1 Introduction

Smart home refers to a residence integrated with network communication, security, and automatic control technology that intend to provide families with a safe, comfortable, and convenient living environment (Kim et al. 2017). Generally, smart home services are multifunctional services which include lighting control, household appliance control, intelligent security, scene management, and time management, aiming to provide residents with flexible and efficient remote control services (Alam et al. 2012).

Data indicate that global smart home market revenue has reached 48.71 billion dollars in 2018, and the revenue is predicted to grow to 121.96 billion dollars by 2022 with an annual growth rate of 25.8% (Zhang and Liu 2021). With the expansion of the smart home market, scholars have also gradually attached significance to the research on smart homes. However, most of the previous literature about smart home services have focused on the perspective of engineering or technology; few studies have discussed the acceptance of consumers toward smart home services (Yang et al. 2017; Mashal and Shuhaiber 2019). In addition, most of the published research has emphasized the functional or egocentric benefits that smart home services can provide (Park and Lee 2014). Only a limited amount of research has conceptualized smart homes as eco-friendly, while published literature has largely ignored the environmental protection function of smart homes (Schill et al. 2019).

Actually, smart home services promote environmental protection (Marikyan et al. 2019). The core parts of smart homes include smart electricity meters, smart energy appliances, and energy management functions. In addition, smart homes are also the terminal use node of the smart energy system, which can reduce carbon emissions and save energy by promptly shutting down household appliances when they are not in use (Alexandra et al. 2012; Wilson et al. 2015). The reduced energy consumption contributes to the formation of a low-carbon society. Such smart home services that facilitate the effective use of resources and reduce environmental degradation are called eco-friendly smart home services (ESHS) (Schill et al. 2019).

Despite the ability of ESHS to facilitate the lives of people and its huge market value, the rapid and widespread adoption of ESHS have fallen short of expectations (Mashal and Shuhaiber 2019). Public understanding toward ESHS is still in its conceptual stage. Smart home services have not been widely accepted (Yang et al. 2018). Moreover, global environmental problems have become increasingly prominent. As a typical representative of innovative technology, ESHS can substantially contribute to environmental protection (Schill et al. 2019). The exploration of consumer acceptance toward smart home services is a relatively new research filed (Mashal and Shuhaiber 2019), and few scholars highlight the environmental protection function of smart homes. On the basis of the above points, the influencing factors that may influence consumer intention to adopt ESHS must be explored to accelerate the adequate adoption of these services.

Scholars have mostly employed the theory of planned behavior (TPB) and the theory of technology acceptance model (TAM) as the research frameworks in the analysis of the adoption intention toward technologically innovative products or services (Cheng 2018). ESHS are typical representatives of the application of IoT. In the existing literature about the adoption of smart home services, scholars tend to use TAM rather than TPB (Pal et al. 2018; Shin 2018; Hubert et al. 2019; Shuhaiber and Mashal 2019). However, the present study highlights the environmental protection function of smart home services. TPB is also applied widely in the research of consumer acceptance of green purchasing behaviors (Montano and Kasprzyk 2015). ESHS combine the characteristics of technology and environmental protection. And few scholars use TPB theory to study consumers’ behaviors of smart home services. Thus, adopting TPB instead of TAM is appropriate in the research on adoption intention toward ESHS.

Despite the widespread application of the TPB model in the research of consumer acceptance toward environmental technological products or services, previous studies have pointed out that the model still has some limitations. The TPB model assumes that consumers are rational and reasoned (Wang et al. 2014). However, each consumer has a different amount of knowledge, which may have significant effects on the decisions of consumers to some extent. Published literature that has employed the TPB model for the research of pro-environmental behaviors has also neglected the role of environmental knowledge. Considering that ESHS are beneficial to environmental protection, previous studies have proven that lacking knowledge on environmental issues is likely to create apathy among residents (Mohiuddin et al. 2018). Thus, the variable of environmental knowledge must be used as the influencing factor to analyze consumer intention toward ESHS. Published works have also confirmed that the personality of consumers influences their decisions when they have other choices (Wang et al. 2018b). Although adopting ESHS provides an efficient, intelligent, and safe living experience and plays a role in protecting the environment, most consumers are not aware of these benefits and remain ambiguous or unwilling to accept such services (Geissdoerfer et al. 2017). Few scholars have explored the role of ambiguous tolerance on consumer intention to adopt certain services. Thus, ambiguous tolerance to TPB must be applied to explore consumer intention toward ESHS. Moreover, external factors should also be integrated to the TPB model to provide a comprehensive view of the research problem (Wang et al. 2014). ESHS must be connected to other products seamlessly. Existing studies have proven that compatibility is an important reminder for technology acceptance (Lee et al. 2011). Thus, compatibility is a key consideration. In addition, the living habits of consumers vary. Previous literature has proven that living habits are considered a decisive factor that affects the energy-saving behavior of residents (Wang et al. 2014). Few studies have considered the effect of living habits in the research on the adoption intention of consumers and the TPB model. Thus, living habits should be integrated into TPB to explore the adoption intention of consumers.

The contributions of this study are mainly reflected in the following three aspects. First, this study takes into account the environment-friendly functions of smart home, which have been ignored by most previous studies. Second, previous studies on smart home mainly focus on technology optimization. This study innovatively explores consumers' intention to adopt ESHS, expanding the scope of research in the field of smart home. Third, this study innovatively uses the TPB theory to explore consumers' intention to adopt ESHS, and includes the two variables of environmental knowledge and ambiguity tolerance to expand the TPB theory.

The remaining sections of this article are shown below. Section 2 describes the relevant theoretical background and hypothesis development. Section 3 presents the data and research methodology. In Sect. 4, the data analysis and results are introduced by using structural equation modeling. Section 5 provides and discusses the conclusions, implications of the findings, limitations, and future research.

2 Theoretical background and hypothesis development

2.1 Theory of planned behavior

The TPB model proposed by Ajzen (1991) is widely applied in the field of pro-environmental and technology acceptance. It consists of the variables of attitude, subjective norm, and perceived behavioral control, which have significant effects on consumer intention. Attitude is expressed as the negative or positive evaluation of an individual to practice a particular behavior. Published literature has proven that attitude is related to consumer intention (Ajzen 1991). For instance, Yadav and Pathak (2016) confirmed that if an individual holds a positive attitude to purchasing green products, then they are likely to accept green products. Wang et al. (2018a) also proved that positive attitude is associated with the strong intention of consumers to choose green hotels. Subjective norm is regarded as an individual’s perceptions of subjective pressures from important people or groups to conduct or not conduct a certain behavior (Ajzen 1991). Wang et al. (2014) confirmed that if residents perceive high subjective norm in Beijing, then they are likely to conduct energy-saving behavior. Ru et al. (2019) verified that if residents perceive high subjective norm, then they are likely to reduce PM2.5. Perceived behavioral control is defined as the individual’s perception of the extent of ease or difficulty in performing a certain behavior, which includes influencing factors, such as time, money, and knowledge (Ajzen 1991). Yazdanpanah and Forouzani (2015) verified that if Iranians perceive high behavioral control, then they are inclined to buy organic food. Chen and Hung (2016) confirmed that when the perceived behavioral control of consumers is high, then they are willing to accept green products. Dou et al. (2022) proved that subjective norm and perceived behavioral control significantly and positively influence consumers’ intention toward inoculating COVID-19 vaccine.

Within the context of ESHS, we can infer that if the attitude of consumers toward ESHS is positive, they are more enthusiastic about ESHS and more willing to adopt ESHS. When consumers perceive high pressure from loved ones (e.g., relatives or friends), they may feel reluctant to refuse or have a herd mentality and prefer to adopt ESHS. In addition, when consumers perceive the ease in conducting eco-friendly behavior, they have more time, opportunities and funds to adopt ESHS, and they are more likely to accept ESHS. Thus, the following assumptions can be assumed:

H1: Attitude exerts a significant and positive effect on consumers’ intention to adopt ESHS.

H2: Subjective norm exerts a significant and positive effect on consumers’ intention to adopt ESHS.

H3: Perceived behavioral control exerts a significant and positive effect on consumers’ intention to adopt ESHS.

2.2 Extended theory of planned behavior model

2.2.1 Environmental knowledge

Environmental knowledge is an understanding of the factors that affect the health and sustainability of the ecosystems we live in Diekmann and Preisendörfer (2003). The impact of environmental knowledge on consumer intention has two distinct views. Chen and Peng (2012) further proved that consumer attitude is inspired by their environmental knowledge. If the environmental knowledge of consumers is higher, consumers care more about the environment. Fu et al. (2020) suggested that an individual’s high level of environmental knowledge reflects the awareness of environmental issues. Liu et al. (2020) proved that residents with high environmental knowledge are likely to take a positive attitude toward household energy-saving behavior and adopt such behavior. Previous studies have also verified that environmental knowledge plays a role in the environmental behavior of consumers (Sun et al. 2018). Mohiuddin et al. (2018) pointed that environmental knowledge exert a positive influence on the attitude and intention of consumers toward purchasing green vehicles in emerging countries. Moslehpour et al. (2022) also proved that environment knowledge is positively correlated with green purchase intention. However, some scholars doubt the influence of environmental knowledge in pro-environmental behavior intention (Flamm 2009). For instance, Maichum et al (2016) confirmed that environmental knowledge fails to influence consumer intention to purchase green products. Li and Wu (2020) confirmed that the pro-environmental behavior intention of consumers is not affected by environmental knowledge. Under the ESHS background, we prefer to assume that environmental knowledge is a stimulation to encourage consumers to employ ESHS. If consumers increase their knowledge of environmental protection, then their attitude and intention to adopt ESHS are strengthened. The hypotheses can be presented as follows:

H4: Environmental knowledge exerts a significant and positive effect on consumers’ attitude to adopt ESHS.

H5: Environmental knowledge exerts a significant and positive effect on consumers’ intention to adopt ESHS.

2.2.2 Ambiguity tolerance

Ambiguity situation occurs when a person is unable to construct or organize an environment because of the lack of adequate criterion (Hammond et al. 2017). As a personality trait, ambiguity tolerance generally refers to the extent of the threat or discomfort people perceive that an ambiguous situation brings (Budner 1962). Generally, consumers with high ambiguity tolerance experience little pressure from ambiguous information or situations because they can have a strong understanding of the information, whereas consumers with low ambiguity tolerance are less likely to accept ambiguous information or situations (Xu and Tracey 2015). Previous studies have investigated the association of ambiguous tolerance with the attitude and behavioral intention of consumers (Matsumoto et al. 2017; Wang et al. 2018b). For instance, Jena and Sarmah (2015) proved that ambiguous tolerance immediately affects the attitude and intention of consumers to pay an additional charge for returning used products. Wang et al. (2018b) also verified the positive relationship between ambiguous tolerance and the attitude and intention of consumers toward remanufactured products.

In the context of ESHS, consumers may perceive that unknown risks due to ESHS are new and high-tech for consumers. Previous literature has identified three main risks in the context of ESHS, namely, time risk, security risk, and performance risk (Yang et al. 2017). The three risks of adopting ESHS usually leave consumers in an ambiguous situation. Consumers with high ambiguity tolerance are more comfortable and confident to face distrust and poor understanding than those with low tolerance (Hazen et al. 2012). Thus, consumers are likely to shape a positive attitude and intention toward ESHS. If the tolerance of consumers for ambiguity is low, they are unwilling to be in a nondeterministic situation, and consumers are unlikely to shape a positive attitude and are reluctant to adopt ESHS. Thus, the hypotheses are shown as follows:

H6: Ambiguity tolerance exerts a significant and positive effect on consumers’ attitude to adopt ESHS.

H7: Ambiguity tolerance exerts a significant and positive effect on consumers’ intention to adopt ESHS.

2.2.3 Compatibility

Compatibility refers to the extent to which the innovative products or services are regarded as consistent with the existing systems (Chong et al. 2009). In the area of technology acceptance, compatibility is an important driver of adoption intention because consumers usually tend to avoid changes. ESHS demand the high connectivity and fluent communication between different appliances (Shin et al. 2018). Thus, the compatibility of smart home applications with the lifestyles of potential users must be reconciled. Previous studies have proven that compatibility positively affects consumers’ adoption intention of innovation (Wang et al. 2020). Susanto and Goodwin (2013) verified that compatibility is the predictor of the intention of users in innovation. Hubret et al. (2019) proved that if the technology of smart home applications is compatible with current systems, users are likely to adopt it. Hidayat-ur-Rehman et al. (2022) also indicated that compatibility is the key factor affecting the use of mobile wallet. Based on the above analysis, under the ESHS context, if the compatibility of ESHS is high, then customers are more likely to adopt ESHS. Thus, the following assumption can be proposed:

H8: Compatibility exerts a significant and positive effect on consumers’ intention to adopt ESHS.

2.2.4 Living habits

Living habits are usually associated with comfortable behaviors that individuals are unwilling to change (Wang et al. 2014). Living habits may be regarded as a momentous role of consumer intention toward energy-saving behavior. Wang et al. (2014) confirmed that living habits significantly and negatively affect consumer intention to save energy. Liu et al. (2015) also conducted that if additional consideration is given to the living habits of residents in residential energy-saving renovation, the effectiveness for measures of energy-saving can be improved. Under the ESHS context, the introduction of new technologies is a lifestyle disruption, which may bring about a major lifestyle change among consumers. Consumers who do not like or are not willing to change living habits may be reluctant to embrace it. Thus, the following hypotheses are proposed:

H9: Living habits exert a significant and negative effect on consumers’ intention to adopt ESHS.

Hence, we develop the conceptual framework as shown in Fig. 1.

Fig. 1
figure 1

The conceptual framework for the present study

Full size image

3 Data and research method

3.1 Sample and data collection

Data analyzed in this research are collected from a questionnaire survey. To make the study sample representative, we cooperated with Aqara smart home experience company in Hefei, which has a smart home customer base scattered across China. Data collection unit asked Internal staff of Aqara company to help us send out questionnaires to their smart home customers through professional questionnaire survey (http://www.sojump.com). Some of their company's customers base have used smart home services, others have not. The front of the questionnaire clearly explained that this survey would be conducted for academic research purposes only. Demographic information provided by respondents included gender, age, educational background, income level, and adoption experience. The survey data were collected from mid-October to mid-December 2019. A total of 816 questionnaires were issued in this study. The questionnaire collection team rejected those samples with missing and highly repetitive answers and discarded the questionnaires that were completed in less than three minutes. In the end, 629 questionnaires met the standards with an overall response rate of 77.08%.

The demographic characteristics of respondents are presented in Table 1. The full survey contained 629 eligible representative participants, including 342 females (54.4%) and 287 males (45.6%). The age ranged widely from under 20 years old to older than 50 years old. Respondents who were 20–29 years old formed the largest group (53.9%). In terms of education level, 52.6% of the respondents earned a junior college or university degree, while 26.4% had a master’s degree or PhD education background. The monthly income of 34% of the respondents ranged from RMB 2,000 to RMB 5,000 (USD 283 to USD 707), while the monthly income of 29.7% of the respondents ranged from RMB 5,001 to RMB 10,000 (USD 707 to USD 1,414). The respondents participating in the interview with experience in ESHS adoption accounted for 47.4% of the sample. The remaining 52.6% had no experience with ESHS adoption.

Table 1 Demographic profile of respondents
Full size table

3.2 Measures

The measurement items of this research were sourced from existing studies. The measurement of environmental knowledge was from the study of Dunlap et al. (2000). The measurement of ambiguity tolerance was based on the works of Furnham and Marks (2013) and Wang et al. (2018b). The measurement of compatibility was adopted from the research of Mohagheghi et al. (2013). The measurement of living habits was sourced from Wang et al. (2014). Then, the measurement of attitude, subjective norm, and perceived behavioral control came from the study of Taylor and Todd (1995) and Bhattacherjee (2000). Lastly, the measurement of intention was derived from the work of Davis et al. (1989). This study employs a five-point Likert scale to measure the items, that is, 1 = strongly disagree, 2 = disagree, 3 = unsure, 4 = agree, and 5 = strongly agree. “Appendix A” shows the detailed measurement scale items.

To test the validity and reliability of this questionnaire design, 50 questionnaires were circulated to conduct a pilot investigation. Questionnaire collection group randomly distributed questionnaires on Lei street, one of the busiest night markets in Hefei, Anhui province on Saturday and Sunday. The timing and location of questionnaire distribution were carefully considered. The questionnaire collection group randomly interviewed the population whether they were willing to fill out the questionnaire. After receiving a positive response from passers-by, we sent the questionnaire to the respondent. After the respondents completed the questionnaire, we asked them if they had any questions about the questionnaire. The measurement items were slightly revised for clarity on the basis of the feedback of the respondents. The results indicated that the validity and reliability of the pilot survey all met the criteria.

4 Data analysis and results

Common method variance (CMV) might be a potential threat in the hypothesized relationships because the constructs in this research were assessed through self-administered questionnaire (Buckley et al. 1990). Thus, Harman’s one-factor test was used to determine the CMV. The results confirmed that all of the measurement items were separated into six factors with eigenvalues higher than 1.0. The first single accounted for 27.8% of the variance that was below the reference value of 50.0%, indicating that CMV was not a serious issue in this study. The values of Kaiser–Meyer–Olkin (KMO) are higher than the standard value of 0.5 and the significances of Bartlett’s test are less than 0.01, indicating that there is a significant correlation between measure items, which can be used for factor analysis. Subsequently, partial least squares (PLS) structural equation analysis was introduced to check whether the survey data was in agreement with the conceptual model and to verify the hypotheses further by using Smart-PLS 3.0 and SPSS 25.0 software. PLS analysis consists of measurement and structural models, where the former is first used to confirm the reliability and validity of the constructs, and the latter is employed to examine the relationship among constructs.

4.1 Measurement model analysis

Confirmatory factor analysis is used to determine the reliability and validity of the measurement model. The reliability of the constructs are measured by the values of Cronbach’s alpha and composite reliability (Fornell and Larcker 1981). The results are shown in Table 2. The values of Cronbach’s alpha range from 0.840 to 0.921, which exceed the benchmark of 0.70. Moreover, the values of composite reliability meet the minimum standard of 0.70. Accordingly, the reliability of the construct is confirmed. Average variance extracted (AVE) and factor loadings are adopted to examine convergent validity. Fornell and Larcker (1981) proved that if the values of AVE exceed 0.50, the convergent validity is good. Table 2 indicates that the minimum value of the factor loading is 0.771. The values for AVE range from 0.694 to 0.863, verifying the convergent validity of the constructs. Furthermore, with regard to discriminant validity, the correlation between constructs should be less than the square root of the AVE. On the basis of the test results (Table 3), we conclude that the discriminant validity of this research is satisfactory.

Table 2 Results of confirmatory factor analysis
Full size table
Table 3 Mean and standard deviation of the constructs
Full size table

4.2 Structural model analysis

The structural equation model was used to investigate the relationship among the constructs. Path analysis was conducted to assess the path coefficients in this model through Smart-PLS 3.0. Empirical results are shown in Table 4. The results indicate that attitude (\(\beta\) = 0.190, p < 0.001) exerts significant and positive effect on consumers’ intention toward ESHS, thus, H1 is supported. Perceived behavioral control (\(\beta\) = 0.227, p < 0.01) significantly and positively influences consumers’ intention toward ESHS, H3 is also consistent with the original judgment. However, subjective norm fails to play a role in consumers’ intention, H2 is not in the expected direction. Environmental knowledge positively affects consumers’ attitude (\(\beta\) = 0.349, p < 0.001) and intention (\(\beta\) = 0.084, p < 0.01) toward ESHS. Accordingly, H4, H5 are supported. In addition, ambiguity tolerance also positively exerts influence on consumers’ attitude (\(\beta\) = 0.276, p < 0.001) and intention (\(\beta\) = 0.147, p < 0.001) under ESHS context. Thus, H6 and H7 are in the expected direction. And compatibility (\(\beta\) = 0.154, p < 0.001) exerts positive effect on consumers’ intention toward adopting ESHS, H8 is confirmed. However, living habits (\(\beta\) = − 0.300, p < 0.001) negatively influence consumers’ intention toward adopting ESHS, H9 is also supported.

Table 4 Results of structural model testing
Full size table

5 Discussion, implications and limitations

5.1 Discussion

The research uses the extended TPB model to explore consumer intention to adopt ESHS. The results confirmed that consumer intention to adopt ESHS is predicted by attitude and perceived behavioral control toward ESHS. The results are in line with previous studies, which have verified the positive influences of attitude and perceived behavioral control on consumer intention (Yadav and Pathak 2016; Chen and Hung 2016; Wang et al. 2018b). If users hold a positive evaluation toward ESHS, then they are likely to embrace such services. However, subjective norm does not play an important role in the adoption intention of consumers as hypothesized. This finding is inconsistent with published works that have verified the positive role of subjective norm on the adoption intention of consumers (Wang et al. 2014; Ru et al. 2019). ESHS are relatively private, and the opinions of important people are less likely to influence consumer decisions compared with public goods or services. Moreover, the sample of this questionnaire survey mostly consists of young people, and they are less bound by group pressure.

The results also indicate that environmental knowledge exerts a positive influence on the attitude and intention of consumers toward ESHS, which is line with published literatures (Steg and Vlek 2009; Pothitou et al. 2016; Mohiuddin et al. 2018). If consumers have knowledge of environmental protection, they can determine that smart home services are eco-friendly solutions that mitigate environmental issues. Egoistic consumers are less likely to switch altruistic values to eco-friendly behavioral intentions and behaviors. Under the context of ESHS, if consumers have more environmental knowledge, they are more likely to adopt ESHS. Furthermore, this research verifies a significant and positive effect of ambiguity tolerance on consumer intention toward adopting ESHS. The high ambiguity tolerance of consumers increases their likelihood to adopt ESHS. Eco-friendly smart homes are new high-tech services, and many consumers have not been previously exposed to such services. Thus, they may perceive time risk, security risk, and performance risk toward adopting ESHS. If consumers’ ambiguity tolerance is higher, they have a higher tolerance for the perceived uncertainty associated with using ESHS, the more likely they are to adopt ESHS.

This study confirms that compatibility positively influences consumer intention toward ESHS, and the result verifies the claims of previous studies (Susanto and Goodwin 2013; Hubert et al. 2019). The consistency of ESHS with existing systems is an important consideration for users to adopt such services. Moreover, compatibility reduces unnecessary installations and use risks of ESHS, thus increasing the psychological trust of consumers. If the compatibility of ESHS is higher, there are less barriers for consumers to use ESHS. However, living habits play a negative role in consumers’ adoption intention under ESHS context. ESHS are novel for most consumers, and the new adoption will bring changes to their living habits. If they resist lifestyle changes, consumers are less likely to accept ESHS.

5.2 Implications

This research provides a new perspective on the adoption intention of consumers instead of the technological aspects ESHS. On the basis of the model verification results and research discussion in this study, several implications can be presented to ESHS providers to promote the application of ESHS and to create economic and social values. Attitude exerts significant and positive effect (\(\beta\) = 0.190, p < 0.001) on consumers’ intention toward adopting ESHS. Governments can introduce smart equipment into public places to increase the publicity and guidance of smart home life concepts. ESHS suppliers can also establish smart home experience hall, and increase the advertising of smart home, which is conducive to consumers’ formation of good attitude toward smart home.

Perceived behavioral control exerts positive effect (\(\beta\) = 0.227, p < 0.01) on consumers’ intention under ESHS context. Governments can adopt reasonable policy measures to set guiding prices for ESHS. Reasonable prices and wider access promote the increased behavioral control of consumers. In addition, ESHS manufacturers can expand the supply channels of smart home and increase the consumers' contact with smart home. And whether it is online or offline sales channels, suppliers should guide consumers to use ESHS. This series of measures help to the enhancement of consumer's perceived behavioral control of ESHS.

Environmental knowledge significantly and positively influences (\(\beta\) = 0.084, p < 0.01) consumers’ intention toward adopting ESHS. Government are supposed to increase the publicity of environmental protection knowledge and encourage more consumers to adopt energy-saving products or services. These initiatives help to increase the environmental knowledge of consumers. At the same time, ambiguity tolerance also positively affects (\(\beta\) = 0.147, p < 0.001) consumers’ intention toward adopting ESHS. Considering the uncertain situation that consumers may face, providers can ease the resistance of consumers to ESHS by providing reliable products and excellent after-sale services. In addition, the perceived risks and insufficient understanding of consumers toward eco-friendly products and services are the sources of ambiguity. Providers should improve technical safety to eliminate consumer concerns.

Given that compatibility is an important consideration for consumers to adopt ESHS (\(\beta\) = 0.154, p < 0.001), providers should strive to improve the level of adaptability of ESHS to existing systems. In addition, living habits negatively influence (\(\beta\) = − 0.300, p < 0.001) consumers’ intention to adopt ESHS. Consumers who are used to their existing lifestyle and state are more reluctant to try new products or services. To encourage more consumers to adopt ESHS, living habits and preferences of potential consumers should be fully investigated and understood to reduce consumers’ discomfort in the use of such services.

This paper also has some theoretical significance. First, this study is the one of the first to explore consumers’ intention toward adopting ESHS in China, which expands the regional research on consumers’ acceptance toward ESHS. Second, this study innovatively uses TPB theory to explore consumers' intention to adopt ESHS, while previous studies mostly use TAM theory to investigate consumers’ adoption intention. Third, TPB has been criticized for being a rational theory, the integration of environmental knowledge and into TPB can further consider the impact of irrational factor on consumers’ acceptance toward new products or services. In addition, there are few studies exploring the influence of ambiguity tolerance on consumers’ decision-making, the integration of ambiguity tolerance also includes the personal characteristic into TPB, which can further enhance the explanatory power of TPB. Moreover, this study provides a reference for further exploring the consumer behavior mechanism toward adopting ESHS in future.

5.3 Limitations and further research direction

This study provides a comprehensive picture in understanding consumer intention toward ESHS. However, several limitations need to be further addressed in future research. First, 53.9% of the respondents are between the ages of 20 and 29, while 52.6% of the respondents have earned a junior college or university degree. The distribution of age and educational background should be further balanced in later studies. Second, this study only explores the adoption intention of consumers toward ESHS. Although intention is an important indicator of behavior, a gap exists between the behavioral intention and actual behavior. Third, additional influencing factors should be explored more deeply than be limited to the variables added in this article. Finally, if this research can be conducted in different countries, scholars can further explore cultural differences in the context of ESHS adoption intention.