Live Shopping Interactivity, Social Presence and Sustainable Consumer Purchase Intention: Based on TAM Model

2.1 Relationship between interactivity, perceived ease of use, perceived usefulness, social presence, and purchase intention

Interaction is very effective for the transmission of information, so that consumers do not have to spend too much effort to understand the information transmitted by the anchor, which improves the effectiveness of communication, reduces the communication barriers brought about by the different knowledge backgrounds of communication subjects, and makes it easier for consumers to understand and master live streaming shopping [4]. Cyr et al. [5] also found that for most consumers, the ability to get a better sense of interactivity is associated with higher ratings of features such as ease of use of the website. Islam et al. [6] found that high levels of interaction positively influenced consumer perceptions and that the virtual shopping environment was more useful, easier to use, and enjoyable. Therefore, the following hypothesis is proposed:

H1: Interactivity has a positive impact on perceived ease of use in live streaming shopping.

In the live streaming shopping scenario, anchors provide timely feedback and targeted answers based on users' questions, and users share product details and usage experiences through comments among themselves, and the effective information interaction among them promotes the sharing of product knowledge and information [7], reduces users' anxiety about product uncertainty, and enhances users' perception of the usefulness of live streaming shopping [8]. Zhou et al. [9] also found that effective interaction between users and anchors not only increases consumers' interest in shopping, but also satisfies consumers' sense of involvement, and the pleasant experience obtained through interaction helps to maintain a good relationship between anchors and consumers and form perceived usefulness of consumers. For most consumers, a better sense of interactivity will increase consumers' evaluation of the usefulness and ease of use of the website [10], which will also lead to consumer loyalty to the website. Therefore, the following hypothesis is proposed:

H2: Interactivity has a positive effect on perceived usefulness in live streaming shopping.

At the level of interpersonal interactions, social presence is described as a warm sense of socialization that the medium brings to the user [11], i.e., a psychological perception that the user interacts with others in the virtual space and perceives the presence of others. In the bidirectional interaction scenario of live streaming shopping, anchors can display products in multi-dimensional and multi-angle, and exchange product information with users to create a face-to-face shopping scenario, allowing users to experience the reality of being there, seeing is believing, which can gradually instill a sense of warmth and socialization between people and create a strong atmosphere of social presence [12], giving users a good immersive experience [13]. It can also create an "immersive" shopping situation and a unique "virtual touch" perception [9]. Ou et al. [14] also found that interactions promote social presence and to some extent influence consumer trust building and repeat transaction forming. Therefore, the following hypothesis is proposed:

H3: Interactivity has a positive impact on social presence in live streaming shopping.

The instant interaction through pop-up information during the live broadcast provides a channel for the anchor to understand users' consumption needs, and the anchor consciously adjusts the recommended products to meet their consumption needs according to the users' demands. Users can also send pop-ups to communicate with other users to ask questions or express their opinions about the product. The bidirectional information exchange between users and anchors and between users and users is not bound by time and space, which can arouse users' desire to participate [15]. Online shopping is a social process. During the interaction, users can obtain information, experience and advice from others to complete their shopping decisions, which stimulates consumers' purchase intentions and triggers their purchase behavior [16]. Therefore, the following hypothesis is proposed:

H4: Interactivity has a positive impact on purchase intention in live streaming shopping.

The shopping site itself is a form of information technology, and for consumers who visit online stores, their perception of the system determines their willingness to try out this new way of shopping. The cognitive factors of both perceived ease of use and perceived usefulness are derived from the technology acceptance model, where perceived ease of use influences perceived usefulness [17], and both perceived usefulness and perceived ease of use influence intention to use through attitudes in the technology acceptance model [18]. Convenient plays an important role in helping to achieve marketing goals, and Internet experience and Convenient are drivers of Internet use [19, 20]. Technology acceptance pioneers who maintain a positive attitude towards new technologies and services are more like pragmatists, people tend to focus more on ease of use [21] and users are more likely to shift their goals when the content and functionality are too complex [22]. Therefore, the following hypothesis is proposed:

H5: Perceived ease of use has a positive impact on the perceived usefulness of live streaming shopping.

H6: Perceived ease of use has a positive effect on purchase intention for live streaming shopping.

H7: Perceived ease of use mediates the effect of interactivity on purchase intention.

When customers visit a website, the information, pictures, videos, website helpers, online reviews or online customer service, etc., all trigger different levels of social presence. If consumers can experience a high-quality website presence and feel relaxed, easy, free or stress-free when browsing products, this will inevitably make consumers feel that the website is real and useful [11]. Consumers' perceptions of shopping websites are no longer limited to time and cost savings; their needs have increased to the level of social presence such as friendliness, sensitivity, friendliness, and sociability. The ability to make consumers perceive a shopping website not only as an emotionless computer system but also as a sentient and independent individual can greatly influence consumers' purchase intentions [23]. Fan et al. [24] point out that social presence makes users of live streaming rooms feel connected to others, satisfies their need for social interaction, and gives them a satisfying live streaming viewing experience, which in turn promotes their willingness to continue using it. Ou et al. [14] found that interactions promote social presence and to some extent influence consumer trust building and repeat transaction formation. Therefore, the following hypothesis is proposed:

H8: Social presence has a positive effect on the perceived usefulness of live streaming shopping.

H9: Social presence has a positive effect on purchase intention for live streaming shopping.

H10: Social presence mediates the effect of interactivity on purchase intention.

Live streaming shopping offers consumers a wealth of product information, competitive product prices, a large variety of products, and flexible shopping options. All features that increase the benefits or reduce the costs of shopping for consumers are relative advantages offered by online shopping, which can make consumers perceive the usefulness of live streaming shopping and thus increase their intention to shop online [25]. The perceived usefulness of live streaming shopping emphasizes the shopping experience of consumers in the shopping process and the results of accomplishing shopping goals, and the perceived Convenient of live streaming shopping places more emphasis on the shopping process of achieving shopping goals, i.e., the ease of use of live streaming shopping. The easier consumers perceive the operation of online shopping, the more likely they are to have the intention to shop [26]. Also, the stronger the consumers' perception of ease of use of online shopping, the more their perception of usefulness of online shopping increases [27]. Therefore, the following hypothesis is proposed:

H11: Perceived usefulness has a positive effect on purchase intention for live streaming shopping.

H12: Perceived usefulness mediates the effect of vividness on purchase intention.

2.2 Research model

Based on the above literature review and hypotheses, the following theoretical model was constructed in this study. As shown in the Figure 1 below, interactivity is the variable of external stimuli. The model is used to test the mediating effects of perceived usefulness, perceived ease of use, and social presence. In addition, the model identifies the relationship between perceived ease of use, perceived usefulness, and social presence.

1.png

Figure 1. Conceptual framework of this study

Table 1. Reliability analysis

Note: I is Interactivity, PEU is Perceived Ease of Use, PU is Perceived Usefulness, SP is Social Presence, and PI is Purchase Intention.

Table 2. Structural validity analysis

Note: I is Interactivity, PEU is Perceived Ease of Use, PU is Perceived Usefulness, SP is Social Presence, and PI is Purchase Intention.

Table 3. Model fit indices

In this study, the reliability and validity of the recovered samples were analyzed using SPSS24.0 software. Then, AMOS24.0 was used to construct a structural SEM to test the relationship between variables and verify the hypotheses. Finally, the reasons affecting the relationship between variables were revealed and the results were analyzed to find a scientific solution.

4.1 Reliability and validity

As shown in Table 1, the Cronbach's α coefficients under different dimensions were all above 0.8, indicating good dimensional reliability results. Among them, the correlations between the corrected terms of the question items and the total are all greater than 0.6, indicating that there is a correlation between the question items. Based on this analysis, it indicates that the reliability of the questionnaire is good.

As shown in Table 2, Different questionnaire items belonging to the same concept are clustered in the same common factor as predicted by the theory. The variance of the 6 common factors is all greater than 0.4, and the concepts are clearly delimited. Factor loads for all measured items were between 0.5 and 0.95, indicating that each measured variable could effectively reflect factor construction. To sum up, the questionnaire designed in this study has good structural validity.

As shown in Table 3, the Kaiser-Meyer Olkin measure is 0.929, which is already much greater than 0.7, and can prove that there is a strong correlation between the questions in the test; it can also be seen that the approximate chi-square value obtained by Bartlett's test is 2913.023, and the Bartlett's test significance sig<0.001, which rejects the original hypothesis. It indicates that it is suitable for factor extraction. In this study, factors with eigenvalues greater than 1 were extracted using Principal Component Analysis(PCA), 5 common factors were extracted, and the Cumulative rotated square sum was 72.627%, which was greater than 60%. After rotation by orthogonal rotational method, the 21 question options can be divided into 5 categories of factors, and the loading of each item is greater than 0.5, which indicates that each measured variable can effectively reflect the factor construction. Based on this analysis, it indicates that the questionnaire designed in this study has good structural validity.

As shown in Table 3, the ratio of chi-square values to degrees of freedom was 1.183, indicating a good model fit. GFI = 0.885, AGFI = 0.860, NFI = 0.888, TLI = 0.940, CFI = 0.946, and RMSEA = 0.051, indicating a good model fit. Based on this analysis, it indicates that all the indicators of the exploratory factor analysis in this study met the criteria and the overall model fit was good.

In order to further verify the rationality of dimensions, it is necessary to test the measurement model before evaluating and verifying the structural model in the SEM operation process, so as to ensure whether all items in the measurement model can accurately reflect the required factors. As shown in the Figure 2 below, All factor loads were greater than 0.6, and SMC was greater than 0.36

2.png

AVE is used to measure the convergent validity and discriminant validity of the latent variables, reflecting the changes in the latent variables due to errors and revealing the explanatory power of the questionnaire items on the variables [36]. The larger the AVE, the smaller the measurement error. As shown in Table 5, the AVE values of each dimension are greater than 0.5, and the square root of AVE is greater than the correlation coefficient between the dimensions, based on this analysis, it indicates that the scale has good discriminant validity.

Table 5. Discriminant validity analysis table

Note: I is Interactivity, PEU is Perceived Ease of Use, PU is Perceived Usefulness, SP is Social Presence, and PI is Purchase Intention.

4.2 Structural equation

3.png

The mediating effects are shown in Table 7, and Bootstrapping test was adopted to test all mediating paths in the study model. In the analysis 5000 samples were repeated and 95% confidence intervals were designed. If the p is higher than 0.05, 0 is not included between the upper and lower intervals of the test results, which means that the mediation effect is significant. In addition to the three mediator variable of perceived ease of use, perceived usefulness, and social presence that hold, the chain mediating effects of interactivity in turn through perceived ease of use, perceived usefulness to purchase intention and interactivity in turn through social presence, perceived usefulness to purchase intention are both significant, suggesting that interactivity indirectly affects purchase intention.