of “ease of consumption”, while their package was evaluated as superior to that of their Greek counterparts. Finally, while the effect of ethnocentrism on consumers’ beliefs towards food products is unquestionable, no equivalent effect was observed in the purchasing frequency habits towards products originating from foreign countries. Cross-tabulation of purchasing frequencies with the two clusters demonstrate few trivial differences between ethnocentric and non-ethnocentric consumers, but none of them was found statistically significant. Thus, it appears that ethnocentrism does not apparently affect the final purchasing behaviour, as this is reflected through the number of purchases of foreign products. As it was mentioned earlier, Wall et al. (1991) found that the country of origin relates to products’ quality evaluations but it is of trivial importance when it comes to evaluation of purchasing likelihood. Similar findings arose by Rahman (2000) who postulated that COO effect influences consumers’ product evaluations but not necessarily their final purchasing behaviour, as the latter is affected by other more powerful predictors of behaviour, such as price sensitivity.

Conclusions – managerial implications The current study argues that middle-aged Greek consumers (age of 40 or older) exhibit a moderate ethnocentric attitude in the evaluation of food products, whereas younger consumers (around 35 years old) are much less ethnocentric. Focus on the three products under evaluation and separately for the two clusters of ethnocentric and non-ethnocentric consumers revealed the set of food attributes that consumers seem to mostly appreciate. CE and COO effect are linked together, but the stimulus that activates their link differs between the two clusters. In the ethnocentric cluster, the mere mentioning of a foreign COO suffices to create a negative bias against foreign products, which subsequently leads to a non-favourable evaluation of their attributes, as compared to domestic products. On the other hand, in the non-ethnocentric cluster COO effect appears as an expression of ethnocentrism only in the level of product attributes, resulting to a favourable but selective evaluation of only specific attributes. Simple mentioning of the COO (between Greek and foreign foods) or the type of a product (between foreign origin foods) is not enough to activate the link between ethnocentrism and COO effect. Ethnocentric consumers evaluated more favourably all three Greek products. Given that their ethnocentric beliefs are activated at the level of the country a food product is originated in, it is easily understandable that a successful marketing strategy should focus on bringing out the Greek origin of products. However, in the case of non-ethnocentric consumers such a strategy will not be effective since the superiority of Greek products is limited to specific only attributes per product category. The present study suffers the limitation of the sample not being representative of the Greek population. All respondents were residence of the capital city (Athens), while the sample is purposively biased towards relatively younger and more educated consumers. It is not possible thus to generalise the findings for the entire population with various demographic profiles. Results may be different in rural areas, where educational level as well as familiarity with foreign products is lower. Moreover, the present study focused on the influence of ethnocentric beliefs in food products evaluation (COO effect) and not on their impact on final purchasing behaviour. It would be challenging for future research to examine if the current findings apply to a broader set of products and countries of origin, including countries that differ significantly from EU countries in cultural and economic terms. Also, in light of the current results, future research should consider products with no domestic competitive alternatives, as well as more than two foreign countries of origin for each product, so as to examine whether a country is regarded as the “benchmark” for each product type or are all foreign products are being evaluated similarly. Another promising area would be the further examination of COO effect with reference to brand names in order to investigate their possibility of moderating or intensifying the phenomenon. Finally, it would be useful to approach the issue of COO effect considering more factors of psychological, political and technological nature, such as the perceived risk and level of development of the country of origin. References Ahmed, S. and d’Astus, A. (1995), “Comparison of country of origin effects on household and organizational buyers’ product perceptions”, European Journal of Marketing, Vol. 29 No. 3, pp. 35-51.

Ethnocentric beliefs and COO effect 1541

EJM 41,11/12

1542

Ahmed, S. and d’Astus, A. (1996), “Country-of-origin and brand effects: a multi-dimensional and multi-attribute study”, Journal of International Consumer Marketing, Vol. 9 No. 2, pp. 9-31. Al-Sulaiti, K. and Baker, M. (1998), “Country of origin effects: a literature review”, Marketing Intelligence & Planning, Vol. 16 No. 3, pp. 150-99. Bailey, W. and Pineres, S. (1997), “Country of origin attitudes in Mexico: the malinchismo effect”, Journal of International Consumer Marketing, Vol. 9 No. 3, pp. 25-41. Balabanis, G. and Diamantopoulos, A. (2004), “Domestic country bias, country-of-origin effects, and consumer ethnocentrism: a multidimensional unfolding approach”, Journal of the Academy of Marketing Science, Vol. 32 No. 1, pp. 80-95. Brunning, E. (1997), “Country of origin, national loyalty and product choice”, International Marketing Review, Vol. 14 No. 1, pp. 59-74. Cordell, V. (1992), “Effects of consumer preferences for foreign sourced products”, Journal of International Business Studies, Vol. 23 No. 2, pp. 251-69. de Ruyter, K., van Birgelen, M. and Wetrels, M. (1998), “Consumer ethnocentrism in international services marketing”, International Business Review, Vol. 7 No. 2, pp. 185-202. Gurhan-Canli, Z. and Maheswaran, D. (2000), “Determinants of country-of-origin evaluations”, Journal of Consumer Research, Vol. 27 No. 1, pp. 96-108. Han, C.M. (1989), “Country image: halo or summary construct?”, Journal of Marketing Research, Vol. 26, May, pp. 222-9. Han, C.M. (1990), “Testing the role of country image in consumer choice behaviour”, European Journal of Marketing, Vol. 24 No. 6, pp. 24-39. Herche, J. (1994), “Ethnocentric tendencies, marketing strategy and import purchase behavior”, International Marketing Review, Vol. 11 No. 3, pp. 4-16. Javalgi, R.G., Pioche Khare, V. and Gross, A.C. (2005), “An application of the consumer ethnocentrism model to French consumers”, International Business Review, Vol. 14, pp. 325-44. Juric, B. and Worsley, A. (1998), “Consumers’ attitudes towards imported food products”, Food Quality and Preference, Vol. 9 No. 6, pp. 431-41. Kaynak, E. and Kara, A. (2002), “Consumer perceptions of foreign products: an analysis of product-country images and ethnocentrism”, European Journal of Marketing, Vol. 36 Nos 7/8, pp. 928-49. Kaynak, E., Kucukemiroglu, O. and Kara, A. (1994), “Consumers’ perceptions of airlines: a correspondence analysis approach in a global airline industry”, Management International Review, July. Khachaturian, J. and Morganosky, M. (1990), “Quality perceptions by country of origin”, International Journal of Retail & Distribution Management, Vol. 18 No. 5. Knight, G. and Calantone, R. (2000), “A flexible model of consumer country-of-origin perceptions: a cross-cultural investigation”, International Marketing Review, Vol. 17 No. 2, pp. 127-45. Kucukemiroglou, O. (1997), “Market segmentation by using consumer lifestyle dimensions and ethnocentrism”, European Journal of Marketing, Vol. 33 Nos 5/6, pp. 470-87. Lantz, G. and Loeb, S. (1996), “Country of origin and ethnocentrism: an analysis of Canadian and American preferences using social identity theory”, in Corfman, K.P. and Lynch, J. (Eds), Advances in Consumer Research, Vol. 23, Association of Consumer Research, Provo, UT, pp. 374-8. Lee, D. and Ganesh, G. (1999), “Effects of partitioned country image in the context of brand image and familiarity”, International Marketing Review, Vol. 16 No. 1, pp. 18-39.

McLain, S. and Sternquist, B. (1991), “Ethnocentric consumers: do they buy American?”, Journal of International Consumer Marketing, Vol. 4 Nos 1/2, pp. 39-57. Maheswaran, D. (1994), “Country of origin as a stereotype: effects of consumer expertise and attribute strength on product evaluations”, Journal of Consumer Research, Vol. 21 No. 2, pp. 354-65. Moon, B.J. and Jain, S.C. (2002), “Consumer processing of foreign advertisements: roles of country of origin perceptions, consumer ethnocentrism, and country attitude”, International Business Review, Vol. 11 No. 2, pp. 117-38. Okechuku, C. (1994), “The importance of product country of origin”, European Journal of Marketing, Vol. 24 No. 4, pp. 5-19. Orth, U. and Firbasova, Z. (2003), “The role of consumer ethnocentrism in food product evaluation”, Agribusiness, Vol. 19 No. 2, pp. 137-53. Papadopoulos, N. (1993), “What country images are and are not”, in Papadopoulos, N. and Helsop, L. (Eds), Product Country Images, International Business Press, New York, NY. Peterson, R. and Jolibert, A.J.P. (1995), “A meta-analysis of country-of-origin effects”, Journal of International Business Studies, Vol. 26 No. 4, pp. 883-900. Piron, F. (2000), “International outshopping and ethnocentrism”, European Journal of Marketing, Vol. 36 Nos 1/2, pp. 189-210. Rahman, S. (2000), “Country of origin effect in a developing country: does it really matter?”, dissertation, University of Western Sydney, Sydney. Rawwas, M.Y.A. and Rajendran, K.N. (1996), “The influence of worldmindedness and nationalism on consumer evaluation of domestic and foreign products”, International Marketing Review, Vol. 13 No. 2, pp. 20-38. Roth, M. and Romeo, J. (1992), “Matching product category and country image perceptions: a framework for managing country-of-origin effects”, Journal of International Business Studies, Vol. 23 No. 3, pp. 477-97. Schaefer, A. (1995), “Consumer knowledge and country of origin effects”, European Journal of Marketing, Vol. 31 No. 1, pp. 56-72. Shimp, T. and Sharma, S. (1987), “Consumer ethnocentrism: construction and validation of the CETSCALE”, Journal of Marketing Research, Vol. 24, August, pp. 280-9. Showers, V. and Showers, L. (1993), “The effects of alternative measures of country of origin on objective product quality”, International Marketing Review, Vol. 10 No. 4, pp. 53-67. Smith, W. (1993), “Country-of-origin bias: a regional labelling solution”, International Marketing Review, Vol. 10 No. 6, pp. 4-12. Steenkamp, J.-B.E.M. and Baumgartner, H. (1998), “Assessing measurement invariance in cross-national consumer research”, Journal of Consumer Research, Vol. 25 No. 1, pp. 78-90. Steptoe, A., Polland, T.M. and Wardle, J. (1995), “Development of a measure of the motives underlying the selection of food: the food choice questionnaire”, Appetite, Vol. 25 No. 3, pp. 267-84. Stoltman, J.J., Lim, Y.K. and Morgan, F.W. (1991), “The effect of country of origin, product familiarity and ethnocentrism on the acceptance of foreign products”, Marketing Theory and Applications, Academy of Marketing Winter Educators Conference, pp. 82-9. Supphellen, M. and Rittenburgh, T. (2001), “Consumer ethnocentrism when foreign products are better”, Psychology and Marketing, Vol. 18 No. 9, pp. 907-27. Verlegh, P. and Steenkamp, J.B. (1999), “A review and meta-analysis of country of origin effect research”, Journal of Economic Psychology, Vol. 20 No. 5, pp. 521-46.

Ethnocentric beliefs and COO effect 1543

EJM 41,11/12

1544

Wall, M., Liefeld, J. and Heslop, L. (1991), “Impact of country of origin cues on consumer judgements in multi-cue situations: a covariance analysis”, Journal of the Academy of Marketing Science, Vol. 19, Spring, pp. 105-13. Wang, C. and Chen, Z. (2004), “Consumer ethnocentrism and willingness to buy domestic products in a developing country setting: testing moderating effects”, Journal of Consumer Marketing, Vol. 21 No. 6, pp. 391-400. Watson, J. and Wright, K. (2000), “Consumer ethnocentrism and attitudes towards domestic and foreign products”, European Journal of Marketing, Vol. 34 Nos 9/10, pp. 1149-66. Yarpak, A. and Baughan, C. (1991), “The country of origin effect in cross-national consumer behaviour: emerging research avenues”, Proceedings of the 5th Bi-annual World Marketing Congress of the Academy of Marketing Science, pp. 263-9. Further reading Tan, J.S., Lee, S.K. and Lim, H.G. (2001), “Warranty and warrantor reputations as signals of hybrid product quality”, European Journal of Marketing, Vol. 35 Nos 1/2, pp. 110-32. About the authors George Chryssochoidis is Assistant Professor in Food Marketing at the Agricultural University of Athens, Laboratory of Agribusiness Management. Athanassios Krystallis is Senior Researcher, Food Marketing and Consumer Behaviour at the National Agricultural Research Foundation (NAGREF), Athens, Greece. Athanassios Krystallis is the corresponding author and can be contacted at: [email protected] Panagiotis Perreas is an MBA Graduate at the Agricultural University of Athens, Laboratory of Agribusiness Management.

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints