Analysis of Influencing Factors of Deviation of Consumer Willingness and Behavior in Popular Tea Consumption Ping Chen 1, 2, Jiangfan Yang 1 1 College of Economic, Fujian Agriculture and Forestry University, Fuzhou 350002, China 2 College of Horticulture, Fujian Agriculture and Forestry University, Fuzhou 350002, China Abstract Based on the survey data of 1419 respondents from general public, this paper uses logistic regression analysis to identify the key influencing factors of deviations of consumer willingness and behavior in popular tea consumption. It finds that compared with those with consistent popular tea consumption willingness and behavior (unwilling and not buying popular tea), consumers may be inconsistent in popular tea consumption willingness and behavior due to such factors as gender, age, education received, family income, understanding of tea, its price, brand, and quality. In view of this, enterprises and governments can popularize tea-related knowledge and strengthen the regulation of tea quality and price to break the asymmetric information between consumers and tea enterprises, and to provide more standardized and rational consumer choices. Keywords: Popular Tea, Purchase Intention, Consumer Behavior, Behavior Deviation. 1. INTRODUCTION Popular tea is the tea of affordable prices for the public, and is also known as parity tea or folk tea. Based on previous studies, the price of the popular tea is between 101-500 yuan / 0.5kg. This paper takes the popular tea consumption as an example to empirically test the deviation of consumer preference and consumption behavior, as well as the influencing factors. Consumers willingness to consume can reflect their descriptive preference, but the willingness to consume may not be enough to effectively predict consumer behavior, that is, the willingness to consume may not be effectively converted to the actual purchase behavior, hence the deviation. The existence of such deviation may lead to that in market demand deduced and policies developed. Although many scholars have studies consumer spending willingness and behavior, few examined this kind of deviation. In view of this, exploring the deviations between consumer willingness to consume and consumers actual consumption behavior is conducive to accurately grasping the market demand and creating a good market environment, which will also greatly promote the commercialization of tea. The reason for such deviation should be analyzed and understood by relevant agricultural departments and tea enterprises, which will, on this basis, put forward scientific and reasonable countermeasures and suggestions to correctly guide consumer behavior in the popular tea market. 2. DATA SOURCES The paper uses two kinds of questionnaire surveys: field survey and online survey, with samples all randomly selected. The formal questionnaire survey started in August 2016 and the one-month survey involved 9 cities in Fujian Province, such as Fuzhou, Xiamen, Quanzhou and Zhangzhou. The online survey last for 3 months and targeted at the general public in China. The main objective of it was to cover as many areas as possible with limited funds, so as to ensure the representativeness of samples and the results of follow-up empirical research. Finally, a total of 1618 questionnaires were distributed, and 908 and 710 pieces were collected from online surveys and field surveys respectively. After the questionnaire data integrity check, 1419 copies were valid. Respondents of the valid questionnaires were consumer groups of different age, income, gender, education and occupation. 3. MEASUREMENT MODEL CONSTRUCTION AND VARIABLE SELECTION 3.1 Measurement model construction 93
The analysis of consumer s willingness and behavior deviation adopts the multivariate disorder Logistic regression model, and refer to the maximum likelihood method. The model is: z i = α + β i X i + μ (1) Among them, β i is the matrix of estimated coefficients, X i, the explanatory variable matrix representing consumer s personal characteristics, socio-economic characteristics, cognition and evaluation of the popular tea, and μ is the error term. The probability of consistency of public tea consumption willingness and behavior is: The estimation formula isl p i = F(z i ) = F(α + β i X i + μ) = 1 1+e z i = 1 1+e (α+β i X i ) (2) n ln( p i ) z 1 p i = α + i=1 β i X i + μ (3) i The respondents are first put into two groups according to their willingness or unwillingness to purchase the popular tea. Then check if there is inconsistency between their willingness and behavior with the type of tea they usually purchase (popular tea and non-popular tea). According to the results, the respondents are ultimately put into four groups, with the first being those of consistent willingness and behavior, namely, unwilling to and not buy the popular tea, the second, those of consistent willingness and behavior, willing to and buy the popular tea, the third, those of inconsistent willingness and behavior, willing to but not buy the popular tea, and the fourth, those of inconsistent willingness and behavior, unwilling to but buy the popular tea (it is not specifically analyzed for the lack of this type of samples and it is not the purpose of research here). The occurrences in the model are that of the second type (willing to and buy the popular tea), the third type (willing to but not buy the popular tea) and the first type (unwilling to and not buy the popular tea). Table 1 The Potential Influencing Factors of Consuming Will of Popular Tea and the Explanation Name of the variable Symbol Explanation of the Variable lperson.il Characteristic Age Age 1 = <18;2 = 18-25;3 = 26-35;4 = 3^45;5 = 46-60;6 = X50 Gender Gender l=male;2=feniale Education Edu 1 = junior high school or below;2 = high school or secondary school;3 = college;4 = undergraduate; 5 = master or above Enplovment Employ 1 = civil servants; 2 =institutional staff;3 = enterprise and corporate managements; 4 = enterprise and company general staff; 5 = farmer; 6 = students;7 = others Marital Status Marital 1 = unmarried (divorced, widowed);2 = married Number of the Children Children 1 = no child;2 = 1 child;3 = 2 childrea;4 = 3 children;5 = 4 children or more The Number of People 1 = nearly all don't like;2 = few people like;3 = some people Surround around Who Drink Tea like;4 = many people like 2 Consumption Motive Factor Motiv 1 = refreshing and quench their thirst;2 = health care 3 Cognitive Factors Tea Cognition Level Know 1= don't know;2hcnow some;3=kno\v well Subjective Index Subject 1 =price: 2=quality; 3=brand Quality Identification Identi l^)rice;2=taste I store;3=the recommendation of others;4=brand;5=other 4 External Emiionmental Factors Choice of Tea Packaging Popular Tea Consumption Channels Packing Channel 1 =small separate package, iron (paper) box in bulk;2=with packing box;3=don, t care 1 = tea wholesale market;2 =tea chain store;3 =gift from others;4 =counter in si^ermarket or store; 5=small tea shop;6=consimie via Internet 94
As is shown in Table 2, among the consumers willing to buy the popular tea, 929 consumers buy it and 119 do not. Of those unwilling to buy the popular tea, 265 do not buy it and 26 buy it at last (it is not specifically analyzed for the lack of this type of samples and it is not the purpose of research here). Table 2 The Relationship between Consumption Intention of the Popular Tea and Actual Consumption Behavior Purchase Intention Consumer Behavior Consumer Behavior Not Consumer Behavior total Purchase Intention 929 199 1128 Not Purchase Intention 26 265 291 total 955 464 1419 2.2 Willingness and behavior deviation analysis of popular tea consumption samples There are mainly four reasons for unwillingness to buy popular tea. As is shown in Figure 1, low price and worry about the quality is the major factor for unwillingness to buy, as for most consumers, the price indicates the quality and they may judge whether the tea is good or not by looking at its price. For those willing to buy the popular tea, the three most influencing factors are "price", "quality" and "brand", showing that when making purchase decisions, consumers first consider "price" and "quality", and then "brand" and "safety". Therefore, consumers who do not stress brand are more likely to buy the popular tea. D Few people around to buy this kind of tea 18% C Don't want to buy without big brand 19% A Can't buy this kind of tea which never heard of 18% B Be afraid of low cost 45% Figure 1. Survey Results of the Reasons for Not Purchasing Popular Tea Consumer behavior shows obvious preference in popular tea consumption. In the survey, types of tea consumed mainly includes black tea, green tea, oolong tea, dark tea, etc. in terms of fermentation, or top-quality products, first-class products, and quality products in terms of quality, and price is less considered in the consumption. We can see that consumer willingness and purchase behavior of popular tea are separate and not entirely consistent. This is because of the complex factors that affect purchasing decisions, such as cognitive level, purchasing power, etc., as is analyzed below. Willing but not buying, 14.02% Not willing but buying, 1.83% no buy and no will, 18.68% buy with willness 65.47% Figure 2. Survey Results of the Deviation of Consumption Intention and Consumption Behavior of Popular Tea Consumption 95
As is shown in Figure 2, 14.02% of consumers are willing to consume but not actual buy it, 65.47% is consistent in purchase intention and behavior, and 18.68% are not willing to and so do not buy the popular tea. 4. IDENTIFICATION AND DISCUSSION OF KEY INFLUENCING FACTORS Based on the above theoretical and statistical analysis, we use SPSS19.0 to evaluate the logistics model. The regression results are shown in Table 3. According to the estimation results, the variables of "gender", "brand first" and "education" are significant at 10% level while the variables of "age", "price priority" and "monthly family income" are significant at 5% level. First, we analyze the factors that affect the ratio of the second type of incident (willing to and buy popular tea) to the first type of incident (unwilling to and not buy the popular tea). Of consumers personal characteristics, the gender regression coefficient is positive, and significant at the level of 10%, and age regression coefficient, negative and significant at the level of 5%, indicating that the older the consumers are, consumer spending willingness is more possibly to be consistent with consumer behavior. The main reason is that with the growth of age, consumers are more clear about their own purchasing needs and preferences, their experience in consumption is richer, their spending power higher, and are more easily to convert their consumption willingness to purchase behavior. In contrast, young consumers are easily influenced by people around, so is their purchase intention. The level of education is divided into five groups, with reference to graduate students and above. Except for the negative coefficient of primary education, the coefficients of other education levels are all positive, indicating that the higher the level of education is, the willingness and behavior of consumers will be more likely to be consistent. That s because people that have received a higher level of education are equipped with stronger ability to accept and process information, and the ability of logical judgment and ideal spending power. As far as socio-economic characteristics are concerned, the monthly income of consumers is divided into five groups, with high income (more than 8001 yuan) as a reference. The regression coefficients of all these groups are positive, of which the high-income groups (5001-8000 yuan) are significant at the level of 5%, showing that household income is an important factor that affects the consumption willingness and behavior of popular tea. Consumers with higher incomes will pay more attention to the quality and the destination of consumption in daily expenses of their families. As they have higher spending power, their consumption willingness are more likely to be consistent with their purchase behavior. The impact of tea knowledge on the deviation between the willingness and behavior of consumption shows that the consumer s basic evaluation of popular tea shows a significant and positive effect on the statistical level of 10%. This shows that unfavorable evaluation may lead to high inconsistency in consumption willingness and behavior. Among them, the price and brand are significant at the level of 5% and quality, significant at the level of 10%, indicating that consumer is the most sensitive to the price, followed by brand. Table 1 The estimated result of popular tea consumption willingness and behavior deviation (with reference to the first category) Table 3 Model Fitting Results of the Consumption Intention and Behavior Difference of the Popular Tea Consumption (Reference type1) variable buy with willness B SE Gender 0.367* 0.013 Age -1.229** 0.115 Marital -0.307 0.558 Knowledge 0.983 0.547 Subjective1 1.141** 0.238 Subjective2 1.017** 0.225 Subjective3 1.682* 0.454 Education1 0.491 0.214 Education2 0.055 0.003 Education3 1.34 0.842 Education4 1.212 0.368 Education5 Employ1 Employ2-2.725 0.457 Employ3 2.001 0.44 96
Employ4-3.514 0.78 Employ5-1.505 0.492 Employ6-1.26 0.133 Packing2 0.235 0.114 Packing3 1.854 0.562 Children 0.079 0.11 Surrounding 0.199 0.063 Income1 Income2 0.786 0.145 Income3 0.586* 0.316 Income4 0.613* 0.31 Income5 0.640*** 0.751 Channel1 Channel2-0.071 0.122 Channle3 0.652 0.205 Channel4-2.991 0.97 Channel5-1.413 0.675 *, **, *** are significant at 10%, 5% and 1% respectively Second, we analyze the factors that influence the occurrence of the third type of incident (willing to but not buy the popular tea) and the first type of incident (unwilling to and not buy the popular tea). In terms of personal characteristics, the regression coefficient of age is negative, and significant at the level of 5%, indicating that the greater the age, the greater the possibility of consistency in consumption willingness and behavior. The main reason is that with the growth of age, consumers are more clear about their own purchasing needs and preferences, their experience in consumption is richer, their spending power higher, and are more easily to convert their consumption willingness to purchase behavior. In contrast, young consumers are easily influenced by people around, so is their purchase intention. The level of education is divided into five groups, with reference to graduate students and above. Except for the negative coefficient of primary education, the coefficients of other education levels are all positive, indicating that the higher the level of education is, the willingness and behavior of consumers will be more likely to be consistent. That s because people that have received a higher level of education are equipped with stronger ability to accept and process information, and the ability of logical judgment and ideal spending power. As far as socio-economic characteristics are concerned, the monthly income of consumers is divided into five groups, with high income (more than 8001 yuan) as a reference. The regression coefficients of all these groups are positive, of which the high-income groups (5001-8000 yuan) are significant at the level of 5%, showing that household income is an important factor that affects the consumption willingness and behavior of popular tea. Consumers with higher incomes will pay more attention to the quality and the destination of consumption in daily expenses of their families. As they have higher spending power, their consumption willingness are more likely to be consistent with their purchase behavior. The impact of tea knowledge on the deviation between the willingness and behavior of consumption shows that the consumer s basic evaluation of popular tea shows a significant and positive effect on the statistical level of 10%. This shows that unfavorable evaluation may lead to high inconsistency in consumption willingness and behavior. Among them, the price and brand are significant at the level of 5% and quality, significant at the level of 10%, indicating that consumer is the most sensitive to the price, followed by brand. Table 4 Model Fitting Results of the Consumption Intention and Behavior Difference of the Popular Tea Consumption (Reference type1) variable Willing but not buying B SE Gender 0.822* 0.54 Age -0.208** 0.689 Marital -1.603 0.022 Knowledge 0.073 0.269 Subjective1 0.754** 0.576 Subjective2 0.517* 0.036 Subjective3 1.643 0.05 Education1-1.721* 0.022 Education2 1.351 0.207 97
Education3 1.223 0.568 Education4 0.604 0.466 Education5 Employ1 Employ2-2.28 0.458 Employ3 1.345 0.917 Employ4-2.212 0.727 Employ5-1.414 0.116 Employ6-1.181 0.737 Packing2-2.331 0.898 Packing3-1.237 0.586 Children 0.609 0.626 Surrounding -2.331 0.898 Income1 Income2-1.237 0.586 Income3 0.982 0.068 Income4 1.105 0.576 Income5 1.172** 0.304 Channel1 Channel2 0.237** 0.613 Channle3 0.297 0 Channel4-1.237 0.586 Channel5 1.548 0.107 *, **, *** are significant at 10%, 5% and 1% respectively 5. CONCLUSION AND RECOMMENDATIONS While exploring consumer willingness and behavior on popular tea consumption, this paper probes into influencing factors on the deviation of consumer willingness and behavior and discusses those restrict the conversion of consumer willingness to actual consumption behavior, which in turn limits the popularity and market expansion of popular tea. Multivariate Logistic results show that compared with those with consistent popular tea consumption willingness and behavior (unwilling and not buying popular tea), consumers may be inconsistent in popular tea consumption willingness and behavior due to such factors as gender, age, education received, family income, understanding of tea, its price, brand, and quality. The paper reveals that targeted tea popularization, especially the knowledge of tea, and the increase of consumer awareness of tea price composition and quality appraisal can help consumers make a more rational choice. For tea safety and quality labeling (traceability), we can supervise the implementation by major Chinese tea enterprises and the government should set up regulatory agencies (like Tea Quality Inspection Center) to make the information disclosure system open and transparent and to fully protect the consumers right to information and the consumer welfare. The government should also promote and popularize popular tea that accords with the mass consumption habits and economic affordability, break the asymmetric information pattern between consumers and tea enterprises, provide a more standardized and rational consumption choice, and protect consumers right to know and to choose through the communication of authoritative information between consumers and the public. ACKNOWLEDGEMENTS This word is supported by Chinese Oolong Tea Collaborative Innovation Project, 2011: Research on Innovation of Internet + Oolong Tea (No. K8015H01C). REFERENCES Chen C., Shi C.Y., Zhan J.T., Lu X.Y. (2013). Analysis on Bias between Statement Preferences and Purchase Behavior Regarding Genetically Modified Foods: Based on Edible Oil Consumption of Urban Residents, Issues in Agricultural Economy, (6), 82-88, 112. Chen P., Wang Y.T., Guan X. (2016). Fujian residents tea consumption behavior survey, Forestry economic problems, 36 (4), 369-372. 98
Chen S.N. (1984). International tea market research: Study on the behavior of tea buyers, Tea in Fujian, (02), 39-42. Cook A., Kerr G., Moore K. (2002). Attitudes and Intentions towards purchasing GM Food, Economic psychology, 50-52. Ge W., Jiang H.C. (2015). The Judgment and Hint of Tea Necessity and Luxuries Since Economy, China Tea Processing, (1), 48-52. Guan X. (2011). Tea consumption is driven by brand bulk tea, China Tea, (11), 10-11. Guan X., Yang J.F., Xie X.Y. (2011). Research on the Status of Chinese Tea Consumption and Factors of Tea Consumer's Behavior, Journal of Tea Science, 31 (6), 546-551. Li D.H. (2009). Economics Analysis of Tea Industry in China, China Agriculture Press Liu J.J., Li D.H., Tan Z.W. (2015). Empirical Analysis on the Influencing Factors of Consumers' Purchase Intention of Tea-Based on Jiangxi Province Survey Data, Rural Economy and Technology, (9), 98-101. Liu Q.J. (2014). On the rational regression of tea industry and commodity popularization, The Book of Tea Indestry, Social Sciences Academic Press, 30-32. Noussair C., Robin S., Ruffieux B. (2002). Do consumers not care about biotech foods or do they just not read the labels, Economics Letters, 75, 47-53. 99