The Influence of Participant Subject Factors on Collaboration Effects in the Protection of China’s ICH: The Mediating Role of Relationship Quality

Abstract

Intangible cultural heritage (ICH) is a keystone of sustainable development which, in recent years, has received growing attention from scholars and governments. Previous studies on ICH collaborative protection have focused on ICH connotations and practice, but relatively few studies have focused on the interrelationships among the influencing factors of the collaboration effect. This study introduces a new variable, relationship quality, in an effort to describe the collaborative relationship among participants. The collaborative relationship consists of three dimensions: trust, communication, and fairness. Regarding relationship quality as an intermediary variable, we construct a research framework embracing participant subject factors (collaboration attitude and collaboration ability) and collaboration effects in a case study on collaborative protection of ICH in Regong Thangka, China. A structural equation model, for which data were collected via a questionnaire, is used to assess the strength of the relationships among variables. A total of 416 questionnaires were distributed, and 365 valid completed questionnaires were returned. The findings indicate that relationship quality has a significant influence on collaboration effects in the collaborative protection of ICH. Within the study’s context, collaboration attitude has a significantly positive impact on relationship quality, and it indirectly influences collaboration effects via relationship quality. Collaboration ability has a significantly positive impact on relationship quality and collaboration effects, and it could also impact collaboration effects indirectly via relationship quality. This paper is the first to introduce the concept of relationship quality into the ICH collaborative protection research field, and its findings may have valuable theoretical and practical implications.

1. Introduction

Intangible cultural heritage (ICH) is an important cornerstone of human civilization and a valuable asset of mankind. ICH is defined as oral traditions and expressions, the performing arts, social practices, rituals and festive events, knowledge and practices concerning nature and the universe, and traditional craftsmanship, as well as the sites and spaces in which culturally significant activities and events occur [1]. Tangible cultural heritage is the heritage that has been preserved in the form of concrete objects that embody cultural aspects [2]. The intangibility and tangibility of heritage are relative concepts because their emphasis is different. Tangible cultural heritage is a “static” cultural relic, while ICH is a “living” skill. Therefore, ICH protection is a complex matter which demands input from a wide constituency of stakeholders from various disciplines. Some studies have shown that ICH protection requires effective collaboration among all social groups [3,4], especially stakeholders [5,6]. It is generally agreed that ICH protection relies on multi-subject collaboration, and that such collaborative protection is a vital prerequisite for the sustainable development of ICH [7].

The collaborative governance theory refers to “a governing arrangement where one or more public agencies directly engage non-state stakeholders in a collective decision-making process that is formal, consensus-oriented, and deliberative and that aims to make or implement public policy or manage public programs or assets” [8]. Thus, ICH collaborative protection refers to the collective action of participating subjects. Such action entails establishing communication, consultation, and consensus, thus forming an efficient multi-subject collaboration network for realizing the joint protection of ICH. The means for achieving ICH collaborative protection include the collaborative production of ICH assets, for example, the collaborative development of ICH tourist attractions and other resources, and encouraging collaborative activities carried out by multi-subjects in the promotion of sustainable development of ICH. For example, the collaboration of multiple subjects to protect the ICH of Regong Thangka is a typical case. Regong Thangka is the epicenter of Regong art, one of the first traditional Chinese art forms to be inscribed on UNESCO’s Representative List of the ICH of Humanity. It is a kind of folk scroll painting in the style of New Mantang painting, which takes Tibetan Buddhism, biographies of important Tibetan figures, and major historical events as its creation content. Regong Thangka is regarded as the sacred and traditional art of Tibetan Buddhism and is revered by people who follow Tibetan Buddhism, as well as by art lovers. Regong Thangka originated in the 13th Century and can be found in the villages of Wutun, Niandouhu, Guomari, and Gashari in Tongren City, Huangnan Tibetan Autonomous Prefecture, Qinghai Province, China. In these villages, there is a saying that “family has a painter, household painting Thangka”, meaning that every family has someone to draw Thangka. The protection of Regong Thangka presents a collaborative mode of “government-led, market operation and social participation,” forming the development trend of interdependent collaboration among government, enterprises, and inheritors [7]. In other words, in light of the increasing emphasis on ICH collaborative protection in recent times, there remains some debate as to how to establish an efficient multi-subject collaboration network and about which factors can influence collaboration effects. These unresolved issues constitute the focus of the present study.

At present, most research studies in this field have been focused on the definitions and implementation of the collaborative protection of ICH, and there have been relatively few studies concentrating on the internal mechanisms of collaboration. Some scholars have argued that, in order to enhance the effectiveness of such collaboration, it is necessary to establish a collaborative governance model based on community-led universal participation; however, they have not explored in depth the interactions among the factors which can determine the collaboration effects [9]. Regarding collaboration effects, collaboration theory holds that the attitude and ability of the participants will dictate the effects of collaboration [10]. Nevertheless, it fails to explain the formation mechanisms of the participants’ attitudes and abilities and their roles in producing the collaboration effects. Some scholars have claimed that the degree of communication, trust, and fairness during collaboration will determine the extent to which the collaboration is effective [8]. However, to date they have only verified the influences of different factors on collaboration effects via a unidimensional perspective, rarely considering the interaction among different factors. In short, an integrated framework for analyzing the collaboration effects does not currently exist. In essence, communication, trust, and fairness in collaboration jointly serve as descriptors of the participant’s collaborative relationship quality, in that the collaboration attitude and ability are qualities specifically ascribed to the subject (i.e., participant). During the actual collaboration process, participant subject factors could substantially impact the collaborative relationship, and the interaction of various factors may eventually have an impact on the collaboration effects. In summary, there is a lack of systematic research on the role of, and factors associated with, collaboration effects in the collaborative protection of ICH. In order to explain the formation mechanisms of collaboration effects and explain the interaction between various factors, it is necessary to present an integrated research framework of collaboration effects.

Therefore, based on the collaborative governance theory, this study introduces a comprehensive variable-relationship quality that describes the collaborative relationship among participants. This variable is composed of three dimensions: trust, communication, and fairness. A research framework is constructed where relationship quality is regarded as an intermediary variable, showing the relationship between participant subject factors (i.e., collaboration attitude and collaboration ability) and collaboration effects during the multi-subject collaborative protection of ICH in the context of China. This study takes collaborative protection of the ICH of Regong Thangka, China, as the research object, and uses the structural equation model as the basis for empirical research.

The main contributions of this paper to the field are as follows: (i) it provides an analysis of the influence path of “participant subject factors, relationship quality, and collaboration effects,” and (ii) it expands on the existing collaborative governance theory by providing practical suggestions for the protection and inheritance of ICH. Another contribution of our study is that it breaks through the limitations of one-dimensional research by introducing a comprehensive relationship quality variable to describe the collaborative relationship among participants. In addition, this paper extends the canon of knowledge on relationship quality research, rendering the relationship quality theory more widely applicable.

2. Literature Review and Hypotheses Development

2.1. Theory of Planned Behavior

2.1.1. The Definition and Dimensions of Relationship Quality

Relationship quality is a concept which appears frequently in studies of relationships in both the business-to-customer market and the business-to-business market in the field of business management [11,12]. Relationship quality refers to the appropriateness of a relationship to meet the needs of each party [13], and it is therefore a measure of the quality of the relationship among the participant subjects in the relationship [14]. Song et al. (2006) stated that the term refers to the extent to which the parties in the relationship are involved in a positive, long-term working relationship, including cooperation and conflict resolution [15]. Tran (2020) recently argued that it refers to the degree to which both parties participate in an active, long-term working relationship in the business-to-business market [16]. In the present study, relationship quality is defined as the quality of the interactive relationship among participants involved in ICH collaborative protection as described by a comprehensive evaluation criterion which addresses the needs of different participants in the existing relationship.

However, the conceptualization of relationship quality does not have a unified overall framework, and its dimensions vary among different studies [17,18]. For example, Leonidou et al. (2006) based their measure of relationship quality on commitment, adaptation, information, benefits, satisfaction, trust, and understanding [19]. In their study of relationship quality in e-commerce, Liang et al. (2011) claimed that the dimensions of relationship quality should include satisfaction, trust, and commitment [20]. Li et al. (2019) studied the impact of customer involvement on the correlation between relationship quality and performance in the business-to-business market, basing their measurement of relationship quality on trust, commitment, and transaction-specific investment [21]. In the research field of the collaborative protection of ICH, scholars generally believe that trust, communication, and fairness are the primary elements of a multi-subject relationship [22,23]. Therefore, relationship quality in this study includes three dimensions: trust, communication, and fairness. Trust refers to one’s perception that the other participants’ intentions are positive and stable. Communication refers to the sharing of formal and informal information, as well as the joint communication and consultation on specific matters, among the actors involved in the collaboration. Fairness refers to the equality of participation in decision making, risk taking, and benefit distribution.

2.1.2. Relationship Quality and Collaboration Effects

Relationship quality has an important influence on the effects of the multi-subject collaborative protection of ICH. In this study, relationship quality describes the quality of the collaborative relationship between participants, and consists of the three dimensions: trust, communication, and fairness. The term “collaboration effects” is defined as the final results of the collaborative governance, which in turn can be divided into objective and subjective aspects. Objective aspects refer to the tangible results of the collaboration (in this case, the collaboration on the protection of Regong Thangka), such as an increase in the size of the workforce involved in protecting the ICH of Regong Thangka. The subjective aspects includes the partners’ subjective evaluations of whether their coordination ability, management ability, and resource integration ability have improved [24]. Sun, Zhu and Yang (2021), who researched collaboration among stakeholders involved in agricultural supply chains, pointed out that the perceptions of fairness and trust can affect the quality of relationships and have an impact on the final results [25]. First, in a relationship in which there is a high degree of trust, the participants are more willing to participate in social communication and collaborative interaction [26] and are thus more likely to realize information sharing and improve communication efficiency [27]. Second, effective communication can promote participants’ communication on common goals, collaborative tasks, power allocation, risk distribution, etc., thus reducing the risk of conflict, and facilitating progress towards the collaboration goals [28]. Yang (2018), who studied the collaborative governance mechanism of ICH protection in the era of the Internet, pointed out that adequate communication and sharing among multiple subjects leads to positive collaboration effects [29]. Fynes et al. (2004) claimed that frequent and timely communication are conducive to resolving conflicts and coordinating perspectives and expectations [30]. Finally, fair relations tend to yield less conflict and more trust, stability, and collaboration. The perception of fairness can affect the levels of satisfaction, commitment, and willingness among collaborative participants. Fairness can ensure mutual trust, reduce the occurrence of conflicts, and help maintain stability in the collaborative relationship [31].

In summary, the higher the degree of trust, communication, and fairness among the participants involved in ICH collaborative protection, the better the quality of relationships. Some scholars have mentioned that trust, communication, and flexibility simultaneously exert a positive impact on relationship quality [32]. Moreover, scholars generally agree that relationship quality has a positive effect on the final results. For instance, Tajvidi et al. (2021) noted that relationship quality is positively related to the consumers’ intention to create brand value together on social business sites [33]. Goetz et al. (2021) found that, in the field of education, higher relationship quality is associated with stronger positive emotions [34]. Therefore, in various fields, the higher the level of relationship quality, the more positive the final outcome. Based on the previous findings, the following hypothesis is proposed:

Hypothesis 1 (H1).

There is a positive correlation between the quality of the relationship among multi-subjects and the collaboration effects in ICH collaborative protection.

2.2. Participant Subject Factors

2.2.1. Collaboration Attitude and Collaboration Effects

“Collaborative attitude, as a kind of potential will, is often manifested by the willingness, ability and values of individuals and organizations” [24].

Collaborative attitude can be condensed into three aspects:

First, the degree of interest in the collaborative content [24]. The more interested the participants are in the collaborative content, the more they tend to take the initiative to participate in the collaborative communication, which not only contributes to relationship quality, but also promotes collaboration effects. For example, Kong and Wu (2005) pointed out that in supply chains, a fully collaborative attitude is helpful for communication between partners, and has a positive impact on the integrated production process and the final result [35].

Second, the degree of support in collaboration, which includes material and spiritual support [24]. The higher the participants’ support of the collaboration, the more energy, time, and resources they will invest, which ultimately leads to a high-quality partnership and better collaboration effects. Chae and Ryoo (2017) studied the impact of the collaborative process on the results of collaboration in a cultural heritage disaster-safety work project [36]. Their conclusions showed that the trust, reciprocity, communication, and information-sharing between partners, and the resulting high-quality partnership established, together had a positive impact on the final results. In other words, when a high degree of collaborative support exists among participants, they are more likely to actively participate in setting collaboration goals, and are more communicative and diplomatic, and thus more likely to reach a consensus. Good communication will promote the establishment of mutual trust, ensure fairness during the collaboration process, and lead to positive collaboration effects.

Finally, the degree of participation in the collaborative process [24]. The higher the degree of participation, the more resources will be invested, and the more stable the collaborative relationship will be.

In summary, there exists a positive collaborative attitude when the participants show a keen interest in the collaborative content, are more supportive of one another, and more enthusiastically participate in the collaboration process.

A positive collaborative attitude has a positive impact on the dimensions of trust, communication, commitment, and fairness in the criterion of relationship quality, which in turn has a positive impact on the final collaboration effects. In other words, collaborative attitude has an impact on trust, communication, fairness, etc., between partners, and thus dictates relationship quality and indirectly determines the collaboration effects. At the same time, collaborative attitude can directly affect the collaboration effects. For example, in a case study of community residents, Qiu et al. (2019) pointed out that there were significant positive correlations between the residents’ cognition, attitude towards ICH value, and travel intentions, which together influenced the development of ICH tourism [37]. In addition, some scholars in the field of supply chains management have shown that collaboration attitude directly affects the effectiveness of emergency planning. Further, they claim that collaboration attitude plays an important role in the actual collaboration environment, and that the right attitude will lead to lower transaction costs and have a positive impact on the final collaboration effects [38]. From the perspective of the development prospects of cultural heritage protectors, Ogden (2007) reported that positive attitudes such as understanding, respect, and collaboration are essential for the protection and development of ICH [39]. Based on the above arguments, a positive collaborative attitude can promote the generation of high-quality relationships, thereby reducing the cost of collaboration and promoting the realization of collaboration effects.

Accordingly, the following three hypotheses are constructed:

Hypothesis 2 (H2).

Collaboration attitude has a positive influence on relationship quality in ICH collaborative protection.

Hypothesis 3 (H3).

Collaboration attitude positively influences collaboration effects via the mediator of relationship quality in ICH collaborative protection.

Hypothesis 4 (H4).

Collaboration attitude has a positive influence on collaboration effect in ICH collaborative protection.

2.2.2. Collaboration Ability and Collaboration Effects

“Collaboration ability refers to the participant’s ability to communicate, coordinate, and integrate resources, as well as the experience of collaboration, etc.” [24].

First, a high level of collaboration ability exists when the participants have strong communication and coordination skills that in turn can promote information sharing, reduce conflict, increase mutual trust, and thus reduce transaction costs [40]. Second, the stronger the participants’ abilities to integrate resources, the more capable the group is of pooling the advantageous resources of different participants, and the better the collaboration effects will be. Finally, previous collaboration experience will also enhance the collaboration ability of the participants, rendering them more competent in identifying effective solutions and realizing collaboration goals. According to Hoof and Thiell (2014), higher collaboration ability leads to more innovative solutions and the successful realization of collaboration effects [41].

In summary, the stronger the participant’s abilities to communicate, coordinate, and integrate resources, the stronger their collaboration ability. In addition, rich experience in collaboration among participants leads to high-quality communication, coordination, and integration capabilities, as well as the ability to resolve disagreements during collaboration. Moreover, the stronger the collaboration ability, the more effective the collaboration. For example, Hirsenberger et al. (2019) have claimed that team coordination skills are essential for tackling the challenges and risks involved with cultural heritage protection projects [42]. The stronger their coordination ability, the more capable the participants are of perceiving and addressing risks faced during collaboration, and the better the collaboration effects. According to Ott et al. (2015), “The ability to integrate resources is an important basis for collaborative planning of educational interventions on ICH [43]”. Lee and Roh (2021) noted in their research that the collaboration ability of partners can affect the level of communication between them, promote the sharing and collaboration of information, and thus have an impact on the final market performance [44]. Ergo, collaboration ability can have an impact on the communication dimension of relationship quality and can ultimately dictate collaboration effects. In their study of the medical field, Ansa et al. (2020) reported that communication skills are the core competency of cross-professional collaboration and they can substantially affect the final results [45]. Communication ability directly affects the communication dimension of relationship quality, and ultimately influences collaboration effects.

Based on the above arguments, it can be said that when the participants possess strong collaborative ability—characterized by outstanding communication and coordination ability, resource integration ability, and collaboration experience—they are more capable of developing high-quality relationships with the collaboration partners, leading to positive collaboration effects. In view of the above findings, the following three hypotheses are constructed:

Hypothesis 5 (H5).

Collaboration ability has a positive influence on relationship quality in ICH collaborative protection.

Hypothesis 6 (H6).

Collaboration ability positively influences collaboration effects via the mediating factor of relationship quality in ICH collaborative protection.

Hypothesis 7 (H7).

Collaboration ability has a positive influence on collaboration effects in ICH collaborative protection.

This study in the field of ICH collaborative protection sets out to explore the relationship between participant subject factors (collaboration attitude and collaboration ability), relationship quality, and collaboration effects. The research model portraying the network of relationships between relationship quality, its antecedents (collaboration ability and collaboration attitude), and consequences (collaboration effects) is provided in Figure 1. More importantly, this study is the first to empirically investigate relationship quality in the context of ICH collaborative protection.

3. Methodology

3.1. In-Depth Interview Design

The purpose of using semi-structured in-depth interviews in this paper is to gain an insight into the reality and background of the Regong Thangka ICH collaborative protection, as well as to test the seven hypotheses guiding this study. The semi-structured interview included two open-end questions:

Q1—“Do you have any experience participating in Regong Thangka collaborative protection, and what are the reasons for your participation or non-participation?”

Q2—“What factors do you think will affect the collaboration effects of Regong Thangka collaborative protection?”

We interviewed ten respondents, including five inheritors of Regong Thangka, three entrepreneurs who are inheritors of Regong Thangka, and two government officials responsible for the development of Regong Thangka ICH preservation. All respondents were Regong Thangka stakeholders and had personal reasons for participating in the Regong Thangka collaborative protection efforts.

3.2. Questionnaire—Measurement Criteria

There are many factors that affect collaboration effects in the collaborative protection of ICH. Several important factors that affect the collaboration effects have been identified in the literature review. The questionnaire was divided into two parts: the first part investigated participants’ demographic characteristics, including their gender, age, educational level, annual household income, and occupation; the second part of the questionnaire contained 22 questions used to measure collaboration attitude, collaboration ability, relationship quality (i.e., trust, communication, fairness) and collaboration effects. To that end, a seven-point Likert scale (from 1, ‘strongly disagree’ to 7, ‘strongly agree’) was employed.

The methods for measuring collaboration attitude, collaboration ability, and collaboration effects were adapted from Bo et al. (2019) [24] and Provan and Milward (2010) [46]. We adapted the scale developed by Lee et al. (2020) [47] and Fynes et al. (2004) [30] to measure relationship quality. The relationship quality was conceptualized as a second-order factor consisting of three different but related dimensions (trust, communication, fairness). In addition, to ensure sufficient validity and reliability of the questionnaire, a small-scale pilot survey was conducted with a sample of 30 participants in Tongren City, China. The survey results showed that the questionnaire met the requirements. The questionnaire items are detailed in Appendix A.

3.3. Data Collection

The survey was conducted from 8 May to 1 October 2020. There were 300 offline questionnaires distributed door-to-door to local residents in each village. In addition, an online survey was conducted in China using Questionnaire Star (https://www.wjx.cn/) (8 May to 1 October, 2020), a professional questionnaire survey platform, whereby 116 questionnaires were distributed and later collected. A link to the electronic questionnaire was directed via email to the target population. Both online and offline research questionnaires included filtering questions that not only ensured that participants fit the target group, but also provided more data on the target group, more truly reflecting reality.

A total of 416 completed questionnaires were received, of which 365 were considered valid (87.7% response rate). Participants were all residents of Tongren City and slightly more likely to be male (52.9%). The proportion of males is higher than that of females, which is in line with the local Regong Thangka tradition of males rather than females inheriting the skill. However, with the emergence of schools, enterprises, and other inheritance channels, women also began to participate in the protection and inheritance of Thangka. The largest group or respondents were under the age of 30, constituting 38.1% of the sample. Since Thangka apprentices usually embark on their training when they are about 10 years old, the inheritors in the sample are generally young, and the research sample is considered representative of the population of Thangka inheritors. In terms of educational qualifications, the largest group of respondents had completed junior high school (slightly less than half the sample, constituting 46.0%), while those completing elementary school or below comprised 15.1% of all participants. The education level of the respondents is generally low. On the one hand, the survey site was located in the minority areas of northwest China, where development is relatively backward, and the educational level of the masses is relatively low. On the other hand, the inheritance of Regong Thangka skills emphasizes “starting from the children.” In the in-depth interviews, many inheritors indicated that they could not give consideration to both school education and Thangka learning, so they chose to drop out and continue to learn Thangka. For the above reasons, the education level of the respondents is generally low. As per occupation, the largest group consisted of inheritors of ICH, constituting 48.2%. The majority of reported yearly family incomes fell below RMB 100,000. Complete sample demographics are provided in

Table 1. Descriptive Statistics of the Respondents’ Profile (N = 365).

Measure	Item	n	Percentage (%)
Gender	Male	193	52.9
	Female	172	47.1
Age	Under 30	139	38.1
	31–40	111	30.4
	41–50	74	20.3
	51–60	25	6.8
	61 and over	16	4.4
Educational Qualification	Elementary school and below	55	15.1
	Junior high school	168	46.0
	High school	91	24.9
	Undergraduate	32	8.8
	Master’s degree and above	19	5.2
Occupation	Inheritor of ICH	176	48.2
	Inheritor and Enterprise	40	11.0
	Inheritor and Government official	10	2.7
	Other	139	38.1
Yearly Family Income	Below RMB 10,000	16	4.4
	RMB 10,000–100,000	181	49.6
	RMB 100,000–500,000	131	35.9
	RMB 500,000–1,000,000	25	6.8
	RMB 1,000,000 and above	12	3.3

.

4. Statistical Results

4.1. Data Screening and Measurement Model

In order to process the questionnaire data, a structural equation model was constructed. Specifically, a two-step model was used: a measurement model based on confirmatory factor analysis (CFA) was employed in order to test the reliability and validity among the items and constructs; structural equation modelling (SEM) was used to evaluate the model fit, and for testing the hypotheses.

Prior to testing the measurement model, the data were initially screened using the SPSS statistical software package. No missing or outlier data were found in the sample. In standard practice, kurtosis and skewness values should be less than ±10 and ±3, respectively. For this study, the kurtosis value ranges from −0.711 to +0.282, and the skewness value of the data ranges from −0.506 to +0.278, thus indicating that the data meet the requirements of normal distribution. Next, we used Harman’s single factor test to detect common method bias (CMB). We applied exploratory factor analysis to all items in the data, using the no-rotation method, to identify the resolution of the first factor in the obtained factors. If the test result does not exceed 50%, CMB is considered to fall within an acceptable range. The resolution of the first factor obtained in this study was calculated to be 32.500%, which is below the recommended 50% threshold. Thus, CMB is not considered to be a problem in this study. Finally, because the relationship quality is conceptualized as a second-order factor consisting of three different but related dimensions, “trust, communication, and fairness,” it is necessary to test the multicollinearity of each dimension. It is accepted that if the value of the variance inflation factor (VIF) falls within the 0 to 10 range, there is no multicollinearity. Our results show that the VIF values of all dimensions range from 1.799 to 4.363, and that the tolerance is greater than 0.1; therefore, there is no multicollinearity among the dimensions in this study.

Prior to the structural equation modelling, we conducted confirmatory factor analysis (CFA). CFA of all items was conducted simultaneously to evaluate the validity of the items and the underlying factors. Overall, the fit of the measurement model to the data was found to be good: the chi-square (χ2) result is 438.638 (p = 0.000), and χ2 /df equals 2.193, which is below the suggested threshold of 3.0 (Kline, 1998). The root mean square error of approximation (RMSEA) result is 0.057, below the recommended threshold of 0.08 (Hu and Bentler, 1999). The goodness-of-fit index (GFI) is 0.903, the normalized fix index (NFI) is 0.904, the Tucker–Lewis index (TLI) is 0.93, the comparative fit index (CFI) is 0.945, and the incremental fit index (IFI) is 0.945; all these results exceed the threshold of 0.9 recommended by Bagozzi and Yi (1988) [48].

The validity test addresses convergent validity and discriminant validity. Convergent validity is a measure of whether items effectively reflect the corresponding construct, whereas discriminant validity is a measure of whether two factors are statistically different. As shown in

Table 2. Confirmatory Factor Analysis of the Measurement Items.

Construct and Item	Standardized Loading	SMC	Cronbach’s α	AVE	C.R
Collaboration attitude
A1	0.780	0.609	0.752	0.503	0.751
A2	0.673	0.453
A3	0.669	0.448
Collaboration ability
B1	0.824	0.678	0.857	0.609	0.860
B2	0.867	0.752
B3	0.761	0.579
B4	0.653	0.426
Trust
C1	0.732	0.535	0.827	0.545	0.827
C2	0.740	0.547
C3	0.744	0.553
C4	0.737	0.544
Communication
D1	0.846	0.715	0.928	0.766	0.929
D2	0.903	0.816
D3	0.870	0.757
D4	0.880	0.775
Fairness
E1	0.773	0.598	0.859	0.677	0.863
E2	0.809	0.654
E3	0.883	0.780
Collaboration effects
F1	0.676	0.457	0.851	0.598	0.856
F2	0.833	0.695
F3	0.817	0.668
F4	0.758	0.574

NB. CR = composite reliability; AVE = average variance extracted.

, Cronbach’s α was used to measure the reliability among the items of each variable in the questionnaire. The Cronbach’s α coefficient value ranges from 0.752 to 0.928, which meets the cutoff value of 0.7 and higher. Similarly, the composite reliability (CR) of the latent variables ranges from 0.751 to 0.929. The standardized factor loading of items ranges from 0.653 to 0.903. All three results are statistically significant (p < 0.001). The average variance extracted (AVE) result of the latent variables ranges from 0.503 to 0.766. This finding suggests that a large portion of the variance can be explained by the items, and that convergent validity is satisfactory. The requirement of discriminant validity is satisfied if the AVE result is greater than 0.500, and the correction coefficient among latent variables is lower than the squared root of the AVE. The AVE of each variable ranges from 0.503 to 0.766, which meets the suggested criterion of 0.500 and higher.

As shown in

Table 3. Correlation data among the variables.

Correlation	Collaboration Ability	Collaboration Attitude	Collaboration Effects	Fairness	Communication	Trust
Collaboration ability	0.780
Collaboration attitude	0.482	0.709
Collaboration effects	0.530	0.504	0.773
Fairness	0.219	0.249	0.310	0.823
Communication	0.340	0.387	0.481	0.327	0.875
Trust	0.280	0.319	0.396	0.269	0.418	0.738

NB. The bold diagonal values in italics represent the square root of the AVE.

, the square root of the AVE of the latent variables ranges from 0.709 to 0.875, while the correlation coefficients among the latent variables fall between 0.219 and 0.530. The square root of the AVE of each variable is higher than its correlation’s value, which indicates adequate discriminant validity [49].

4.2. Structural Model: Model Fit and Hypothesis Testing

4.2.1. Model Fit

The reliability, convergent validity, and discriminant validity test results for the criteria of the measurement model were found to be satisfactory. The goodness-of-fit indices of the theoretical framework were assessed using the structural model. The SEM output shows that the proposed theoretical framework represents a good data fit. As shown in

Table 4. Fit Indices for the Structural Model.

Structural Model	Absolute Fit		Incremental Fit				Parsimonious Fit
	χ2/df	RMSEA	NFI	IFI	TLI	CFI	PNFI
Model	2.193	0.057	0.904	0.945	0.936	0.945	0.783
Fitted value	<3.0	<0.08	>0.9	>0.9	>0.9	>0.9	>0.5

NB. Root mean square error of approximation = RMSEA; normalized fix index = NFI; incremental fit index = IFI; Tucker–Lewis index = TLI; comparative fit index = CFI; parsimonious normalized fit index = PNFI.

, χ2 equals 438.638, χ2/df equals 2.193 (<3.0), RMSEA equals 0.057 (<0.08), NFI equals 0.904 (>0.9), IFI equals 0.945 (>0.9), TLI equals 0.936 (>0.9), CFI equals 0.945 (>0.9), and PNFI equals 0.783 (>0.5). These findings indicate that the model fits the data well.

4.2.2. Hypothesis Testing

Table 5. Structural Equation Modeling Results.

Path	Coefficient	T-Value	Relationship
Relationship quality → Collaboration effects (H1)	0.499 ***	4.480	Supported
Collaboration attitude → Relationship quality (H2)	0.408 ***	3.369	Supported
Collaboration ability → Relationship quality (H5)	0.281 *	2.408	Supported
Collaboration attitude → Collaboration effects (H4)	0.120 (0.176)	1.352	Not Supported
Collaboration ability → Collaboration effects (H7)	0.234 **	2.905	Supported

Note: * p < 0.05, ** p < 0.01, *** p < 0.001.

and Figure 2 detail the regression results for the hypotheses testing. H1 is supported, as relationship quality was found to have a significantly positive influence on collaboration effects. The regression pathways of collaboration attitude and collaboration ability to relationship quality were found to be significantly positive, and so H2 and H5 are supported. Collaboration ability was found to have a significantly positive effect on collaboration effects, a finding which lends support to H7. However, H4 is not supported, as collaboration attitude was found to have a moderately positive, insignificant influence on collaboration effects.

4.2.3. Mediation Effect

The bootstrap method was used to study the mediating effect in this study. As shown in

Table 6. Mediating Effect.

Path	Effects		Bias-Corrected Percentile 95% CI		Mediator
			Lower	Upper
Collaboration attitude → Relationship quality → Collaboration effects	Total effect	0.323	0.079	0.504	Yes
	Direct effect	0.120
	Indirect effect	0.204
Collaboration ability → Relationship quality → Collaboration effects	Total effect	0.374	0.025	0.429	Yes
	Direct effect	0.234
	Indirect effect	0.140
Relationship quality → Collaboration effects	Total effect	0.499	/	/	/
	Direct effect	0.499
	Indirect effect	0.000

, the direct effect, the indirect effect, and the total effect of collaboration attitude on the collaboration effects are 0.120, 0.204, and 0.323, respectively. Regarding the indirect effect of collaboration attitude on collaboration effects, the lower limit of confidence interval is 0.079, while the upper limit is 0.504. The whole confidence interval exceeds 0, indicating the existence of an intermediary effect [50]. The influence of collaboration attitude on collaboration effects transmitted through relationship quality is significant; thus, H3 is supported. The results for the direct effect, indirect effect, and total effect of collaboration ability on the collaboration effects are 0.234, 0.140, and 0.374, respectively. Regarding the indirect effect of collaboration ability on collaboration effects, the lower limit of confidence interval is 0.025, while the upper limit is 0.429. The entire confidence interval exceeds 0, indicating the existence of the intermediary effect. The influence of collaboration ability on collaboration effects transmitted through relationship quality is found to be significant; therefore, H6 is supported.

5. Discussion

Regong Thangka is a traditional art with aesthetic value. It records the changes of ethnic minority spirit, ethnic minority culture, ethnic minority history, ethnic minority cohesion, and centripetal force, and it reflects the economy, culture, religious beliefs, and folk customs of ethnic minorities through historical changes. The protection and inheritance of Regong Thangka will help to continuously develop culture, enhance aesthetic research, boost social cohesion, etc., thereby promoting the sustainable advancement of human development. This study regarded collaborative protection of the ICH of Regong Thangka in China as the research object and attempted to distinguish the relationships between the variables of participant subject factors (collaboration attitude and collaboration ability), relationship quality, and collaboration effects in the context of ICH collaborative protection. Taking relationship quality as the intermediary variable, we constructed a research framework of participant subject factors and collaboration effects in the collaborative protection of ICH. In particular, ours is the first study which includes relationship quality as a factor in ICH collaborative protection. As described above, we can verify the significance of the relationship between participant subject factors and collaboration effects. In addition, on the basis of the collaborative governance theory, the findings in this paper explain the formation mechanisms behind the collaboration effects in ICH collaborative protection. It is hoped that the findings presented in the previous section offer theoretical contributions to the field of ICH collaborative protection, from which recommendations for management practices might be formulated.

First, this study has empirically tested the direct positive effect of relationship quality on collaboration effects. In reference to H1, it is found that relationship quality has a significant impact on collaboration effects, specifically in the context of the multi-subject collaborative protection of China’s ICH. This finding is consistent with studies in other areas, such as education [51], business [52], supply chain collaboration [53], etc. In short, there is a positive connection between the quality of the relationship among participants and the final result. Relationship quality consists of trust, communication, and fairness. When a high level of trust exists among participants, they are more active in communicating with each other, which in turn facilitates information sharing. In the interviews, some respondents articulated that they would place more trust in partners who were related by blood and kinship, that their relationships with them were closer, and that collaboration effects were often better. In addition, the respondents also indicated that they were more likely to trust people who shared their religious beliefs (i.e., Tibetan Buddhism); specifically, they hold the belief that by painting in the Regong Thangka tradition, they are serving Buddha; therefore, they are less likely to cheat each other. Many respondents hold the opinion that better communication among the participants reduces the risk of conflict occurring during collaborative protection projects, thereby increasing trust and fairness, improving the relationship quality, and positively impacting collaboration effects. The findings of the present study are consistent with those of Sun, Zhu, and Yang (2021), who researched collaboration among stakeholders involved in agricultural supply chains, and who pointed out that fairness perception and trust perception can affect the quality of relationships and have an impact on the final results [25]. In the interviews, respondents indicated that regular communication, especially informal communication on a daily basis, can improve the relationship quality among participants and help them identify and solve problems in a timely manner; thus, regular communication can have a positive impact on collaboration effects. Fairness in this context is reflected in the equity of decision making, risk distribution and benefit distribution. Fairness in collaborative ICH protection is conducive to improving the relationship quality among participants [24]. This finding is consistent with that of Lee et al. (2020), who pointed out that distributive fairness and informational fairness lead to a high-quality relationship [47]. During the interviews, it was evident that respondents believe fairness is an essential prerequisite for a good collaborative relationship, particularly fair participation in decision making, and that the risks and benefits should be shared fairly among participants. When the process of collaboration is fair, few contradictions or disagreements will arise, and this helpful climate will lead to positive collaboration effects. In short, trust, communication, and fairness in collaborative ICH protection can improve the relationship quality among participants and have a positive impact on the collaboration effects of ICH protection.

Second, this study’s results show that collaboration attitude is an antecedent variable of relationship quality, but not necessarily an antecedent variable of collaboration effects. According to the statistical results corresponding to hypotheses H2, H3, and H4, collaboration attitude in the context of ICH collaborative protection of the Regong Thangka tradition has a positive impact on relationship quality, but does not have a significant impact on collaboration effects. Further, collaboration attitude indirectly influences collaboration effects via relationship quality; that is, relationship quality plays a mediating role between collaboration attitude and collaboration effects. It is worth noting that collaboration attitude was found to have a slightly positive but insignificant impact on collaboration effects in our study, which is inconsistent with the conclusions of previous studies. For example, in their study on attitudes and collaboration among stakeholders in enterprises, Itani et al. (2019) reported that relationship quality partially mediates the relationship between attitudes and the final results in marketing campaigns [54]. They also found that attitudes can have a direct impact on the final results in the marketing field. Chong and Kim’s (2017) study in the field of collaborative child protection services showed that collaboration attitude can directly dictate collaboration effects [55]. These contrasting results could be ascribed to differences in research context. The protection of Regong Thangka is a complex matter. Besides the positive collaboration attitude of the participants, there must be other factors, such as collaboration ability, the availability of start-up funds, and access to core technology. For example, some respondents indicated that they have a positive attitude towards the collaborative protection of Regong Thangka, but their communication skills are poor and they can only speak Tibetan, so a communication barrier will arise with people who cannot speak Tibetan.

In addition, it is found that collaboration attitude has a significantly positive impact on relationship quality. This could be explain by the fact that Regong Thangka is an art form specific to Tibetan Buddhism, and most of its inheritors are followers of Tibetan Buddhism. Shared religious beliefs can improve the level of trust among participants and is an important source of social trust [56]. Some studies have alluded to a positive correlation between religious beliefs and trust [57]. Where a high level of trust exists, the participants have a greater inclination to collaborate [7], and this stronger collaborative attitude enhances the level of trust, strengthens communication, and promotes fairness of collaboration, which together engender a relationship of excellent quality. During the interviews, it was also confirmed that some respondents tend to place greater trust in people within their collaboration with shared religious beliefs. Where there is mutual trust, the more positive the collaboration attitude, and the better the relationship quality. Relationship quality affects collaboration effects, and relationship quality plays an intermediary role between collaboration attitude and collaboration effects.

Finally, the results of our study confirm that collaboration ability is an antecedent variable of relationship quality and collaboration effects, and that relationship quality plays a mediating role in the relationship between collaboration ability and collaboration effects. Regarding H5, H6, and H7, in the collaborative protection of China’s ICH, collaboration ability likely has a significantly positive impact on relationship quality and collaboration effects and can also impact collaboration effects via relationship quality. This assumption is consistent with that of Hoof and Thiell (2014), who reported in their study of supply chain collaboration that the collaborative ability of participants positively influences collaboration effects [41]. Lee and Roh (2021) also stated that the collaborative ability of project partners directly affects collaboration effects [44]. The collaborative ability of the participants directly and significantly impacts collaboration effects. At this point it should be noted that collaborative ability consists of communication ability, coordination ability, resource integration ability, and collaboration experience. The richer the participants’ previous collaboration experience, the better they can handle problems that arise in the business of collaborative protection, and the smoother the collaboration process will be. As mentioned, collaboration ability can also have an indirect effect on collaboration effects via relationship quality. The stronger the collaborative ability of the participants, the better the quality of the relationship, and the stronger the positive impact on collaboration effects. Our finding in this regard is consistent with that of Bo et al. (2019). In their quantitative research study in the field of risk management, they argued that collaboration ability has a direct effect on relationship quality and an indirect effect on final collaboration effects via the mediator of relationship quality [24]. In the interviews in the present study, some respondents also opined that strong collaboration ability, such as communication skills, resource integration skills, etc., leads to a good collaborative relationship among the participants, and in turn to beneficial collaboration effects.

In addition, authors of previous studies have claimed that collaboration attitude cannot directly affect collaboration effects. In view of this, it might be argued that in the field of ICH collaborative protection, collaboration ability is more important than collaboration attitude. This may be due to the fact that participants generally have a positive collaboration attitude due to their shared interest in Regong Thangka, but that their educational attainment is generally low, which limits their collaboration ability. In the interviews, the majority of respondents said that their poor collaborative ability might to a large extent explain their poor collaboration effects. Regong Thangka is practiced in Western China among minority ethnic communities whose economic and education statuses are relatively backward, and who might not be proficient in soft skills such as teamwork and negotiation. Therefore, collaboration ability training for ICH stakeholders should be prioritized in future programs for preserving the Regong Thangka tradition.

6. Conclusions and Recommendations

Firstly, this paper proposed the new concept of relationship quality to describe the relationship among collaboration participants. Relationship quality has a direct positive impact on collaboration effects in the field of ICH collaborative protection. Therefore, effective measures should be taken to establish a good quality relationship among multiple subjects. As mentioned, the three dimensions of relationship quality are trust, communication, and fairness. Therefore, in the collaborative protection of ICH, it is necessary to establish effective communication channels, achieve timely information sharing, and improve the level of trust, thus improving the quality of relationship among participants. In the concrete construction of trust and communication systems among collaborative participants, particular emphasis should be placed on interpersonal and institutional trust. On the one hand, the principal aims of constructing interpersonal trust are to establish a variety of communication channels within the collaboration system, facilitate the communication and consultation among participants, encourage participants to share information, and improve the level of trust between the participants. On the other hand, construction of an institutional trust system mainly involves improving various rules and regulations within the collaboration system, optimizing the design of various processes, and making the collaboration process transparent so as to increase the level of trust among participants [22,24]. In addition, it is necessary to ensure equal rights among the participants, especially the right of decision making. This is an important premise that guarantees the fairness of decision making, risk bearing, and interest distribution. This requires the design of collaboration rules. At the beginning of the collaboration, it is necessary to extensively solicit the participants’ ideas on collaboration decision making, collaboration risk sharing, and collaboration benefit distribution. Based on the understanding of each participant’s interest demands and value preferences, collaboration rules should be jointly formulated and effectively implemented.

Secondly, in the context of ICH preservation, collaboration attitude has a positive impact on relationship quality, and indirectly influences collaboration effects via its effect on relationship quality. Therefore, it is necessary to foster a positive collaborative attitude among multiple subjects. “Collaboration attitude” as a term refers to each participant’s attention towards and support for collaboration and the perceived attitude of other participants that affect relationship quality. Thus, participants involved in collaborative efforts to preserve the Regong Thangka tradition should self-evaluate their own behavior and provide feedback on the behaviors of other participants. This requires a sound communication system, i.e., both formal and informal channels of communication. For example, regular symposia or meetings should be held to strengthen communication among different participants and to reduce information asymmetry and conflicts during collaboration. At the same time, informal communication, such as during team-building sessions, can also strengthen ties between participants and improve the level of trust across the group. As mentioned earlier, Regong Thangka is an art form specific to Tibetan Buddhism, so a large proportion of participants involved in conservation efforts will be adherents of Tibetan Buddhism. Therefore, the most obvious means by which participants can sustain and improve the quality of communication and trust is to participate in religious activities together. In addition, the government can support Regong Thangka collaborative protection activities by running campaigns using media channels including television, radio, and newspapers, and perhaps funding and establishing exhibitions, to encourage more people to participate in multi-subject collaboration and cultivate stronger positive attitudes.

Finally, within the field of ICH collaborative protection, collaboration ability has a positive impact on relationship quality and collaboration effects and influences collaboration effects through the mediator of relationship quality. Therefore, various measures should be taken to improve collaborative abilities among multiple participants. For example, it is necessary to provide relevant training to participants and to help them improve their communication, coordination, and resource integration skills. Stakeholders involved in Regong Thangka collaborative protection include government officials, inheritors, and local enterprises. Therefore, improved collaboration ability can be cultivated from two aspects. On the one hand, training should be directed towards improving the abilities of non-government participants such as local inheritors and entrepreneurs. As mentioned, educational attainment among many participants involved in the Regong Thangka collaborative protection is relatively low. One possible solution is for the government to popularize nine-year compulsory education and strengthen the provision of vocational education. The government could also improve laws, regulations, policies, and measures related to the protection of ICH elements, such as Regong Thangka, and provide professional training in conflict management, teamwork, and problem-solving to improve the collaborative ability of the participants. However, of greater importance is the need for non-government participants such as inheritors to take the initiative to learn about and improve their collaborative abilities. In addition, government workers should be professionally skilled communicators. They should be encouraged to visit, investigate, and learn lessons from successful ICH collaborative protection projects from the past, as well as attend management and teamwork training courses, with the aim of improving their ability to deal with collaboration issues.

The study has certain limitations that should be addressed in future studies. First, the research object of this study, as an example of ICH, is Regong Thangka, which is one of a great many forms of traditional art. Therefore, caution should be applied when extending our study’s findings and their implications to other ICH projects. Future research should consider a wider range of regional ICH projects and traditions. Second, in this study we considered only the core dimensions of relationship quality (trust, fairness, and communication). Other dimensions of relationship quality, such as intimacy, interdependence, and satisfaction, may yield different participant evaluations. Further, we did not take into account the interrelationships between different elements and entities. We also failed to make a unified comparative analysis between intangible cultural heritage and cultural heritage. Future research can be carried out focusing the above aspects. Finally, the research framework developed in this study has principally focused on the influences of participant subject factors and relationship quality on collaboration effects, without considering external factors. In future studies, researchers should consider external factors, such as government support, economic development, and cultural atmosphere, which could more comprehensively explain the factors influencing collaboration effects.