Homestays in China: Mediator effects of brand perception upon host-guest interaction and Tourist behavior intention

Abstract

Background

As for homestays in China, the embedded distinctiveness is enriched by China's uniqueness in geographical conditions and cultural connotations. Consequently, for the sake of enhanced influences on the homestay market, an ascending number of peculiar homestay brands are forged and developed to raise their identification, recognition, and attraction.

Purpose

This paper probes into the relevance involving host-guest interaction, brand perception, and tourist behavior intention.

Methodology

The social exchange theory and brand relationship theory are adopted for the questionnaire survey to collect the data which are processed and verified through principal component analysis and linear regression.

Conclusion

First, host-guest interaction of homestays casts positive effects on tourist behavior intention. Secondly, host-guest interaction of homestays positively impacts brand perception. Thirdly, homestay brand perception imposes mediator effects on host-guest interaction and tourist behavior intention.

Contribution/Value: First, more emotional interpersonal communication is to be input. Secondly, the homestay brand perception that could never be replicated is to be shaped.

1. Introduction

Homestays, represented by Bed & Breakfast in Great Britain, originating from Europe and the United States with the initial definition going as “the accommodation product for experience in a touristic environment”. A homestay refers to an accommodation facility that is transformed by a homestay host from an idle residential resource in a rural area to provide tourists with leisure, recreation, and a wide range of local culture [1].

China, a country with a long history and abundant culture, has become a magnet to myriads of tourists from home and abroad. Those tourists living in homestays instead of traditional hotels could not only access cozy accommodations but even a deep experience in the customs and culture of China. On this land, each tourist attraction is coupled with local distinctiveness in history and culture. Whether unique ethnic customs or royal gardens in ancient China, diversity and peculiarity are, likewise, bestowed upon homestays in China, let alone archaic riverside towns and mysterious rustic villages. The year 2016 has marked the inception of significant growth in China's homestay industry. As shown from the data in A Survey Report of Development Analysis and Investment Prospects on China Homestay Industry (2023–2030), the average accommodation rate in 2019 hit 64 %, while declined to 29 % in 2020 during the outbreak of COVID-19. Notwithstanding this, the industry quickly came back to life as the epidemic was contained. According to the further statistical data in A Report of Big Data on Homestay Industry in Chinese Mainland and Taiwan (2023), Chinese Mainland has been deluged with a total of 97,730 homestays, respectively up by 32.1 % from the year 2019 and 42.1 % from 2022. What's more, the data in An Insights Report of Market and Consumption on China's Online Hotel Accommodation suggest that the majority of online accommodation reservations are targeted at hotels, homestays, apartments, and serviced apartments, of which the first one is topped at the largest percentage (42 %), and followed by homestays (34 %). judging from the data above, it's safe to say that despite the of impacts the epidemic, both resilience and recovery are extraordinarily manifested in the homestay industry which has enjoyed an ascending position in the overall market of tourist accommodation. As inferred from the comparison between traditional hotels and homestays in market shares, a growing competition position in this market has been bestowed upon homestays [2]. Consequently, homestays have served as an integral and highlighted part of tourism and accommodation in China.

The following grounds unveil the pivotal role of tourism in China's economy. The World Travel & Tourism Council has anticipated that a total contribution of CNY 9.9 trillion would be generated by travel and tourism in China to its GDP, up by 150 % from the same period in the previous year. Behind this big number is highlighted China's rapid recovery in travel and tourism within a short time, and the surging percentage of contributions to the economy of the whole nation as well. Second, according to the data released by China Tourism Academy, the aggregate of domestic travel in the first half year of 2023 amounts to 2.384 billion with revenues reaching CNY 2.3 trillion in this regard, respectively up by 63.9 % and 95.9 %. Based on the two data, the momentum of tourist revival is further ascertained. In particular, the fold increase in tourist revenues manifests the strong aspiration and high confidence that consumers cherish regarding travel. It's safe to say that tourism stands in an ascending position in China's economy and plays a pivotal role in boosting domestic consumption and economic growth, as enormous contributions flow to GDP accompanied by the year-on-year growth of tourism revenues [3,4].

As for homestays in China, the embedded distinctiveness is enriched by China's uniqueness in geographical conditions and cultural connotations. Consequently, an ascending number of peculiar homestay brands that echo distinctive geographical locations and regional cultures are forged and developed in terms of identification, recognition, and attraction. However, despite the attention paid to homestays from aspects of business model, tourist experience, and tourist satisfaction, few types of research probe into homestay brand perception. Of existing literature documents, the majority of perceptions of China homestay brands are undertaken from such perspectives as business model, tourist experience, tourist satisfaction, local cultures, and community participation [[5], [6], [7], [8], [9], [10]]. However, few of them thoroughly probe into the perception of China homestay brands, though only Zu et al. design customized platforms under the multi-information fusion sensor network, in which the important role that customized marketing plays in homestay marketing is underscored [11].

As benefits from the travel of significant activities in modern communities, not only do residents in tourist destinations and those players in the tourism market harvest economic revenues, but the tourists and residents could even seize opportunities for in-depth interaction and communication. Host-guest interaction, referring to communication and exchanges between hosts and tourists during tourist activities, involves common service communication, but even, from a deeper perspective, contains interaction concerning culture, customs, and values. As the consequences of behaviors and choices that tourists decide during tourist activities, tourist destinations are vividly manifested, and even, their deep knowledge and expectations concerning tourist service and tourist experience are unveiled [12]. Hence, the insights into host-guest interaction, tourist behavior intention, and driving forces behind them will facilitate understanding and optimization of homestay development.

In the circumstance of customers’ aspiration for customized tourist experience, brand perception has recently played an extremely vital role in homestay kind of accommodation with distinctive culture and local experience. A brand, more than a name or emblem, conveys the core value and peculiarity that are embedded in homestays, thus casting effects upon the choice and satisfaction of tourists [13]. Given the above, both brand building and marketing strategies would be understood and optimized in more effective ways if tourist behavior intention and its driving forces could be thoroughly analyzed, especially from the perspective of the homestay brand perception. For this purpose, this paper aims to discuss the mediator effects of brand perception in this regard upon host-guest interaction and tourist behavior intention. As this paper penetrates the impact mechanism of homestay brands, brand strategies, and interaction tactics are tailored for homestay businesses to enhance both the satisfaction and loyalty of tourists. This paper centers on three main tasks, including: (1) identification of the mechanism that how homestay host-guest interaction affects tourist behaviors. (2) understanding of effects cast by homestay host-guest interaction upon homestay brand perception. (3) analysis of mediator effects of homestay brand perception upon homestay host-guest interaction and tourist behaviors. Firstly, the study will scrutinize the influence mechanisms of host-guest interactions on tourist behaviors (loyalty and recommendation), with a nuanced exploration of both online and offline dimensions. Then, the research aims to deepen the understanding of how host-guest interactions impact homestay brand perception, including an examination of brand identification and brand experience. Finally, by exploring how brand perception acts as a mediator, the study seeks to uncover the intricate dynamics between the quality of interactions, the perception of the brand, and the overall perception of tourist behaviors, shedding light on the underlying mechanisms at play.

This paper, based of the analysis on relationships between brand perception, host-guest interaction, and tourist behavior intention in China's homestays, theoretically and practically sheds light on both the sustainable development of China's homestays and the enhancement in international competitiveness of China's tourism. Players and decision-makers in tourism must understand the mechanism of tourists' behaviors and choices before the planning of effective marketing strategies and service optimization. Furthermore, the tourist experience would be enhanced with the aid of research findings, as result of which sustainable tourism development would be pushed ahead. Meanwhile, a new perspective for literature documents regarding tourist behaviors is provided in this research. In other words, as the deep penetration into the emotional experience and policy-making of tourists is manifested in this paper, how homestay brand perception impacts tourists is revealed. Thanks to the analysis of the relationships between tourists' emotional experience and decisions concerning homestay brands, current literature documents in social science are enriched concomitantly with a more full and accurate provision of methods and knowledge.

2. Literature review and theoretical framework

2.1. Basic theory

In this paper, the social exchange theory and brand relationship theory are applied to the discussion about relationships between homestay brand perception, host-guest interaction, and tourist behavior intention.

Social Exchange Theory was initially proposed by Homans in the 1960s. according to this theory, people-to-people communication and interaction belong to exchange behaviors. In other words, a person exchanges with other individuals for the acquisition of some benefits. This theory comprises six major propositions: (1) success. A person is more willing to repetitively take a particular action that will bring remunerations. (2) stimulus. If a person has previously acquired remunerations after taking an action under a stimulus, this person stimulated similarly will be more likely to take the same or similar action. (3) value. If a person could harvest greater benefits from consequences generated by an action, this person will be more prone to take it. (4) deprivation-satiation. If a person frequently accesses a remuneration or reward in a short period, the same remuneration or reward will, afterward, appear less valuable to this person. (5) aggression-approval. If an action taken by a person fails to bring a reward or punishment as expected, this person will be irritated to take an aggressive action. Conversely, if more rewards than expectations are acquired from an action taken by a person, this person will be so elated that he or she will take an action endorsed by others. (6) rationality. From various action plans, the action that brings the maximal value and the largest probability of success is frequently selected by a person. Since then, the growing importance has been attached to the application of this theory in research on tourism. As proposed by Laxson, the host-tourist gap in a tourist destination will result in distrust that, hence, imposes detriments upon the development of relationships between the two [14]. Given the above, the affinity between the host and the tourist could reinforce the tourist's empathy for the host, and thus, the tourist could acquire perception from interaction to affect behaviors and satisfaction concerning accommodation in a homestay [15].

In its core idea, it's contended that we unceasingly attempt to maximize benefits at minimized costs during social interaction which is stimulated by our aspiration for reward acquisition and retribution evasion. The key concept of Social Exchange Theory is composed of such main constituents as rewards and costs, and comparison levels and its substitutes. Rewards and costs stand for two major factors that individuals take into account during social interaction, the former of which may either be embodied in material like money, or non-material ways like recognition or compliments. Costs might refer to the loss of materials or inputs of time and energy. Comparison level, based on previous experience and social orientation, represents satisfaction that an individual hankers to acquire from a relationship, while comparison level substitutes, as individuals think, refer to optimal rewards that they could acquire from other relationships. This theory is broadly applied in numerous fields of social science, like organizational behavior, family studies, and intimate relationship studies.

Brand relationship theory is the second theoretical foundation for this paper. The brand theory is a significant issue in management and marketing, in which the brand relationship theory plays a major role. As contended by Blackston, the brand relationship refers to the interplay between a brand's attitudes toward consumers and those of consumers toward brands. For one thing, a brand shows its image in front of consumers (namely, objective brand); for the other, a consumer shapes their attitudes toward the brand image (namely, subjective brand). The brand relationship precisely stands for the interaction between a subjective brand and an objective one. According to Kervyn et al. (2012), a consumer perceives a brand in the same way as the perception of other persons. In turn, the brands as intentional agents framework was proposed. That is, a consumer's perception and capabilities of brand evaluation will catalyze a variety of emotions that will, thus, generate different brand behaviors [16,17].

Concerning the brand relationship theory, its core idea is concentrated on the rigidity and quality of relationships between a consumer and a brand. As proposed in this theory, a brand, more than an emblem, is a medium on which a consumer could an erect emotional connection. Its key concept comprises such major aspects as brand personality, brand emotion, and brand resonance. Brand personality refers to personality traits embedded in a brand, like “youth and vitality” and “sedation”, all of which would impose effects upon the emotional response and loyalty of a consumer. Brand emotion stands for the emotional response that a consumer takes to a brand, like confidence, love, and respect [18]. As the result of brand the deep emotional connection between a consumer and a brand, both loyalty and recommendation intention are catalyzed at a high level. This theory is widely applicable to marketing, advertising, and brand management. As deep relationships between brands and consumers are understood and developed, the enhancement of a brand may be accomplished in terms of market shares, loyalty, and word-of-mouth recommendations.

In this paper, it's believed that both social exchange theory and brand relationship theory are extremely applicable to tourism, especially the homestay brands for this paper. The former theory focuses on the perception and behavior selection stemming from people-to-people communication and interaction. In other words, the host-guest interaction between homestay hosts and tourists imposes effects upon homestay brand perception and behavior intention of tourists; by contrast, brand relationship theory emphasizes the role of brands in the recognition and consuming behaviors of consumers. That is, the homestay brand perception affects the decision of a tourist on accommodation in a homestay, and in turn, the communication between the tourist and the homestay host. Therefore, these two theories are employed as the theoretical foundations in this paper to clarify the relationships involving homestay brand perception, homestay host-guest interaction, and tourist behavior intention.

2.2. Literature review and research hypotheses

2.2.1. Effects of Homestay host-guest interaction on tourist behavior intention

Research on homestay host-guest interaction, has elicited that such interaction plays an indispensable role in the perception and satisfaction of a tourist. In the dimension of the economic exchange theory, Bimonte and Punzo proposed the viewpoint that homestay host-guest interaction stands for the exchange between resource space and revenues [19]. From this economic model, it is disclosed that the owner's interaction and experience with tourists may impose effects on the perception and payment will of tourists [20]. Fan et al. held the idea that host-guest interaction may tremendously affect tourist perception [21]. In particular, the community-oriented contact between homestay hosts and tourists is beneficial to the enhancement of the tourist perception of cultures in tourist attractions, which, however, belongs to the deep perception that a tourist could never shape during the travel alone [22]. From their studies on tourist accommodation satisfaction in two aspects of external homestay environment and host-guest interaction, Lin et al. concluded that the external environment casts little effect upon tourist satisfaction and revisit intention, while accommodation satisfaction and revisit intention are under significant impacts from experience in interaction between homestay hosts and tourists [1,23].

In research on tourist behavior intention, the research made by Tavitiyaman et al. indicated that familiarity with tourist destinations may affect upon the tourist's behavior intention of choosing tourist destinations [24]. If tourists collect more information about tourist destinations on the Internet and hold a stronger perception of destination images, they would present a higher behavior intention of travel [25]. In terms of studies on tourist behavior intention of revisit, Jeong and Shin probed into the effects of intelligent tourist technologies on tourist experience, tourist satisfaction, and tourist intention of revisit, from which it is concluded that only three aspects (information, interaction, and customization) embedded in tourist technologies are the key factors that may influence the tourist satisfaction and the tourist intention of revisit [26]. Soliman probed into the tourist invention of revisit by building the extension model for the theory of planned behavior [27]. He, through the investigation of 302 guests, concluded the research finding that the tourist invention of revisit is significantly influenced by tourist motivation, destination images, and familiarity with destinations [28].

Judging from the scrutiny of research on homestay host-guest interaction and tourist behavior intention, most of them are rooted in ways that both the perception and satisfaction of tourists are raised via homestay host-guest interaction. Deeply as these researchers probe into this regard, they never incisively discuss the relationships between the two. Some of the researches are even destitute of further penetration into ways that such interaction casts effects upon tourist behavior intention, though the significant influences of homestay host-guest interaction upon accommodation satisfaction and revisit intention of tourists are incorporated in these researches.

As indicated in the social exchange theory, homestay hosts purposefully interact with tourists. A tourist expects more information and services to be acquired, while a homestay host craves revenues from the tourist's accommodation at this homestay [29]. The strong interaction between a homestay host and a tourist, like a tourist's quick acquisition of local tourist information from the homestay host, may imbue the tourist with positive feelings about the homestay host [30,31]. Consequently, the effects of homestay host-guest interaction on tourist behaviors are based on the mechanism, that is trust establishment--mutual dependence--mutual benefits. First, more frequent and deeper interaction between homestay hosts and guests, like the provision of local tourist consulting services and other valuable information, facilitates the two to establish trust which underpins the long-standing relationships between homestay hosts and tourists [32,33]. Thus, tourists in frequent interactions more deeply rely on recommendations and services provided by homestay hosts who, likewise, wouldn't survive without the accommodation and feedback of tourists. At this moment, interdependence between the two takes shape [34]. Eventually, tourists may be motivated to repay those homestay hosts (for example, favorable feedback, revisit, or positive recommendation) who provide useful service or information [35]. Judging from the above, the stronger interaction between homestay hosts and guests will more easily instill positive feelings into tourists for higher satisfaction with the accommodation experience. Consequently, tourists will come up with stronger intentions of revisiting and recommendations. Given the above, the hypothesis hereunder is proposed.

H1

Host-guest interaction of homestays imposes positive effects on tourist behavior intention.

2.2.2. Effects of Homestay host-guest interaction on Homestay brand perception

A major issue entangling the academic community is how homestay owners take advantage of host-guest interaction for precise orientation in the market and effective promotion approaches. According to the questionnaire survey undertaken by Jamal et al. on 353 tourists accommodating in homestays, it's concluded that the value of tourist perception depends on functions, emotions, and experience (host-guest interaction) embedded in homestays [36]. Therefore, it's recommended homestay operators place stress on the value of homestays from aspects of functions and experiences. In the circumstance of multicultural tourism, Tasci and Severt probe into three kinds of feedback: homestay hosts' feedback upon themselves, their feedback on guests, and tourists' feedback upon homestays [37]. As inferred from differences in three dimensions, homestay hosts present the most rigorous feedback about themselves, but such self-perception may impact reciprocal perception between tourists and homestay hosts.

In research on homestay brand perception, the brand asset model proposed by Aaker and Equity comprises such factors as brand image, brand recognition, and brand affection, with a focus on the significance of brand perception in customer loyalty and brand selection. It is elicited from the research that a hotel with a stronger brand perception is endowed with more capabilities of attracting customers of high loyalty for its strength in market competition [38]. Mody and Hanks posed the idea that homestays play the role of the alternative to hotels and tend to enjoy higher popularity. Judging from the comparison between hotel brands and homestay brands, homestays are excavating customers with love and loyalty to brands via actuality in brands, existence, and individuals, while hotels only make use of brand actuality [39].

Judging from the above-mentioned review of literature documents, the majority of current research is focused on the ways that homestay host-guest interaction influences satisfaction and revisit intention of tourists, and those studies on brand perception are targeted at hotels and large-sized tourism enterprises. As for the homestay norm of growing popularity, however, the studies on brand perception still seem to remain in stagnation at the primary stage. It is the precise opportunity that we have to seize for more research in this regard. In other words, the studies mentioned above never present in-depth penetration into relationships between homestay host-guest interaction.

In light of the non-standardized residential service provided by homestays, homestay brand building plays a vital role, given that targeted tourists would be hooked on diversity, specialty, and differentiation in homestays only after the homestay brand orientation is clearly defined. As for tourists, the first step to becoming aware of the homestay brand lies in the information search on the Internet which may give birth to the preliminary acquaintance by reservation and consultation with homestay hosts. It's safe to say that the online interaction between homestay hosts and tourists acts as the first phase before the homestay brand perception is shaped. During this period, a tourist will decide whether or not to finally choose this homestay, and thus, deeper brand awareness and experiences will take shape [40,41]. A tourist accommodated at a homestay will communicate with the homestay host face-to-face. According to the brand relationship theory, a tourist in the context of communication and connection with homestay hosts will subjectively cherish emotional perception of the homestay brand, which may, then, cast impacts upon their perception of the homestay brand along the route as follows: online interaction-preliminary brand perception-accommodation check-in-offline interaction - brand experience-deep brand perception [41,42]. Judging from the above, the host-guest interaction of homestays is seen as an influencing factor in the tourist's perception of brands. If by interaction, tourists are impressed by homestay hosts for their extreme thoughtfulness and hospitality, this positive emotional feeling will enhance their perception of homestay brand images. Accordingly, this paper presents another hypothesis as follows.

H2

Host-guest interaction of homestays imposes positive effects on brand perception.

2.2.3. Mediator effects of Homestay brand perception on host-guest interaction and tourist behavior intention

Regarding the effects of host-guest interaction on homestay brands, the study made by Zhang et al. penetrate such effects upon homestay guests from aspects of perception value, customer satisfaction, and word-of-mouth advertising intention, recommending that a person engaged in homestay implement solutions of management and marketing, concerning expectation and preferences of guests [43]. As concluded from 30 intensive interviews made by Casais et al. with a Portuguese landlord of homestays on Airbnb, the close marketing relationships between the host and guests could be established during accommodation via their intimate interaction [2].

As far as the effects of homestay brand perception on tourist behavior intention, the research was made by Lee and Kim to investigate the effects of tourists upon the relationship between Airbnb homestay brands and tourist participation; moreover, the interaction effect was also employed to examine how genders affect participation of a tourist [44]. According to the research findings, tourist participation is impacted by the personalities of tourists, and meanwhile, female tourists display higher participation than males. According to the research of Yoo and Donthu on the effects of hotel brand assets on customer behavior intention, hotel brand assets (brand awareness, brand association, brand knowledge, and brand loyalty) significantly impose positive effects on customer behavior intention (intention for next-time reservation and reputation) [45]. From research made by Kim et al. on the relationship between brand assets and the performance of deluxe hotels or chain restaurants, it is inferred that the positive relationship lies between brand assets and hotel performance, in which hotel brand assets cover brand awareness, brand connection, brand quality and brand loyalty [44].

Through the review of the above-mentioned theses, it's elicited that the question still lies in the ways for the strong marketing relationships between landlords and guests to be transformed into positive brand perception, though the establishment of such relationships via intimate interaction between the two is discussed in previous studies. Notwithstanding the relationships between homestay brand perception and tourist behavior intention analyzed in previous studies, the role of host-guest interaction in this regard is never expressly conveyed therein. In a word, these studies exposed the destitution of analysis on how homestay brand perception plays a mediator role between host-guest interaction and tourist behavior intention.

From the perspectives of the brand relationship theory and the social exchange theory, the first step for tourists to become aware of the homestay brand lies in the online information search, as the result of which the preliminary perception of brands takes shape using online interaction; upon accommodation in a homestay, tourists will make face-to-face conversations with homestay hosts. Thus, the subjective emotional perception will be generated to impose effects upon tourist perception of homestay brands [46,47]. In turn, the tourist perception of brands, like satisfaction with homestay facilities, homestay environment, and service provided by homestay hosts, will trigger the tourist behavior intention in different forms. For instance, with stimulation from the tourist satisfaction about brand perception (like service, facilities, and environment), tourists will give preference to accommodation in the same homestay for the subsequent travel in a locality where the homestay is situated. The roadmap of effects is described as follows: host-guest interaction - homestay brand perception (like services, facilities, and environment) - behavior intention. As inferred by the above, the tourist perception of homestay brands is seen as the mediator variable of the effects of host-guest interaction and tourist behaviors. Hence, the following hypothesis is put form in this paper.

H3

Homestay brand perception casts mediator effects on host-guest interaction and behavior intention.

Therefore, the research model is Fig. 1.

Fig. 1.

Research model.

3. Methodology

3.1. Research design

Given that appropriate research methods and tools play a vital role in the incisive analysis of homestay host-guest interaction, homestay brand perception, and tourist behavior intention, the questionnaires are employed in this study for data collection. The following table overviews the research methods adopted in the study and summarizes their corresponding relations with the design and application of questionnaires (see Table 1).

Table 1.

Research methods and questionnaire design.

Step	Description	Completion via Questionnaire Survey
Design	Determine research objectives, define research variables, and finalize data types	Ensure the alignment of each question with the research objective during the questionnaire design
Development	Scrutinize related literature documents, consult experts in related fields, and develop the first draft of the questionnaire	Use online survey tools to prepare questionnaires, and finish the initial testing
Testing	Formally carry out testing in greater groups of samples	Hand out questionnaires in targeted groups, and collect data
Application	Gain research conclusions from numerically-controlled analysis	Analyze questionnaire data with statistics software, and gain research conclusions from comparison with research hypotheses

3.2. Questionnaire design

For the stringent questionnaire design, this study begins with a review of the previous literature on host-question interaction, brand perception, and tourist behavior intention. Then, the pre-measurement questionnaires are designed based on this research object, concerning the well-development scales devised by authoritative scholars [21,48,49]. Eventually, the science-based questionnaires are formally weaved based on efforts to improve the questionnaires, including consultation with supervisors, professors and doctoral students in the field of tourism to ensure items are normative and logically justifiable, integration with similar items, and enhancement in textual expression.

For this paper, the whole questionnaire is composed of four parts. The first part involves the measurement of host-guest interaction in homestays from aspects like online and offline interaction [21,41,48]; the second is to measure the homestay brand perception from such perspectives as brand awareness and brand experience [48,50,51]; in the third one, the tourist behavior intention is measured in the dimensions of tourist loyalty and tourist recommendation [49,52]; the fourth past contains the basic information about respondents, such as sex, age, occupation, educational background and average monthly salary.

In the questionnaire, all other parts than the basic information about respondents are measured based on the 5-point Likert scale: 1 is for “utterly disagreed”, 2 for “slightly disagreed”, 3 for “not sure”, 4 for “slightly agreed”, and 5 for “utterly agreed”.

3.3. Data collection

Given the manpower and financial capacity, the questionnaires for this paper are collected in online manners. After being generated on “wjx.cn”-a website specializing in surveys, the questionnaire link is quickly shared on WeChat, QQ, and other social media via colleagues, friends, families, and alumni. Based on reviewing questionnaire answers before receiving lucky money, a total of 711 valid questionnaires were returned eventually, in which 8 samples of tourists never showed any experience in accommodation in homestays. After the elimination of the 8 above-mentioned samples according to questionnaire questions, altogether 703 valid samples are finally acquired, with a sample validity rate of 98.87 %. Furthermore, on account of the total acquired samples at over 33 times as many as measured items, the quantity of samples matches with both the questionnaire survey and the requirement to ensure robustness in model verification. Therefore, altogether 703 valid data samples are screened out for this study (see Table 2).

Table 2.

Demographic characteristics of samples.

Characteristics		Sample Quantity	Proportion (%)
SEX	Male	341	47.96
	Female	370	52.04
AGE	Under 18	35	4.92
	19–25	214	30.10
	26–30	178	25.04
	31–45	222	31.22
	46 and above	62	8.72
EDUC	Senior secondary education and below	66	9.28
	Junior college	156	21.94
	Undergraduate	409	57.52
	Master and above	80	11.25
CARE	Student	38	5.34
	Company staff	339	47.68
	Civil servant	57	8.02
	Self-employed entrepreneur	75	10.55
	Teacher or doctor	41	5.77
	Freelancer	126	17.72
	Other	35	4.92
Monthly Income	2000 and below CNY	36	5.06
	2001-4000 CNY	29	4.08
	4001-6000 CNY	65	9.14
	6001-8000 CNY	191	26.86
	8001and above CNY	390	54.85

Based on the thorough analysis of 703 valid data samples, this paper demographically analyzes each detail of the basic characteristics of respondents, comprising sex, age, occupation, educational background, and average monthly income.

From the perspective of sex percentage, 370 female tourists are contained, a bit higher than the number of males (341), which implies that females might be prone to be accommodated in homestays during travel.

If it is looked at in the dimension of age, the majority of tourists who are determined to be accommodated in homestays during travel are aged between 19 and 45. It is to be particularly noted that 31.22 % of total tourists are constituted by those aged 31 to 45, which appears in line with the high consumption capacity embodied in this group of tourists. Accordingly, their aspiration for expenditures on high-quality tourist experience is unveiled.

As shown from the perspective of educational background, the majority of respondents have graduated from higher education institutions, taking up 57.52 % of the total. Accordingly, the results, in the following table cater to their aspiration for expenditures on high-quality tourist experience is unveiled. Accordingly, most of the respondents are well-educated and might be close in their aspiration for expenditures on high-quality tourist experience.

Of occupations, company staff take up the highest percentage at 47.68 %, which, perhaps, correlates to financial standings and tourist selection rights enjoyed by this group of consumers.

Finally, those respondents with an average monthly income above 8000 CNY account for 54.85 % of the total, which, further, discloses that accommodation during travel is oriented to those at medium or high levels of consumption.

On balance, the samples collected for this study are highly matched with the major characteristics of consumers in China's domestic tourism market. Thus, solid data support is provided for the follow-up parts of this study.

3.4. Variables

In this paper, research subjects are defined elaborately to more clearly expound the investigation field and the research object of this paper.

3.4.1. Host-guest interaction

Host-guest Interaction (HGI) refers to the service provided by homestay operators to tourists, the information exchange between homestay operators and tourists, and all kinds of interaction and communication necessary for the establishment of friendship between homestay operators and tourists. It comprises such major aspects as information communication, service exchange, and emotional intercourse. During information communication, a homestay operator provides a tourist with various information like accommodation, catering, and tour routes, as well as replies to questions posed by the tourist. It may involve response time to online consultation, homestay hosts' proactive liaison before accommodation and daily communication during accommodation. A service exchange is focused on the service rendered by the homestay operator to the tourist-based on his or her customized demands, like distinctive food, recommended tour routes, and transport arrangements. Emotional intercourse represents the emotional connection, trust, and understanding between a homestay operator and a tourist. For example, homestay operators are enabled to establish true friendships with tourists by understanding and caring for their feelings, demands, and expectations. This paper employs the ways adopted by Fan et al. to measure the host-guest interaction, and also the ways to differentiate online host-guest interaction from offline forms [21], to measure HGI largely from these aspects: the response time to online consultation, homestay hosts’ proactive liaison before accommodation, acquaintance of homestay hosts with guest requirements on accommodation, communication between homestay hosts and guests after check-in, proactive recommendation of tour routes by homestay hosts and service attitudes of homestay hosts [41,43].

3.4.2. Brand perception

Brand Perception (BP) represents the awareness and experience in all respects concerning brand information. Hence, this paper measures the homestay brand perception in two dimensions: brand awareness and brand experience. First, brand perception, equivalent to knowledge and comprehension about brands, involves understanding specific attributes or qualities embedded in brands. As far as homestays, brand perception may pertain to customized service, auxiliary room facilities, architectural decoration impression, landscape environment features, facilities and equipment, and auxiliary surrounding facilities. Second, the brand experience could be measured in three dimensions, including emotional experience, behavioral experience, and relevance experience. Emotional experience refers to affection and emotion instilled into consumers during their interaction with brands. In terms of homestays, it stands for the perception of affiliation, or coziness and relaxation emanating from accommodation. Behavioral experience equals the behavioral response taken by consumers under interaction with brands. For example, a homestay operator may roll out novel service activities that attract the participation and interaction of tourists. Relevance experience is reckoned as significance and connection that are established between consumers and brands. The relevant experience about homestays may relate to the folk custom experience or the experience in social communication with other tourists. This paper, concerning the brand perception measurement taken by Sheth et al. [48], measures the homestay brand identification in the two central dimensions of this study (homestay brand identification and brand experience) from perspectives of the following: customized homestay service, auxiliary room facilities, architectural decoration impression, landscape environment features, facilities and equipment, and auxiliary surrounding facilities, along with the brand experience measured based on a perception of homestay affiliation, accommodation moods, novel services and activities, folk custom experience and social experience [50,51].

3.4.3. Tourist behavior intention

Tourist Behavior Intention (“BI”) refers to the behavior intention of repurchasing, recommendation, or advertising that is held by tourists after their subjective perception and assessments of services and experience during travel are framed. Tourist behavior intention, more than an easy purchase decision, is composed of various factors ranging from repurchase to word-of-mouth recommendation and even attitudes towards price changes or service matters. To be specific, it is embodied in two indicators: loyalty and recommendation of tourists. The former unveils a tourist's inclination to choose again or keep choosing a tourist destination, hotel, or other service. Loyal tourists, more often than not, hold a connection with brands or destinations deeply and emotionally insomuch that they may be willing to neglect other choices which, though, overshadow those they have made in some respects. Second, in addition to direct repurchase, recommendations to others, or experience sharing with others projects the tourist behavioral intention. Whether word-of-mouth recommendations or experience sharing on social media, such recommendation originates from the satisfaction, experience, and feelings of tourists. This paper, concerning the measurement of tourist behavior intention taken by Boulding et al. [49], measures the tourist behavior intention in two dimensions concerned (tourist loyalty and tourist recommendation) from these aspects: tourist intention of subsequent accommodation in the same homestay, tourist attitudes toward price increase, tourist solutions to encountered service problems, possibilities of tourist's recommendation to others, and possibilities of tourist's sharing on social media [53] (see Table 3).

Based on the delineation and analysis of the research variables provided above, Table 3 summarizes the specific test items for each variable. The independent variable, Host-Guest Interaction, is measured along two dimensions: Online Interaction and Offline Interaction, totaling 6 test items. The mediating variable, Homestay Brand Perception, is assessed through the dimensions of Brand Identification and Brand Experience, with a total of 6 test items. The dependent variable, Tourist Behavior, is measured across the dimensions of Loyalty and Recommendation, comprising a total of 12 test items (see Table 3).

Table 3.

Summary of test items for variables.

Variable	Code	Items
Online Interaction	ONI1	Online inquiries receive prompt responses
	ONI2	After booking the homestay, the host initiates communication
	ONI3	The host inquiries about my check-in preferences
Offline Interaction	OFI1	There is interactive communication with the host, creating a relaxed atmosphere
	OFI2	The host enthusiastically recommends local tourist routes and delicacies
	OFI3	Homestay staff undergo professional training, displaying patience, friendliness, and respect in their service attitude
Loyalty	VL1	If I visit the area again, this homestay is my first choice
	VL2	Even if the homestay raises its prices, I would still choose it
	VL3	If I encounter service issues at the homestay, I would communicate with the host
	VL4	If I encounter service issues at the homestay, I would provide feedback online
Recommendation	VR1	If relatives, friends, or colleagues plan to travel, I would recommend this homestay to them
	VR2	I would praise this homestay online or on social media
Brand Identification	BA1	The homestay host caters to personalized needs, creating a warm and homey atmosphere
	BA2	The homestay rooms are clean and tidy, with well-prepared bedding and toiletries
	BA3	The architectural and decorative style of the homestay leaves a positive visual impression
	BA4	The courtyard landscape and surroundings of the homestay are appealing
	BA5	The facilities provided by the homestay meet my satisfaction
	BA6	The transportation, recreational activities, shopping, and dining around the homestay are convenient
Brand Experience	BE1	The homestay gives me a strong sense of belonging
	BE2	The living experience at the homestay helps me relax both physically and mentally
	BE3	The homestay offers novel services and activities
	BE4	The homestay allows me to experience local customs and traditions
	BE5	The homestay provides opportunities for interaction with others
	BE6	The homestay experience introduces me to new friends

3.5. Building of research models

In this study, the regression model, instead of the structural equation model, is employed for two reasons: (1) data requirements. If the structural equation model were adopted, a greater sample quantity would be entailed to acquire estimates stably and reliably. Insufficient samples may bring instability or inaccuracy to SEM results. As far as the sample quantity in this study, it may lag behind SEM preconditions. (2) the regression analysis applies better to this study which is largely aimed to probing into and validating relationships between several key variables, rather than establishment of an entire latent structure. Regression analysis, being more direct and straightforward, aligns more effectively with the core objectives of the study, enabling a more efficient understanding and interpretation of causal relationships among variables.

In this paper, Hypothesis 1 stands for the effects of host-guest interaction of homestays upon tourist behavior intentions. Model 1 is built to verify this hypothesis.

$$BI={\alpha }_0+{\alpha }_{1}HGI+{\alpha }_{2}SEX+{\alpha }_{3}AGE+{\alpha }_{4}EDU$$

$$+{\alpha }_{5}CAR+{\alpha }_{6}INO+{\alpha }_{7}FT$$

(Model 1)

This paper contains Hypothesis 2 pertaining the effects of host-guest interaction of homestays upon brand perception. Model 2 is built to verify this hypothesis.

$$BP={\beta }_0+{\beta }_{1}HGI+{\beta }_{2}SEX+{\beta }_{3}AGE+{\beta }_{4}EDU$$

$$+{\beta }_{5}CAR+{\beta }_{6}INO+{\beta }_{7}FT$$

(Model 2)

Hypothesis 3

is presented in this paper to refer to the mediator effects of homestay brand perception upon host-guest interaction and customer behavior intention. Model 3 is built to verify this hypothesis.

$$BI={\gamma }_0+{\gamma }_{1}HGI+{\gamma }_{2}BP+{\gamma }_{3}SEX+{\gamma }_{4}AGE+{\gamma }_{5}EDU$$

$$+{\gamma }_{6}CAR+{\gamma }_{7}INO+{\gamma }_{8}FT$$

(Model 3)

In the model mentioned above, HGI stands for the host-guest interaction of homestays, BP is the abbreviation of brand perception, and BI represents tourist behavior intention. Other variables are control variables, including “SEX” for sex, “AGE” for age, “EDU” for educational background, “CAR” for career, “INO” for monthly income, and “FT” for travel ways.

4. Data analysis

4.1. Principal component analysis of questionnaire

4.1.1. Analysis of reliability

In this study, Cronbach's $\alpha$ coefficient is applied to test the heterogeneity of each question in questionnaire scales. To enhance the accuracy of scales, those questions that impair measurement reliability are eliminated with Corrected-Item Total Correlation (CITC). For this study, SPSS24.0 is employed to analyze the reliability of sub-variables involving such variables as homestay host-guest interaction, tourist behavior intention, and brand perception. The results are indicated as below in Table 4.

Table 4.

Cronbach's $\alpha$ coefficients and CITC coefficients of questions in scales.

Variable	Code	Overall Relevance Coefficients of Questions	$\alpha$ Coefficient after Question Elimination	$\alpha$ Coefficient
Online Interaction	ONI1	0.690	0.727	0.816
	ONI2	0.689	0.728
	ONI3	0.631	0.784
Offline Interaction	OFI1	0.637	0.740	0.802
	OFI2	0.691	0.762
	OFI3	0.615	0.682
Loyalty	VL1	0.631	0.700	0.768
	VL2	0.658	0.719
	VL3	0.623	0.685
	VL4	0.500	0.743
Recommendation	VR1	0.645	–	0.784
	VR2	0.645	–
Brand Identification	BA1	0.614	0.833	0.853
	BA2	0.611	0.834
	BA3	0.711	0.815
	BA4	0.695	0.818
	BA5	0.642	0.828
	BA6	0.563	0.842
Brand Experience	BE1	0.483	0.865	0.858
	BE2	0.691	0.827
	BE3	0.717	0.822
	BE4	0.700	0.825
	BE5	0.694	0.826
	BE6	0.619	0.840

As shown in Table 4, each variable is endowed with Cronbach's $\alpha$ coefficient respectively at 0.816, 0.802, 0.768, 0.784, 0.853, and 0.858, all of which are greater than 0.7. It's suggested that high congruity lies in questions of each variable, conforming with the basic requirements of questionnaire design.

The results of CITC coefficient testing disclose that those questions with relevance to total questions below 0.6 include CITC coefficient 0.500 at VL4 in the dimension of loyalty; CITC coefficient 0.563 at BA6 in the dimension of brand identification; CITC coefficient 0.583 at BE1 in the dimension of brand experience. Given that the elimination of these three questions will raise Cronbach's $\alpha$ coefficient in each dimension, these three questions will be excluded from the final questionnaire.

4.1.2. Analysis of validity

Before the measurement of 21 test items based on the factor analysis in terms of three major variables in this study (host-guest interaction of homestays, brand perception, and tourist behavior perception), both KMO test and Bartlett's test of sphericity are required for each test item to verify the correlation between each of them, so as to determine whether or not these items are fit for the factor analysis. In the context below, Table 5 represents the analysis KMO test and Bartlett's test of sphericity. Judging from the data in the Table, Online Interaction (ONI) shows KMO statistical magnitude at 0.712, Offline Interaction (OFI) with KMO statistical magnitude at 0.702, Brand Identification (BA) with KMO statistical magnitude at 0.815, Brand Experience (BE) with KMO statistical magnitude at 0.845, tourist Loyalty (VL) with KMO statistical magnitude at 0.683, and tourist Recommendation (VR) with KMO statistical magnitude at 0.650, all of which are greater than 0.6. The results unveil the high level of each variable, indicating variables are fit for the factor analysis. Furthermore, each variable has the remarkable Bartlett's statistical magnitude at a 1 % significance level. Thus, the strong correlation lies between each variable, which tells all of the variables are fit for the factor analysis (see Table 5).

Table 5.

KMO and Bartlett's test of sphericity for test items.

Variable	KMO Measure of Sampling Adequacy	Bartlett's Test of Sphericity
		Approximate Chi-square	Degree of Freedom	Significance
ONI	0.712	737.865	3	0.000
OFI	0.702	673.423	3	0.000
BA	0.815	1411.257	10	0.000
BE	0.845	1594.436	10	0.000
VL	0.683	476.571	3	0.000
VR	0.650	376.176	3	0.000

4.1.3. Factor analysis of questionnaire

4.1.3.1. Extraction of common factors

After both the questionnaire reliability and the validity of test items of variables are tested, the principal component analysis is employed to extract common factors from each variable. Eventually, those factors with characteristic value greater than 1 are extracted from test items. Table 6 hereunder shows the total variance explained by questionnaire factors.

Table 6.

Total variance explained of factors.

	Component	Initial Characteristic Value	Variance Percentage	Cumulative Percentage (%)	Total	Variance Percentage of Sum of Squares of Extracted Weights	Cumulative Percentage (%)
Online Interaction	ONI1	2.197	73.232	73.232	2.197	73.232	73.232
	ONI2	0.452	15.066	88.298
	ONI3	0.351	11.702	100.00
Offline Interaction	OFI1	2.148	71.609	71.609	2.148	71.609	71.609
	OFI2	0.484	16.154	87.763
	OFI3	0.367	12.237	100.00
Loyalty	VL1	1.985	66.157	66.157	1.985	66.157	66.157
	VL2	0.559	18.626	84.782
	VL3	0.457	15.218	100.000
Recommendation	VR1	1.645	82.230	82.230	1.645	82.230	82.230
	VR2	0.355	17.770	100.00
Brand Identification	BA1	3.075	61.509	61.509	3.075	61.509	61.509
	BA2	0.748	14.962	76.471
	BA3	0.481	9.628	86.099
	BA4	0.374	7.481	93.580
	BA5	0.321	6.420	100.000
Brand Experience	BE2	3.255	65.101	65.101	3.255	65.101	65.101
	BE3	0.616	12.313	77.415
	BE4	0.463	9.257	86.672
	BE5	0.359	7.190	93.862
	BE6	0.307	6.138	100.000

Extraction method: principal component analysis.

Each value in the table is explained as follows: (1) a characteristic value represents the original variance that could be explained with this factor. (2) a variance percentage unveils the proportion of a variance explained with this factor. (3) a cumulative variance percentage equals the proportion of a variance that could be explained with all factors. Next, an explanation of the specific procedures employed in Principal Component Analysis (PCA): Multiple instances of PCA were conducted on six sub-dimensions derived from three main variables. Factor rotation was applied during the analysis to enhance the interpretation of principal components within these dimensions.

4.1.3.2. Host-guest interaction (HGI)

The common factor extracted from Online Interaction (QNI) has the characteristic value of 2.197, with the explained variance at 73.232 %, which tells that this common factor could explain 73.232 % of the total sample.

The common factor extracted from Offline Interaction (OFI) has the characteristic value of 2.148, with the explained variance at 71.609 %, which tells that this common factor could explain 71.609 % of the total sample.

4.1.3.3. Tourist behavior intention (BI)

The common factor extracted from Loyalty (VL) has the characteristic value of 1.985, with the explained variance at 66.157 %, which tells that this common factor could explain 66.157 % of the total sample.

The common factor extracted from Recommendation (VR) has the characteristic value of 1.645, with the explained variance at 82.230 %, which tells that this common factor could explain the total sample.

4.1.3.4. Homestay brand perception (BP)

The common factor extracted from Brand Identification (BA) has the characteristic value of 3.075, with the explained variance at 61.509 %, which tells that this common factor could explain 61.509 % of the total sample.

The common factor extracted from Brand Experience (BE) has the characteristic value of 3.255, with the explained variance at 65.101 %, which tells that this commonly factor could explain 65.101 % of the total sample.

Judging from characteristic value and variance percentages in the table, high explainability is bestowed upon these three common constructed major factors. Thus, it's established that the factor analysis adopted for this study is endowed with reliability in terms of the explainability of common factors.

4.1.3.5. Calculation of factor scores

The following Table 7 indicates the component matrix and the component score matrix after common factors are extracted from primary variables in each dimension. Each value in the table is explained as follows: (1) a factor loading refers to the coefficient of relevance between an observed variable and a specific factor, describing the rigidity of relevance between the two. The absolute value of factor loading” gets closer to 1, and the stronger relationships between the observed variable and its corresponding factor will be signified. (2) a component score coefficient is used to transform an observed value into the factor score. In turn, the observed value of a principal factor is acquired. The equation hereunder is adopted based on Table 5 to calculate the total score of each principal factor.

Table 7.

Component matrix and component score matrix.

Online Interaction	Factor Loading	Component Score Coefficient	Offline Interaction	Factor Loading	Component Score Coefficient
ONI1	0.869	0.395	OFI1	0.840	0.391
ONI2	0.867	0.395	OFI2	0.873	0.406
ONI3	0.831	0.378	OFI3	0.825	0.384
Loyalty	Factor Loading	Component Score Coefficient	Recommendation	Factor Loading	Component Score Coefficient
VL1	0.835	0.420	VR1	0.907	0.551
VL2	0.819	0.413	VR2	0.907	0.551
VL3	0.786	0.396
Brand Identification	Factor Loading	Component Score Coefficient	Brand Experience	Factor Loading	Component Score Coefficient
BA1	0.756	0.246	BE2	0.784	0.241
BA2	0.771	0.251	BE3	0.833	0.256
BA3	0.848	0.276	BE4	0.831	0.255
BA4	0.806	0.262	BE5	0.827	0.254
BA5	0.734	0.239	BE6	0.756	0.232

Extraction method: principal component analysis.

Table 7 presents the factor loading matrix and component score coefficient matrix derived from factor analysis for six dimensions of three primary variables. The factor loading represents the strength of the relationship between each variable and each factor. The second column of component score coefficients indicates the contribution of each observation to each factor.

Based on the results from Table 7, the formula for calculating factor scores is derived. This computation method integrates contributions from various variables to each factor, yielding the scores for each observation across different factors. The weights are determined by the common factors explaining the variance in the sample (characterized by eigenvalues >1), and the coefficients for each observation are obtained from the component score coefficient matrix in Table 7.

4.1.3.6. Host-guest interaction (HGI)

ONI = (0.395*ONI1+0.395*ONI2+0.378*ONI3)/0.73232

OFI = (0.391*OFI1+0.406*OFI2+0.384*ONI3)/0.71609

HGI = ONI + OFI

4.1.3.7. Tourist behavior intention (BI)

VL = (0.420*VL1+0.413*VL2+0.396*VL3)/0.66157

VR = (0.551*VR1 + 0.551*VR2)/0.82230

BI = VL + VR

4.1.3.8. Homestay brand perception (BP)

calculate the final results of brand perception in two dimensions.

BA = (0.246*BA1+0.251*BA2+0.276*BA3+0.262*BA4+ 0.239*BA5)/0.61509

BE = (0.241*BA2+0.256*BA3+0.255*BA4+0.254*BA5+0.232*BA6)/0.65101

BP = BA + BE

4.2. Regression analysis

4.2.1. Effects of Homestay host-guest interaction on tourist behavior intention

For Model 1, regression is applied to verify the effects of homestay host-guest interaction upon tourist behavior intention, with results shown in the following Table 8. Model 1 shows adjusted R² at 0.351, illustrating the applicability of Model 1 in the explanation of the variance at 35.1 %. In turn, the satisfactory goodness of fits bestowed upon Model 1 is manifested. Its statistical value F at 55.35 is displayed with value P below 0.001, indicating the holistic significance of this regression model (see Table 8).

Table 8.

Empirical regression analysis.

			Part_A	Part_B	Part_C
			Hypothesis 1	Hypothesis 2 BP	Hypothesis 3
Independent Variable	HGI	Coefficient	0.581	0.699	0.218
		T value	19.36	21.22	6.89
		P value	<0.001	<0.001	<0.001
Mediator Variable	BP	Coefficient			0.519
		T value			18.28
		P value			<0.001
Control Variable	SEX	Coefficient	0.034	−0.132	0.103
		T value	0.46	−1.58	1.65
		P value	0.647	0.115	0.100
	AGE	Coefficient	−0.024	0.108	−0.081
		T value	−0.31	1.24	−1.23
		P value	0.759	0.215	0.218
	EDU	Coefficient	0.160	−0.061	0.192
		T value	1.70	−0.59	2.49
		P value	0.089	0.552	0.013
	CAR	Coefficient	−0.061	−0.046	−0.037
		T value	−0.78	−0.54	−0.58
		P value	0.434	0.587	0.565
	INC	Coefficient	0.049	0.154	−0.030
		T value	0.50	1.44	−0.38
		P value	0.614	0.150	0.701
	FT	Coefficient	0.051	0.129	−0.015
		T value	0.68	1.54	−0.24
		P value	0.496	0.123	0.808
AdjustedR2			0.351	0.392	0.561
ANOVA	Test Value F		55.35	65.88	113.40
	P value		<0.001	<0.001	<0.001
N			703	703	703

In Table 8, Part_A represents the regression pertaining correlation between the explanatory variable (home-guest interaction of homestays, HGI) and the explained variable (tourist behavior intention, BI), in which the regression coefficient ${\alpha }_1$ is demonstrated at 0.581, and test value T at 19.36 with value P below 0.001. That is to say, the home-guest interaction of homestays significantly casts positive effects upon the tourist behavior intention at the 1 % significance level: the stronger home-guest interaction of homestays will be more likely to catalyze the tourist emotion and the high tourist satisfaction with accommodation. Thus, the intentions of both revisit and recommendation will be generated at a higher level. That is to say, Hypothesis 1 in this paper is verified, therefore.

As inferred from the result of the control variable-education background “EDUD” (regression coefficient 0.160, test value T 1.70, P < 0.10), this control variable at the 10 % significance level is positively correlated with tourist behavior intention. Thus, a tourist with a higher educational background may be more likely to recommend and accommodate a homestay next time.

4.2.2. Effects of Homestay host-guest interaction on Homestay brand perception

For Model 2, regression is applied to verify the effects of homestay host-guest interaction upon brand perception, with results shown in the following Table 8. Model 2 shows adjusted R² at 0.392, illustrating the applicability of Model 2 in the explanation of the variance at 39.2 %. In turn, the satisfactory goodness of fits bestowed upon Model 2 is manifested. Its statistical value F at 65.88 is displayed with value P below 0.001, indicating the holistic significance of this regression model.

In Table 8, Part_B represents the regression pertaining correlation between the explanatory variable (home-guest interaction of homestays, HGI) and the explained variable (homestay brand perception, BP), in which the regression coefficient ${\beta }_1$ is demonstrated at 0.699, Test Value T at 21.22 with value P below 0.001. As inferred from the results above, the home-guest interaction of homestays significantly casts positive effects upon the homestay brand perception at the 1 % significance level. That is to say, tourists would show higher satisfaction with homestay brand perception, as homestay hosts more frequently communicate with guests, and more professionally render services. Hence, Hypothesis 2 in this paper is verified.

4.2.3. Mediator effects of Homestay brand perception on Homestay host-guest interaction and tourist behavior intention

Regression is applied to Model 3 to verify the mediator effects of homestay brand perception, with results shown in the following Table 8 Model 3 shows adjusted R² at 0.561, illustrating the high goodness of fits bestowed upon Model 3. As the mediator variable is vested upon adjusted R² at 0.35 in Model 1 (Hypothesis 1), adjusted R² rises to 0.561. It's safe to say that the model with the mediator variable accounts for the variance at 56.1 %, thus significantly enhancing its goodness of fit. Furthermore, its statistical value F is displayed at 113.40 with P < 0.001, which proves the holistic significance of this regression model.

In Table 8, as Part_C is combined with the mediator variable (brand perception, BP) based on Hypothesis 1, the result shows the regression coefficient ${\gamma }_1$ (the coefficient of tourist host-guest interaction) at 0.218 with test value T 6.89 and value P below 0.001; meanwhile, it shows the regression coefficient ${\gamma }_2$ (the coefficient of brand perception, also the mediator variable) at 0.519 with test value T 18.28 and value P below 0.001. Therefore, the inference from the above proves the hypothesis that brand perception is the major tool for home-guest interaction of homestays to impose its effects on tourist behavior intention. Thus, it's safe to say that more pleasant communication between tourists and homestay hosts will be more likely to affect homestay brand perception via positive emotion. For example, the enhanced satisfaction with homestay facilities, homestay environment, and services provided to homestay hosts will, thus, stimulate the decision on subsequent accommodation in this homestay next time or the recommendation of this homestay to colleagues, friends, or families. Hence, Hypothesis 3 in this paper is verified.

The regression analysis is applied to verify hypotheses in this study, with results shown as below in Table 8.

H1

Host-guest interaction of homestays imposes positive effects on tourist behavior intention.

H2

Host-guest interaction of homestays imposes positive effects on brand perception.

H3

Homestay brand perception casts mediator effects on host-guest interaction and behavior intention.

5. Discussion and conclusion

5.1. Discussion

This paper centers on three main tasks, including (1) identification of the mechanism that how homestay host-guest interaction affects tourist behaviors. (2) understanding of effects cast by homestay host-guest interaction upon homestay brand perception. (3) analysis of mediator effects of homestay brand perception upon homestay host-guest interaction and tourist behaviors.

First, the study on the correlation of host-guest interaction of homestays to tourist behavior intention leads to the conclusion that host-guest interaction of homestays imposes positive effects on tourist behavior intention, which echoes the outcomes of research made by Bimonte and Punzo, Lin et al. and Jeon and Shin on tourist will of payment and revisit. According to the viewpoint proposed by Bimonte and Punzo, interaction and experiences between homestay hosts and tourists may impact tourists’ feelings and their will to pay. As proposed by Lin et al. and Jeon and Shin, experiences in host-guest interaction of homestays are significantly and positively correlated to tourist satisfaction and tourists will revisit. In this paper, it is established by the empirical analysis that the stronger host-guest interaction of homestays will make it easier for tourists to shape positive emotions and high satisfaction with accommodation, and will, thus, give birth to the strong tourist aspiration to revisit. For the study on tourist behavior intention in this paper, the investigation and analysis concerning tourist recommendations are incorporated, given that we currently focus on and depend on social media [1,19]. According to this paper, the effect of recommendation is generated as tourists log on to personal accounts on social media and share their feelings about accommodation in homestays for their travel. Therefore, this paper further analyzes the effect of recommendation in terms of tourist behavior and presents the result of the empirical analysis: host-guest interaction of homestays may impose positive impacts upon tourist recommendation.

Second, in this paper, the study on the correlation of host-guest interaction of homestays to homestay brand perception leads to the conclusion that host-guest interaction of homestays imposes positive effects upon homestay brand perception. The review of previous studies on homestays discloses that no research centers on the correlation between host-guest interaction of homestays and homestay brand perception. However, according to the research conclusion made by Tung et al. [52], the negative reports from public media about host-guest interaction impose influences upon assessments of prospective tourists about homestay images. Hence, Jamal proposes the viewpoint that homestay operators are required to place stress on the value of homestays in functions and experiences, which triggers the author's thinking pertinent to the significance of homestay brand building [36]. As contended in this paper, the homestay brand perception refers to the measured tourist perception of homestays in the tourist industry, the special place of abode in this specific industry. Before the tour to a locality where a homestay is situated, tourists will resort to the Internet to search and grasp information about this homestay, and to some extent, shape awareness of this homestays through communication with the host of the homestay which they have reserved; upon accommodation in the homestay above and the face-to-face communication with the homestay host, tourist will further perceive of the brand in terms of awareness and emotion. According to this paper, therefore, upon host-guest interaction (namely, when tourists communicate and connect with homestay hosts), the subjective emotional perception will strike tourists, which will, in turn, influence tourist perception of homestay brands. Eventually, it was discovered based on the empirical analysis that Tourists would hold a higher perception of homestay brands if homestay hosts more frequently communicated with guests, and rendered services more professionally.

Third, along with the study herein on the mediator effects of homestay brand perception upon host-guest interaction and tourist behavior intention, this paper unleashes the viewpoint that homestay brand perception casts the mediator effects upon host-guest interaction and tourist behavior intention. The review of previous studies on brand perception unveils the framework of brands as intentional agents proposed by Kervyn et al. [54]. In other words, consumer's perception and capabilities of brand evaluation will engender different emotions that will, thus, lead to diverse brand behaviors. With the application of this conclusion to this paper, it is deduced by the author that tourist perception of homestay brands will influence tourist behavior intention. From the review of this study based on the previous two empirical results, it is inferred that, the first step says the subjective emotional perception will be bestowed upon tourists via their online information search and zero-distance communication with homestay hosts to cast impacts upon tourist perception of homestay brands. During the second step, tourist perception of homestay brands, like satisfaction with homestay facilities, homestay environment, and services provided by homestay hosts, will trigger different tourist behavior intentions. For instance, a tourist may decide to choose accommodation services provided by the same homestay during the travel in the locality where this homestay is nested or recommend this homestay to colleagues, friends, or families. Following the above-mentioned research idea, we take further empirical analysis and discover that homestay brand perception imposes mediator effects upon host-guest interaction and tourist behavior intention. As a result, the roadmap of effects takes shape as host-guest interaction - homestay brand perception (brand identification and brand experience) - behavior intention [[55], [56], [57], [58]].

Notwithstanding the foregoing, further in-depth research afterward is suggested for the deep analysis of homestay hosts which, unfortunately, is not incorporated into this paper on account of inadequacy of time and competence. In this paper, the study on host-guest interaction in homestays is largely based on tourist feelings, without any real perception of homestay hosts taken into consideration. As a participant in host-guest interaction, however, homestay hosts take different actions for the sake of their perception of tourists. Therefore, a thorough interview with homestay hosts is required for the study to penetrate more facets of problems in host-guest interaction of homestays.

5.2. Research conclusion

This paper probes into the relevance involving host-guest interaction, brand perception, and tourist behavior intention. This paper centers on three main tasks, including: (1) identification of the mechanism that how homestay host-guest interaction affects tourist behaviors. (2) understanding of effects cast by homestay host-guest interaction upon homestay brand perception. (3) analysis of mediator effects of homestay brand perception upon homestay host-guest interaction and tourist behaviors. Based on a thorough analysis of research data, the statistical results of the verification of each hypothesis are elaborately summarized as follows.

Based on the findings of this paper, three key insights have been identified. First, homestay host-guest interaction has a positive impact on tourist behaviors. Specifically, as the intensity of interaction between hosts and guests increases, tourists are more likely to develop positive emotions and higher levels of satisfaction with their accommodation, consequently leading to a higher likelihood of revisiting and recommending the accommodation. Second, there is a positive influence of host-guest interaction on brand perception, indicating that increased communication between hosts and guests, coupled with the provision of professional services, enhances tourists' satisfaction with the perception of the homestay brand. Third, brand perception of homestay has a mediating effect on the correlation between host-guest interaction and tourist behaviors. That is, the more pleasant the communication process between tourists and hosts, the more they can positively influence their perception of the homestay brand, subsequently influencing behaviors such as choosing to revisit, as well as recommending the homestay to colleagues, friends, or family (see Table 9. Hypothesis Verification Statistics).

Table 9.

Hypothesis verification Statistics.

Hypothesis		Statistic Result	Verification
H1	Host-guest interaction of homestays imposes positive effects on tourist behavior intention	α1 = 0.581, T = 19.36, P < 0.01	Verified
H2	Host-guest interaction of homestays imposes positive effects on brand perception	β1 = 0.699, T = 21.22, P < 0.01	Verified
H3	Homestay brand perception casts mediator effects on host-guest interaction and behavior intention	γ1 = 0.218, T = 6.89, P < 0.01; γ2 = 0.519, T = 18.28, P < 0.01	Verified

6. Summary and inspiration

6.1. Contributions

6.1.1. Theoretical contributions

First, this paper, based on the social exchange theory and the brand relationship theory, projects a new perspective for those researchers on tourism and clarifies internal relationships between the homestay host-guest interaction and tourist behavior intention. Second, with deep penetration into relationships between host-guest interaction and tourist behavior intention, this paper further expands the application scope of social exchange theory in tourism. Third, the “behavior-perception-behavior” theoretical model proposed in this paper serves as the structured framework of research on relationships between host-guest interaction, brand perception, and tourist behavior intention.

6.1.2. Practical contributions

This paper, first, clearly defines the positive effects imposed by homestay host-guest interaction upon tourist behavior intention and brand perception, providing recommendations on homestay practices. Second, while unveiling the mediator role of brand perception between host-guest interaction and tourist behavior intention, it places stress on brand building at the core of homestay business. The insights gained from guest behavior intentions and brand perception emphasize the need for strategic planning that integrates positive host-guest interactions, effective communication, and service excellence to cultivate a strong and favorable homestay brand.

6.2. Recommendations

By penetrating the relationships between host-guest interaction, brand perception, and tourist behavior intention, this paper provides inspiration and recommendations on management for the development of the homestay industry.

First, more emotional interpersonal communication is to be input. This study elicited the finding that as the stronger interaction lies between homestay hosts and guests, tourists would more easily cherish the positive emotion and the high satisfaction with accommodation, which would, in turn, catalyze the strong intentions of revisit and recommendation, and bring to tourists stronger homestay brand perception from offline interaction (for example, face-to-face communication with homestay hosts, and experiences like participation in homestay activities) than online interaction. Based on the analysis above, homestay hosts are required to communicate with tourists more proactively and sincerely, like acquaintance with desired travel of tourists, recommending to tourists the time-saving tour routes, and introducing folk customs and culture to tourists during the accommodation of tourists. Both the charisma of homestay hosts and host-guest interaction with more emotion play vital roles in the attraction of tourists.

Second, the homestay brand perception that could never be copied is to be shaped. As inferred from the research findings of this paper, the higher tourist satisfaction with customized homestay accommodation, auxiliary room facilities, architectural decoration styles, courtyard design, facilities and equipment, and service experience would stimulate the tourists' greater willingness to take action proactively, like a decision on accommodation in the same homestay again, communication with homestay hosts on exposed problems, and recommendation of this homestay to friends or on social platforms. In addition, tourists care more about novel homestay activities and experiences like acquaintance with a greater number of friends than about homestays, environment, and supporting facilities. Therefore, homestay hosts, while enhancing ambient atmosphere, sanitary situation, and hardware facilities, are suggested to focus eyes on the provision of better service experiences, such as consultation and recommendation, customized services (children's slippers, fruit juice, and beverages) and interactive experiences (like handcraft activities), which is the homestay brand logo that could never be replicated.

6.3. Implication and limitations

6.3.1. Research implication

This paper, unlike previous research, proposes a new theoretical structure and analytical methods from the perspective of homestay brand perception. In addition to the effects of tourist behavior interaction on brand perception, the effects imposed by tourist's homestay brand perception on tourist behavioral decisions are further discussed in this paper. Furthermore, with emphasis on the internal motive of tourist behaviors and the reasons behind choices, this study ends with the conclusion that their behavioral decision is occasionally stricken by impacts from emotion and circumstances, rather than adamantly rooted in economic or material benefits.

6.3.2. Limitations and the future research

As for this study, further subsequent researches are recommended on account of its limitations in two following aspects arising from insufficiency of time and competence.

The broad concept of homestay is to be segmented further. Homestays, from the perspective of regions, may be divided into urban and rural homestays. Given the big differences between the two, especially different brand perceptions of tourists toward these two types, it's recommended to analyze the two separately during research.

Host-guest interaction research lacks deep analysis of homestay hosts. The researches on host-guest interaction are based on tourist feelings, with no real perception of homestay hosts taken into account. As participants in such interaction, homestay hosts take different behaviors against their perception of tourists. Given the above, intensively interviewing homestay hosts is recommended to more comprehensively know matters arising from homestays in host-guest interaction.

Undiversified data collection ways impose detriments to research accuracy. A questionnaire survey is never exempt from limitations (like response bias and inaccurate self-reports), though it is frequently adopted for data collection. Furthermore, some crucial information may be omitted for the reason of reliance on undiversified data collection ways.

Given the research limitations mentioned above, the recommendations for subsequent research are proposed hereunder. First, comparative studies on homestay host-guest interaction in different regions and cultural backgrounds are to be undertaken, to discuss the effects of cultural differences on homestay host-guest interaction and tourist behavior intention. Second, longitudinal researches are recommended for more knowledge about the ways secular time changes affect homestay host-guest interaction and tourist behavior intention. Third, diversified data collection ways (like intensive interviews or focus group discussions) are applicable to deeply excavate the feelings and experiences of tourists.

Ethics approval

The studies involving human participants were reviewed and approved by University of Ethics Committee. Written informed consent was not required to participate in this study in accordance with the national legislation and the institutional requirements.

Consent to participate

The studies involving human participants were reviewed and approved by Ethics Committee of University. The participants provided oral informed consent to participate in this study.

Funding

Major Program of Anhui University Scientific Research Project in China. [Grant Number No.2023AH040053].

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

CRediT authorship contribution statement

Hui Chen: Writing – original draft, Visualization, Validation, Software, Resources, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Shuai Fu: Writing – original draft, Visualization, Validation, Software, Resources, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Bei Lyu: Writing – original draft, Writing – review & editing, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Bei Lyu reports financial support was provided by Huaibei Normal University. Bei Lyu reports a relationship with Huaibei Normal University that includes: employment. Author “Bei Lyu” is noted as an AE of this journal “Heliyon". If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Biographies

Bei Lyu (Ph.D.), Assoc. Prof. at School of Economics and Management, Huaibei Normal University of China and Panyapiwat Institute of Management in Thailand, concentrated his research on business and management and psychology. Bei Lyu is the corresponding author and can be contacted at: lyubei@pim.ac.th

Hui Chen, a doctor and lecturer in Law School, Panzhihua University of China, concentrated his research on business and management and psychology. Email:Aaron.chen@ad-vantage.cn

Shuai Fu, co-first author, a doctor in Chinese Graduate School, Panyapiwat Institute of Management, Thailand, concentrated his research on business and management and psychology. Email:fushuai1638@126.com

Appendix A. Supplementary data

The following is the Supplementary data to this article.