Trust, Perceived Benefits, and Purchase Intention in C2C E-Commerce: An Empirical Testing in China

The discussion on trust has been involved in fields such as psychology, philosophy, sociology, economics, and organizational theory (Purwandari et al., 2022). The importance of trust in commodity trading is even more evident. Especially with the progress of e-commerce, both parties' anonymity and information asymmetry in the transaction have highlighted the importance of trust, which hasdramatically affected the transaction behavior (Antwi-Afari et al., 2022). In the current research, no unified definition of trust can cover all its meanings at different levels. This is because trust is an abstract and multidimensional concept used interchangeably with other concepts, such as trustworthiness and reliability in practical research. It also includes dimensions such as cognition, emotion, and behavior (Kim et al., 2020). However, the definition of trust mainly has two perspectives: cognitive perspective and behavioral perspective (Hou et al., 2021). From a cognitive perspective, trust is a belief, willingness, and attitude that primarily focuses on the trustworthy attributes of trading partners, such as goodwill, integrity, and ability, and is established based on these attributes (Sánchez et al., 2021). From the behavior perspective, trust is believed to be a behavior taken by the trusting party to satisfy consumer interests or even harm their interests in situations of risk and uncertainty (Lee, 2022). In the C2C environment, trust has different definitions. Figure 1 presents the specific content.

Figure 1.

Definition of trust in the C2C environment

In Figure 1, trust in the C2C environment can be divided into trade relations and trading environment. Specifically, online trust in a trade relationship refers to the relationship between consumers and sellers. Online trust in the context of a transaction refers to the relationship between the consumer and the transaction website.

In the C2C business environment, perceived benefits mainly refer to consumers' perception and evaluation of transaction-related factors, so it can be considered that perceived benefits are the reputation of merchants (Schwob et al., 2023). The specific content of perceived benefits includes five points, and Figure 2 shows the details.

Figure 2.

Specific content of perceived benefits

The five perceived benefits in Figure 2 include transaction security, product quality perception, perceived service quality, value perception, and social perception.

Perceived benefits are closely related to trust. Perceived benefits can lead to the generation of trust, but the generation of trust does not rely solely on perceived benefits (Gomes et al., 2020). In online transactions, the separation of the parties involved makes it more difficult to form trust (Zahara et al., 2021). It should be noted that different people have different perceived benefits for C2C transactions. In C2C transactions, merchants should continuously improve their services and quality to meet the needs of more consumers (Cerdan and Darcy, 2021). Consumer behavior research is widely valued in marketing, covering multiple fields such as sociology, psychology, advertising, marketing, and communication (Le et al., 2021). Understanding and studying consumers' psychological states and behavioral patterns during market transactions is significant for developing better marketing strategies (Yeap et al., 2022). Consumer purchase intention belongs to the category of consumer behavioral intention, and behavioral intention is an essential concept in attitude theory (Qin et al., 2021). The attitude theory believes that cognition, emotion, and intention constitute attitude. Cognition includes an individual's knowledge and beliefs, emotions refer to their feelings, and intentions refer to their willingness to act (Kusuma et al., 2020). Figure 3 is a theoretical model.

Figure 3.

Theoretical model of purchase intention

The main components of the purchase intention model in Figure 3 are behavioral beliefs, behavioral consequences evaluation, following motivation, behavioral attitudes, subjective norms, behavioral intentions, and practical actions.

In e-commerce transactions, when consumers cannot touch the product and have difficulty understanding relevant product information, the perceived benefit evaluation mechanism can help consumers identify and select sellers (Castillo-Sotomayor et al., 2023). Consumers will choose sellers with good reputations, believe that the seller has goodwill and integrity, can complete the transaction, form a purchasing intention, and finally take purchasing action (Qu et al., 2022).

Perceived benefits can be defined as the evaluation given by a buyer to the user's trading behavior in a C2C e-commerce transaction based on the user's past trading behavior and feedback from other sellers within a certain period. The evaluation mechanism of the transaction platform reputation is applied to classify the user's trading behavior based on information such as store level, dynamic rating, and seller reviews (Feng and Chen, 2022). It is divided into three dimensions: accumulated member credit, dynamic store score, and online comment (Mameri et al., 2020). Table 1 displays the specific meanings of each dimension.

This work also divides trust into three dimensions: goodwill trust, integrity trust, and ability trust (Trehan and Sharma, 2021). Table 2 displays the specific content.

The definition of purchase intention is the possibility for consumers to purchase goods in online stores using the internet as a carrier. The stronger the consumer's willingness to buy is, the greater the likelihood of purchase is (Zhang et al., 2022). Based on the above information, this work proposes the following hypotheses. Table 3 shows the specific content.

This work mainly explores the relationship between trust, perceived benefits, and purchase intention in the market. Thereby, the survey subjects are especially customers with some online shopping experience and consumers who have transacted on the Taobao trading platform (Ye and Chen, 2021). The main part of the questionnaire survey adopts a seven-level scoring method. The respondents rate the situation described in the questionnaire from “completely disagree” to “completely agree” according to their browsing experience of a store based on purchasing goods on Taobao (Ansari and Sanayei, 2020). Meanwhile, corresponding scores are assigned: 1 point, 2 points, 3 points, 4 points, 5 points, 6 points, and 7 points (Moriuchi and Takahashi, 2023). The questionnaire consists of 33 questions, including four parts. Table 4 presents the specific content.

SPSS has powerful statistical analysis capabilities and a wide range of applications. As a professional statistical software, SPSS provides a wealth of statistical methods and analysis tools, which can meet the needs of this study. With SPSS, statistical operations, such as reliability, validity, correlation, and regression analysis can be performed to verify the relationship between sample structure and variables to further reveal the influence mechanism between trust, perceived benefit, and purchase intention.After the data collection, this work mainly uses the statistical software SPSS25.0 as an analysis tool. Reliability, validity, correlation, and regression analyses are adopted to analyze sample structure and the relationships between variables (Kuhi et al., 2020).

400 questionnaires were distributed, and 376 questionnaires were collected. After removing 55 invalid questionnaires, 321 valid questionnaires are obtained, with an effective rate of 85.4%. The proportion of males and females participating in the questionnaire survey is balanced, with 49.8% of males and 50.2% of females. Young people aged 18-27 account for over 95% of the sample size. Cronbach’s α coefficient is selected as the reliability indicator for testing. When the coefficient is more significant than 0.7, it indicates that the reliability of the questionnaire is high. When the coefficient is between 0.6 and 0.7, it is also acceptable. Figure 4 displays the reliability of the questionnaire.

Figure 4.

Reliability analysis

In Figure 4, A represents accumulated member credit, B represents dynamic store scores, C represents online reviews, D represents goodwill trust, E represents integrity trust, F represents ability trust, and G represents purchase intention. Figure 4 shows that Cronbach’s α coefficients are all greater than 0.8 and can reach a maximum of 0.93, indicating that this questionnaire has high reliability and acceptable internal consistency. Validity analysis uses Kaiser Meyer Olkin (KMO) values and Bartlett’s test. It is generally believed that it is suitable for factor analysis when the KMO value is more significant than 0.7. When it is between 0.6 and 0.7, factor analysis can be performed. When the KMO value is below 0.6, it is unsuitable for factor analysis. Factor analysis can be achieved when the significance probability of Bartlett's test value is less than or equal to the significance level. Figure 5 displays the experimental results.

Figure 5.

Validity analysis

The analysis in Figure 5 shows that the KMO values of perceived benefit, trust, and purchase intention are 0.868, 0.852, and 0.749, respectively, all above the recommended threshold of 0.7. This indicates that the sample data have good usability and adaptability for factor analysis. In addition, by observing the value of the significance probability, it can be found that the significant probability of perceived benefit, trust, and purchase intention are all 0, which is less than the set significance level. This means that there is a significant correlation among perceived interest, trust, and purchase intention in the sample data with other variables, which is consistent with the study requirements. In summary, the data of perceived benefit, trust, and purchase intention have good availability and adaptability in factor analysis, and there is a significant correlation with other variables. This lays a foundation for further exploring the relationship among perceived benefit, trust, and purchase intention and provides reliable data support for subsequent regression analysis.

Correlation analysis is a standard statistical method for studying uncertain relationship between variables. The description is mainly based on the correlation coefficient, which is significant for analyzing the relationship between variables. Table 5 displays the correlation analysis results of trust, perceived benefits, and purchase intention.

The data in Table 5 shows that, first, the correlation coefficients among cumulated member credit, dynamic store rating, and online reviews and goodwill trust are 0.47, 0.31, and 0.64, respectively. The correlation coefficient is highest between online reviews and goodwill trust. The correlation coefficients among cumulated member credit, dynamic store rating, and online reviews and integrity trust are 0.3, 0.33, and 0.74, respectively. Online reviews have the highest correlation coefficient. The correlation coefficients with ability trust are 0.35, 0.29, and 0.46, respectively, with online reviews having the highest correlation coefficient.

Second, there is a significant positive correlation between perceived benefits and trust. Specifically, the correlation coefficients among goodwill trust, integrity trust, and ability trust and purchase intention are 0.63, 0.64, and 0.5, respectively. The correlation coefficient between integrity trust and purchase intention is the highest, followed by the correlation coefficient between goodwill trust and purchase intention. The last is the correlation coefficient between ability trust and purchase intention. These results validate the hypotheses H1a, H1b, H1c, H2a, H2b, H2c, H3a, H3b, and H3c. The correlation coefficients of cumulated member credit, dynamic store score, and online review and purchase intention are 0.33, 0.2, and 0.46, respectively. This reveals a positive correlation between perceived value and purchase intention.

A pairwise correlation between the variables involved is tested through correlation analysis. The direction of the relationship between variables is indicated through regression analysis, indicating whether there is a causal relationship between variables. Figure 6 presents the regression analysis results of perceived benefits and trust.

Figure 6.

Regression analysis results of perceived benefits and trust

In Figure 6, P&G refers to perceived benefits and goodwill trust. P&I refers to perceived benefits and integrity trust. P&A refers to perceived benefits and ability trust. Figure 6 shows, first, that at a significance level of 0.01, cumulated member credit and online reviews have a significant impact on goodwill trust. This indicates a positive correlation between member points, online reviews, and goodwill trust. Dynamic store ratings have no effect on goodwill trust, while online reviews have a higher impact on goodwill trust than cumulated member credit. The results of regression analysis verify the hypothesis H1a and H3a.

Second, at a significance level of 0.05, dynamic store ratings have a significant impact on integrity trust. The impact of online reviews on integrity trust is significant at a significance level of 0.01. The impact of online reviews on integrity trust is more significant than that of store dynamics, while accumulated member credit has no impact on integrity trust. The results of regression analysis verify the hypothesis H2b and H3b.

Third, at a significance level of 0.01, cumulated member credit and online reviews have a significant impact on ability trust. Online reviews have a greater impact on integrity trust than member credit, while dynamic store scores have no impact on ability trust. The results of regression analysis verify the hypothesis H1c and H3c.

Figure 7 shows the regression analysis results of purchase intention, perceived benefits, and trust.

Figure 7.

Regression analysis results of purchase intention, perceived benefits, and trust

Figure 7 reveals that all three trust variables enter the regression equation. That is, the explanatory variables in the regression equation include goodwill trust, integrity trust, and ability trust. The significance probability corresponding to the variable T value is 0, indicating a significant linear relationship between goodwill trust, integrity trust, ability trust, and customer purchase intention. The established regression equation is valid. Besides, the regression equation shows that integrity trust has the most significant impact on customer purchase intention, followed by goodwill trust, while ability trust has a more minor impact. The regression analysis results also validate the previous research hypotheses H4a, H4b, and H4c. In the regression analysis of purchase intention and perceived benefits, the significance probability corresponding to the variable T value is 0. It can be considered that there is a significant linear relationship between accumulated member credit, dynamic store score, online comments, and customer purchase intention. The established regression equation is valid. Moreover, the regression equation shows that online reviews have the most significant impact on customer purchase intention, followed by dynamic store reviews, while the accumulated member credit has a more minor impact. Hypotheses H5a, H5b, and H5c are validated.