3. Research Methods
3.5 Measurement reliability Test
國
立立 政 治 大
㈻㊫學
•‧
N a tio na
l C h engchi U ni ve rs it y
The choice of analyzing wine purchase intention can be explained as follows:
previous research have focused on products that are, as explained by Usunier (2005), over-represented in the work of country of origin image, namely consumer electronics, cars, household appliances, etc. However, no research has integrated studies about wine which, moreover, are very limited (Usunier , 2005). Indeed, as the product category modulates the country of origin effect (Roth and Romeo, 1992;
Insch and McBride, 1998; Piron , 2000), it could explain the low significance of the effect. Finally, we note that the influence on purchase intent emerged in the work of Piron (2000) which focused on several types of products, including wine. The author had noticed that for product such as wine, the "country of origin" effect on purchase intention was stronger compared to other products of basic necessities.
3.5 Measurement reliability Test
The measurement items used in this study have been adapted from previous research and therefore are validated. The measurements have internal and external validity and internal consistency reliability (Cronbach's α). The table 3.1 shows the measurement reliability test conducted for this analysis in order to prove the reliability of multiple-item scales of this study. The table presents the the multi-item scales used in the survey with their factor loadings values and Cronbach's α values.
The Cronbach's α for the five scales are acceptable to good reliability with all the Alpha values higher that 0.5 more preciselly between 0.5 < α > 0.8 (Leech, Barrett, Morgan; 2005). The Factor loadings values for the five scales have a high validity with scores higher that 0.6, between 0.7 < Factor loading > 1 (Kline, 2004).
•‧
國
立立 政 治 大
㈻㊫學
•‧
N a tio na
l C h engchi U ni ve rs it y
Table 3.1 : Factor Loadings and Reliabilities of Measurement Scale.
Measurement Items Factor Loadings Cronbach's α
Country Image
France is economically developed 0.831
France has a rich culture 0.815
French people are hard working 0.808 0.681
French people are friendly 0.832
Overall opinion about France 0.706 Knowledge about wine
Knowledge about wine 0.815
Interest in wine 0.689 0.831
Others ask your advice on wine
0.755
Since all the measurement items used in this study are examined to assure appropriate reliability, I also constructed composite measures by summating the multiple indicators of each essential research variable. Then, these composite measures are used in the final regression analysis.
•‧
國
立立 政 治 大
㈻㊫學
•‧
N a tio na
l C h engchi U ni ve rs it y Chapter IV. Data Analysis
Of the 202 respondents, 93 (46%) were male and 109 (53%) were female. The average age in the highest respondent group is between 21 and 29 years old. All income and education groups were represented in the sample.
Table 4.1 to 4.4 provide the exact numbers of the respondents demographic data.
Table 4.1 : Gender of respondents
Frequency Percent Cumulative Percent
Female 93 46 46
Male 109 54 100
Total 202 100
Table 4.2 : Age of respondents
Frequency Percent Cumulative Percent
18 to 20 years old 6 3 3
21 to 29 years old 81 40 43
30 to 39 years old 74 36.6 80
40 to 49 years old 32 16 96
50 to 59 years old 7 4 99
60 years old and over 2 1 100
Total 202 100
•‧
Table 4.3 : Highest completed education level
Frequency Percent Cumulative Percent
High school and below 27 13.4 13.4
Undergraduate 121 59.9 73.3
Graduate and above 54 26.7 100
Total 202 100
Table 4.4 : Average monthly income in New Taiwan Dollar
Frequency Percent Cumulative Percent
< NT$20,000 24 11.9 11.9
20,000 - 49,999 64 31.7 43.6
40,000 - 59,999 65 32.2 75.7
60,000 - 79,999 37 18.3 94.1
80,000 - $99,999 6 3.0 97
> 100,000 6 3.0 100
Total 202 100
Table 4.5 : Wine consumption frequency
Frequency Percent Cumulative Percent
Daily 6 3 3
Few times a week 49 24.3 27.2
Few times a month 89 44.1 71.3
Once every few descriptive socio-demographic data was explored using descriptive analysis. In order to test the research hypotheses, a hierarchical multiple regression analysis was used to examine the significance of the three independent variables--socio-demographic
•‧
factors, the perception about France (country image) and the knowledge about wine
—in predicting the purchase intention of French wine among Taiwanese consumers.
The hierarchical multiple regression analysis contains three types of predictors:
socio-demographics, country image and knowledge ability about wine.
Table 4.6 Hierarchical multiple regression analysis of all the factors (N = 202)
Model
β
t Sig.(p)•‧
國
立立 政 治 大
㈻㊫學
•‧
N a tio na
l C h engchi U ni ve rs it y
Hypothesis 1 : Influence of gender factor on country image.
Based on the literature review, the research about demographics factors influencing country of origin effect shows that younger women with higher education level are the most responsive to the French origin of the wine during the purchase. According to the result of regression analysis presented on Table 4.6, gender is significantly predicting the consumers' country image at a low level of (β=0.155, p<0.05)， while age and education level are not significant predictors for
this study. Hypothesis H1 is partially supported as gender showed significant influence but the two other socio-demographic variable of age and level of education are not significant in this study. We can conclude that women are more sensitive to the French country image of the wine than men. Therefore hypothesis 1 is partially supported as women's purchase intent is more influenced by the French country of origin of the wine than men's.
Hypothesis 2 : Country Image
Country image perception was tested with a five-item index, with all items measured on a five-point Likert scale (from 1 = strongly disagree to 5 = strongly agree). The five measurement items evaluated France's economical development and cultural level, French people's friendliness and work capabilities. The dependent variable is the likeliness to buy French wine.
The hierarchical multiple regression test for hypothesis H2 examines the influence of all country image measurement items on the final purchase intent. The result indicates that model 2 is statistically strongly significant. The composite measure of
•‧
國
立立 政 治 大
㈻㊫學
•‧
N a tio na
l C h engchi U ni ve rs it y
the five country image items is significantly related to the purchase intent of French wine. The result shows that respondents evaluated France with an overall positive country image. The result of this analysis also shows that the perception of France is a significant predictor of the final purchase intent of French wine in Taiwan. In this case, the relationship between country image and purchase intent of French wine is strongly significant (β=0.467, p<0.001). Hypothesis 2 is therefore supported:
consumers with overall positive country image have greater purchase intention of French wine.
Hypothesis 3 : Expertise about wine
Wine expertise is measured using a three item measurement index from the respondent's indication of level of expertise on their level of knowledge, interest and if friends and relatives are asking for their advice when purchasing wine.
The path coefficient of this variable, expertise about wine, is also significant (β=0.161, p<0.01). This data analysis confirms the prediction that consumers will use their country image to evaluate the wine. Therefore, the more positive is the country image, the more favorable is the purchase intention on French wine. The more knowledgeable consumers are about wine, the more they will be likely to buy French wine because of their positive country image. Hypothesis H3 is therefore supported.
•‧
suggestions for the purpose of future research on country imageTable 5.1 : Summary of Research Hypothesis and Outcomes
Research Hypothesis Outcome
H1: Socio-demographic factors will influence Taiwanese consumers purchase intention of French wine such that women, higher educated, and younger ones will have greater purchase intention toward French wine
Partially supported P<0.05 H2: Consumers who have a more favorable country image about
France will have greater purchase intention of French wine. Supported P<0.001 H4: The more the consumer is knowledgeable about wine, the more
favorable is the purchase intention of French wine. Supported P<0.01
5.1. Country of Origin effect on purchase intent of French wine.
The discussion of results is done through the analysis of the impact of country of origin on Taiwanese consumers' attitudes toward French wine. The analysis showed that the overall impact of country of origin on the purchase intention of French wine is positive. It would appear that a positive image of the country of origin of the wine is a positive determinant to final purchase decision.
The report table 4.5 on wine consumption frequency shows that 96% of respondents are wine consumers with a consumption of at least one time every few months.
Because a part of the respondents were selected from the email database of a wine shop's clients we can understand the high level of wine consumers.
•‧
國
立立 政 治 大
㈻㊫學
•‧
N a tio na
l C h engchi U ni ve rs it y
Finally, our results were not much influenced by socio-demographic variables except for the gender of respondents.
On the managerial level, this work provides a real answer to the question on the impact of image brought to the country of origin of a product. Such study tries to educate professionals on the risks and benefits of country of origin effects. It gives a general track of actions to play with when exporting wine in Asia.
The main goal of this research was to understand Taiwanese consumers perception of France and its impact on their purchase intention.
View from abroad, France has always had a specific image. Sometimes admired for its cultural influence, envied or criticized for its unconditional commitment to its lifestyle, pointed at for its endemic social issues or just misunderstood because of its way to do things differently, namely, in "the French way."
France's influence is taken in a very general sense, and in this definition, stereotypes about France are often strong. France's foreign policy, its economic and social model, its cultural influence, or the its social and economic weight are all factors that can influence public opinion. On the other hand, it is difficult to compare the actual level of influence of a country image in the international scale. The goal is to understand if its influence is seen as mainly positive or negative by the public opinion.
The test results for this study have shown that Taiwanese people have a overall positive image of France. Only 2,5% of respondents reported having a negative opinion about France with almost 58% of respondents having a positive image. Even though as much as 72% of the respondents have never visited France before and 32%
feel not knowledgeable about France, it shows that Taiwanese people's positive
•‧
國
立立 政 治 大
㈻㊫學
•‧
N a tio na
l C h engchi U ni ve rs it y
opinion about France comes from their perceived image. Thanks to such positive overall image, France posses advantages at the international level. Such advantages are used by companies to expand in new markets which is one of the reason why French wine is forcing exportations in Asia.
If we consider the fact that products made in France own a strong quality image worldwide it is interesting to notice that Taiwanese respondents expressed a very high opinion on French products. Indeed, 64% of respondents are having positive opinion on Made in France products (14% very positive). In regard to their opinion of French wine, the results are even higher with 44% of respondents having a very positive opinion about French wine (45% positive). Such high opinion on a single product is extremely encouraging knowing that 39% of the respondents don't feel knowledgeable about wine. The study shows us that the country of origin of wine remains a very important factor of quality evaluation for Taiwanese consumers, although can be price is also a meaningful factor when purchasing wine for personal purpose. Indeed, the results of H6a/b show that Taiwanese consumers attach importance to the country of origin of the wine when purchasing for social purpose rather than for private consumption. This finding goes in line with Yang's (1989) research explaining that Chinese culture consumers are in low-involvement purchase when buying for private purpose and therefore they will be first influenced by the price.
Taiwanese wine market is growing but remains small in terms of consumption per capita. The fact that Taiwan market it not yet mature for wine can explain somehow the lack of familiarity with French wine which helps understand the strong influence of the wine's country of origin cues when at the time of purchase.
•‧
國
立立 政 治 大
㈻㊫學
•‧
N a tio na
l C h engchi U ni ve rs it y
As predicted and analyzed on the test novice consumers having an overall positive opinion about France have expressed a strong likeliness to purchase French wine (M
= 4). In the other hand, experts about wine are not directly influenced by the French origin of the wine which could be explained by their knowledge about wine in general. It does not mean they will not buy French wine but instead evaluate the other attributes of the wine such as the grape used to make the wine or the year of bottling. We can infer that despite the clear influence of country of origin of the wine, country of origin is not the only predictable we have to take into account. Moreover, only 8% of the respondents indicated that they had an expert level which is a very low sample, not sufficient to analysize effectively the influence country image on the level of expertise.
5.2 Limitations
In our review of the literature on the "country of origin" effect, we presented several models to understand the influence of the image of the country of origin on consumer evaluations. Among these models, there was the one of Obermiller and Spangenberg (1989) and of Samiee (1994). Both models incorporate a multitude of variables including two more significant for this research. The first comes from the model of Obermiller and Spangenberg (1989) which refers to the importance of the country of origin, ie the identification and use of the label indicating the origin of the product by the consumer. The second variable is derived from the model Samiee (1994) which also constitutes its input variable, namely the consciousness of the country of origin or the importance accorded to it by the consumer. This author also states that when the consumer is not sensitive to information about the country of
•‧
國
立立 政 治 大
㈻㊫學
•‧
N a tio na
l C h engchi U ni ve rs it y
origin of the products, the model becomes irrelevant. In addition, we have also seen that the problem of the consumer's sensitivity to information on the origin of a product was raised in the surveys conducted by COFREET and CTCOE in the textile sector, which show that the origin of the products directly attracted the attention of consumers for 36% of them and only when compared to other product information.
Therefore, the fact that we only address in this study the influence of country of origin information on the wine without taking into account the other importent criteria necessary for the evaluation of a wine may seem limited.
This study also shows limitations in the following: the questionnaire survey was not conducted in an actual purchase situation, instead, it was a survey conducted on the internet. This situation can have several bias because we can't be sure that in a real context of purchase, the respondent would proceed in the same way to evaluate the wine.
Firstly, the lack of physical wine bottle in the study could have affected the results of the study. Indeed, all evaluations of the respondents relied on the analysis of their opinion on the characteristics of French wine. However, the literature on the "country of origin" effect shows that in the absence of stimulus (product) during the study, the
"country of origin" effect is generally stronger than when product is present (Peterson and Jolibert, 1995). This is especially important in terms of assessments of perceived product quality. It is possible that some of our results were affected by the lack of physical product. Obviously, given the mode of administration of our investigation (online) but also for reasons of high cost in an experiment with real products, we were not able to offer a physical product to our members in the survey.
•‧
國
立立 政 治 大
㈻㊫學
•‧
N a tio na
l C h engchi U ni ve rs it y
In regards to the analysis limitations, each independent variable was treated separately from other independent variables. However, this approach might prevents to highlight the joint effect of all independent variables on the dependent variable.
Our findings might have been different or more nuanced. However, as it was the first empirical work on the influence of country image on wine purchase intention in Taiwan, this study was foremost an exploratory approach.
The sample of respondents for this study may have, in some ways limitations.
Admittedly, the sample respondents possess the necessary characteristics to be considered actual or potential consumers of wine. Nevertheless it raises the question of the representativeness of the sample to the target population. In addition, several characteristics of the respondents proved to be too homogeneous, which did not allow us to operate certain treatments, including their level of knowledge about France and familiarity with products made in France. These two variables are usually taken into account in the work on the "country of origin" effect as they can influence the purchase behavior of individuals. Therefore, extrapolation of our results to the entire population of consumers of wine could be problematic. Other research to new respondent audiences should be conducted to confirm these findings.
5.3 Suggestions for future research
This study can give rise to different research approaches that are worth exploring.
First, it aims to overcome some limitations that we have just stated, and second, to propose new analysis of the impact of the country of origin of wine on consumer's shopping behavior.
•‧
國
立立 政 治 大
㈻㊫學
•‧
N a tio na
l C h engchi U ni ve rs it y
For future research we suggest to take into consideration other factors than the country of origin of the wine. In this research, we have considered only one criteria for the assessment of the wine: the country of origin. As exposed previously in the limitation it could be interesting to expand and take into account other aspects than the origin, such as grape wine, the year of bottling , the region of production...etc.
Moreover, as presented in the limits of this research, we noted that the study did not use test product physically present and did not place respondent's in an actual purchase situation. We relied on the literature including the research of Peterson and Jolibert (1995) to explain that in the absence of physical product, the "country of origin" effect might be overestimated. For their part, Liefeld (1993) and Samiee (1994) they argue that the consumer pays less attention to information on the country of origin in real purchasing situation. It could be interesting to investigate respondents' attitude in real or alike purchase situation in future research on wine purchase behavior.
•‧
國
立立 政 治 大
㈻㊫學
•‧
N a tio na
l C h engchi U ni ve rs it y
References
Adler, N.J., R.Doktor, and S.G.Redding. (1986). From the Atlantic to the Pacific Century: Cross Cultural Management Review.’ Journal of Management. vol.
12, 2:pp. 295-318.
Ahmed, S. A., d’Astous, A. and S. Zouiten (1993). Personality Variables in the Made-in Concept. in N.G. Papadopoulos and L.A. Heslop, Ed, Product country images: Impact and role in international marketing. New York:
International Business Press, 197- 222.
Bilkey W, and Nes E. (1982). Country-of-Origin Effects on Product Evaluations.
Journal of International Business Studies, Vol. 13, Issue 1, pp. 89-100.
Chih-Kang, W., & Lamb Jr., C. W. (1983). The Impact of Selected Environmental Forces Upon Consumers' Willingness to Buy Foreign Products. Journal Of The Academy Of Marketing Science, 11(1), 71.
Crawford, J. C., & Lamb Jr., C. W. (1981). Source Preferences for Imported Products. Journal Of Purchasing & Materials Management, 17(4), 28-33.
Dewald, B. (2003). Wine consumption in Hong Kong. International Journal of Wine Marketing, Vol. 15 No. 1, pp. 54-68.
Erickson, G. M., Johansson, J. K., & Chao, P. (1984). Image variables in multi- attribute product evaluations: Country-of-origin effects. Journal of
Erickson, G. M., Johansson, J. K., & Chao, P. (1984). Image variables in multi- attribute product evaluations: Country-of-origin effects. Journal of