The Impact of Taobao's Negative Comments on Consumer Willingness

. Based on the theory of trust orientation and perceived risk, this study explores the impact of Taobao's negative evaluation on consumer behaviour, and then conducts research. According to the theory of trust tendency, this paper discusses how consumers view the negative evaluation of goods. Through data analysis, it is found that consumers will have lower purchase intention when they browse negative comments. The psychological mechanism of this is that consumers are affected by perceived risks, resulting in lower brand identity. The subjects of this study are 70 Taobao users, and the hypothesis proposed is tested through correlation analysis.


Introduction
With the development of the Internet era, more and more consumers are shopping online.As with traditional WOM communications, online consumer reviews are important for driving the actions of consumers.(Jumin Lee a, Do-Hyung Park, Ingoo Han, 2006) 1 Product reviews can influence consumers' consumption decisions.(Haoxing Tang and Hui Cao, 2020) 2 Most of the studies on the negative evaluation of goods focus on Amazon and other overseas retail brands.However, China's shopping software Taobao is also an indispensable and important research object.

Research background
Gao Bo and Song Yanping 3 introduced the consumer behaviour model in the analysis of consumer behaviour, that is, consumer behaviour is one of the most common and important activities of human beings.In his article on the discussion of cross-border e-commerce consumer behaviour mode and marketing strategy, Liu Li mentioned that the consumer behaviour process is understood from both psychological and behavioural perspectives, including the psychological activities and actual purchase behaviour related to purchase decisions.Through consulting the domestic and foreign literature, we know that most of the domestic academic research focuses on the relationship between advertising and consumer behaviour.For example, Zhuo Jun's article 4 on online marketing studies the impact of online advertising on consumers' external behaviour.Zhang Peipei 5 wrote many similar studies on the impact of short video ads on women's consumer behaviour.However, the research on consumer behaviour from the perspective of commodity evaluation is less involved.For example, Bai Li's 6 analysis of the impact of commodity customization on consumption evaluation behaviour in e-commerce platform and Wang Hui's 7 empirical analysis of the impact of social factors on online shopping consumption behaviour -taking online evaluation of digital goods as an example were mentioned in two articles.The articles published by domestic scholars in foreign journals also have some relevant research, such as the reference groups and their language styles in the research of Jinhua Zhang and Yongsheng Huang's 8 positive online comment brand attitude towards consumers, and the summary of research on counterfeit goods detection by Hao Xingtang and Hui Cao. 2 The research on this aspect by foreign scholars has increased significantly compared with that in China, mainly focusing on foreign shopping software such as Amazon.Boban Melovi ć aDamir Š ehovi ć aVesna Karad ž i ć aMarina Dabi ć b cDragana Ć irovi ć A. 9 The determinants of millennial online shopping behavior -the impact on consumer satisfaction and e-commerce development has been mentioned a little.Agostino Vollero, Domenico Sardanelli, Alfonso Siano 10 also explored the role of Amazon effect on customer expectations from the perspective of merchants: analysis of user generated content in consumer eretailing, and Farhad Khoshbakht, Atena Shiranzaei, S. M. K 11 Comparison performance evaluation of search items and experience items in the Amazon customer reviews of Quadri.Based on the above information, I found that few scholars at home and abroad have studied the negative evaluation of Taobao, a shopping software product.Therefore, I will take Taobao, a Chinese ecommerce platform, as an example to study the impact of negative evaluation on consumer behaviour.

Theoretical development
With the development of the Internet era, more and more consumers are shopping online.Product reviews can influence consumers' consumption decisions.(Haoxing Tang and Hui Cao, 2020 2 ) Since most scholars at home and abroad focus on positive evaluation, negative evaluation is the key to affect consumers' purchase intention.Therefore, this paper will start from the negative evaluation of consumer behaviour.There are two theoretical bases to support this paper.First, consumers tend to trust other consumers.The tendency of trust is a general rather than a specific tendency, that is, showing belief in human nature and taking a trusting attitude towards others (McKnight et al., 1998)  12 .Second, consumers often have perceived risks before purchasing goods.Glover and Benbasat (2010) 13 believed that perceived risk was an important determinant of consumers' unwillingness to buy online.When consumers take action under uncertainty, trust will play a role (Kim, D., Ferrin, D., & Rao, H. 2008 14 ).Then they will choose to trust some comments left by consumers, especially some negative comments.Based on these two theories, this paper conducts a more indepth and comprehensive study of consumer behaviour and consumer psychology, hoping to fill the gap in this field of academia through research.

Participants and Study Design
I created an online questionnaire using a questionnaire star to enable me to collect online data.this The questionnaire is opened and filled by scanning the QR code.Participants are 70 Taobao users.
Almost everyone in China uses Taobao for online shopping, so you can choose convenient sampling for research.

Measures
In order to evaluate the trust tendency (McKnight et al.,  1998), the answer is based on the 5-point scale, 1 point=very inconsistent, 5=very consistent, and answer the question.
To determine the browsing order, use sorting for evaluation.
In order to understand the perceived risk (Kim, D., Ferrin, D., & Rao, H. 2008), the written answers are used for further evaluation 3 Data

Reliability Analysis
The data obtained from the questionnaire survey has a certain stability, and can give consistent results.This stability is the reliability of the satisfaction survey results, also known as reliability.The higher the reliability is, the more successful the survey is.That is, the data reflects the true opinion of the survey object.In general, the Kronbach coefficient is between 0 and 1.When the coefficient is less than 0.6, the reliability is not enough, while when it is greater than 0.8, it indicates that the reliability is good.This paper uses SPSS22.0 to analyse the reliability of the 3.2questionnaire: According to the table, the Kronbaha Alpha coefficient of the survey results is 0.711, indicating that the reliability of the questionnaire is higher than the standard, and the internal reliability is reasonable.

validity analysis
validity is to measure the internal consistency of the questionnaire.When the internal consistency is high, it means that the views represented by the questionnaire are more unified.In this paper, SPSS26.0 is used to test the KMO and Bartlett sphericity of the questionnaire structural variables: according to the research analysis, when the KMO value is less than 0.6, it is not suitable for factor analysis, while when it is greater than 0.6, it is suitable for factor analysis.From the following table, it can be seen that the KMO value is 0.723, The questionnaire can be used for subsequent analysis   According to the description of the use of Taobao, the rating of browsing Taobao every day is 3.13, which means that the daily browsing of Taobao is not very obvious.In the survey where you will see product reviews and negative reviews will reduce your purchasing desire every time you buy something, the score is above 4 points.This indicates that evaluation and negative evaluation in shopping are the direction of great concern for consumers.Every time a product evaluation is written, the score is 2.20, which means that consumers do not often write comments.

Frequency analysis
When to write comments Among those who can write comments, 48.57% are uncertain, 34.29% are satisfied, and 17.14% are dissatisfied.There is a significant difference between different genders in each evaluation (p<0.05), and there is no difference in other dimensions.Gender shows 0.01 level significance (t=-8.500,p=0.000) for each commodity evaluation.The average value of men (1.00) is significantly lower than that of women (2.24).The frequency of evaluation of male groups will be significantly lower than that of female groups.

Correlation analysis
The correlation analysis mainly uses the person coefficient to test the degree of correlation between the data.If the correlation coefficient of two variables is positive, it indicates that there is a positive correlation between the two variables; if it is negative, it indicates that there is a negative correlation between the two variables.In the correlation analysis, it was found that the monthly income and the tendency to believe in positive or negative reviews had significant effects.The annual consumption amount and whether it is more inclined to add up or commodity evaluation have a relevant impact.Among them, the monthly income can be explained as the higher the monthly income, the more likely the group is to believe in bad reviews.The higher the annual consumption amount, the more likely the group will believe in the evaluation content of the goods.

Results
My purpose in this study is to study the impact of negative evaluation on consumer behaviour.As predicted, negative evaluation is the focus of consumers.Consumers will reduce their purchase desire after seeing negative comments.However, there are too few male samples in the sample, so it is uncertain whether male samples can bring the same results.In addition, groups with higher monthly income tend to believe in poor reviews.In other words, high-income people have a stronger ability to perceive risks.Because high income buys life safety action (Daniel Kahneman and Angus Deaton, 2010) 15 .

Research Significance, Limitations and Future Directions
Many studies have shown that word of mouth communication affects consumers' attitudes towards various products and services (Klement Podnar, Pina Javernik, 2012). 16Consumers have a tendency to trust, and the probability of failure in online shopping is reduced.My findings show that negative comments will largely affect consumers' purchase desire.Therefore, many Taobao merchants have chosen to spend money to buy praise to attract more consumers.However, the data shows that 15.71% of consumers are still willing to believe the few bad comments.Therefore, the development direction of merchants should be changed from changing the direction of commodity evaluation to improving product quality.

Figure 1 .
Figure 1.Occurrence of the above events

Figure 2 .
Figure 2. When to write a review

Table 3 .
Frequency analysis result total of 70 final samples were selected as research objects in this survey.Among them, there are 16 people under 18 years old, accounting for 22.86%, 48 people between 19 and 25 years old, accounting for 68.57%, and 6 people between 26 and 40 years old, accounting for 8.57%.In the distribution of monthly income, there are 10 people with income of less than 3000 yuan, accounting for 14.29%, 22 people with income of 3001-5000 yuan, accounting for 31.43%,27 people with income of 5001-10000 yuan, accounting for 38.57%, 8 people with income of 10001-20000 yuan, accounting A for 11.43%, and 3 people with income of 20001 yuan and above, accounting for 4.29%.In the distribution of different genders, there are 2 males, accounting for 2.86%, and 68 females, accounting for 97.14%.In the distribution of educational background, there are 3 persons in senior high school and below, accounting for 4.29%, 63 persons in undergraduate/junior college, accounting for 90%, and 4 persons in graduate and above, accounting for 5.71%.The distribution summary of annual consumption amount, the highest proportion is below 5000 yuan, accounting for 65.71%, followed by 5001-10000 yuan, accounting for 28.57%.

Table 4 .
Results of t-test analysis by gender