The value–attitude–behavior (VAB) hierarchy theory, introduced by Homer and Kahle (1988), offers a structured framework for understanding how personal values influence attitudes and ultimately behavior. Values are defined as trans-situational, enduring beliefs about desirable end states or modes of conduct (Rokeach, 1973). These constructs guide perception and decision-making over time, shaping attitudes and behavioral tendencies (Bergman, 1998).

In the context of the sharing economy, user motivations have been linked to utilitarian, hedonic, and symbolic values (Hepola et al., 2020; Hwang and Griffiths, 2017; Lee and Kim, 2018; Miao et al., 2014). Utilitarian values are rooted in function and efficiency: consumers guided by these values often lead simpler lifestyles, and seek practical benefits from consumption (Wang and Lin, 2009). In the sharing economy, such practical advantages include cost savings (Hamari et al., 2016). For example, convenience and economic efficiency are dominant in car-sharing platforms (Bardhi and Eckhardt, 2012). Airbnb offers hosts a source of income and provides customers with more affordable and unique experiences compared to hotels (Lee and Kim, 2018; Guttentag, 2015).

Hedonic value, by contrast, involves emotional and experiential rewards. It reflects a consumer’s desire for novelty, pleasure, and enjoyment (Tsou et al., 2019). Staying in unique accommodations such as igloos, castles, or treehouses adds experiential richness (Miao et al., 2014), while renting luxury attire for special events enhances emotional value (Pantano and Stylos, 2020).

Symbolic value refers to non-material meanings embedded in consumption, often tied to self-image, identity, and social affiliation (Hasan, 2022). Consumers may choose sharing services to express pro-environmental values or project socially responsible identities. For example, by ride-sharing to reflect sustainability concerns (Gadeikiene and Svarcaite, 2021).

The theory of reasoned action (TRA) (Ajzen and Fishbein, 1980) and the theory of planned behavior (TPB) (Ajzen, 1991) explains attitude as a person’s evaluation of a behavior as positive or negative. Attitudes are shaped by beliefs and previous experiences and play a central role in predicting intentions (Ajzen and Fishbein, 1980; Lo et al., 2020). Attitudes have been identified as one of the strongest predictors of sharing economy participation (Akande et al., 2020), and their influence may vary depending on prior experience (Tanveer et al., 2025).

Building on the main ideas of these three theories (VAB, TRA, and TPB), and on existing research in the sharing economy, this study follows the logic of the value–attitude chain. Value is regarded as an antecedent of attitude in various fields (Chang et al., 2020). We propose that utilitarian, hedonic, and symbolic values are the main factors influencing the user’s attitude. Therefore, we expect these three core values to shape attitudes toward the benefits of the sharing economy, which can encourage increased participation.

Thus we propose:

Subjective norm refers to an individual’s perception of social pressure to perform or avoid a particular behavior (Ajzen, 1991). These norms stem from normative beliefs about how significant others (e.g., friends, family, or peers) expect them to behave, and the extent to which individuals are motivated to comply (Park, 2000; Moser, 2016).

In the context of the sharing economy, particularly Airbnb, three subjective norms have been identified as influential: peer influence, external influence, and word-of-mouth (Tajeddini et al., 2021). Peer influence is rooted in the behavior and opinions of one’s social group (Schiffman and Kanuk, 2007). For instance, individuals may follow friends’ recommendations when booking shared accommodation (Barnes and Mattsson, 2017). External influence refers to credible input from strangers (e.g., expert reviews or media) that shapes perceptions (Cohen et al., 2014). Word-of-mouth often comes through online reviews, where users evaluate others’ feedback to guide their own decisions (Xu, 2020; Lee and Wong, 2021). These social cues help shape behavioral expectations (Małecka et al., 2022).

In contrast, personal norms are internally held moral obligations (Thøgersen, 2009; de Groot et al., 2021). When individuals view certain behaviors as consistent with their ethical standards or identity, personal norms strongly predict intention. For example, those with high moral identity may reduce car use or prefer sustainable alternatives like bike-sharing (Ricci, 2015; Culiberg et al., 2023). Rental users (e.g., Uber or Airbnb users) often express pro-sustainable identities rooted in personal values (Shukla et al., 2024). While subjective norm represents external social pressure, personal norm reflect internal moral duty (Roos and Hahn, 2019). Research shows that the more robust a person’s personal norm related to pro-environmental behavior, the stronger their corresponding intentions and actions tend to be (Joanes, 2019; Onwezen et al., 2013).

In our conceptual framework, following the prior research, subjective norm is the key variable for customer participation (Blut and Wang, 2025), and personal norms that are separate from social norms across various economic contexts (Bašić and Verrina, 2024). Perceived moral obligation may influence the subjective norm and personal norm differently (Fauzi et al., 2024). Regarding the sharing economy, the values can also indirectly encourage behavior by shaping subjective and personal norms (see Figure 1).

Thus, we hypothesize:

Both subjective and personal norms influence behavior not only directly but also through their impact on attitudes (Smetana and Adler, 1980). According to the theory of reasoned action, intentions are shaped by attitudes and subjective norm (Ajzen and Fishbein, 1980). Multiple studies confirm that attitude is a key predictor of behavioral intention in the sharing economy (Akande et al., 2020; Chen and Ryu, 2024).

Empirical work further shows that enhancing consumer perceptions and experiences improves reuse intention (Elnadi and Gheith, 2022), and both attitude and social norms influence consumer decisions in services like fashion rental (Lee and Chow, 2020). Additionally, studies in transportation-sharing contexts have demonstrated a strong positive impact of attitude on behavioral intention (Huang et al., 2021; Jing et al., 2021). Carsharing users often view it as an eco-friendly transportation option because it helps reduce many negative impacts of automobility (Vejchodská et al., 2024), and provides a sense of belonging (Schaefers, 2013).

In peer-to-peer accommodations, moral obligations (i.e., personal norms) also influence intention alongside attitudes and subjective norms (Shin et al., 2023). Previous studies have shown that personal norm positively influences intention (Tandon et al., 2023; Wang et al., 2023). In this study, we integrate subjective norm, personal norm, and attitude into a unified research framework. Attitude refers to the consumer’s overall evaluation of sharing economy participation, including its social, environmental, and economic benefits. In line with circular economy principles, these attitudes are expected to play a central role in predicting intention.

We therefore hypothesize:

This study explores engagement in the sharing economy to provide insights for both academic and business sectors. We administered an online survey targeting consumers involved in the sharing economy across the U.S. and China. Identical surveys were distributed to participants in both countries, with localized language for the Chinese context. The survey comprised seven sections designed to assess respondents’ understanding of the sharing economy and their utilitarian, hedonic, and symbolic values, as well as their subjective norm, personal norm, attitude, and intention to participate.

The data for this study were gathered in September 2017 in the U.S. and in May 2020 in China, during the COVID-19 pandemic. The pandemic caused significant disruptions across industries, yet China emerged as the first major economy to recover, with most cities returning to normal by the second or third quarter of 2020 (Wang et al., 2022). An Overview of the Development of China’s Sharing Economy from 2017 to 2020 (State Information Center, 2021) reported year-on-year growth of approximately 2.9% in 2020, despite a deceleration compared to the previous year. Therefore, the data collection years reflect meaningful points in each country’s sharing economy development.

Independent datasets were collected using two online platforms: Amazon Mechanical Turk (MTurk) in the U.S., and Wenjuanxing, a professional online survey provider, in China. In both countries, participants spent approximately 7 min completing the survey.

After data collection, we conducted data analysis using SPSS 27.0 and AMOS 28. To evaluate the validity and reliability of the latent constructs, Confirmatory Factor Analysis (CFA) was performed on the final measurement model, which included 23 items. To test the proposed hypotheses, Structural Equation Modeling (SEM) was used. SEM is well-suited for multivariate path analysis and enables examination of complex relationships among variables beyond the capabilities of traditional regression analysis.

All measurement items used in this study were adapted from previously validated scales (see Table 1). Because the original items were developed in English, the Chinese version of the survey was translated using the back-translation method to ensure linguistic and conceptual equivalence.

The value dimensions (utilitarian, hedonic, and symbolic) are based on Hwang and Griffiths (2017). Utilitarian value was measured with a three-item scale assessing perceived economic benefit (e.g., “Collaborative consumption saves me money”). Hedonic value was measured with two items reflecting enjoyment and emotional satisfaction (e.g., “I enjoy collaborative consumption”). Symbolic value was assessed with three items addressing sustainability and identity (e.g., “Collaborative consumption makes me feel smart”).

Subjective norm and personal norm were measured using scales from Roos and Hahn (2017). Subjective norm was measured with four items reflecting social pressure and perceived expectations from significant others (e.g., “Most people who are important to me think that I should consume collaboratively”). Personal norm was measured with three items assessing internalized moral obligation (e.g., “I strongly feel a personal obligation to consume collaboratively”).

Attitude was measured using a three-item scale adapted from Roos and Hahn (2017), evaluating cognitive, emotional, and behavioral responses toward collaborative consumption (e.g., “All things considered, I find participating in collaborative consumption to be a wise move”).

Behavioral intention was measured using four items from Bhattacherjee (2001), capturing future engagement likelihood (e.g., “All things considered, I expect to continue collaborative consumption often in the future”).

All items used a 7-point Likert scale ranging from 1 (“strongly disagree”) to 7 (“strongly agree”). Table 1 summarizes the measurement items and their sources.

Finally, a total of 302 responses were collected from U.S. participants, comprising 47.68% males (n = 144) and 52.32% females (n = 158). In China, 319 individuals responded; after removing questionnaires with extremely long completion times and uniform responses (indicative of disengagement), 305 valid responses remained. The gender distribution in the Chinese sample was also balanced, with females representing 52.46% (n = 160) and males 47.54% (n = 145). Table 2 present the demographic information of the samples.

Hair et al. (2018) suggest that a factor loading of at least 0.5 is a general guideline for practical significance (Hair et al., 2018). To evaluate convergent validity, all standardized factor loadings were checked and found to exceed the minimum threshold of 0.50. Internal consistency was assessed using Cronbach’s alpha, Composite Reliability (CR), and Average Variance Extracted (AVE). Additionally, the calculation of CR indicated that the CR values in both countries are within the acceptable range, meeting Hair et al.’s (2018) criterion that CR should be above 0.7. In the U.S. sample, CR values ranged from 0.867 to 0.952, and in the China sample from 0.800 to 0.869, all exceeding the 0.70 threshold. For convergent validity, the average variance extracted (AVE) should exceed 0.5 (Fornell and Larcker, 1981). In the study, AVE values ranged from 0.619 to 0.831 in the U.S., and from 0.506 to 0.690 in China, meeting the recommended 0.50 minimum. These results indicate that the measurement model demonstrates sufficient convergent validity in both samples.

Discriminant validity was assessed using Fornell and Larcker’s (1981) criteria. The square root of each construct’s AVE exceeded its correlations with other constructs, indicating acceptable discriminant validity and no concerns with multicollinearity in the U.S. sample (see Table 3). However, in the Chinese sample (Table 3), hedonic value, attitude, and intention showed some overlap, suggesting slightly weaker discriminant validity (Anderson and Gerbing, 1988).

A commonly used guideline for the Comparative Fit Index (CFI) is values above 0.9 (Bentler, 1990). Model fit was evaluated using common indices: χ²/df < 3, GFI > 0.80, AGFI > 0.80, CFI > 0.90, RMSEA < 0.08. These indices suggested that the model showed a good fit to the data, with the normed chi-square was below the recommended 3.0 level (Hu and Bentler, 1999). The U. S. model achieved good fit: χ² = 423.707, df = 152, χ²/df = 2.788, GFI = 0.882, AGFI = 0.821, CFI = 0.969, RMSEA = 0.077. The China model also showed good fit: χ² = 385.521, df = 153, χ²/df = 2.520, GFI = 0.882, AGFI = 0.821, CFI = 0.912, RMSEA = 0.071.

Maximum likelihood estimation was used to determine the parameters of the proposed structural relationship. Structural Equation Modeling (SEM) verified that the overall model fit was acceptable (see Table 4), enabling hypothesis testing based on the proposed path relationships. The structural models demonstrated a good fit, with the overall goodness-of-fit indices meeting the required thresholds.

In the U.S. sample, results indicated that both utilitarian and hedonic values had significant positive effects on attitude toward the sharing economy. Specifically, utilitarian value (β = 0.415, p < 0.001) and hedonic value (β = 0.656, p < 0.001) were both significant, thereby supporting hypotheses H1a and H1b. However, symbolic value exhibited a non-significant negative effect on attitude (β = −0.110, p = 0.201), resulting in the rejection of H1c.

With regard to the influence of values on norms, utilitarian value did not significantly predict either subjective norm (β = 0.128, p = 0.174) or personal norm (β = −0.044, p = 0.203), leading to the rejection of H2a and H2b. Similarly, hedonic value did not have a significant effect on subjective norm (β = 0.181, p = 0.061) or personal norm (β = 0.071, p = 0.428), resulting in H3a and H3b also being unsupported. In contrast, symbolic value showed a significant positive impact on both subjective norm (β = 0.457, p < 0.001) and personal norm (β = 0.764, p < 0.001), providing support for hypotheses H4a and H4b.

The model also tested the effects of attitude and norms on behavioral intention. Attitude and subject norm significantly influenced behavioral intention (β = 0.739, p < 0.001), (β = 0.158, p = 0.009), supporting hypotheses H5 and H6. However, personal norm did not have a significant effect on behavioral intention (β = 0.060, p = 0.308), so H7 was not supported. These findings suggest that in the U.S. context, while utilitarian and hedonic values are critical in shaping attitudes, symbolic value plays more significant role in shaping both types of norms. Furthermore, attitude remains the most powerful predictor of behavioral intention.

In the Chinese sample, the results similarly supported hypotheses H1a and H1b, indicating that both utilitarian value (β = 0.237, p = 0.004) and hedonic value (β = 0.918, p < 0.001) had significant positive effects on attitude toward the sharing economy. However, contrary to expectations, symbolic value had a significant but negative effect on attitude (β = −0.263, p = 0.032), resulting in the rejection of H1c. Regarding the influence of utilitarian value on norms, the results did not show significant effects on either subjective norm (β = 0.092, p = 0.167) or personal norm (β = 0.059, p = 0.368), leading to the rejection of H2a and H2b. In contrast, hedonic value significantly influenced both subjective norm (β = 0.391, p < 0.001) and personal norm (β = 0.450, p < 0.001), thereby supporting H3a and H3b. Similarly, symbolic value was found to significantly predict both subjective norm (β = 0.475, p < 0.001) and personal norm (β = 0.552, p < 0.001), supporting H4a and H4b. When examining the predictors of behavioral intention, attitude (β = 0.551, p < 0.001) and personal norm (β = 0.327, p = 0.009) had significant effects, confirming H5 and H7, respectively. However, subjective norm did not significantly affect behavioral intention (β = 0.146, p = 0.174), resulting in the rejection of H6. These findings suggest that in the Chinese context, hedonic and symbolic values play a more prominent role in shaping both norms and attitudes, and that personal moral obligations may be more influential than external social expectations in driving behavioral intention (see Table 5).

This study contributes to the literature on the sharing economy by emphasizing the role of individual-level values and norms in shaping participation. It underscores the importance of integrating values into models of consumer behavior to better understand sharing intentions. It also provides new evidence for applying the TRA and the TPB across different cultural contexts.

First, we provide a comparative assessment of how different value dimensions influence attitudes in both countries. The results show that utilitarian and hedonic values significantly shape consumer attitudes in the U.S. and China, which aligns with the core proposition of the VAB framework that values drive attitude (Homer and Kahle, 1988; Milfont et al., 2010). This reinforces the notion that the efficient use of resources is a universally appealing value. For example, shared mobility services in China attract users by reducing transaction costs (Shao et al., 2023). Similarly, previous research in collectivist cultures has shown that functional benefits like cost-efficiency and resource optimization predict green purchase intentions (Cheung and To, 2019). In the U.S., utilitarian motivations such as affordability and convenience also influence attitudes, paralleling mechanisms found in other domains (Fu et al., 2014).

Additionally, consumers in both countries report deriving enjoyment from their participation, confirming the role of hedonic value. This finding is consistent with previous studies that also emphasize the hedonic motives behind participation on sharing platforms (Kozlenkova et al., 2021; von Richthofen, 2022). However, symbolic value did not influence attitudes in either country. That is an outcome consistent with Hwang and Griffiths’s (2017) findings. This may reflect the utilitarian nature of sharing platforms, where users are motivated more by functional or emotional benefits than by identity signaling. For instance, Bardhi and Eckhardt (2012) found that Zipcar users value on-demand access and cost savings over symbolic associations, and that such motivations may affect attitude indirectly through sustainability perceptions (Lange et al., 2023). In the context of peer-to-peer accommodation, symbolic value negatively influenced continuance intention (Ofori et al., 2023). These empirical observations collectively support the basic idea that a person’s real behavior is influenced by both values and attitudes (Tajeddini et al., 2021).

Second, the influence of cultural context becomes especially apparent when examining how values affect norms. In both countries, utilitarian value did not significantly affect either subjective or personal norm. Symbolic value, by contrast, shaped both types of norms in both contexts. Notably, hedonic value influenced neither type of norm in the U.S., but had a strong positive effect on both subjective and personal norms in China. These results illustrate how cultural divergence moderates the value–norm link, which aligns with prior research, where values influence behavior via more specific and tangible constructs like subjective norms (Izquierdo-Yusta et al., 2022).

Third, the findings point to key differences in how norms influence behavioral intention. In the U.S., subjective norm significantly predicted intention, whereas personal norm did not. This finding aligns with prior research showing that social influence and trend affinity shape Airbnb users’ intentions in Western contexts (So et al., 2018). Conversely, in China, subjective norm had no significant effect on intention, while personal norm did. This may be because Chinese consumers rely more on information-based social influence, such as online reviews from strangers, than on normative influence from friends or family (Wang and Hassan, 2024). These dynamics reduce the predictive power of subjective norm and elevate the role of internalized moral obligation in shaping behavioral intention.

This study also offers practical guidance for sharing economy platforms seeking to adapt to different cultural environments. Understanding the role of values and norms can help firms tailor their marketing and platform design strategies.

In the U.S. market, enhancing subjective norm is crucial. Platforms should leverage social proof mechanisms, such as reviews, influencer endorsements, and peer referrals, to amplify social influence. And emphasize social media as a potent platform for spreading normative messages via user-generated content (Liu and Lapinski, 2025). They should also create emotionally engaging experiences that align hedonic value with symbolic identity, for instance, by promoting narratives like “sharing is freedom” or “smart and sustainable choices.”

In the Chinese market, activating personal norms through responsibility-based messaging is more effective. Firms should emphasize narratives around environmental protection, community benefit, or national development. Moreover, platforms should enhance the social aspects of consumption. For example, firms could incorporate features that support group participation or sharing stories to strengthen the relationship between hedonic and symbolic value and personal norms (Song et al., 2025).

While this study contributes novel insights into cross-cultural sharing behavior, it also presents several limitations that should be addressed in future research.

First, the analysis focuses exclusively on the U.S. and China. Expanding the model to include consumers in other regions (e.g., Europe or Southeast Asia) would improve generalizability and enable broader comparisons of cultural influences. Additionally, future studies could explore sharing behaviors across diverse sectors, including consumer goods, media, and communications (Lim, 2020). Additionally, we intend to use a mixed-methods approach, combining online surveys with offline interviews and field studies, to reduce sampling bias and improve data representativeness. Lastly, utilizing longitudinal tracking data will help confirm the causal links between values, attitudes, norms, and behavioral intentions.

Second, although participants were real consumers in the sharing economy, we did not examine how demographic characteristics affect the relationships in the model. Prior studies suggest that social norms may influence behavior differently depending on the consumer segment (Sands et al., 2020). Future research could examine how factors such as age, gender, or lifestyle shape the relationships among values, norms, and behavioral intention.

Finally, our model focused specifically on the role of values, attitudes, and norms. While these are important personal-level predictors, other well-documented factors (e.g., trust, risk perception, or reputation) were outside the scope of this study. Incorporating these additional constructs in future studies could provide a more comprehensive understanding of what drives sharing behavior. Moreover, using qualitative methods such as interviews or ethnographic observation may offer richer insights into the psychological and contextual drivers of sharing economy participation.