The Effects of Emotional and Interaction Factors on Consumer Behavior in Virtual CSR Co-Creation: The Mediating Role of Social Presence

Abstract

In the context of the global environment, the rapid advancement of Internet technology has facilitated a transition from traditional offline to online modes for corporate social responsibility (CSR) activities, leading to the emergence of virtual CSR co-creation initiatives. A crucial challenge for enterprises engaging in such activities lies in enhancing consumer participation and retention rates. This study constructs a mediation effect model incorporating emotional factors, interaction factors, social presence, and consumer willingness to participate, grounded in the Stimulus–Organism–Response (S-O-R) model. Based on data collected from 232 consumers in China, this study reveals that identification, enthusiasm, and satisfaction exert positive influences on consumer willingness to participate. Additionally, social presence positively affects willingness to participate and acts as a mediator between community platform interaction, offline interaction, and willingness to participate.

1. Introduction

Climate change and environmental degradation are exacerbating extreme weather events and placing significant pressure on enterprises to reduce carbon emissions and pursue sustainable development. Concurrently, persistent social issues, such as wealth inequality, educational disparities, and unequal employment opportunities, highlight the need for businesses to foster social equity and inclusiveness through CSR co-creation activities [1]. The rapid development of digital technologies, such as the Internet, social media, big data, and artificial intelligence, has provided new platforms and means for the co-creation of CSR, facilitating a transition in its activities, shifting from conventional offline formats to online platforms. This shift has led to the emergence of virtual CSR co-creation initiatives, exemplified by Alipay’s “Ant Forest” and “Ant Manor”, as well as WeChat’s “Donate Steps” and other similar endeavors. These innovative and consumer-friendly activities offer participants a novel and thrilling experience while allowing them to engage in cost-free and accessible public welfare activities, thereby attracting a substantial number of consumers. Virtual CSR co-creation refers to the activities in which companies strategically plan and encourage valuable consumer participation through Internet platforms and social media technologies [2,3]. The unique attributes of virtual CSR co-creation lie in its virtuality, interactivity, ecological focus, and emotional orientation. It also differs from other forms of online consumer engagement, such as social media interactions or online reviews, by focusing specifically on CSR activities and fostering a sense of collective responsibility and purpose among participants. The process leverages the collective intelligence and resources of consumers by providing a more accessible, scalable, and engaging way for individuals to contribute to societal welfare. As co-creation becomes increasingly prevalent in virtual CSR initiatives, companies are confronting a multitude of challenges. First, consumers have a limited understanding of virtual co-creation, resulting in a low willingness to participate. Second, consumer trust in companies is relatively low, particularly during the virtual co-creation process, where insufficient information transparency further diminishes consumer engagement. Additionally, companies lack effective interaction mechanisms, which fail to fully stimulate consumer enthusiasm. These challenges not only affect the effectiveness of virtual co-creation but also hinder the realization of CSR. And, while offline CSR initiatives offer face-to-face interactions and tangible rewards, online platforms offer greater accessibility and scalability but often struggle with user engagement and retention. This necessitates a deeper understanding of the factors that motivate consumers to participate in virtual CSR co-creation activities. Emotional factors, such as identification, enthusiasm, and satisfaction, play a crucial role in consumer willingness to participate in such activities. Online environments lack the immediate social cues and physical rewards present in offline settings, making emotional connections and satisfaction with the activity itself even more important. Identification with the CSR values and goals can foster a sense of belonging and purpose, while enthusiasm and satisfaction can mitigate the potential for boredom or disengagement that may arise in virtual settings. Furthermore, interaction factors, including community platform interaction and offline interaction, also influence consumer willingness to participate.

In recent years, as environmental issues have gained significant attention, firms across various industries and countries are expected to implement green initiatives [4]. By leveraging virtual communities to engage consumers in co-creation activities, enterprises not only foster a positive brand image and enhance their reputation but also draw attention to pressing social issues [5]. Prior research highlights that multiple stakeholder pressures can serve as catalysts, motivating firms to adopt greener production practices [6,7,8]. The existing research on CSR co-creation predominantly focuses on specific industries and regions, limiting its generalizability. For example, Cai and Zhou (2014) examined green innovation in Chinese enterprises, with limited applicability to other economies [7]. Additionally, most studies concentrate on manufacturing, neglecting green practices in other sectors like services. A key limitation is the lack of attention to how emotional and interaction factors affect consumer participation in CSR activities, especially given the growing reliance on virtual platforms. Although social presence in online environments is acknowledged [9], its mediating role between the emotional/interaction factors and consumer willingness to participate in virtual CSR co-creation remains underexplored.

Recognizing the limitations of solely focusing on internal resources, many firms have turned their attention to open innovation [10,11,12,13,14]. For instance, Alipay’s Ant Forest initiative urges individuals to care for the environment, reduce carbon emissions, and preserve natural landscapes. Keep’s public welfare running activities realize the co-creation of social value through the combination of sports and public welfare. For example, the “Dream Stadium” project provides children in remote areas with sports facilities by exchanging consumers’ running miles for charity materials. This model not only enhances consumers’ participation in public welfare but also conveys positive social values through sports. In today’s business landscape, sustainability is an area of growing interest for most organizations [15,16]. Several studies indicate that environmental performance correlates with higher financial performance [17,18]. These activities provide participants with a novel and thrilling experience, allowing them to engage in cost-free and accessible public welfare efforts, thus attracting substantial consumer participation.

Consumer participation in virtual CSR co-creation activities inherently represents a collaborative process of social value creation. According to the theory of value co-creation, “consumer participation” denotes a cooperative effort between consumers and enterprises, highlighting consumers’ pivotal role as the primary agents driving this process. Consumer participation can be categorized into narrow and broad dimensions: narrow participation refers exclusively to tangible actions, while broad participation encompasses emotional involvement as well [19]. Gamified reward mechanisms (e.g., behavior-based vs. result-based incentives) align with risk-sharing and transparency principles. Building on Jun et al. (2020), we argue that experiential interaction paths (e.g., social storytelling, real-time feedback) fulfill consumers’ needs for autonomy and competence, driving sustained co-creation [20]. In addition, drawing from Infovision iCommunity’s case studies, virtual CSR platforms act as “digital commons”, where global and local values intersect (e.g., self-transcendence vs. self-enhancement cultural norms). Cognitive appraisal theory emphasizes that an individual’s cognitive appraisal of a situation influences his or her affective responses and behavioral decisions. In virtual CSR co-creation, emotional factors (e.g., vitality, sense of belonging) may influence consumers’ participation behavior by affecting their cognitive appraisal. Consumers’ willingness to participate fluctuates with their emotional responses throughout the participation process. In real-world contexts, consumer satisfaction with the content and outcomes significantly influences their willingness to participate. Additionally, their identification with the activity’s content and value, coupled with their enthusiasm for dedicating time, also shape their participation. The heightened environmental consciousness among consumers further motivates enterprises to reduce their negative environmental impacts [21,22]. Furthermore, the intricate themes that involve enterprises, consumers, and other stakeholders impart a social aspect to these activities. Stakeholders are recognized as individuals or groups that can influence the decisions, practices, or objectives of the organization [23,24]. It is particularly crucial to systematically communicate CSR outcomes to external stakeholders through corporate reports, as they lack access to private information from firms [24]. Depending on consumers’ interactions with others during these activities, they can be divided into two categories—community platform interaction and offline interaction—both of which impact the consumer willingness to participate and the subsequent behaviors.

To address challenges, our research focuses on understanding the factors that influence consumer willingness to participate in virtual CSR co-creation activities. Specifically, we propose that emotional factors (identification, enthusiasm, and satisfaction) and interaction factors (community platform interaction and offline interaction) play a crucial role in determining consumer willingness to participate. This connection is grounded in the Stimulus–Organism–Response (S-O-R) model, which posits that external stimuli affect consumer behavior by influencing their psychology. In the context of virtual CSR co-creation activities, emotional factors, such as identification, enthusiasm, and satisfaction, are critical as they shape consumers’ emotional responses and attitudes towards the activities. Identification, enthusiasm, and satisfaction reflect affective evaluations rooted in Lazarus’s (1991) transactional theory of stress and coping. Identification fosters a sense of belonging [25], enthusiasm activates intrinsic motivation [26], and satisfaction aligns with goal attainment theory [27]. When consumers feel a sense of unity and belonging (identification), enjoy participating (enthusiasm), and are satisfied with the outcomes (satisfaction), they are more likely to persist in their engagement. Similarly, interaction factors, both within the virtual community platform and offline, foster a sense of mutuality and camaraderie, enhancing consumer willingness to participate. According to the S-O-R model, external stimuli exert an impact on consumer behavior by affecting consumer psychology. We conceptualize enterprise-led virtual CSR co-creation activities as the primary stimulus. Such stimuli function as environmental cues that initiate cognitive and affective processing, akin to algorithmic nudges in social media [28]. Notably, enterprise-led virtual CSR co-creation activities primarily occur online, devoid of face-to-face interactions among consumers and between consumers and enterprises [29]. Social presence, in this context, pertains to the capacity of consumers to convey a sense of personal existence and social warmth through virtual communities and online messaging platforms. As posited by Swan and Shih (2005), social presence embodies the ability of consumers to present their “true selves” on online social media platforms during participation, thereby influencing reciprocal behaviors [9]. Consequently, this study posits that social presence mediates the influence of emotional and interactive factors on consumer willingness to participate. This mediation is grounded in the dual-process model [30], where social presence amplifies affective salience and reduces transactional uncertainty.

2. Research Questions

This research explores CSR’s complex dynamics and strategic adaptability in the “virtual village” era, focusing on three core questions:

• RQ1: How do emotional factors (identification, enthusiasm, and satisfaction) and interaction factors (community platform engagement and offline interaction) shape consumers’ willingness to participate in virtual CSR co-creation?
• RQ2: To what extent does social presence mediate the pathways from emotional and interaction stimuli to participation willingness in algorithm-driven virtual environments?
• RQ3: How do platform affordances (e.g., gamification, real-time collaboration tools) moderate the relationship between interaction design and social presence in sustaining engagement?

3. Literature Review and Hypotheses Development

3.1. Emotional Factors and Consumer Willingness to Participate

Identification pertains to the sense of unity and belonging that consumers experience towards virtual CSR co-creation activities. Analogous to how a stronger identification with one’s profession propels personal career advancement, research indicates that individuals possessing heightened senses of belonging and identity are more prone to cultivating positive attributes, such as self-confidence and optimism, within an organization [31]. Enthusiasm denotes the degree of enjoyment and interest consumers have in virtual CSR activities. A higher level of enthusiasm translates into a greater willingness to invest time and effort into these activities. Furthermore, enthusiasm, being a positive emotion, mitigates consumer fatigue during participation in virtual CSR co-creation endeavors and stimulates an inherent positive psychological state [32]. Gamification research shows that real-time feedback loops (e.g., impact badges) trigger dopamine-driven engagement, which is validated in CSR apps [33,34]. Satisfaction represents the alignment between consumers’ post-participation feelings and their expectations concerning virtual CSR co-creation activities. Studies have shown that experiential elements significantly impact consumer participation in virtual CSR activities. When consumers are satisfied with such activities, they develop a profound emotional connection, triggering positive and upbeat emotions that encourage further engagement. Satisfaction and enthusiasm are recognized CSR drivers; their conflation obscures mechanistic insights: satisfaction reflects ex-post outcome appraisal, peaking in structured tasks, whereas enthusiasm captures ex-ante intrinsic motivation, thriving in open-ended co-creation. Social presence uniquely bridges this duality by simulating communal embodiment, a gap unaddressed in transaction-centric models [35]. Given these premises, the following hypotheses are formulated:

H1a. 

The consumer identification of virtual CSR co-creation activities positively influences the consumer willingness to participate.

H1b. 

Consumer enthusiasm for virtual CSR co-creation activities positively influences the consumer willingness to participate.

H1c. 

Consumer satisfaction with virtual CSR co-creation activities positively influences the consumer willingness to participate.

3.2. Interaction Factors and Consumer Willingness to Participate

Interactions within and outside the platform can foster a sense of belonging and trust, which in turn enhance participation [36,37]. Furthermore, social exchange theory posits that individuals are more likely to engage in activities that provide them with social benefits, including the satisfaction derived from interactions [38]. Interaction factors encompass both the exchanges that consumers have with others participating in the same virtual CSR co-creation activities through the virtual community and their communications with familiar individuals in their everyday lives. In an environment conducive to easy communication, consumers are more inclined to share information and forge long-term relationships, which in turn fosters the development of a positive psychological state [36]. Users feel empowered when they have the autonomy to make choices, the ability to influence outcomes, and access to transparent information. Hedonic adaptation theory [39] considers adapting to digital contexts, where a dynamic CSR task variety prevents satiation [40]. Several studies identified fundamental issues surrounding sustainability communications, such as consumers’ limited awareness of firms’ sustainability practices and their underlying attitudes toward CSR communications [41,42]. Engaging in such interactions cultivates a sense of mutuality and camaraderie within the community, creating a favorable participation experience that bolsters consumers’ sense of well-being and subsequently heightens their willingness to engage. Social interaction theory emphasizes that consumer interactions within virtual communities can enhance their sense of belonging and identity within the community. This interaction includes not only the exchange of information but also emotional support and cultural identity, thereby promoting an increase in consumer willingness to participate [43]. The experiential value co-creation theory posits that offline interactions have the capacity to augment consumers’ perceived value of a brand by furnishing them with direct experiences of products and services. This experience has been shown to enhance consumers’ understanding of the product, as well as strengthen their brand loyalty through emotional interaction [44]. In view of this, the following hypotheses are formulated:

H2a. 

Community platform interaction positively influences consumer willingness to participate.

H2b. 

Offline interaction positively influences consumer willingness to participate.

3.3. The Mediating Role of Social Presence

Compared with traditional CSR activities, virtual CSR co-creation distinguishes itself by engaging consumers not merely as participants, but as active contributors responsible for design, participation, and improvement. Consumers devote substantial time and energy within the virtual community, propelling the activities forward. Within this digital realm, interactions occur seamlessly, whether with acquaintances or strangers. The frequent exchanges foster deeper connections among consumers, fulfilling their need for immediate feedback, nurturing intimate relationships, and ultimately cultivating a sense of social presence. Participation in the task facilitates more frequent and meaningful interactions, thereby enhancing the sense of social presence. Bridge disciplines integrate communication theory (social presence) with behavioral psychology (self-efficacy), addressing calls for interdisciplinary VR/CSR research [45]. Virtual CSR activities have been shown to break the time limit of interactions, and frequent interactions provide individuals with more opportunities to interact with others, thus enhancing their sense of social presence. The concept of physical distance versus visual contact holds particular relevance in the realm of virtual CSR activities. Despite the absence of actual physical distance between participants, the visual design of the interactive platform can effectively simulate a sense of “proximity”. When consumers perceive that their actions matter and see tangible results from their efforts, they are more likely to continue participating. By carefully crafting the visual elements of these virtual environments, a feeling of closeness and engagement can be fostered among participants, enhancing their overall experience in virtual CSR initiatives. For instance, certain virtual activities make participants feel as if they are in the same space, through high-definition videos, real-time interactions, and other features, thus enhancing their sense of social presence. Integrated with Dual-Process Models, Evans (2008) said that social presence bridges heuristic processing (e.g., quick trust in interactions) and systematic processing (e.g., deliberative evaluation of CSR outcomes) [30]. Consequently, it is imperative for companies to select the most appropriate medium for disclosing and communicating their CSR initiatives [46]. In recent years, social media has ascended as a potent and indispensable marketing platform, demonstrating a significant correlation with sustainable consumption practices [47]. This sense of social presence imbues consumers with a feeling of direct communication and the warmth of social interaction, thereby fostering a sustained desire to participate [48]. It is evident that consumers who possess a pronounced sense of social presence are likely to experience a profound sense of interconnectedness with the other participants in the event, as well as with the company itself. This heightened sense of identification with the event is a significant predictor of active participation and the contribution of ideas and efforts towards the realization of its objectives. In light of these observations, the following hypotheses are formulated:

H3. 

Social presence positively influences consumer willingness to participate.

H4a. 

Social presence mediates the relationship between identification and consumer willingness to participate.

H4b. 

Social presence mediates the relationship between enthusiasm and consumer willingness to participate.

H4c. 

Social presence mediates the relationship between satisfaction and consumer willingness to participate.

H4d. 

Social presence mediates the relationship between community platform interaction and consumer willingness to participate.

H4e. 

Social presence mediates the relationship between offline interaction and consumer willingness to participate.

This work aims to fill this research gap by constructing a model that incorporates emotional factors (identification, enthusiasm, and satisfaction), interaction factors (community platform interaction and offline interaction), social presence, and consumer willingness to participate, operationalizing “willingness” through validated scales (e.g., OCAS by Zhang et al., 2020) to capture both the intent and persistence [49]. By focusing on the consumer perspective and the mediating role of social presence, we provide a more nuanced understanding of the factors that influence consumer participation in virtual CSR co-creation activities. Based on the theory of value co-creation, cognitive appraisal, and the S-O-R model, this study investigates the effects of emotional factors and interaction factors on consumer willingness to participate, with social presence as the mediator.

Table 1. Correspondence between theories and S-O-R model.

Theory	Application in the S-O-R Model
Social Identity Theory	Explaining how consumers form identification with CSR activities through group belonging and comparison processes, influencing their participation willingness.
Goal-Setting Theory	Analyzing how CSR activities stimulate consumers’ participation motivation by setting clear goals and providing feedback.
Gamification Principles	Exploring how CSR activities enhance consumers’ interactive experience and sense of participation through gamified designs (e.g., points, rewards).

presents a theoretical clarification.

Identification, framed as a self-concept alignment stimulus through algorithmic identity cues (e.g., personalized impact dashboards), is supported by recent work on identity-based engagement in CSR apps [50]. Enthusiasm/Satisfaction were positioned as hedonic stimuli based on gamification design principles [33].

The conceptual model is shown in Figure 1.

4. Research Methodology

4.1. Questionnaire Design

Based on the conceptual model, this work undertook a thorough analysis of the scales employed in prior research and developed a questionnaire to assess consumers’ intention to persistently engage in virtual CSR co-creation activities. The questionnaire encompasses seven variables—identification, enthusiasm, satisfaction, community platform interaction, offline interaction, social presence, and willingness to participate—totaling 26 items. To reduce Common Method Bias (CMB), clear instructions are provided at the beginning of the questionnaire explaining the purpose of the survey and how it is to be completed to help respondents understand the questions correctly.

The questionnaire is structured into three sections: Firstly, it elucidates the background and subject matter, stating that the questionnaire is filled out anonymously, promising not to disclose information about the content. During the data collection and analysis process, the participants’ personal information is anonymized to ensure that individuals cannot be identified through the data. Secondly, it collects the demographic information, including age, gender, monthly income, and educational background, of the respondents. Finally, it presents the measurement items for the relevant variables. The variable measurements in this study are grounded in well-established scales from both domestic and international research, and each item is evaluated using a five-point Likert scale. The scores, ranging from 1 to 5, correspond to Strongly Disagree, Disagree, Fairly Agree, Agree, and Strongly Agree, respectively, indicating the level of agreement of the respondents with each item. The main motivations for social media consumers’ engagement include social connections, self-expression and identity, entertainment and leisure, as well as information acquisition and practicality. The measurement items are detailed in

Table 2. Measurement items.

Construct	Item Code	Measurement Items	References
Identification	ID1	When someone praises this activity, I perceive it as a compliment to myself.	[25]
	ID2	I hold a positive attitude towards participating in this activity.
	ID3	I believe that engaging in this activity is beneficial.
	ID4	I see merit in being involved in this activity.
Enthusiasm	EN1	I dedicate a significant amount of time to this activity.	[52]
	EN2	I strive to participate in this activity as frequently as possible.
	EN3	I integrate this activity into my daily routine.
	EN4	I am filled with enthusiasm for participating in this activity.
Satisfaction	SAT1	My experience of participating in this activity is satisfying.	[53,54]
	SAT2	My involvement in this activity is enjoyable and seamless.
	SAT3	The outcomes of participating in this activity are fulfilling to me.
Community platform interaction	INT1	I greatly enjoy interacting with others within the community of this program.	[37,55]
	INT2	I am inclined to assist others when I am an active member of the community.
	INT3	I take pleasure in initiating conversations within the community.
	INT4	I am an avid participant in discussions within the community.
Offline interaction	CON1	I enjoy participating in this activity with my friends.	[37,56]
	CON2	I find the activity more enjoyable when I am in the company of others.
	CON3	The presence of others participating in the activity enhances my enjoyment.
Social presence	SP1	I sense a connection with others when I engage in this activity.	[9,46]
	SP2	Participating in this activity allows me to feel my “true self”.
	SP3	Through my participation, I have gained insight into the attitudes of other consumers.
	SP4	I have realized that there are other consumers interested in this activity when I participate.
Willingness to participate	CPI1	I am currently inclined to actively participate in the activity.	[53,54]
	CPI2	I am willing to actively engage in this program in the future.
	CPI3	I am committed to keeping track of the development of this activity.
	CPI4	I would recommend this platform’s activity to others around me.

. Follow-up calls and mailings were made to improve the response rate [4,51].

4.2. Data Collection

In this study, the questionnaire was distributed in the form of an online questionnaire via multiple social media platforms, namely QQ, WeChat, Weibo, and RED (Xiaohongshu). We shared the survey link through WeChat Moments, WeChat groups, or mutual fill groups, using personal networks for dissemination, yielding 232 valid responses. To improve the reach and engagement of the target audience, we must clarify the characteristics of the target audience (such as age, gender, occupation, etc.) and choose recruitment channels that match them, and then design attractive questionnaire titles and introductions, using eye-catching images or videos alongside the questionnaire link to enhance users’ desire to click. Participants were recruited via Prolific, filtered for prior engagement with CSR platforms (e.g., Ant Forest, Alipay Charity). We excluded users with less than three months of platform activity to ensure familiarity. While our sample is not fully population-representative, it accurately reflects the actual user base of virtual CSR platforms, ensuring ecological validity for the studied phenomena.

Table 3. Statistical analysis of basic information.

Variable	Option	Frequency	Percentage
Sex	Male	96	41%
	Female	136	59%
Age	Below 20 years old	31	13%
	20–25 years old	150	65%
	26–30 years old	27	12%
	30–35 years old	10	4%
	Above 35 years old	14	6%
Academic qualification	High school and below	22	10%
	College degree	13	5%
	Undergraduate degree	174	75%
	Postgraduate and above	23	10%
Average monthly income	3000 and below	151	65%
	3000–6000	42	18%
	6000–10,000	22	9%
	10,000 or above	17	8%

illustrates that among those participating in corporate virtual CSR co-creation activities, 96 are males and 136 are females, accounting for 41% and 59% of the total, respectively, which indicates a greater proportion of female consumers. As for the age demographics of the respondents, 65% fall into the 20–25 age group, emphasizing the dominance of this younger age range. Younger individuals may have a higher acceptance and willingness to participate in virtual CSR co-creation activities, while older demographics may exhibit lower levels of engagement in such activities. Therefore, the higher proportion of young people in the sample may lead to research findings that are biased towards the behaviors and attitudes of the younger group. Regarding their educational attainment, 75% of the participants possess a bachelor’s degree. Coupled with their age, this indicates that the main demographic engaging in corporate virtual CSR co-creation activities consists of students and recent workforce entrants. In terms of the respondents’ average monthly income, 151 individuals (65%) report earnings of RMB 3000 or less, 42 (18%) earn between RMB 3000 and 6000, 22 (9%) earn between RMB 6000 and 10,000, and 17 (8%) earn above RMB 10,000. To conclude, the primary participants in virtual CSR co-creation activities are young women with a higher education.

5. Construct Reliability and Validity

As shown in

Table 4. Results of questionnaire scale reliability test.

Construct	Item Code	Adjusted Total Correlation (CITC Value)	The α Factor for Which the Item Has Been Removed	Cronbach’ α
Identification	ID1	0.653	0.881	0.880
	ID2	0.774	0.834
	ID3	0.767	0.836
	ID4	0.774	0.834
Enthusiasm	EN1	0.743	0.870	0.893
	EN2	0.780	0.856
	EN3	0.781	0.856
	EN4	0.757	0.866
Satisfaction	SAT1	0.796	0.840	0.890
	SAT2	0.776	0.850
	SAT3	0.792	0.839
Community platform interaction	INT1	0.821	0.867	0.905
	INT2	0.678	0.913
	INT3	0.839	0.858
	INT4	0.825	0.864
Offline interaction	CON1	0.727	0.821	0.863
	CON2	0.732	0.817
	CON3	0.764	0.785
Social presence	SP1	0.729	0.805	0.854
	SP2	0.700	0.816
	SP3	0.770	0.782
	SP4	0.609	0.849
Willingness to participate	CPI1	0.763	0.828	0.875
	CPI2	0.706	0.850
	CPI3	0.744	0.837
	CPI4	0.721	0.845

, the α coefficients of all the variables are greater than 0.8, signaling a very high level of scale reliability. The CITC value of each item is above 0.3, denoting strong correlations among the items and good reliability. Furthermore, the “α coefficient when an item is deleted” for each item is lower compared to the overall α coefficient of its corresponding variable, suggesting that each item contributes positively to the variable’s α coefficient, thereby affirming the rationality of their inclusion. Collectively, the questionnaire employed in this study demonstrates excellent reliability, with the utilized variables and their respective measurement items exhibiting high internal consistency. This satisfies the reliability standards for empirical research and supports further in-depth analysis.

As shown in

Table 5. Results of KMO and Bartlett’s test.

KMO Value		0.934
Bartlett’ sphericity test	Approximate chi-square	4636.016
	df	325
	p-value	0.000

, the KMO value is 0.934, exceeding the threshold of 0.7. The results of Bartlett’s test of sphericity are significant at the 0.001 level, indicating a high degree of correlation among the measurement items and their suitability for further factor analysis.

This study employs confirmatory factor analysis (CFA) to assess the structural validity of the conceptual model. This study comprises 26 analysis items, with an effective sample size of 232, which is nine times the number of items, indicating a sufficient sample size for conducting factor analysis.

As illustrated in

Table 6. Results of exploratory factor analysis.

Item Code	Factor Loading							Communality
	Factor 1	Factor 2	Factor 3	Factor 4	Factor 5	Factor 6	Factor 7
ID1	0.734							0.722
ID2	0.780							0.788
ID3	0.737							0.794
ID4	0.805							0.793
EN1				0.769				0.790
EN2				0.716				0.778
EN3				0.795				0.830
EN4				0.674				0.768
SAT1							0.594	0.776
SAT2							0.757	0.854
SAT3							0.575	0.757
INT1		0.721						0.814
INT2		0.655						0.705
INT3		0.782						0.843
INT4		0.818						0.840
CON1					0.744			0.767
CON2					0.766			0.753
CON3					0.799			0.829
SP1						0.673		0.756
SP2						0.605		0.710
SP3						0.696		0.805
SP4						0.566		0.689
CPI1			0.758					0.805
CPI2			0.742					0.723
CPI3			0.760					0.757
CPI4			0.717					0.768
Explanatory rate of variance of eigenvalues	3.306	3.265	3.250	3.155	2.818	2.344	2.077	-
Variance explanation rate	12.714	12.555	12.499	12.134	10.838	9.014	7.987	-
Cumulative variance explanation rate	12.714	25.270	37.770	49.903	60.741	69.755	77.742	-

, the rate of the cumulative variance explained is 77.742%, indicating that the factors of the seven dimensions selected by the model can explain the scale information well. The factor loading coefficients for all items exceed the threshold of 0.4, and within each dimension, the items demonstrate a high correlation with the corresponding factors. This indicates that the measurement items designed for the seven variables, namely identification, enthusiasm, satisfaction, community platform interaction, offline interaction, social presence, and willingness to participate, can be effectively categorized into seven corresponding factors. Overall, the scale exhibits a strong structural validity.

As shown in

Table 7. Results of confirmatory factor analysis.

Construct	Item Code	Factor Loading	AVE	CR
Identification	ID1	0.702	0.6547	0.8829
	ID2	0.843
	ID3	0.859
	ID4	0.823
Enthusiasm	EN1	0.784	0.6767	0.8932
	EN2	0.845
	EN3	0.817
	EN4	0.843
Satisfaction	SAT1	0.875	0.7332	0.8918
	SAT2	0.825
	SAT3	0.868
Community platform interaction	INT1	0.855	0.7147	0.9085
	INT2	0.711
	INT3	0.914
	INT4	0.887
Offline interaction	CON1	0.820	0.6804	0.8646
	CON2	0.804
	CON3	0.850
Social presence	SP1	0.790	0.6081	0.8606
	SP2	0.789
	SP3	0.844
	SP4	0.688
Willingness to participate	CPI1	0.829	0.6397	0.8765
	CPI2	0.767
	CPI3	0.806
	CPI4	0.796

, the factor loading coefficients of most variable items exceed 0.7, with a few, such as the factor loading coefficient of 0.688 for SP4, closely approaching the threshold. The AVE values for the variables are 0.6547, 0.6767, 0.7332, 0.7147, 0.6804, 0.6081, and 0.6397, all surpassing the standard of 0.5. Similarly, the CR values are 0.8829, 0.8932, 0.8918, 0.9085, 0.8646, 0.8606, and 0.8765, respectively, each exceeding the benchmark of 0.7. Overall, the scale demonstrates good convergent validity.

As shown in

Table 8. Pearson’s correlation coefficient and AVE square root value.

	Identification	Enthusiasm	Satisfaction	Community Platform Interaction	Offline Interaction	Social Presence	Willingness to Participate
Identification	0.809
Enthusiasm	0.634	0.823
Satisfaction	0.774	0.733	0.856
Community platform interaction	0.431	0.682	0.630	0.845
Offline interaction	0.592	0.531	0.724	0.631	0.825
Social presence	0.595	0.716	0.720	0.738	0.690	0.780
Willingness to participate	0.558	0.638	0.664	0.628	0.585	0.792	0.800

Note: bolded numbers are AVE square root values for each factor.

, the AVE square root values for most variables surpass the highest absolute correlation coefficients between these variables and others. Only the AVE square root value for social presence falls slightly below the absolute correlation coefficients with other variables, yet it remains very close. In summary, these results demonstrate that the scale possesses good discriminant validity.

6. Data Analysis and Results

In this study, the theoretical model undergoes further validation through the application of structural equation modeling (SEM), with the analysis conducted using AMOS 28.0. The model fit indices, such as the absolute fit index (e.g., GFI, RMSEA), value-added fit index (e.g., CFI, IFI, NNFI), and parsimonious fit index (e.g., CMIN/DF, PGFI, PNFI), are selected for validation and evaluation. The model can be considered as well fitted if the individual indices satisfy the standard requirements [57]. As illustrated in

Table 9. Results of model fitness evaluation.

Fitting Indicator Name Options		Criteria or Thresholds for Adaptation	Metric
Absolute Fitness Index	GFI	>0.80 or more	0.833
	RMSEA	<0.08 or less	0.071
Value Added Fitness Index	CFI	>0.80 or more	0.928
	IFI	>0.80 or more	0.929
	NNFI	>0.80 or more	0.875
Simplicity Fitness Index	CMIN/DF	1 < X2 < 3	2.168
	(X2 degree of freedom ratio)
	PGFI	>0.50 or more	0.660
	PNFI	>0.50 or more	0.749

, all model fit indices satisfy the criteria, indicating that the model exhibits a good fit.

As shown in Figure 2 and

Table 10. Hypothesis testing.

Path Relationship	Standardized Estimates	Unstandardized Estimates	S.E. Standard Error	C.R. Critical Ratio	p
Identification → Social presence	0.081	0.242	0.085	0.880	0.379
Enthusiasm → Social presence	0.242	0.109	0.075	2.559	0.010
Satisfaction → Social presence	0.109	0.324	0.101	0.884	0.377
Community platform interaction → Social presence	0.324	0.230	0.053	3.805	***
Offline interaction → Social presence	0.230	0.055	0.082	2.580	0.010
Identification → Willingness to participate	0.055	0.046	0.108	0.554	0.010
Enthusiasm → Willingness to participate	0.046	0.133	0.097	0.441	***
Satisfaction → Willingness to participate	0.133	0.057	0.128	1.007	***
Community platform interaction → Willingness to participate	0.057	−0.019	0.070	0.595	0.552
Offline interaction → Willingness to participate	−0.019	0.602	0.107	−0.192	0.848
Social presence → Willingness to participate	0.602	0.709	0.141	5.012	***

Note: *** represents a significance level of p < 0.001.

, the hypotheses proposed in this study all posit positive effects, thus a hypothesis is considered valid if the standardized path coefficient β falls within the range (0, 1) and the significance level p is less than 0.05. According to the results in Table 10, it can be seen for H1a that β equals 0.055 and is significant at the p < 0.05 level, confirming the validity of H1a. For H1b and H1c, the β values are 0.046 and 0.133, respectively, both with p-values below 0.05, validating these hypotheses as well. For H2a, while β is 0.057, the p-value of 0.552 exceeds 0.05, rendering H2a invalid. Similarly, H2b shows a β of −0.019 and a p-value of 0.848, also exceeding 0.05, thus invalidating H2b. Lastly, H3 demonstrates a β of 0.602 with a p-value less than 0.05, confirming its validity.

This study hypothesizes that social presence plays a mediating role, and the hypothesis is valid when the significance level of p < 0.05. According to

Table 11. The mediating effect of social presence.

Path Relationship	Efficiency Value	S.E. Standard Error	Bias-Corrected 95% CI
			Lower	Upper	p
Identification → Social presence → Willingness to participate	0.049	0.120	−0.137	0.301	0.586
Enthusiasm → Social presence → Willingness to participate	0.146	0.113	−0.012	0.373	0.073
Satisfaction → Social presence → Willingness to participate	0.065	0.177	−0.239	0.371	0.510
Community platform interaction → Social presence → Willingness to participate	0.195	0.093	0.065	0.487	0.003
Offline interaction → Social presence → Willingness to participate	0.139	0.099	0.007	0.385	0.010

, the p-values for H4a, H4b, and H4c are all greater than 0.05, thus the hypotheses are not supported; whereas, the p-values for H4d and H4e are 0.003 and 0.01, respectively, both falling below 0.05, confirming the validity of the hypotheses.

7. Discussions

This study integrates and innovates upon existing research on the influencing factors of enterprise virtual CSR co-creation activities. It introduces five independent variables and one mediator variable, constructing a conceptual model that considers the consumer willingness to participate as the dependent variable. Based on this model, corresponding measurement scales and questionnaires are designed. The questionnaires are distributed online, and data are collected accordingly. The hypotheses are then verified and analyzed using SPSS 27.0 and AMOS 28.0, leading to the following conclusions and our reflections.

7.1. Effect of Independent Variables on Consumer Willingness to Participate

All the variables of the emotional factors, namely identification, enthusiasm, and satisfaction, exert a positive influence on consumer willingness to participate. Among these, enthusiasm has the most significant impact on consumer willingness to participate. Conversely, interaction factors do not notably influence consumer willingness to participate. This suggests that the primary determinant of whether consumers participate in an activity is their emotional response when engaging in enterprise virtual CSR co-creation activities. Higher levels of identification, enthusiasm, and satisfaction increase the likelihood of participation. Conversely, if consumers are dissatisfied with the content of a company’s virtual CSR co-creation activities, or if the activities’ content and values fail to resonate with them, their willingness to participate diminishes accordingly, reducing the likelihood of their engagement. This aligns with the existing literature on consumer behavior, which consistently highlights the importance of emotional connections in driving engagement [58]. Enthusiasm, while important, may be a transient emotional state that is easily influenced by external factors, such as marketing campaigns or social influence. Future research should explore the stability of enthusiasm as a predictor of long-term consumer participation in virtual CSR activities. Regarding H2a and H2b, which hypothesized that community platform interaction and offline interaction positively influence consumer willingness to participate, the non-significant results may stem from several factors. First, the virtual nature of CSR co-creation activities might limit the perceived impact of offline interactions compared to direct, physical involvement. Second, the study focused on Chinese consumers, who may have a stronger preference for online interactions due to cultural and technological factors. Third, the measurement of these interactions might not fully capture their nuances, leading to potential biases. Further research is needed to explore these factors across different contexts and cultures to provide a more comprehensive understanding. The implications are that enterprises should consider the specific cultural and technological environment when designing virtual CSR activities to optimize consumer engagement. A more critical approach would involve a cross-cultural comparison to validate whether these findings are specific to the Chinese context or applicable to other regions [59]. Additionally, this study’s reliance on self-reported data may introduce biases, as consumers’ perceptions of interactions may not accurately reflect their actual experiences [60].

7.2. The Influence of Social Presence on Consumer Willingness to Participate

Based on the aforementioned result regarding social presence and willingness to participate, it is evident that the extent to which consumers perceive their “self” within a virtual community largely determines their ultimate decision to participate in virtual CSR co-creation activities. Consequently, a higher level of social presence increases the likelihood of participation, indicating a significant positive impact of social presence on consumer willingness to participate. Creating immersive and interactive virtual environments is crucial for consumer engagement. This finding supports the broader trend of leveraging digital technologies to enhance consumer experiences [61]. Nevertheless, our conceptualization of social presence is somewhat narrow. It focuses primarily on the perception of “self” within a virtual community, neglecting other dimensions, such as social support, trust, and community norms, that may also influence consumer behavior [62]. A more comprehensive understanding of social presence could provide deeper insights into the factors that drive consumer engagement in virtual CSR activities.

7.3. The Mediating Role of Social Presence

According to the results of the mediating effect of social presence, it serves as a mediator between both community platform interaction and offline interaction and willingness to participate. During the process of interaction, consumers are more likely to perceive their “self”, thereby substantially enhancing the probability of participation. Social presence not only acts as a key intermediary between emotional and interaction factors but also underscores the importance of fostering an immersive and interactive virtual environment to enhance consumer engagement. This highlights the need for enterprises to prioritize the creation of inclusive and supportive virtual communities that resonate with consumers’ emotional needs, ultimately driving sustained participation in virtual CSR co-creation activities. The conclusion highlights the importance of creating a more immersive and interactive participation experience for consumers, providing enterprises with a new approach to conducting virtual CSR co-creation activities. The problem is that the reliance on a single mediator (social presence) may oversimplify the complex dynamics of consumer engagement in virtual CSR activities. Other potential mediators, such as perceived value or social identity, could also play significant roles in influencing consumer willingness to participate. Future research should consider a more nuanced model that incorporates multiple mediators to provide a comprehensive understanding of the relationships between interaction factors, social presence, and consumer engagement.

8. Conclusions

This study examined the effects of emotional and interaction factors on the consumer willingness to participate in virtual CSR co-creation activities, with a particular focus on the mediating role of social presence. Our results indicate that emotional factors, including identification, enthusiasm, and satisfaction, positively influence consumer willingness to participate in virtual CSR co-creation activities. Among these factors, enthusiasm emerged as the most significant predictor, underscoring the importance of fostering enjoyable and engaging experiences for consumers. Secondly, our study reveals that social presence plays a crucial role in mediating the relationship between interaction factors (both community platform interaction and offline interaction) and consumer willingness to participate. This finding highlights the significance of creating a sense of social connection and belonging within virtual CSR communities. Thus, this work extends the Stimulus–Organism–Response (S-O-R) model by incorporating social presence as a mediator in the context of virtual CSR co-creation. Furthermore, it provides empirical evidence on the relative importance of emotional and interaction factors in shaping consumer willingness to participate, with implications for the design and promotion of virtual CSR initiatives.

This study innovatively applies the S-O-R framework to virtual CSR co-creation, identifying algorithmically mediated stimuli as novel environmental cues and revealing social presence’s unique mediating role. It also parses emotional factors into identity-driven and experience-driven pathways, offering fresh insights for affective computing in non-transactional contexts.

9. Implications

As a relatively novel public welfare model, virtual CSR co-creation activities are favored by the public for their entertainment and convenience. However, enterprises face numerous challenges in attracting consumer attention, designing engaging activity content, and sustaining the consumer willingness to participate. Enterprises need to continually explore and refine strategies to boost participant numbers while enhancing consumer satisfaction, thereby strengthening the cohesion and influence of the established virtual CSR communities. Based on our findings, we offer the following concrete recommendations for enterprises engaging in virtual CSR: (1) Enhance emotional engagement by fostering a strong sense of identification, enthusiasm, and satisfaction among consumers through personalized content and interactive features. (2) Prioritize building social presence by encouraging consumers to share their experiences and feelings within the virtual community, fostering a sense of belonging. (3) Focus on the quality of community platform interactions, ensuring they are engaging and meaningful, as they significantly mediate consumer willingness to participate. These strategies, tailored to our findings, can help enterprises effectively attract and retain consumers in their virtual CSR activities. Through empirical analysis, this study draws research conclusions on the factors influencing consumer participation in virtual CSR co-creation activities and offers corresponding countermeasures and suggestions. These insights not only position virtual CSR co-creation activities as a vital tool for enhancing corporate reputation and word-of-mouth but also assist enterprises in attracting more consumers in the process of fulfilling corporate social responsibility.

10. Limitations

Although the findings in this research contribute to both the literature and practice, several limitations should be noted. First, our study is limited by its focus on Chinese consumers, which may affect the generalizability of findings. Additionally, the measurement of interaction factors may not fully capture their complexity. For future research, we suggest investigating these factors across diverse cultural and technological contexts. Non-significance may stem from interaction overload (e.g., fragmented discussions on WeChat/Weibo reducing focus). Conversely, structured platforms, like Ant Forest’s gamified feeds, show a higher efficacy, suggesting that the platform design moderates effects. Furthermore, longitudinal studies could provide insights into the dynamic nature of the consumer willingness to participate over time. We aim to address these limitations and explore new avenues in future endeavors.