臺歐消費品貿易之擴展空間 - 政大學術集成

全文

(1)國立政治大學亞太研究英語碩士學位學程 International Master’s Program in Asia-Pacific Studies College of Social Sciences National Chengchi University. Master’s Thesis. 學. ‧ 國. 政治大立碩士論文 ‧ sit. y. Nat. er. io. 臺歐消費品貿易之擴展空間. al. n. v i n C hgoods trade –U Space for expansion? Taiwan-EU consumer engchi. Student: Martin Kovář 柯馬汀 Advisor: Professor Hsiao-hung Nancy Chen 陳小紅教授. 中華民國 104 年 7 月 July 2015.

(2) 臺歐消費品貿易之擴展空間 Taiwan-EU consumer goods trade – Space for expansion? 研究生: 柯馬汀. Student: Martin Kovář. 指導教授: 陳小紅教授. Advisor: Professor Hsiao-hung Nancy Chen. 國立政治大學亞太研究英語碩士學位學程政治. 立. 大. 碩士論文. ‧. ‧ 國. 學 A Thesis. y. Nat. er. io. sit. Submitted to International Master’s Program in Asia-Pacific Studies National Chengchi University. al. n. v i n Ch In partial fulfilment Requirement i U e n g cofhthe For the degree of Master in China Studies. 中華民國 104 年 7 月 July 2015.

(3) 論文摘要國際貿易統計數據與平日消費者體察示意台灣和歐盟的消費品貿易有明顯的擴展空間。本研究集注消費品進口以期找出某些品類較少進口數量的原因。根據其他的進口國家－日本和韓國－貿易結構數據比較分析，挑選較小市場佔有率的品類，其後使用問卷調查消費者對此類歐盟進口產品的看法。本研究發現台灣消費者對歐盟產品與歐盟廠商有一定的認知，且消費. 治政大者相當讚賞歐盟產品的品質，此為購買意向增加之原因。而商品的原立. 產地形象顯然與消費者之購買意向沒有直接關係。較高歐盟產品見識. ‧ 國. 學. 度貌似跟購買意向有正相關，因此本研究建議進口商使用恰當的招攬. ‧. 活動，例如產品介紹會或試用會。. sit. y. Nat. 並且，藉由移除貿易障礙以及放寬貿易協定來改善台灣與歐盟的經濟. n. al. er. io. 關係，提供互相貿易之擴展機會。. Ch. engchi. i n U. v.

(4) Abstract Statistical data on international trade as well as daily consumer experience suggest that there are several areas in the trading relationship between Taiwan and the European Union, where there is a lot of space for further expansion. This study focuses on the imports of consumer goods in an effort to find the reasons why products of some categories are imported in relatively smaller quantity to Taiwan. The specific product categories, which have a relatively low share on the total value of imported goods, are chosen on the basis of comparison of international trade statistics of Taiwan and. 政治大. other comparable markets of the East Asian region – Japan and Korea –. 立. and subsequently. a questionnaire-based survey. of consumer attitude. ‧ 國. 學. towards the chosen product categories is conducted.. The research finds that products made in European Union and the maker. ‧. brands are quite well-known among Taiwanese consumers and their quality. y. sit. Nat. is perceived as being quite high, which seems to increase purchase intention. io. er. greatly. Country-of-origin image does not seem to play an important role in the purchase intention a shaping. The high levels of familiarity with the EU. n. v i l n C h correlation with products appear to have positive the final purchase intention engchi U and a suggestion is therefore made, that the producers and importers should try to promote it by suitable marketing activities, such as product presentation and trials. Development of economic ties between Taiwan and the European Union by the means of removal of trading barriers or preferential trade agreements also seems to provide a good opportunity for the expansion of the mutual trade..

(5) TABLE OF CONTENTS 1 Introduction ..................................................................................................................... 1 2 Literature review and Research Design .......................................................................... 9 2.1 Scope of the research .................................................................................................... 9 2.2 Research design .......................................................................................................... 10 2.2.1 Conceptual model .............................................................................................. 10 2.2.2 Product and brand familiarity ............................................................................ 11 2.2.3 Country of origin image .................................................................................... 13 2.2.4 Quality perception ............................................................................................. 15 2.2.5 Value perception................................................................................................ 15 3 Methodology ................................................................................................................. 16 3.1 Questionnaire design................................................................................................... 16 3.1.1 Product and brand familiarity ............................................................................ 16 3.1.2 Country of origin image .................................................................................... 18 3.1.3 Quality perception ............................................................................................. 19 3.1.4 Value perception................................................................................................ 20 3.1.5 Purchase intention ............................................................................................. 20 3.1.6 Personal data...................................................................................................... 20 3.2 Data Collection ........................................................................................................... 21 4 Research findings .......................................................................................................... 24 4.1 The brand familiarity variable .................................................................................... 24 4.2 Country of origin image variable ................................................................................ 25 4.3 Country-wise observations ......................................................................................... 26 4.4 Analysis ...................................................................................................................... 27 4.4.1 Bivariate relationships analysis ......................................................................... 30 4.4.2 Regression analysis ........................................................................................... 32 5 Conclusions ................................................................................................................... 38 6 Limitations and potential future research ...................................................................... 39 7 Appendices .................................................................................................................... 40 7.1 Appendix 1: Trade defence measures taken by EU against Taiwan ............... 40 7.2 Appendix 2: Trade defence measures taken by Taiwan affecting EU members . 41 7.3 Appendix 3: Import structure statistical comparison – Japan, Korea & Taiwan . 41 7.4 Appendix 4: The focus product categories – details ............................................ 43 7.5 Appendix 5: Food products evaluation criteria .................................................... 44 7.6 Appendix 6: The brand familiarity pre-test results .............................................. 45 7.7 Appendix 7: The questionnaire – English version ............................................... 46 7.8 Appendix 8: The questionnaire – Chinese version .............................................. 49 7.9 Appendix 9: Correlations – Cars.......................................................................... 52 7.10 Appendix 10: Correlations – Bags .................................................................... 52 7.12 Appendix 11: Correlations – Cheese ................................................................. 53 7.13 Appendix 12: Correlations – Shoes ................................................................... 53 7.14 Appendix 13: Country selection combinations data.......................................... 54 7.15 Appendix 14: COO image-income cross-tab analysis ...................................... 55 8 Bibliography .................................................................................................................. 56. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.

(6) List of Tables Chapter 1 – Introduction: Table 1 - Import structure statistics in comparison – Japan, Korea and Taiwan ................ 4 Table 2 - Main trade indicators - Product: 8703 Cars (incl. station wagon) ....................... 5 Table 3 - Main trade indicators - Product: 4202 Trunks, suitcases, handbags etc, ............. 6 Table 4 - Main trade indicators - Product: 0406 Cheese and curd...................................... 7 Table 5 - Main trade indicators - Product: 6403 Footwear, upper of leather...................... 8. Chapter 2 – Literature review and Research Design: Table 6 - Shares of different product categories in total import value ............................. 10. Chapter 3 – Methodology: Table 7 - Category captions .............................................................................................. 16 Table 8 - Brands chosen for the brand familiarity testing ................................................ 17 Table 9 - COO image research overview .......................................................................... 18 Table 10 - Phase 1 sample structure ................................................................................. 21 Table 11 - Phase 2 sample structure ................................................................................. 22 Table 12 - The final sample .............................................................................................. 23. 政治大. 立. Chapter 4 – Research findings:. ‧. ‧ 國. 學. Table 13 - Brand familiarity scores .................................................................................. 24 Table 14 - Structure of the COO image variable responses.............................................. 25 Table 15 - Country-specific data....................................................................................... 26 Table 16 - Univariate descriptive analysis ........................................................................ 27 Table 17 - Relation between "purchase intention" and "purchase intention of EU products" variables ... 29 Table 18 - Bivariate correlations with the dependent variable ......................................... 30 Table 19 - Correlations for hypothesised bivariate relationships ..................................... 31 Table 20 - Regression analysis results - Cars ................................................................... 32 Table 21 - Regression analysis results - Bags ................................................................... 33 Table 22 - Regression analysis results - Cheese ............................................................... 34 Table 23 - Regression analysis results - Shoes ................................................................. 35 Table 24 - The relative importance of brand prestige aspect of the COO image variable ............ 37. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. List of Figures Chapter 1 – Introduction: Figure 1 - European Union (EU 28)'s imports from Taiwan .............................................. 1 Figure 2 - Taiwan's imports from European Union (EU 28) .............................................. 2 Figure 3 - Top 5 importers from Taiwan ............................................................................ 2 Figure 4 - Top 5 sources of imports to Taiwan ................................................................... 3. Chapter 2 – Literature review and Research Design: Figure 5 - Conceptual Model ............................................................................................ 11.

(7) I would like to thank Professor Chen Hsiao-hung, who was the advisor for this thesis, and her assistant Mr. Lin Jun-hung for their help and guidance. I would also like to thank Ms. Lo Hsien-fang, Ms. Lü Mei-ling and Mr. Chen Jing-jie for their help with the questionnaire translation and distribution.. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v.

(8) 1 Introduction In the last decades, the economies of the Southeast Asia were among the world’s fastest growing ones. The economic and political reforms in many of these. countries. enabled. the growth. of international. trade,. while. the technical and industrial development reshaped its structure, making many East-Asian countries like Taiwan some of the top exporters of manufactured goods and especially electronics and machinery (as Figure 1 illustrates) . Figure 1 - European Union (EU 28)'s imports from Taiwan. 政治大. Electrical, electronic equipment. 14%. Machinery, nuclear reactors, boilers, etc. 立. 3%. ‧ 國. 37%. Tools, implements, cutlery, etc of base metal. Nat. 9%. Optical, photo, technical, medical, etc apparatus. ‧. 6%. Iron and steel. Toys, games, sports requisites. y. 4%. Plastics and articles thereof. 18%. io. Others. n. al. sit. 4%. Articles of iron or steel. 學. 3%. Vehicles other than railway, tramway. er. 2%. i n U. v. Source: Bilateral trade between European Union (EU 28) and Taipei, Chinese (ITC Trade Map 2015). C. h eandn g In the same time, the economic political c h i integration in Europe, resulting in the establishment and augmentation of the European Union promoted the cooperation among the European countries, creating a new economic entity comparable in size and importance to such a big international players as USA or China. The unification of different national economies with their respective traditional industry structures made European Union an important exporter with varied and quite balanced portfolio (as shown in Figure 2).. 1.

(9) Figure 2 - Taiwan's imports from European Union (EU 28) Machinery, nuclear reactors, boilers, etc Electrical, electronic equipment. 19%. Vehicles other than railway, tramway. 28%. Pharmaceutical products Optical, photo, technical, medical, etc apparatus 13%. Organic chemicals. 3%. Miscellaneous chemical products. 4%. Mineral fuels, oils, distillation products, etc 11%. 5% 5%. Beverages, spirits and vinegar. 6%. 6%. Others. Source: Bilateral trade between Taipei, Chinese and European Union (EU 28) (ITC Trade Map). 政治大 In recent years, the bilateral trade between European Union and Taiwan 立 was growing steadily, (as demonstrated in Figure 3 and 4) with the value. ‧ 國. 學. of Taiwan’s exports to EU reaching 19.9 billion EURO, which makes EU the fourth most important destination of Taiwanese exports as of 2014.. ‧. Figure 3 - Top 5 importers from Taiwan. 40000. y. sit er. al. n. 50000. io. 60000. Nat. 70000. Ch. engchi U. v ni. China Hong Kong, China United States of America. 30000 20000. European Union (EU 28) Aggregation. 10000. Singapore. 0. Unit: million EURO Source: List of importing markets from EU 28 for a product exported by Taipei, Chinese (ITC Trade Map). The value of imports from EU reached 18.3 billion EURO in 2014, making EU the fourth biggest source of Taiwanese imports after China, Japan and US. 2.

(10) Figure 4 - Top 5 sources of imports to Taiwan 45000 40000 35000. China. 30000. Japan. 25000 United States of America. 20000 15000. European Union (EU 28) Aggregation. 10000. Korea, Republic of. 5000 0. 政治大. Unit: million EURO Source: List of supplying markets for a product imported by Taipei, Chinese (ITC Trade Map). 立. Nonetheless, bilateral trade is still inhibited by several obstacles which. ‧ 國. 學. include a multitude of punitive measures aiming at trade protection; these measures seem to be to Taiwan’s disadvantage, as there currently are about. ‧. fourteen trade defence duties (of which ten are already expired), reaching. Nat. sit. y. amounts as high as 60% in some cases, imposed on Taiwanese imports. er. io. (see Appendix 1), while, according to information provided by Customs. n. administration of Ministry a of Finance (Appendix 2), v there seem to be only four trade defence groups. i l C n investigations by the Taiwanese h e nrun gchi U. of importers. including. European. countries. authorities against (none. of these. investigations targets at the European Union as a whole, however). The analysis of trade statistical data seems to show, that there are several product categories, traditionally exported from Europe to many other countries in the world, that are imported in much lesser quantity to Taiwan than to other comparable markets of the region (as demonstrated in Table 1). The purpose of this research is to find the reasons for the smaller imports of goods of these categories.. 3.

(11) Table 1 - Import structure statistics in comparison – Japan, Korea and Taiwan. HSC code (4-digit grouping). Product category caption. Taiwan. Japan. Korea. '8703 '4202. Cars (incl. station wagon) Trunks, suit-cases, camera cases, handbags etc., of leather, plastic, textile etc. Cheese and curd Footwear, upper of leather. 8.75% 0.98%. 10.66% 2.17%. 9.59% 1.55%. Difference between the two countries’ average and Taiwan 1.37% 0.88%. 0.10% 0.30%. 0.39% 0.59%. 0.17% 0.36%. 0.18% 0.18%. '0406 '6403. Unit: percent of the total value of goods imported from European Union (EU28) Source: Bilateral trade between EU and Taipei, Chinese (..., Japan, Korea) (ITC Trade Map). 政治大. None of the trade defence measures discussed above affects any of these. 立. product categories, there is, however, a number of preferential tariffs for. ‧ 國. 學. goods originating in certain countries (see Tables 2-5).. ‧. The import tariffs for the product categories in question vary significantly (as shown in tables 2-5). The distribution of preferential tariffs to a large. y. Nat. sit. extent follows trade agreements established especially with many countries. er. io. of Central America, with which Taiwan maintains official diplomatic. n. relationships (Scott et aal. n.d.), furthermore Taiwan v maintains exceptional. i l C n U trading ties with New Zealand, import tariffs for all h which e n genjoys c h i liberalised. products except rice, and this preferential treatment is reciprocated by New Zealand by liberalising all import tariffs for Taiwanese products (MOEA Press Release 2013). For the product categories in question, no EU country is affected by increased tariff measures nor is any included in any reduced tariff program (as further discussed in the next section and the relevant tables).. 4.

(12) Table 2 - Main trade indicators - Product: 8703 Cars (incl. station wagon). Top ten sources of import to Taiwan. Germany Japan USA UK Belgium South Africa Italy India Hungary Sweden. Imported value 2014 (USD thousand). Share in Taiwan's imports (%). Ranking in world exports. Share in world exports (%). Tariff (estimated) applied by Taiwan (%). 1566208 1181293 298002 143802 100980 90849 69515 60618 57270 30398. 41.2 31.1 7.8 3.8 2.7 2.4 1.8 1.6 1.5 0.8. 1 2 3 6 9 23 13 18 14 20. 23.3 12.6 8.8 6 4.3 0.6 1.7 0.8 1.6 0.7. 17.5 17.5 17.5 17.5 17.5 17.5 17.5 17.5 17.5 17.5. 0. 61 186. 0 0. 0 17.4. 100. 6039. 97.8. 17.5. 0. 632. 0. 59.8. Tariff groups 0. 立. 政治大. 學. 0. ‧. ‧ 國. 3802684. er. io. sit. y. Nat. New Zealand El Salvador, Guatemala, Panama Germany, Japan, USA, UK, Belgium + 86 countries French Polynesia, F. S. Antarctic Territories, N. Antilles, Aruba, N. Mariana Isl ands. Source: List of supplying markets for a product imported by Taipei, Chinese (ITC Trade Map). al. n. v i n C h the ten biggest There are six EU countries among e n g c h i U suppliers of cars and related products of the “8703 Cars (incl. station wagon)” category in terms of the value of goods imported to Taiwan: Germany, UK, Belgium, Italy, Hungary and Sweden. Germany, which occupies the first place with imports worth 1.566. billion. USD. is followed. quite. closely. by Japan. with 1.181 billion USD. The average tariff ratio for majority of importers of goods of this group amounts to 17.5 %, which is a value much higher than the tariffs for all the other categories concerned.. 5.

(13) Table 3 - Main trade indicators - Product: 4202 Trunks, suitcases, camera cases, handbags etc,. Top ten sources of import to Taiwan Imported value 2014 (USD thousand). Share in Taiwan's imports (%). Ranking in world exports. Share in world exports (%). Tariff (estimated) applied by Taiwan (%). China Italy France Viet Nam Spain Czech Republic Japan Rep. of Korea Thailand India. 261242 126793 78748 25034 19024 6319 5888 5380 4420 4381. 45.9 22.3 13.8 4.4 3.3 1.1 1 0.9 0.8 0.8. 1 2 3 5 11 15 41 16 17 8. 43.7 11.8 9.6 4.3 1.6 0.6 0.1 0.6 0.5 2. 6.9 6.9 6.9 6.9 6.9 6.9 6.9 6.9 6.9 6.9. Guatemala, New Zealand, El Salvador, Honduras, Nicaragua, Panama Bangladesh, Cambodia, Nepal, Laos, Madagascar + 9 countries Singapore China, Italy, France, Viet Nam, Spain + 93 countries Bosnia and Herzegovina, Belarus, F. Polynesia, Greenland, N. Antilles. 5. 0. 0. 0.3. 3.1. Exporters. Tariff groups. 政治大. 1597. 0.3. 1412. 1026 565471. 0.2 99.1. 13 6347. 0. 724. ‧ 國. 立. 1.1 98.2. 4.7 6.9. n. al. Ch. engchi. er. io. sit. y. Nat 17. ‧. 504. 學. 0. i n U. 0. 14.6. v. Source: List of supplying markets for a product imported by Taipei, Chinese (ITC Trade Map). Four of the ten most important suppliers of products of the category “4202 Trunks, suitcases, camera cases, handbags etc.” are EU member countries: Italy, France, Spain and Czech Republic, however this sector is strongly dominated by China, importing goods worth 261.242 million USD, followed by Italy, which imports goods worth only half of that value – 126.793 million USD. All the top ten importers and all the EU countries are subjected to 6.9 % tariff. 6.

(14) Table 4 - Main trade indicators - Product: 0406 Cheese and curd. Top ten sources of import to Taiwan Imported value 2014 (USD thousand). Share in Taiwan's imports (%). Ranking in world exports. Share in world exports (%). Tariff (estimated) applied by Taiwan (%). USA New Zealand Australia Argentina Italy France Brazil Japan Denmark UK. 40035 37626 13974 13272 5777 5030 3319 2681 2672 2298. 29.5 27.7 10.3 9.8 4.3 3.7 2.4 2 2 1.7. 5 7 13 21 4 2 54 62 6 12. 5.4 4.1 2.2 0.9 9 12.7 0 0 5.3 2.4. 5 0 5 5 5 5 5 5 5 5. New Zealand, Singapore USA, Australia, Argentina, Italy, France + 30 countries Belarus. 37717. Tariff groups. 立. 4.2. 0. 72.4. 1104. 83.5. 5. 2.5. 20. 學. 60. ‧ 國. 98101. 政治大 27.8. 11. Source: List of supplying markets for a product imported by Taipei, Chinese (ITC Trade Map). ‧. Similarly to the previous category, there are four EU countries among the top. y. Nat. ten suppliers of goods of “0406 Cheese and curd” category: Italy, France,. io. sit. Denmark and UK. The values of European imports in this category are. er. surprisingly small, with the leading EU country – Italy (5.7 mil. USD) – not. n. a. v. even achieving half lthe n i biggest importer – Cvalues of the previous Argentina (13.2 mil.. hengchi U. USD),. not. to mention. the top. importer. –. USA with imports eight times the size of the Italian ones (40 mil. USD), even as France and Italy surpass all the top four importers to Taiwan on the global scale (France being world’s 2nd top exporter with 12.7 % and Italy world’s 4th top exporter with 9 % share in the world’s exports). Interestingly, this category is one of the few studied in this work, in which the preferential tariffs seem to have a significant effect on the import values, as New Zealand with 0 % import tariff occupies the second place among the top importers. All the other top importers are subject to 5 % tariff.. 7.

(15) Table 5 - Main trade indicators - Product: 6403 Footwear, upper of leather. Top ten sources of import to Taiwan. China Vietnam Italy Indonesia Germany Spain Thailand Bangladesh India Japan. Imported value 2014 (USD thousand). Share in Taiwan's imports (%). Ranking in world exports. Share in world exports (%). Tariff (estimated) applied by Taiwan (%). 90340 61733 52117 18700 8939 5202 4133 3194 2610 2427. 33.7 23 19.4 7 3.3 1.9 1.5 1.2 1 0.9. 1 3 2 7 5 6 25 20 10 59. 21 9.7 15.2 3.7 4.5 3.8 0.6 0.8 3.5 0. 6.3 6.3 6.3 6.3 6.3 6.3 6.3 1.9 6.3 6.3. Tariff groups 1. 立. 11 262074. 0. 127. 0.1. 0. 2.2. 257. 2. 1.9. 0 97.4. 36 3344. 0.2 95.6. 2.5 6.3. 0.1. 187. 學. ‧ 國. 5543. 政治大. ‧. io. n. al. 0.6. 6.5 10. er. 302. sit. y. Nat. New Zealand, El Salvador, Panama Bangladesh, Cambodia, Myanmar, Ethiopia, Haiti Singapore China, Vietnam, Italy, Indonesia, Germany + 64 countries Palau Serbia, Bosnia and Herzegovina, Micronesia. i n U. Ch. v. Source: List of supplying markets for a product imported by Taipei, Chinese (ITC Trade Map). i n gEU In this category, there are only e three among the top importers c hcountries to Taiwan:. Italy, Germany and Spain. Similarly to the “4202 Trunks,. suitcases, camera cases, handbags” category, also this category is dominated by imports from China. All the top importers of this category are subjected to 6.3 % average tariff, except of Bangladesh, which has a lower tariff of 1.9 %. This introductory chapter provides basic information about the Taiwanese import policies for the product categories in question, and it will further be elaborated upon in the conclusion. 8.

(16) 2 Literature review and Research Design 2.1 Scope of the research Consumer preference research has been conducted at different levels – Lin, Marshall & Dawson (2009) look at Taiwanese consumer attitude towards private brand food products, Wang & Heitmeyer (2006) research consumer attitudes toward US made apparel, while several other studies, like Elliott & Cameron (1994),. explore. the perception. of multiple. goods. categories (namely tires, dishwashers, jam, cars, shoes and computers). Roth & Romeo (1992) explain, that the consumer perception of different importers. 政治大. varies significantly among different categories of goods, therefore it seems. 立. that the most balanced approach to conduct a research on consumer. ‧ 國. 學. preferences is to create a representative sample of imported products, which includes several different categories.. ‧. The goal of this research is to discover where the areas for possible trade. sit. y. Nat. expansion might lay by examining the reasons why some products of EU are imported less, therefore attention is paid especially to the types of products,. er. io. which are sold in relatively smaller quantity on Taiwanese market. a. n. v i l n In order to find suitable Cproducts a representative sample, h e n g ctohcreate i U the Taiwanese trading statistics are compared to those of other markets. Due to geographical and cultural proximity and relative similarity of the market size (relatively to other much larger entities present in the region, like China), Korea and Japan seem to be good counterparts for such comparison. The proximity of the three countries in the point of view of economic and business analysis, is also demonstrated by the fact, that they are often discussed together and compared to each other in a number of studies, related to. economic. and. international. trade. Feenstra, Yang & Hamilton (1999), or Lee (1986). 9. research,. like.

(17) On the basis of the statistical comparison, four product categories, which rank the furthest below the average in terms of their share in the total import value, are chosen (see Appendix 3, for list of subcategories see Appendix 4). Table 6 - Shares of different product categories in total import value. HS code. Product group name. Share in total imports value. '8703 '4202. Cars (incl. station wagon) Trunks, suit-cases, camera cases, handbags etc., of leather, plastic, textile etc. Cheese and curd Footwear of leather. 8.75% 0.98%. '0406 '6403. 0.10% 0.30%. Source: Bilateral trade between EU and Taipei, Chinese (..., Japan, Korea) (ITC Trade Map). 2.2 Research design 2.2.1 Conceptual model 立. 政治大. The conceptual model in this study is inspired by the one devised by Lin,. ‧ 國. 學. Marshall, & Dawson (2009) in their research of consumer attitudes toward European retailer’s private brand (PB) food products, however, due. ‧. to the different focus of this research, several adjustments are made:. sit. y. Nat. • In this work, it is hypothesised, that low product familiarity can have. er. therefore a. with product. the brand familiarity is examined together iv l C n familiarity a hypothesis connecting the product h eand ngchi U. n. below),. io. a significant effect on the purchase decision process (as discussed. familiarity directly to purchase intention is added.. • The COO image replaces the “attitude towards PB” (in this work, the COO image is the core of the model, instead of the PB versus national brand contrast in the original work) • Risk perception and attitude towards product are combined in the COO image variable.. 10.

(18) Figure 5 - Conceptual Model. Product/brand familiarity Country of origin image Quality perception. 立. Value perception. 治政 Purchase intention 大. ‧ 國. 學. 2.2.2 Product and brand familiarity. Roth & Romeo (1992) examine the perception of products originating. ‧. in various countries among US, Irish and Mexican respondents and find that the product familiarity may vary significantly across different product. y. Nat. sit. categories (even as the authors do not find it significantly affecting. er. io. the product image perception or the purchase intention within the scope. a of their particular study). n. iv l C n h e nand h i UJain, & Richardson (1995:19) Lin, Marshall, & Dawson (2009:3) g cDick,. look at private brand perception and point out, that low level of brand familiarity can eliminate specific products from the choice options of the consumers and in environments as geographically and culturally distant as Taiwan and the EU are, it might be seen as justified, that the product familiarity can have the same effect and can therefore also directly affect the purchase intention (especially a low level of product and brand familiarity can have a negative effect on the product purchase intention). H1a: Product familiarity may directly affect the purchase intention.. 11.

(19) H1b: Brand familiarity may directly affect the purchase intention. The influence of product familiarity on COO is an important issue in the context of COO research and it is widely discussed in literature: Nagashima (1970:9) looks at US and Japanese consumers’ attitude towards imported products and notes that: "the made-in image" is naturally affected by the familiarity and availability of the country's product, and the stereotype of that country. Some representative products of the country influence the total product image. Such products as Coca Cola, Chevrolet, Ford, IBM, and Sunkist, are forming the Japanese image of "Made in U.S.A.". 政治大 COO effect and. Josiassen, Lukas and Whitwell (2008) discuss the interaction between the product familiarity of product familiarity. and. 立 on purchase-related. the indirect. influence. behavioural intentions through. ‧ 國. 學. the COO image and observe, that there is a significant correlation.. ‧. Josiassen, Lukas and Whitwell (2008) examine the relations between product involvement, product familiarity, COO and quality perception and notice that. y. Nat. sit. “[…] the results also show that product familiarity has a significant and. a lproduct involvement. i v Product involvement n C U negativeh moderating influence on the effect of COO engchi. n. intentions,. er. io. negative moderating influence on the effect of COO image on behavioural as does. has a significant and. image on quality perception, but product familiarity does not.”(Ibid., p. 10). According to Lee and Ganesh (1999:4) several studies, like Johansson and Nebenzahl (1986) and Johansson, Douglas and Nonaka (1985) show “positive interaction between product familiarity and COO effect, with the latter becoming stronger with an increase in the former”. The importance of this relationship also seems to be important in a study of consumer attitudes towards imported food products by Krystallis and Chryssochoidis (2006), where it is concluded, that “the degree of familiarity. 12.

(20) with the product shapes the appearance of the COO effect in the food purchasing context”. The literature listed above seems to constitute strong grounds for the presumption, that the product and brand familiarity have significant effects on the COO image. H2a: Product familiarity has effect on Country-of-origin image. H2b: Brand familiarity has effect on Country-of-origin image.. 2.2.3 Country of origin image. 政治大 of comparison of producers 立. In many existing studies, the COO (Country-of-origin) image is examined in the context. of different. origin,. e.g. Nagashima (1970) compares consumer attitude towards goods imported. ‧ 國. 學. from France, GB, Germany, USA and Japan, while Wang & Heitmeyer (2006). ‧. look at the perception of domestic versus US-made apparel on Taiwanese market. The focus of this study is, however, only on Taiwanese imports from. Nat. sit. y. EU and the attitude of Taiwanese consumers towards products imported. al. n. of this research.. er. io. thereof, thus the comparison with other importers is out of the scope. i n U. C. v. h e n g cconcerning There is an extensive discussion the effect COO image hi has on the perception of product quality by the consumers – for example, Agrawal & Kamakura (1999) compare consumers’ quality perception and objective quality evaluation (drawn from quality rankings in Consumer Reports tests) and conclude, that although the COO image might be one of the several factors considered by the consumers when making an actual purchase decision, it is a valid cue in assessing relative product quality. Papadopoulos et al. (1987; as quoted in Al-Sulaiti and Baker 1998:8) express certainty about the existence of the country of origin effect and the influence of “made in” images on the product evaluation by both consumers and 13.

(21) industrial buyers, while they acknowledge the variety of the COO stereotypes across different product categories. Verlegh & Steenkamp (1999:17) observe, that the “consumers have been found to infer judgments of product quality from product-country images, which contain beliefs about a country’s products, but also about more general characteristics, like its economy, workforce and culture” and furthermore, that “Apparently, there exists a particularly strong link between country of origin and perceived quality” (Ibid. , p.18). H3: Positive Country-of-origin image will increase the product quality perception.. 立. 政治大. Agrawal & Kamakura (1999) discuss the possibility that COO image. ‧ 國. 學. perception is connected with value perception in the sense that producers from countries with positive image perception will be able to charge premium. ‧. prices on their products, whereas producers originating in countries. sit. y. Nat. with negative image perception will have to offer discount prices in order. the pricing. al. n. that. io. conclude. was more. related. er. to maintain the competitiveness of their products and (although the authors to the objective quality. iv. of the products rather thanC the COO image innthe scope of their specific. hengchi U. research). This question seems to also be a point of interest in other studies, like Nebenzahl & Eugene (1997:2), where a price discount is considered one of the possible defence strategies against negative impacts of low country of origin image. In this work, the perception of value will be researched in the context of a scale defined by discounted and premium pricing as opposite extremes and its relation to the COO image will also be assessed. H4: Better COO image will have positive impact on value perception.. 14.

(22) 2.2.4 Quality perception As Lin, Marshall, & Dawson (2009:5) observed, the quality perception is one of the most important factors affecting the process leading to purchase decision, where the consumers are likely to have higher purchase intention towards product or the entire brand they perceive as being of high quality, whereas an unfavourable quality perception might lead to a decline in the purchase intention. H5: High quality perception will strengthen the purchase intention.. 2.2.5 Value perception. 政治大. Lichtenstein, Ridgway, & Netemeyer (1993:2) characterize the perception. 立. of the price as “concern with the ratio of quality received to price paid. ‧ 國. 學. in a purchase transaction”, they discuss two sides of the product value perception – negative role of the price (including the value and price. ‧. consciousness) and also the positive role of the price, which encompasses. y. Nat. the price-quality schema (implying that higher price will be seen as indicator. io. sit. of higher product quality) and the prestige sensitivity (suggesting that. n. a. er. excessive spending can contribute to the personal image and social status perception of the consumer). l. Ch. hi. en. i n U. v. gc Jacoby, Olson, & Haddo (1971:9) marginalise the effect of price on quality perception, stating that “price does not operate as an indicant of product quality. when. embedded. in a multicue. setting”.. Prestige. generated. by the product purchase is already mentioned as one part of the COO image research, therefore the main focus will be put on the value and price consciousness and its effect on purchase intention, since price perception is viewed as having direct effect on purchase intention (Lichtenstein, Ridgway, and Netemeyer 1993:2). H6: Positive value perception will increase the purchase intention. 15.

(23) 3 Methodology 3.1 Questionnaire design The core part of this work is research of consumers’ attitudes based on the distribution of self-administered questionnaires. The questionnaire is designed as a structured questionnaire with 11 closed-ended questions, repeated for each of the four product categories concerned. For the sake of text conciseness and legibility, the four categories will be assigned shorter captions in the questionnaire and simplified one-word captions will be used in the text henceforth (as listed in Table 7). Table 7 - Category captions. 立. Cars Suitcases, handbags, briefcases backpacks, camera/computer bags Cheese Leather shoes. One-word caption Cars Bags. 學 Cheese Shoes. ‧. '0406 '6403. Product group name Cars (incl. station wagon) Trunks, suit-cases, camera cases, handbags etc., of leather, plastic, textile etc. Cheese and curd Footwear, upper of leather. ‧ 國. HS code '8703 '4202. 治政 Questionnaire 大captions. 3.1.1 Product and brand familiarity. y. Nat. sit. The product familiarity is researched in the context of previous purchase. er. io. experience – the respondents are asked to indicate the frequency (or absence). n. a on a four-point scale. i v of their purchase experience. l C n U h i e n g cahproduct Q1 – Have you in the past purchased of the following categories? (Never - Once - Seldom - Several times / Regularly). In order to establish the respondents familiarity with the EU brands, five European brands, producing goods of each of the specific categories, will be offered and the respondents will be asked to choose the brands they are familiar with. This question is analogic to the way Schaefer (1997:15) researches the consumer familiarity with different beer brands among English consumers; it will be formulated as a five-element index and a numerical score will be calculated for each category, assigning one point for each option chosen or correct alternative option inputted by the respondent. 16.

(24) Although this approach to the brand familiarity measurement is precedented in the literature, there seems to be little explanation offered on the choice of the specific brands and no statistical data on the producer-company level seem to be publicly accessible, therefore a pre-test was run, in which the respondents were offered 9 brands for the Cars and Bags categories and 14 for the Shoes category in an open-ended questionnaire. This pre-test gathered 14 responses (detailed results in Appendix 6) and five brands for each category were chosen based on three criteria (in this order): 1. The brands sample should represent as many countries as possible, therefore, only one brand of each country was chosen (where possible).. 政治大. 2. Priority was given to brands from the middle parts of the price. 立. spectrum, so as not to distort the measurement of the value perception. ‧ 國. 學. variable and make the survey relatable to respondents with broad variety of backgrounds (therefore luxurious brands like the British. ‧. Jaguar or German Mercedes-Benz were deliberately left out).. sit. y. Nat. 3. The brands with higher number of votes were chosen over brands with lower number (only brands with at least 1 vote were selected).. er. io. n. Q2 – Please choose which of the following EU brands you have heard a of (not necessarilyl have purchased before)i v. Ch. n U engchi Sweden. Table 8 - Brands chosen for the brand familiarity testing. Cars. Suitcases, handbags, briefcases backpacks, camera/computer bags. Leather shoes. Volvo Volkswagen Land Rover Škoda Citroën Kappa Rimowa Hama Delsey Hedgren Tod's Ecco Camper GEOX Baťa. Germany United Kingdom Czech Republic France Italy Germany Germany France Belgium Italy Denmark Spain Italy Czech Republic. 17.

(25) Cheese is a commodity, which is better identifiable by its type than the specific manufacturer brand (cheese and similar products are also often produced by small-size local producers) and in other research works, focusing on cheese and related products, like Krystallis & Chryssochoidis, (2006) the particular producer brands remain out of the scope of the research. For these reasons the Cheese product category is left out from this question.. 3.1.2 Country of origin image The country of origin image is a complex, multi-faceted concept, examined differently in various research works – Roth & Romeo (1992:6) compare. 治政 and create the following overview: 大立. the dimensions of country of origin image as discussed in different studies. n Narayana (1981). Cattin, Jolibert & Lohnes (1982) Jaffe & Nebenzahl (1984) Johansson & Nebenzahl (1986) Han & Terpstra (1988). Ch. engchi. Technical Advancements, Prestige, Workmanship, Economy, Serviceability. Source: Roth & Romeo (1992:6). 18. y. sit. io. al. Workmanship, Innovation, Prestige. er. Nat. White (1979). Price & Value, Service & Engineering Advertising & Reputation, Design & Style, Consumers' Profile Expensive, Price, Technicality, Quality, Workmanship, Inventiveness, Innovation, Selection, Serviceability, Advertising, Durability, Reliability, Brand recognition Quality, Recognition, Prestige, Production, Form, Expensiveness, Popularity, Functionality Pricing, Reliability, Workmanship, Technicality, Performance Product-technology, Marketing, Price Economy, Status. Production & Marketing Image Dimensions Innovation, Prestige, Design. ‧. Nagashima (1977), (1970). Country Image Dimensions. 學. Study. ‧ 國. Table 9 - COO image research overview. i n U. v. Workmanship, Prestige, Innovation, Design. Workmanship, Innovation. Innovation, Prestige Prestige Innovation, Prestige, Workmanship.

(26) In this overview,. the most. frequently. quoted. country-of-origin image. dimensions are “Prestige”, “Innovation”, “Workmanship”, and “Design”, therefore these four elements will be taken as key aspects of the COO image and the respondents will be asked to choose, which of these aspects they view as the strongest advantage of the different EU-imported product categories. Analogically to the previous question, a numeric score will be calculated, assigning one score point for each option chosen. Q3 – Which of the following qualities do you perceive as the greatest advantage of products originating from EU? (Prestige, Innovation, Workmanship, and Design). 政治大 this 立category will have different. Most of these criteria, except of prestige, cannot be applied to the Cheese category, therefore. set of properties. ‧ 國. 學. for the respondents to choose from in a manner similar to the three other categories. Krystallis & Chryssochoidis (2006) use 32 questions to research. ‧. the consumers’ evaluations of imported and domestic cheese, which can roughly be grouped to three categories by their topic: “Taste \ Consumption. y. Nat. captions. seem. to represent. er. general. io. constructed. sit. pleasure”, “Healthiness”, “High-quality ingredients” (see Appendix 5) – thus well. the universal. n. characteristics of food aproducts and therefore they i v will in turn be used l. n U i e nofgthe as choice options for the evaluation category. c hCheese. Ch. Q3b – Which of the following qualities do you perceive as the greatest advantages of products originating in EU? (Taste/Consumption pleasure, Healthiness, High-quality ingredients). 3.1.3 Quality perception The quality perception variable will be researched directly by means of a semantic differential with seven values ranging from “very low” to “very high”. Q4 – How do you perceive the overall quality of EU-produced goods?. 19.

(27) 3.1.4 Value perception As mentioned before, the value variable will be examined in the context of a seven-point. semantic. differential. of the consumers’. perception. of the product pricing, ranging from “Very cheap” to “Highly luxurious”. Q5 – Do you perceive the products of the following categories as cheap or luxurious?. 3.1.5 Purchase intention The potential purchase intention will be examined by means of two six-point semantic differentials – the first one will be enquiring about the presence of potential purchase intention in general, while the second will be asking. 治政 Q6 – In the future, will you likely purchase 大 a product of one of the following立 categories?. specifically about the intention to purchase products originating in EU.. ‧ 國. 學. Q7 – In the future, if you decide to purchase a product of the following categories, will you likely purchase product imported from the EU?. ‧. 3.1.6 Personal data. In order to balance and achieve equal representativeness of the samples, basic. y. Nat. sit. personal information about the respondents will be collected. This section will a lbe. n. The respondents. of the specific groups:. asked. Ch. i n U. to indicate. engchi. er. io. is added to the end of the questionnaire, as suggested by Babbie (2013:266). their v. belonging. to one. Q8 – Sex: Male, Female Q9 – Age: ≦ 25, 26 - 35, 36 - 45, 46 - 55, 56 - 65, > 66 Q10 – Income group (1000 NTD/month): ≦ 20, 21 - 40, 41 - 60, > 61 Q11 – Education: Elementary, High School, College, Graduate. These groups are constructed in a manner similar to other studies of Taiwanese consumers’ preferences – Lin, Marshall, & Dawson (2009, p. 7) – with one more group added to the upper end of both the age and income scale in order to achieve better precision. Furthermore, education is added to ensure the variedness of the sample – the inclusion of education is common in similar research works (e.g. Josiassen, Lukas, and Whitwell 2008). 20.

(28) 3.2 Data Collection The questionnaire was first designed in English (see Appendix 7) and then translated into Chinese (Appendix 8) in order to ensure its legibility by all the respondents. The questionnaires were distributed on self-administered basis in two phases: 3.2.1.1 Phase 1: Electronic questionnaires During the first phase, the questionnaires were distributed in electronic form, based on Google forms and propagated via virtual social networks such as Facebook. One hundred responses were obtained in this way. The structure. 政治大. of thus obtained sample (see Table 10) was not completely satisfactory,. 立. especially with regard to the balance of sex and education factors, therefore. ‧ 國. 學. the second phase of questionnaire distribution was initiated. Table 10 - Phase 1 sample structure. Age 56 - 65; 5% > 65; 0% 46 - 55; 17%. ‧. Sex. y. Nat. n. 36 - 45; 11%. Female. Ch. 0%. Income 14% 35%. 17%. 0%. iv. 36 - 45 46 - 55 56 - 65 > 65. Education. 14% 6%. ≦ 20 000. Elementary Junior High. 20 001 - 40 000. Senior High. 40 001 - 60 000 34%. Un. 26 - 35; 23%. engchi. ≦ 25; 44%. er. io. Male. al. 26 - 35. sit. 35%. 65%. ≦ 25. 80%. Undergraduate Graduate School. > 60 001. 21.

(29) 3.2.1.2 Phase 2: Printed questionnaires In the second phase, thirty additional questionnaires were printed with the aim to gather. more. responses. especially. from. male. respondents. and. the respondents of older age groups. Twenty-four questionnaires were collected in this way, of which two were incomplete and were discarded. The structure of thus obtained additional sample met the expectations in regards to sex ratio, as 95 % respondents were male, and the age representation structure also encompassed categories underrepresented in the Phase 1 sample, especially the 56-65 year-old and 46-55 year-old. 政治大. groups were well represented (see Table 11).. 立. Table 11 - Phase 2 sample structure. > 65; 5% Male 56 - 65; 37%. Female. n. 11%. 58%. 32%. 26 - 35 36 - 45; 5%. 46 - 55; 32%. 46 - 55 56 - 65. sit 5% 0%. 36 - 45. > 65. er. io. Incomea. 26 - 35; 21%. ≦ 25. y. Nat. 95%. 0%. ≦ 25; 0%. ‧. ‧ 國. Age. 學. Sex. 5%. vEducation i l C n U 11% ≦ 20h 000e Elementary ngchi 26%. 20 001 - 40 000. Junior High. Senior High 40 001 - 60 000. 58%. Undergraduate Graduate School. > 60 001. 22.

(30) The final sample thus contained 122 valid responses. As graphs in Table 12 show, the sex and income ratios are quite well balanced. The distribution of the majority. of the questionnaires. in electronic. form. facilitated. the collection and processing of the data greatly and allowed for a large number of responses to be gathered in relatively short time, on the other hand it might have contributed to the slight overrepresentation of younger age groups and higher-educated groups in the sample. Nevertheless, the usage of electronic forms seem to be a viable option for questionnaire distribution as many respondents can be addressed very easily and the computer-based survey preserves the anonymity of respondents better and might therefore be. 政治大. less intimidating, encouraging the respondents to be more open about their. 立. answers.. Sex. 56 - 65; > 65; 10,1% 0,8%. Income. 21,0%. 29,4%. Ch. y. 36 - 45; 10,1%. n. al. sit. io. Female. v ni. engchi U 0,8% 0,0% 13,4%. ≦ 20 000. er. Nat. 54,6%. ≦ 25; 37,0%. 46 - 55; 19,3%. Male. 44,5%. Age. ‧. ‧ 國. 學. Table 12 - The final sample. 26 - 35; 22,7%. ≦ 25. 26 - 35 36 - 45 46 - 55 56 - 65 > 65. Education 9,2% Elementary Junior High. 20 001 - 40 000. Senior High. 19,3% 30,3%. 40 001 - 60 000. 76,5%. Undergraduate Graduate School. > 60 001. 23.

(31) 4 Research findings 4.1 The brand familiarity variable Table 13 - Brand familiarity scores Category Original Cars. Inputted. Bags. Original. Inputted. Citroën BMW Mercedes Benz Opel Kappa Rimowa Delsey Hedgren Hama Fjällräven Chanel Targus Tod's Ecco Camper GEOX Baťa. 立. Inputted. Responses 107 79 55 51. Percentage 87.70% 64.75% 45.08% 41.80%. 33 6 5 1 78 41 15 15 10 2 1 1 57 41 37 22 5. 27.05% 4.92% 4.10% 0.82% 63.93% 33.61% 12.30% 12.30% 8.20% 0,01639 0.82% Incorrect 46.72% 33.61% 30.33% 18.03% 4.10%. 1. Incorrect. 政治大. Bally. Nat. y. ‧. ‧ 國. Original. Country Sweden Germany UK Czech Republic France Germany Germany Germany Italy Germany France Belgium Germany Sweden France USA Italy Denmark Spain Italy Czech Republic Switzerland. 學. Shoes. Brand Volvo Volkswagen Land Rover Škoda. sit. The brand familiarity scores and percentages (Table 13), together. er. io. with the brand familiarity means (Table 16) demonstrate quite high levels. n. a most of the respondents.i vIn accord with informal of brand familiarity by. l C ncategories are led by Italian U he expectations, the scores for the Bags and Shoes ngchi brands, whereas, surprisingly, the Cars category score is highest for Swedish Volvo, although the German brands are also reaching high scoring values and were among those most frequently inputted by the respondents. The ability of the respondents to input brands from several different EU countries on their own supports the presumption that knowledge of the concept of EU and its structure is well established among the Taiwanese public, as only two answers, incorrectly including brands originating in non-EU countries were recorded.. 24.

(32) 4.2 Country of origin image variable Table 14 shows the structure of the evaluation of the different aspects of the COO image variable. In the Cars, Bags and Shoes categories, the highest scoring aspects are those associated with the product quality – “Workmanship” and “Design”; slightly lower, but still quite high, importance is given to the “Prestige” aspect, while the “Innovation” aspect is ranked on a significantly lower level. Such evaluation is in accord with an image of reliable, high-quality, traditional producer brands. Similar observation of high. quality. perception. can. be. made. in the Cheese. category,. 治政 by “Healthiness”, while the “Prestige” aspect大 does not seem to be highly 立 relevant in this category. 學. ‧ 國. with the “Ingredients” and “Taste” aspects ranking the highest, followed. Table 14 - Structure of the COO image variable responses. al. n. Cheese. Mean N. Prestige .246 30. Ch. Healthiness .369 45. engchi. 25. Workmanship .713 87 .680 83 .746 91. Design .566 69 .607 74 .516 63. vIngredients. Taste .541 66. y. sit. er. io. Shoes. Nat. Bags. Mean N Mean N Mean N. Innovation .139 17 .156 19 .098 12. ‧. Cars. Prestige .541 66 .426 52 .369 45. i n U. .533 65.

(33) 4.3 Country-wise observations The design. of the brand-familiarity. question,. where. each. country. is represented by one brand at most (except for two German brands presented in the Bags category and two Italian brands in the Shoes category), allows for certain indirect observations about the specific countries to be made. Table 15 presents a summary of the average values the respondents, who chose the brands of the respective countries, associated with the different variables (detailed information about the frequencies of the specific country choice combinations can be found in Appendix 13). Table 15 - Country-specific data. 立. 治政 Quality 大. Cars Sweden Germany Czech Republic United Kingdom France. 1.762 107 Bags Italy 1.492 80 Germany 0.959 51 France 0.934 55 Belgium 0.590 33. ‧ 國. Mean 3.975 3.328 2.115 1.959 1.238. N Mean N 107 Bags Italy 2.754 78 80 Germany 1.803 50 55 France 0.607 16 51 Belgium 0.525 15 33. io. Price. al. n. 3.811 107 Bags Italy 3.164 80 Germany 2.025 55 France 1.869 51 Belgium 1.246 33. Ch. 2.582 1.730 0.525 0.500. 78 50 16 15. engchi U. y. Shoes Italy Denmark Spain Czech Republic. sit. 78 50 16 15. er. 1.213 0.852 0.295 0.254. ‧. COO. Nat. Cars Sweden Germany United Kingdom Czech Republic France. Shoes Italy Denmark Spain Czech Republic. 學. Cars Sweden Germany United Kingdom Czech Republic France. Shoes Italy Denmark Spain Czech Republic. v ni. Mean 2.721 1.385 1.205 0.180. N 61 41 37 5. 1.262 0.689 0.557 0.074. 61 41 37 5. 2.664 1.393 1.148 0.180. 61 41 37 5. The mean values seem to indicate certain level of country-specific bias quite clearly: in the Cars category, the mean values for all three variables are highest for Sweden, followed quite closely by Germany (an interesting observation in this category is the United Kingdom having lower COO mean than Czech Republic – this phenomenon may be related to the omission of famous British luxurious brands, like Jaguar); the Bags category is led by Italy, which is loosely followed by Germany; in the Shoes category, Italy exceeds all the other countries by a large margin. 26.

(34) 4.4 Analysis The analysis was performed using the IBM SPSS Statistics software (v. 19). It includes univariate statistics (Table 16) bivariate correlations (Table 18, 18) and regression analysis. Except of two cases (which were discarded), there were no missing data, so the remaining 122 cases were included in all analyses. Table 16 - Univariate descriptive analysis. Minimum 0 0 0 0 0 Brand Familiarity 0 0 0 COO 0 0 0 2 Quality perception 2 1 1 2 Value perception 2 1 1 0 Purchase Intention 0 0 0 0 EU products 1 purchase 0 Intention 0 0 Sex 0 Age 0 Education 1 Income Cars .000 Total Purchase Bags 3.000 Intention Cheese 1.000 Shoes .000 Valid N (all variables, all categories). 政治大. n. engchi. All figures are rounded to at most 3 decimal digits. 27. y. sit. io. Ch. ‧. Nat. al. Mean .30 1.07 1.21 .89 2.75 1.33 1.33 1.96 1.87 1.73 1.69 4.47 4.23 4.14 4.09 4.30 4.06 3.57 3.98 2.80 3.30 3.34 3.05 3.09 3.33 3.39 3.20 .54 1.45 1.34 3.04 8.975 9.959 10.123 9.459 122. er. 立. Maximum 3 3 3 3 7 4 5 4 4 4 4 6 6 6 6 6 6 6 6 5 5 5 5 5 5 5 5 1 5 3 4 15.000 15.000 15.000 15.000. 學. Category Cars Bags Cheese Shoes Cars Bags Shoes Cars Bags Cheese Shoes Cars Bags Cheese Shoes Cars Bags Cheese Shoes Cars Bags Cheese Shoes Cars Bags Cheese Shoes. ‧ 國. Purchase History. i n U. v. Std. Deviation .703 1.026 1.038 1.059 1.688 .940 1.339 .931 .823 .813 .834 .911 .870 .930 .891 .859 .775 .792 .828 1.304 1.127 1.169 1.120 1.253 .974 .975 1.020 .500 1.432 1.103 .521 3.548 2.759 2.876 2.901.

(35) The univariate analysis results (presented in Table 16), and especially the mean values, reveal interesting findings about the purchase behaviour and attitudes of Taiwanese consumers – the purchase history means are noticeably low (relatively to the medium value) and the strength of this phenomenon seems to be inversely related to the price and lifespan of the products of the respective categories. Expectedly, the brand familiarity means are relatively higher for the car brands and lower for bags and shoes. The COO means are those which approach the medium value the closest of all the variables concerned and its standard deviations are among the lowest, which might suggest, that if more categories were included in the set. 政治大. of possible answers for the question measuring the COO image, it could. 立. possibly have measured this variable more accurately. The means for quality. ‧ 國. 學. perception are quite high for all categories, which suggests that the EU products are perceived by the Taiwanese consumers as being of high quality,. ‧. while the relatively low standard deviation seems to demonstrate, that. sit. y. Nat. the levels of such quality perception are quite stable across the sample.. io. er. The means of price perception are high, especially for cars and bags, which is expectable, given that EU is home for many well-known luxurious brands. n. a. iv. l C for cars, Chanel ornLouis Vuitton for bags and (e.g. Mercedes Benz or BMW. hengchi U. shoes); interesting fact is, that the value perception mean is considerably lower for the Cheese category, which seems to reflect the informal everyday consumer observation that cheese and similar products are all rather overpriced in Taiwan, regardless of origin or quality and there is little competition from domestic production of this category. The purchase intention and the purchase intention of EU products are both close to the medium value.. 28.

(36) The analysis of these two variables revealed strong correlation between them (see Table 18), and since both of these variables are related to purchase intention, the decision was taken to merge them in an aggregate variable. This new variable is labelled as “Total purchase intention” and calculated as a sum of the “Purchase intention” (Q6) variable and the “EU product purchase intention” (Q7), where the “EU product purchase intention” is assigned double value, for it is specifically measuring the attitude towards the purchase of products originating in EU, which is the main object of study in this research. The calculation can be expressed by the following formula:. 政治大. 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝ℎ𝑎𝑎𝑎𝑎𝑎𝑎 𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 = 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃ℎ𝑎𝑎𝑎𝑎𝑎𝑎 𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖 + 2 ∗ 𝐸𝐸𝐸𝐸 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝ℎ𝑎𝑎𝑎𝑎𝑎𝑎 𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖𝑖. 立. This decision is supported by the results of Cronbach’s Alpha tests (Table 17). ‧ 國. 學. which were well above the required significance level 0.7 (as recommended in European Social Survey 2013) for all categories. Thus calculated variable. ‧. will be used as dependent variable in all the relevant analyses.. n. al. Ch. engchi. **. Correlation is significant at the 0.01 level (2-tailed). 29. y. sit. Sig. (2-tailed) .000 .000 .000 .000. er. Pearson Correlation .704** .578** .681** .657**. io. Cars Bags Cheese Shoes. Nat. Table 17 - Relation between "purchase intention" and "purchase intention of EU products" variables. i n U. v. Cronbach's Alpha .826 .728 .802 .791.

(37) 4.4.1 Bivariate relationships analysis Table 18 - Bivariate correlations with the dependent variable. Sex Purchase History Brand Familiarity COO Quality perception Value perception Age Education. 立. ‧ 國. Income. Cars -1.044 0.299 .053 .564 .162 .075 .170 .061 .451** .000 .295** .001 -.250** .006 -.167 .066 -.196* .030. Bags 0.112 0.911 .398** .000 .279** .002 .296** .001 .448** .000 .268** .003 -.317** .000 -.096 .293 -.120 .189. 政治大. ** Correlation is significant at the 0.01 level (2-tailed) * Correlation is significant at the 0.05 level (2-tailed) Dependent variable: Total purchase intention. Cheese -0.07 0.944 .417** .000 .046 .614 .503** .000 .103 .259 .006 .943 -.029 .753 -.136 .136. Shoes -1.274 0.205 .234** .009 .301** .001 .138 .129 .281** .002 .022 .810 -.108 .237 .055 .550 -.073 .426. 學. Method T-test: t Sig. (2-tailed) Pearson Sig. (2-tailed) Pearson Sig. (2-tailed) Pearson Sig. (2-tailed) Pearson Sig. (2-tailed) Pearson Sig. (2-tailed) Pearson Sig. (2-tailed) Pearson Sig. (2-tailed) Pearson Sig. (2-tailed). ‧. Table 18 shows the bivariate correlations between the specific variables and purchase. intention).. There. seems. sit. (Total. y. variable. Nat. the dependent. er. io. to be a moderately strong correlation for the brand familiarity and purchase history in all the relevant a categories except for cars – this might possibly. n. v i lpurchase n C h history andU purchase engchi. be related to the low. intention means. for this category. The COO variable does not seem to be strongly significant for any category except bags. The quality variable seems to be significantly connected to purchase intention, as its correlation is exceptionally strong in all four categories and the quality variable correlation coefficients are some of the highest. of the entire. model.. The value. perception. variable. has significant positive correlation only in the cars and bags categories. Among the personal variables, the only ones, which seem to be correlated with the purchase intention are the age in the bags and cars categories and income in the cars category, the correlation is negative in these cases. 30.

(38) Table 19 - Correlations for hypothesised bivariate relationships Pearson Correlation Sig. (2-tailed) Pearson H1b Brand familiarity → Purchase intention Correlation Sig. (2-tailed) Pearson H2a Correlation Purchase history → COO Sig. (2-tailed) Pearson H2b Correlation Brand familiarity → COO Sig. (2-tailed) Pearson H3 Correlation COO → Quality perception Sig. (2-tailed) Pearson H4 Correlation COO → Value perception Sig. (2-tailed) Pearson H5 Correlation Quality perception → Purchase Sig. (2-tailed) intention. H1a Purchase history → Purchase intention. 立. Bags. Cheese Shoes. 0.053. .398**. .417**. .234**. 0.564 0.162. 0.000 .279**. 0.000. 0.009 .301**. 0.075 0.032. 0.002 0.128. 0.728 .236**. 0.161 .216*. 0.009 0.140. 0.017 .343**. 0.007. 0.001 0.071. 0.125 0.129. 0.000 -0.001. 0.943 0.012. 0.434 0.001. 0.157 .451**. 0.991 .448**. 0.896 .503**. 0.993 .281**. 0.000. 0.000. 0.000. 0.002. .295**. .268**. 0.103. 0.022. 0.001. 0.003. 0.259. 0.810. 政治大 Sig. (2-tailed). 0.118 0.197. 0.001 -0.047 0.605 .307**. ‧. ** Correlation is significant at the 0.01 level (2-tailed) * Correlation is significant at the 0.05 level (2-tailed). 學. Pearson H6 Value perception → Purchase intention Correlation. ‧ 國. Cars. sit. y. Nat. Analysis of bivariate correlations relevant to the hypothesised variable. er. io. relationships (Table 19) seems to support H1a and H1b for all categories except cars (cheese isaexcluded from H1b); H2a is not supported for any. n. v i l n C hstrongly for carsU and shoes and moderately category; H2b is supported engchi (on 0.05 level) for bags (cheese is excluded from this hypothesis); H3. is supported only for bags; H4 is not supported at all; H5 is supported for all categories and H6 only for cars and bags (as observed above). The relevance of the relationships revealed by the correlation analysis will be further tested by the means of regression analysis.. 31.

(39) 4.4.2 Regression analysis 4.4.2.1 Cars Table 20 - Regression analysis results - Cars Coefficientsa Unstandardized Standardized t Sig. Model Coefficients Coefficients B Std. Error Beta (Constant) 14.503 1.953 7.427 .000 1 Sex -.475 .656 -.067 -.724 .470 Age -.885 .293 -.357 -3.019 .003 Education -.027 .382 -.008 -.070 .945 Income -1.744 .605 -.256 -2.881 .005 Purchase History .363 .472 .072 .768 .444 Brand Familiarity .451 .190 .215 2.371 .019 (Constant) 13.836 2.135 6.481 .000 2 Sex -.400 .664 -.056 -.603 .548 Age -.835 .300 -.337 -2.780 .006 Education -.070 .386 -.022 -.181 .857 Income -1.689 .610 -.248 -2.766 .007 Purchase History .373 .473 .074 .789 .432 Brand Familiarity .421 .194 .200 2.163 .033 COO .268 .343 .070 .781 .437 (Constant) 5.752 2.729 2.108 .037 3 Sex -.013 .633 -.002 -.021 .983 Age -.439 .296 -.177 -1.486 .140 Education -.166 .362 -.052 -.459 .647 Income -1.554 .572 -.228 -2.715 .008 Purchase History .597 .445 .118 1.342 .182 Brand Familiarity .165 .191 .078 .862 .391 COO .249 .321 .065 .776 .439 Quality 1.224 .388 .314 3.154 .002 Price .518 .383 .126 1.355 .178 Model summary R R Square Adjusted Std. Error Change Statistics R Square of the Estimate R Square Chg. F Chg. df1 df2 Sig. F Chg. .143 2.354 .186 4.373 6 115 .001 1 .431 .186 .140 2.313 .004 .610 1 114 .437 2 .436 .190 .251 2.182 .117 9.419 2 112 .000 3 .554 .307. 立. 政治大. ‧. ‧ 國. 學. n. er. io. sit. y. Nat. al. Ch. engchi. i n U. v. a. Dependent Variable: Total Purchase Intention - Cars. For the Cars category, the regression analysis results seem to only support the significance of Income (negative correlation) and the Quality (positive correlation). variables.. This implies. is the strongest purchase motivator.. 32. that. the quality. of the product.

(40) 4.4.2.2 Bags Table 21 - Regression analysis results - Bags. io. n. al. Ch. engchi. a. Dependent Variable: Total Purchase Intention - Bags. .057 -.288 .060 -.151 .313 .063 .115 .254 .099. ‧. Nat. 1 2 3. R R Square Adjusted R Square .555 .308 .272 .581 .338 .297 .649 .421 .374. 立. 政治大. .015 -.367 .088 -.185 .320 .094 .181. 學. ‧ 國. 3. .002 -.407 .096 -.179 .328 .128. y. 2. (Constant) Sex Age Education Income Purchase History Brand Familiarity (Constant) Sex Age Education Income Purchase History Brand Familiarity COO (Constant) Sex Age Education Income Purchase History Brand Familiarity COO Quality Price. Standardized Coefficients Beta. sit. 1. Coefficientsa Unstandardized Coefficients B Std. Error 12.218 1.412 .010 .453 -.784 .214 .241 .285 -.949 .430 .881 .235 .376 .252 11.197 1.458 .084 .447 -.706 .213 .221 .280 -.978 .422 .860 .231 .277 .252 .608 .268 6.135 1.890 .312 .438 -.555 .205 .151 .265 -.801 .401 .841 .219 .185 .241 .385 .267 .806 .297 .351 .301 Model Summary Std. Error of the Estimate R Square Chg. 2.354 .308 2.313 .030 2.182 .083. Change Statistics F Chg. df1 8.518 6 5.148 1 8.036 2. er. Model. i n U. v. t. Sig.. 8.656 .022 -3.668 .845 -2.209 3.754 1.489 7.680 .189 -3.317 .787 -2.316 3.725 1.102 2.269 3.246 .712 -2.701 .570 -1.998 3.846 .770 1.440 2.710 1.167. .000 .982 .000 .400 .029 .000 .139 .000 .851 .001 .433 .022 .000 .273 .025 .002 .478 .008 .570 .048 .000 .443 .153 .008 .246. df2 115 114 112. Sig. F Chg. .000 .025 .001. As in previous case, the regression analysis results show that the Income variable is negatively associated with the purchase intention, while quality has a significant positive correlation. Furthermore, age seems to have an even stronger negative significance than the income variable – this phenomenon might be connected with the fashion aspect of products of this category, which can be expected to be especially important for the younger generations. The age variable also has a weaker, non-significant negative correlation in Cars (B -0.439, Sig. 0.140) and Shoes (B -0.337, Sig. 0.164) categories. This partial relevance of age in the other two categories might hint 33.

(41) at the presence. of a general. trend. of consumers. of older. generations. manifesting stronger ethnocentrism and preference of domestic products, which is observed in several works related to the research of consumers’ attitudes. towards. imported. products,. like. Han. (1988). or. Chryssochoidis et al. (2007). The plausibility of such explanation seems to be accentuated. by the fact. that. age. is irrelevant. to the formation. of purchase intention in the case of the Cheese category, within which there is virtually no domestic production which could be preferred to the imported products. Another variable, which is important here is the purchase history. 4.4.2.3 Cheese. 政治大. Table 22 - Regression analysis results - Cheese. Coefficientsa Unstandardized Standardized t Model Coefficients Coefficients B Std. Error Beta (Constant) 11.218 1.535 7.310 Sex -.003 .496 .000 -.006 Age -.214 .235 -.106 -.908 Education -.032 .302 -.012 -.107 Income -.727 .477 -.132 -1.525 Purchase History 1.211 .236 .437 5.123 (Constant) 11.204 1.592 7.038 Sex -.003 .498 .000 -.005 Age -.213 .238 -.106 -.892 Education -.034 .306 -.013 -.110 Income -.728 .479 -.132 -1.518 Purchase History 1.210 .240 .437 5.050 COO .011 .301 .003 .036 (Constant) 5.239 1.807 2.900 Sex .210 .444 .037 .473 Age -.009 .216 -.004 -.040 Education -.171 .273 -.065 -.626 Income -.699 .428 -.127 -1.632 Purchase History .948 .227 .342 4.179 COO .069 .268 .020 .259 Quality 1.338 .260 .432 5.145 Price .092 .302 .025 .305 Model Summary R Square Adjusted Std. Error Change Statistics R Square of the Estimate R Square Chg. F Chg. df1 df2 .200 .165 2.628 .200 5.799 5 116 .200 .158 2.639 .000 .001 1 115 .383 .339 2.338 .183 16.732 2 113. 立. n. al. er. io. sit. y. Nat. 2. ‧. ‧ 國. 學. 1. 3. R 1 .447 2 .447 3 .619. Ch. engchi. i n U. v. Sig.. .000 .996 .366 .915 .130 .000 .000 .996 .374 .912 .132 .000 .971 .004 .637 .969 .532 .105 .000 .796 .000 .761. Sig. F Chg. .000 .971 .000. a. Dependent Variable: Total Purchase Intention - Cheese. In the Cheese. category,. the variables. are the Purchase history and Quality variables. 34. with significant. correlation.

(42) The association between the purchase history and purchase intention may be viewed as indirectly supporting the H1a hypothesis implying that the increased level of product familiarity (hereby represented by purchase history) would have a direct positive effect on the purchase intention. The irrelevance of this variable for the other categories may be explained by the low levels of purchase history for these categories, as its significance is inversely related to its means here (cars - mean: 0.3 Sig.:0.182, shoes - mean: 0.89, Sig.: 0.07). 4.4.2.4 Shoes Table 23 - Regression analysis results - Shoes Model. 立. ‧. sit. n. 1 2 3. al. R Square Adjusted R Square .423 .179 .137 .425 .180 .130 .492 .242 .181. er. io. R. y. Nat. 3. 政治大. 學. 2. (Constant) Sex Age Education Income Purchase History Brand Familiarity (Constant) Sex Age Education Income Purchase History Brand Familiarity COO (Constant) Sex Age Education Income Purchase History Brand Familiarity COO Quality Price. ‧ 國. 1. Coefficientsa Unstandardized Standardized t Coefficient Coefficient B Std. Error Beta 11.597 1.618 7.167 -.840 .527 -.145 -1.594 -.471 .241 -.233 -1.954 .043 .350 .016 .122 -.756 .507 -.136 -1.491 .427 .289 .156 1.478 .649 .208 .300 3.114 11.363 1.730 6.567 -.830 .530 -.143 -1.567 -.459 .244 -.226 -1.878 .040 .351 .015 .114 -.747 .510 -.134 -1.466 .440 .292 .161 1.508 .621 .221 .287 2.808 .125 .318 .036 .392 7.885 2.217 3.556 -.593 .520 -.102 -1.141 -.337 .241 -.166 -1.401 -.055 .343 -.021 -.160 -.448 .510 -.080 -.878 .523 .286 .191 1.827 .581 .216 .268 2.694 .089 .309 .025 .287 .933 .312 .286 2.990 -.348 .335 -.099 -1.040 Model Summary Std. Error Change Statistics of the Estimate R Square Chg. F Chg. df1 2.696 .179 4.189 6 2.706 .001 .154 1 2.625 .061 4.538 2. Ch. engchi. a. Dependent Variable: Total Purchase Intention - Shoes. 35. i n U. v. Sig.. .000 .114 .053 .903 .139 .142 .002 .000 .120 .063 .909 .145 .134 .006 .696 .001 .256 .164 .873 .382 .070 .008 .775 .003 .301. df2 Sig. F Chg. 115 .001 114 .696 112 .013.

(43) In the case of the shoes category, the quality also shows strong correlation, which makes the H5 stating, that “High quality perception will strengthen the purchase. intention”. the only. hypothesis undoubtedly. confirmed. by the regression analysis for all four categories. Unlike. the other. categories,. in this category,. the Brand. familiarity. is significant, which is an expectable result, as brand name is an important aspect of product evaluation of shoes, as. demonstrated for example. in Elliott & Cameron (1994:4). The brand name importance for shoes is only accentuated by the fact that. 政治大 for many well-known. Europe (and especially Italy, which is worlds’ number-two exporter. 立. of leather shoes) is home. (in many cases, highly. ‧ 國. 學. luxurious) brands, like Gucci, Prada or Giorgio Armani (as mentioned before). Furthermore, the European shoes are highly-evaluated commodity, as shown,. ‧. for example, in Wall and Heslop (1986; as quoted in Al-Sulaiti and Baker 1998:5) – a study of attitudes of Canadian consumers towards domestic and. y. Nat. sit. imported products, in which the Italian-made shoes were the only product. er. io. category, which was rated highly than all the domestic products.. n. a. iv. l C results, which appears One of the regression analysis n to be rather paradoxical,. hengchi U. is the observation that income is related negatively to the purchase intention for cars and bags (meaning, that the respondents of the lower income groups are more willing to purchase the relatively expensive EU products). This phenomenon might be related to the positive role of price as discussed in (Lichtenstein, Ridgway, and Netemeyer 1993:2-3) and especially the Prestige sensitivity, where the higher price of the product is supposed to affect positively the perception of the social status of the purchaser. It can be expected, that this effect may be particularly desirable for the respondents from the lower income groups. 36.