The effect of consumers' awareness of e-commerce firms' Corporate Social Responsibilities(CSR) activities on consumers' purchase intentions

Abstract

This study aims to empirically examine whether consumers' perceptions of Chinese e-commerce firms's CSR activities, along with psychological construct of company-consumer identification, and corporate trust affect their purchase intention based on theory of reasoned action. The study used a survey method for data collection to confirm research hypotheses with a total of 240 respondents used for final analysis. The results showed that economic responsibility, ethical responsibility, and legal responsibility have a positive effect on consumers' purchase intention. In addition, corporate trust and company-consumer identification were found to mediate the relationship between consumers' CSR perceptions and purchase intention. The result of this study is expected to provide useful theoretical as well as practical implications to advance the current understanding on the effects of consumers' CSR perception on business performance.

본 연구는 합리적 행동이론을 바탕으로 중국 전자상거래 CSR 활동에 대한 소비자의 인식과 기업-소비자 동일시, 기업의 신뢰가 구매의도에 영향을 미치는지 실증적으로 조사하는 것을 목적으로 한다. 본 연구에서 가설을 검증하기 위해 설문조사하였으며 최종 분석에는 총 240부 유휴설문지를 사용하여 통계분석을 실시하였다. 분석결과를 보면 전자상거래 CSR의 네 가지 차원에서 자선적 책임을 제외하고, 경제적 책임, 윤리적 책임, 법적 책임 순으로 구매의향에 긍정적인 영향을 미쳤다. 또한 기업 신뢰와 기업-소비자 동일시가 소비자의 CSR 인식과 구매의향의 관계를 매개하는 것으로 나타났다. 본 연구의 결과는 소비자의 CSR 인식이 경영성과에 미치는 영향에 대한 현재의 이해를 높이는 데 유용한 이론적 시사점과 실무적 시사점을 제공할 것으로 기대된다.

I. Introduction

With the development of society, social responsibility has evolved from a theoretical study to become a practical consideration for an e-commerce company. Many societal issues, such as pollution, the wealth gap, sweatshops, contamination of milk powder, melamine, waste oil, dangerous capsules, clenbuterol, and others, cannot be tackled simply by the government, scholars, and the general public, and must be addressed by corporations. In this era of major social problems and greater public awareness of self-rights protection, the question of whether e-commerce firms are willing to assume social responsibilities is no longer relevant. In countries around the world, the discussion of CSR has evolved from a corporate perspective to a consumer perspective. At present, the focus of the study is on CSR's impact on consumers' evaluation and purchase intentions. The usual research shows that consumers' purchase intentions generally do not correlate with the behavior of firms that do not take social responsibility[1], but they do correlate with the behavior of firms that do take social responsibility[2], and CSR can enable consumers to support the company's products to a certain extent independent of price. Although there is general evidence that CSR will drive consumers' willingness to buy, there are also numerous studies that show that the relationship between the two is complex[3].

Furthermore, relevant research at home and abroad has found that CSR does have a significant effect on consumers' willingness to purchase. However, little research has been conducted on how and why CSR affects consumers' purchase intention and how different aspects of CSR influence purchase decisions. Based on this, this paper attempts to integrate the theory of rational behavior from previous research, and to use corporate trust and company-consumer identification to connect CSR behavior with purchase intentions[4]. This study investigates the relationship between CSR and consumer propensity to buy from the consumer's perspective, and provides guidelines and suggestions for boosting CSR's advantages.

II. Theoretical Background

2.1 CSR perception

CSR refers to whether a company's primary objective is to maximize profits or whether it also has other responsibilities that it is required to fulfill.

CSR has been debated in theory since the concept was first put forward, and scholars have finally concluded that CSR is not just an economic goal of profit maximization. However, the content of CSR is still not unified. The American Economic Development Council defines CSR as involving "three central circles"[5]. Srethi summarized CSR as a three-dimensional model of social restraint, social responsibility, and social response[6].

According to Carroll, CSR is the expectation by society of economic responsibility, legal responsibility, ethical responsibility, and charitable responsibility on the part of enterprises in a certain period of time. Isabelle & David conducted research on CSR in European and American countries, and derived the concept of CSR by regard to five stake holders[7]. Thompson & Zakaria describe CSR as environmental protection responsibility, energy utilization responsibility, product and consumer responsibility, community responsibility, and employee responsibility after researching CSR announcements in Malaysia. Qu defines CSR as a positive obligation an organization has to its employees, consumers, partners, communities, and nation, as well as the response it gives to relevant interest groups[8]. Chen and Han categorized social responsibility into three levels: primary, intermediate and advanced social responsibility[9]. Xu and Yang classified CSR into nine dimensions: economic responsibility, legal responsibility, environmental responsibility, customer responsibility, people-oriented responsibility, public welfare responsibility, employment responsibility, and social stability and progress responsibility[10]. Zhang & Li defined CSR as the responsibilities firms have to stake holders in order to maximize shareholder value and create profits. The purpose of this study is to define CSR as a responsibility that extends beyond economic profit-seeking, meets the requirements of good citizenship, and contributes to society as a whole, including economic, legal, ethical, and charitable responsibilities[11].

2.2 Theory of Reasoned Action (TRA)

Theory of Reasoned Action (TRA) is a model proposed by Fishbein and Ajzen in 1975 to predict the relationship between human behavior and behavior. As a theory emerging from social psychology, it has since been applied to a wide range of fields. As a consumer research method, TRA theory is widely employed and confirmed. Based on the TRA theory, behavior is determined by willingness; behavior is determined by behavioral attitudes and subjective norms. A behavioral attitude consists of two components: the individual's belief in the outcome of the behavior and whether the individual believes that it will be beneficial. There are two components of subjective norm: the individual's opinion as to whether or not the reference object should engage in a certain behavior and the individual's tendency to obey the reference object[12]. This study will discuss the influence of CSR through corporate trust (behavioral attitude) and company-consumer identification (subjective norm) on purchase intentions, applying TRA theory to consumer behavior research.

2.3 corporate trust

Trust is the honest belief that a partner will keep his or her promises and meet the needs of the other party. Alternatively, we believe that the enterprise will be oriented towards long-term interests, it is defined as trust in an exchange partner and confidence in their integrity, There are three components of trust according to many researchers: ability, integrity, and benevolence. Ability refers to skills, abilities, and characteristics specific to a field, and sincerity refers to belief that the object is genuine, reliable information about it, and correspondence between words and actions. It is a desire to protect the interests of the other party, regardless of the outcome. From these three elements, we can see that CSR activities are closely related to ability, perceived quality, honesty, and good deeds.

2.4 company-consumer identification

Wegietl and Camerer believed that company-consumer identification is an effect of the company's past behavior, which was attributed to a variety of economic and non-economic attributes[13]. Fomburn defined company-consumer identification as "the attractiveness to all stake holders of a company based on its perception of past performances and future prospects in comparison with other competitors", which is the perception of the company by its stake holders in the form of net opinions expressed by its customers, investors, employees, and the public in general. Gosi & Wilson described an company-consumer identification as an evaluation of it s past made by stake holders based on their direct experience and other forms of communication and symbols containing information about the company[14]. Based on an extensive review of literatures, Walker defines company-consumer identification as a relatively stable comprehensive perception of an company's past behavior and future prospects by comparing a set of criteria. Gan asserted that company-consumer identification is a kind of social trust and praise the enterprise accumulates over time, including popularity, reputation, and trust. In conclusion, company-consumer identification is largely derived from the characteristics of stakeholder groups, taking their expectations as the standard, the past behavior of the firm, as well as various communication information[15]. As a result, this study defines company-consumer identification as a stakeholder's impression of a firm based on historical behavior and numerous communication and information symbols that may be contrasted to their own expectations. a detailed evaluation[16].

III. Model and hypotheses

CSR can have a good impact on ethical corporate trust and corporate consumption because corporate trust is mostly defined by consumer behavior and consumer identification for a firm is more subjective[16]. In addition, the impact of CSR on purchase intentions has been shown to be moderated by the influence of company-consumer identification. As a result, based on the TRA theory, this study develops a theoretical model of the relationship between CSR and corporate trust (ECT), company-consumer identification(BCT), and purchase intentions (EPI), as illustrated in Figure 1.

Fig. 1. research model

The following research hypotheses are proposed:

H1: CSR is positively correlated to corporate trust (ET).

H2: CSR is positively correlated to company-consumer identification(BCT).

H3: Corporate trust(ET) is positively correlated to consumers’ purchase intentions(EPI).

H4: Company-consumer identification(BCT) is positively correlated to consumers’ purchase intentions(EPI).

H5: Corporate trust(ET) mediates the influence of CSR on consumers' purchase intentions (EPI).

H6: Company-consumer identification(BCT) mediates the influence of CSR on purchase intentions(EPI).

IV. Research method and analysis

4.1 Research Scale and Data Collection

In order to ensure the validity and reliability of the measurements, this paper has followed the following procedure: The measurement indicators used for the research scales were obtained from the existing literature in the first stage; after the second stage, the first stage indicator scales are discussed and analyzed, and in the third stage, the original indicators are revised in light of the analysis results. These measurement indicators of CSR are drawn from Carroll's research, ethical corporate trust are drawn from Zauberman. et. al’s research, company-consumer identification are drawn from Ashforth & Meal, Bhattachary and Sen’s research, consumers’ purchase intention are drawn from Langer’s research. All 25 measures were assessed on a 5-point Likert scale.

The purpose of this paper is to survey the auto products sold by e-commerce firms since the auto industry is closely related to CSR, and there are many auto consumers, and consumers will consider price, fuel consumption, safety, and after-sales service when purchasing a car, brands, etc. and their brands are highly involved. Generally, online questionnaires are used to collect data. Online questionnaires have the advantage of being low cost, quick to collect, non-manual, and have a wide scope of application. Also, to ensure that the data gathered is as representative as possible, this survey includes all social groups and reflects the real feelings of consumers as much as possible. In total, 300 questionnaires were distributed, and 263 were recovered; after screening, 240 valid questionnaires were formed at an effective rate of 80%.

4.2 Reliability Analysis & Validity Analysis

In this paper, we utilize the most commonly used Cronbach’s á coefficient test scale in order to evaluate its reliability. The Cronbach’s á coefficients for each scale are as follows: According to the independent variable 'CSR', the four dimensions-economic responsibility, legal responsibility, ethical responsibility, and charitable responsibility were respectively 0.856, 0.897, 0.888, and 0.881, which indicate that the internal consistency of the data was high. Cronbach’s á measures corporate trust at 0.834, and company-consumer identification at 0.812, both greater than 0.8. The Cronbach’s á value of the purchase intention of the dependent variable is 0.882. This value is greater than 0.8. This indicates that the scale has a high level of internal consistency.

In this paper, factor analysis is used to verify the validity of the scale. Before performing factor analysis, perform a suitability test on the data. According to the test, the KMO value for CSR is 0.871, which is greater than 0.8, and thus is more suitable for factor analysis. The Bartlett sphere test indicates that the significance probability Sig of the chi-square statistic is 0.000, which is less than 0.01, indicating that the data could be analyzed using factor analysis. Based on a factor analysis of the items comprising the CSR scale, four common factors with eigenvalues greater than 1 can be identified. Moreover, the items that are part of the same factor variable are grouped into the same dimension, and the factor loading of each item in the same dimension is greater than 0.5. Therefore, CSR is categorized into four categories: economic responsibility, legal responsibility, ethical responsibility, and charitable responsibility. This dimension is reasonable and valid. According to the validity test of the corporate trust, company-consumer identification, and purchase intention variables, the KMO value for each variable is 0.809, which is greater than 0.5, and the significance probability Sig is 0.000, which is less than 0.01. It is suitable for factor analysis. Due to the fact that the three variables of corporate trust, company-consumer identification and purchase intention are all uni-dimensional variables, the eigenvalues of each variable are one, and they are their common factors. As a result of factor analysis, the public factor of corporate trust explains 65.38% of the variance, the public factor of company-consumer identification accounts for 85.79% of the variance, and the public factor of purchase intention explains 86.42% of the variance, indicating that each variable has good construct validity.

4.3 Confirmatory factor analysis

The purpose of this study was mostly to validate the row factor analysis, perform statistical analysis using AMOS 25.0, and test the construction angle. According to the research results, the model fit is x2=284.477 (df=231, x2/df=1.232*3, p=0.009>0.05), which means that the hypothesis that the data does not fit well to the model cannot be rejected. RMSEA=0.0180.9 CFI=0.994>0.9, NFI=0.971>0.9, TLI=0.993>0.9. these indicators demonstrate that the research data and confirmatory factor analysis model fit well. Overall, through the above tests, it can be concluded that the measurement scale used in this study is highly reliable and valid.

4.4 Hypothesis testing

aumgartner and Homburg demonstrated that data analysis can be conducted using structural equation models only when the sample size exceeds five times the estimated parameters. This model has 25 estimated parameters, and this study collects 240 valid observations, which is larger than the recommended sample size. As a result, the data collected in this study meet the requirements for structural equation modeling.

This study is based on relatively mature scales and scientific research achievements at home and abroad, so a confirmatory factor analysis of this study is necessary. Using AMOS 25.0 software, the modified structural equation model fitting index is shown in the table. The calculated values for all seven indicators are within the fitting range, indicating that the model has a good fit with the sample data. The model can explain consumers' purchase intentions effectively.

Table 1. Model Check Criteria Values

Use AMOS 25.0 software to analyze the structural equation model, verify the proposed research hypothesis, and utilize the maximum likelihood estimation procedure (ML) to incorporate the sample data into the model to calculate the fitting index, and to calculate the normalized regression coefficients and their significance. The parameter estimation results of its structural equation model are displayed in the Fig 2.

Fig. 2. Structural model analysis results

According to the path test results, each path passed the test, and it has a significant impact. The AMOS 25.0 software is useful in analyzing the significance of each path.

4.5 Estimated result.

1)Economic responsibility(ER), legal responsibility(LL), and ethical responsibility(EER) are positively correlated with corporate trust(ET), and the regression coefficients of β= 0.331, 0.215, and 0.261 respectively (among which the impact of ethical responsibility(EER) and economic responsibility(ER) on the corporate trust(ET) is relatively greater and sig <0.001, the relationship is significant. There is, however, no significant relationship between charitable responsibility(CR) and corporate trust(ET). Parts of the Hypothesis1 are valid.

2) Among the three dimensions of CSR: economic responsibility(ER), ethical responsibility(EER), and charitable responsibility(CR), the regression coefficients of β=0.146, 0.17, and 0.199, respectively (where the impact of charitable responsibility on company-consumer identification(BCT) is relatively greater), and Sig <0.001, showing the relationship is significant; however, there is no significant relationship between legal liability(LL) and company-consumer identification(BCT). Parts of the Hypothesis 2 are valid.

3)Purchase intention(EPI) is significantly correlated with corporate trust(ET) based on a regression coefficient of β=0.447 and sig <0.001, the relationship is significant, and Hypothesis 3 is valid.

4)Purchase intention(EPI) is significantly correlated with company-consumer identification(BCT) based on a regression coefficient of β=0.415 and sig<0.001, the relationship is significant, and Hypothesis 4 is valid.

Table 2. Structural Equation-AMOS Model Path Analysis Results

V. Conclusions and Suggestions

While the research context of this paper in a single industry limits its generalisability, it still has important managerial implications. (1)Purchase intentions are positively influenced by CSR. As a result, businesses should create a CSR behavior strategy, engage in CSR actively, and report CSR data in a timely and thorough manner. (2)The beneficial benefits of various dimensions of CSR on prospective purchase intentions vary, with economic responsibility, ethical responsibility, and legal responsibility having the most influence. Companies must concentrate their efforts and sensibly allocate resources in order to obtain the best CSR results. (3)Corporate trust and company-consumer identification have a mediated effect on the implications of CSR on purchase decisions. In order to maximize the positive impact of CSR on corporate trust and consumer trust, companies should, on the one hand, maximize the material benefit of CSR on corporate trust. For this reason, there should be a strong focus on maximizing the positive impact of CSR on corporate trust and the company-consumer identification, moreover, it should maximize consumers' perception scores of product quality and company-consumer identification by implementing, for example, programs to improve product quality and company-consumer identification, as a way of enhancing the market output effect of CSR policy.