E ﬀ ect of Corporate Social Responsibility on Community Resilience: Empirical Evidence in the Nickel Mining Industry in Southeast Sulawesi, Indonesia

: Mining is an important industry in Indonesia. A nickel mining company has operated for almost 45 years. It has managed corporate social responsibility (CSR) programmes in the neighbouring local community. In addition to the environmental conservation and mitigation, as well as socioeconomic enhancement, the CSR is expected to nurture resilience in the local communities. This study’s goal is to examine the e ﬀ ect of CSR on community resilience (COM-R) in the surrounding community. To analyse the e ﬀ ect of CSR practise on COM-R, Partial Least Squares -Structural Equation Model (PLS-SEM) is used. Results show that CSR has a positive e ﬀ ect on and a signiﬁcant relationship with COM-R. Results also indicate that CSR’s contribution to COM-R enhances community collective e ﬃ cacy, community action, and adaptation. Thus, the veriﬁed CSR and COM-R model beneﬁts other researchers, companies, and governments to be further explored.


Introduction
The notion of resilience may start by replacing "sustainability" as the catchy word of policymaking and political rhetoric. To understand how communities react and adapt to environmental, economic, and societal changes, conceptions of marginalized, vulnerability, and resilience have provided an important conceptual framework [1]. Resilience qualities are evident within the notion of adaptive capacity. Researchers use such capacity to examine how systems perform and react to endogenous and exogenous disturbances. This capacity is described as 'a system ability to adjust, moderate effects and cope with disruption' [2,3]. Systems which are characterized by diversity, change potential and connectivity (e.g., feedback and flexibility) and have a good adaptive capacity [4].
The nickel mining industry in Indonesia has been operating for nearly 45 years. It makes social and nonsocial changes. This industry is the main driver of employment and economic development in local communities. Issues are also stood on the disturbance to social, economic, and environmental effects on surrounding communities near mining operating sites. Companies must perform corporate social responsibility (CSR), which is an engagement to improving community well-being (CWB) through reimbursable business practices and corporate resources contribution [5,6]. Therefore, CSR efforts Bronfenbrenner's bioecological theory as a framework to measuring the progress of interventions to improve community resilience. However, this framework focuses within person characteristics for promoting individual resilience as community members. In the context of global change and natural disasters, Cutter et al. [2] reported the dimensions of COM-R assessment, which consist of economic, social, infrastructure, institutional, and community capitals. Stewart et al. [29] backed this idea and said that dimensions of infrastructure durability, economic, and social resilience are the main indicators to raise the robustness of communities that are affected by natural disasters. Berkes and Ross [30] stated nine dimensions of COM-R, which are community infrastructure, social networks, leadership, diverse economies and innovations, participation, knowledge, skills and learning, positive outlook, relationship between individuals, value and trust, and self-advised agency. Walton et al. [22] reported three dimensions in COM-R assessment, which are collectiveness, community action, and adaptation. From the socioecological conception of resilience includes concepts of adaptive capacity-that is, the community ability to transform and inspired proactively for growth, preservation, mitigation, relief, and reorganisation [26]. Community resilience is also portrayed by the community member capabilities to develop effective action to moderate the effect of a problem [31,32]. Therefore, Norris et al. [33] claimed that community resilience as a theory, consists of capacities and strategy for recovery preparedness.
From the sustainable livelihood framework's starting point of view, the community has some assets from which they derive livelihood [34]. An asset is a stock of resources, namely financial, human, physical, natural, or social, which can be acquired, developed, improved, or transferred through generations [28,32,35]. McCrea et al. [36] identified seven community valuable resources; cultural, social, political, intellectual, financial, natural, and built capital that can be mobilized to enhance community resilience and well-being. Steiner and Atterton [37] discussed that companies that operate in the countryside contribute to the community's resilience. Therefore, if CSR is a company's strategy in community development, then CSR has a direct impact to the community's resilience. This study also referred to stakeholder theory, which has been analysed and expanded by Freeman [38] through its article on strategic management. He introduced stakeholder theory in the debate of CSR and stated that a stakeholder implies to any individual or group of people that can be disturbed or adversely affected by the realization of the organization objective. This means that the outcome of the company's operations affects the community.

CSR Practice and COM-R Model
Ismail [39] determined that the majority of CSR-participating companies have applied CSR initiatives since the first 10 years of the millennium and have their core businesses in various sectors. Education programs are the common types of CSR contribution. Brew et al. [40] determined that CSR activities are related to education, health, livelihood, and community aid. Degie and Kebede [41] explained that CSR is an essential interface between local communities and the government. Moreover, they mentioned that CSR could be a standard given that it demonstrates that business corporations can address the needs of communities. They also stated that companies' CSR initiatives improve the capability of communities and dimension of CWB. Sarmila et al. [42] established that CSR projects contribute to economic welfare through sources of income, employment opportunities, and asset financing. Al-Zyoud [43] stated that the philanthropic and ethical influence in CSR significantly affects sustainability development.
Businesses that operate using natural resources have the responsibility and capability to increase community capacity and resilience [44]. Soetanto et al. [20] stated that social responsibility's role is important when trying to protect built environments during floods and other changes. The concept of responsibility becomes important when people realize that they cannot survive potential disasters. They must contribute their part, work together, and participate in their community. The authors also argued that integration between community and corporation helps community leaders to decide well regarding COM-R measures. In 2014, Steiner and Atterton reported that private companies help in social and environmental resilience by identifying the role they play in advancing the quality of life of rural communities in Scotland. They found that resilience can be measured and evaluated by considering socioeconomic dimensions. The "success" of rural communities depends on the development of local resources (including businesses, human capital, land, and natural resources), local control over these resources, and how these resources are used [45]. Gibson and Klinck [46] stated the need for the market to drive the existence of industry and the development in various supporting sectors with a huge effect. The society must negotiate the ways these changes occur, such that society can understand, adapt, and influence. Thus, increasing research on the indicators of endurance and well-being are valuable tools to support key elements and track how benefits and effects are distributed. Buikstra et al. [47] reported that community development and social impact assessments offer insights into the quality and assets of communities that enable them to adapt to major changes or grow successfully. Individual and COM-R components are identified as social networks and support (e.g., learning, positive outlook, environment and lifestyle, initial experience, sense of purpose, infrastructure and support services, differences, innovative and diverse economies, confidence and management) experience changes. These components require interaction among people, infrastructures, communities, the economy, and the environment in the process of building resilience. Rudito [48] stated that the practice of CSR using community development directs a positive change in the sustainability and economic aspects and an effect of CWB [49] and increases COM-R [20,44]. Thus, previous research has shown that direct CSR practices contribute to COM-R. The first hypothesis was expressed on the basis of these findings.

Hypothesis 1 (H1).
A positive and significant relationship exists between CSR practice and COM-R in the Indonesian nickel mining industry.
In building COM-R that has changed because of mining, three important dimensions for COM-R have been identified, namely, collective efficacy, community action and community adaptation in usages of resources, and commitment towards the community and effective relationships among community members [44,50]. These dimensions and their quality also contribute to COM-R [50]. Walton et al. [22] found that the mining industry could strengthen the sustainability of prosperity by enhancing COM-R dimensions, such as community action, community adaptation, and collective efficacy. Rama et al. [19] affirmed that developing the capacity for collective action with CSR implementation can engage community capacity-building [51]. CSR implementation can enhance community adaptive capacity [52]. Consequently, the practice of CSR can improve community relationship [53]. CSR practices relate and contribute to dimensioning COM-R. On the basis of these results, the present study postulates the following hypotheses: Hypothesis 2 (H2). A significant indirect and positive relationship exists between CSR practice in the Indonesian nickel mining industry and collective efficacy dimension.

Hypothesis 3 (H3).
A significant indirect and positive relationship exists between CSR practice in the Indonesian nickel mining industry and the community adaptation dimension.

Hypothesis 4 (H4).
A significant indirect and positive relationship exists between CSR practice in the Indonesian nickel mining industry and the community action dimension.

Corporate Social Responsibility
The CSR concept has been widely used in recent decades as a basis to facilitate business and society. Carroll [54] discussed that businesses have a social responsibility that involves society with regard to the legal, ethical, economic, and discretionary (philanthropic) organizations at some point in time.
The above definition has been used in business and society [14,55,56] on the basis of its worldwide documented applicability. Companies should be held responsible for their effect on communities and the environment where they work [57]. Table 1 shows the definition of CSR in relation to community. Table 1. Definition of CSR in relation to community.

Author (Year) Definition
Frederick [58] CSR is a sign of public posture towards the society's human and economic resources. CSR displays the willingness to utilise resources for broad social ends and not simply for the narrowly circumscribed interests of firms and private people.
Carroll [54] Business social responsibility encompasses legal, economic, discretionary, and ethical expectations that the society has of an organization.
Kotler and Lee [5] CSR guarantees to improve CWB by using philanthropic business practices and corporate resources support.
Hopkins [59], Gallardo-Vázquez et al. [11] CSR focuses on treating firm stakeholders in a socially conscious manner. Stakeholders exist both within and outside of a firm. Social responsibility aims to improve living standards while preserving corporation profitability for its stakeholders, both inside and outside the corporation.
Source: synthesized by the authors.
Carroll [54] proposed the following four dimensions of CSR, that is, economic, legal, ethical, and discretionary responsibilities; the author argued that companies need the basic definition of CSR. Stakeholders are those to whom companies have bonds of responsibilities or dependence. Lyra et al. [60] stated that businesses' main role is to provide services and produce goods for society, emphasising its economic function and prioritising not hurting the principles of responsibility. Carroll [61] discovered that the CSR pyramid holds the whole spectrum of societal expectations regarding responsibilities and upholds the same legal, economic, philanthropic, and ethical dimensions. The author asserted that the most important dimension and the basis for supporting other dimensions is economic dimension. He also stated that, according to the pyramid model, companies must meet the requirements of the market and thus not follow CSR through pure humanity. The author stated that profit should be given priority, followed by other responsibilities.

Community Resilience
Magis [44] stated that COM-R is an individual and shared ability to respond to change. The author discussed how resilient community members intentionally develop individual and shared ability. Members' influence and response to change tolerate and renovate the community and create future community trajectories. Resilience is individuals' adaptive capacity and ability to handle change in a socioeconomic and environmental ecosystem that is continually evolving [62,63]. Social capital is an integrated facet that aids in building COM-R [64,65]. Communities that are resilient are pro-active and independent; thus, such communities are empowered and can affect local life [66,67]. Resilience presents the competence of self-organisation, the capability to absorb disruption and maintain a similar function and the capacity to learn and adapt. Resilient communities are defined as groups that can change [1,15]. Communities need change to live and thrive. Social value fails to survive in social and economic surroundings. Thus, capital should be provided to ensure that a community becomes resilient in the face of change.
Allen et al. [68] claimed that communities have various types of capital; thus, communities are productive, stable, and not weak to external shocks. Wilson [69] reported that COM-R has three types of capital that keep communities together. Social, environmental, and economic capitals are needed for communities to thrive. These capitals are needed in understanding resilience at the community level [1,2]. Certain researchers stated that the dimensions of COM-R include social capital, information and communication, networked resource economic development, and community competence [33]. The subsystems of resilience are economic, ecological, governance, civil society, and physical infrastructure [70]. Both opinions give different emphasis. For the second opinion, ecological and physical infrastructure aspects are equally important as social factors and community competence. Walton et al. [22] argued that community adaptation, collective efficacy, and community action are needed. They argued that the three dimensions are keys to adapting to all future changes. COM-R in real-world applications can be measured by the abovementioned types or dimensions [44]. The COM-R dimensions are as follows in Table 2. Table 2. COM-R Dimensions.

Author (Year) Dimension
Norris et al. [33] A four-sets resources of networked: social capital, economic development, community competence, and information and communication; these resources determine COM-R Stewart et al. [29] A function of the critical infrastructure, supply chain, economic and social resilience Longstaff et al. [71] The resilience function of five subsystems: environmental, economic, governance, civil society, and physical infrastructure Cutter et al. [3] Measured in five domains: economic, infrastructure, social, institutional, and community capital resilience Magis [44] Community resources, its development and engagement, strategic and collective action, active agents, and equity and influence Kulig et al. [72] Leadership and empowerment, non-adverse geography, and community engagement Berkes and Ross [30] Social networks, diverse and innovative economy, community infrastructure, knowledge, leadership, learning, skills, people-place relationship, engaged governance, values and beliefs, positive outlook, and agency and self-organising Collins et al. [73] Community capacity to absorb, react/adjust and recover from disturbance while retaining its essential functions Walton et al. [22] Three aspects of COM-R: community adaptiveness, collective efficacy, and community action Pfefferbaum et al. [65] Caring and bonding, transformative, assets, potential, disaster, information, managing, and interaction Brown et al. [74] Action of the community and community adaptiveness Source: Source: synthesized by the authors.

CSR in Mining Activities
Mining has been part of human history for many years [75]. Since the 1980s, studies on mining companies' CSR have increased [75,76]. The CSR of Canadian mining companies rose in the 1990s and frequently reported on environmental and social issues. The mining industry's investment in CSR increased owing to changes in the decision-making structure and led to community investment [76,77], reflecting wide debates on CSR and contributing business to development [76]. CSR is a key that gives mining a social license to operate. Thus, CSR eases community activism against mining companies [78].
It should be the responsibility of medium-and large-scale mining companies to incorporate CSR to encourage the sustainable development of local communities in nearby mining operations. Franks [77] reported that in 2012, almost $1.7 billion was spent on community investment by 15 of the largest mining companies. The community investment estimates have intensified the scale and scope of CSR programs carried out by mining companies. CSR programmes and activities are hoped to drive local community capacity [79,80]. The CSR of the mining industry must reduce risk in internal and external corporation and push the industry into sustainable future development [81] Development brings about political, social, economic, and environmental changes to enhance the overall quality of life sustainably. Mining companies' concept of development is narrow [82]. Mining activities have been viewed by previous studies as a capitalist process that takes place across time and space [83]. The mining industry has caused great destruction and abandoned areas depleted of valuable economic resources [84], on which a community is highly dependent [32]. Slack [85] stated that a gap exists between CSR's expectations and reality given that CSR is not integrated in companies' operational business models. Moreover, the effectiveness of CSR initiatives in petroleum, gas, and other mineral extracting is increasingly questionable. Multinational mining firms were "rated" as good corporate citizens. However, no clear indication supports how this recognition is given to address issues of sustainability after communities are affected and of whether community development initiatives effectively contribute to sustainable community development [86]. Therefore, CSR practices in the mining industry must be transformed into the development of society with more equity; they must protect natural resources and preserve them to improve life in sustainable development [39]. CSR in mining activities has been increasingly understood by previous studies as a concept of community development. Most CSR initiatives in the mining sector are aimed at protecting the environment, community development and stimulating stakeholder engagement [87] in order to foster resilience and well-being of the community.

Population, Sampling, and Data Collection
The total population of family leaders across 12 surrounding villages at nickel mining operating and processing activities was 6.236. A large size is required for reliable analysis in the SEM model [88]. Roscoe [89] stated that the applicable sample size for research is between 30 and 500 family leaders. Israel [90] proposed that, in addition to the sample size calculation formula, a good sample size is between 200 and 500 respondents if using multiple regression, covariance analysis, and multivariate analysis. Following these guidelines, the sample size was 500 family leaders and selected through stratified simple random sampling technique. Table 3 shows that the population was divided into twelve strata (villages) and stratified sample size of the study. The data was gathered through field surveys, using a tested structured questionnaire [91] with support from trained native enumerators. The family leaders were interviewed face-to-face by researchers with assistance from native enumerators to complete the survey questionnaire. This empirical survey was started in October 2017, at Oko-Oko village, ended in January 2018 at Huko-Huko Village in southeast Sulawesi, Indonesia. Almost 98.0% (490 family leaders) of the survey questionnaires were completed and appropriate for analysis.

Measurement of Constructs
We adopted previous researchers' validated and established measurement items using the same variables. A five-point Likert scale was used to measure CSR and COM-R dimensions. We adopted the CSR measurement from the pyramid of Carroll [61,92] and other researchers (Maignan and Ferell [93]; Perez et al. [94]; Stanaland et al. [95]; Serageldin and Steer [96]). Table 4 shows the CSR dimensions. Walton et al. [22], Magis [44], and Iskandar et al. [49] established the COM-R scale. COM-R has three dimensions as follows. (1) Community actions pertain to respondents considering how their local community responds to CSR practice and nickel mining activities regarding planning, accessing relevant information, having adequate leadership, supporting volunteers, establishing key networks within the community, demonstrating good working relationships between intergroup, preservation, and an overall assessment of how satisfied they are with the manner in which the community responds to change. (2) Community efficacy represents respondents considering how well residents, businesses, governments, and company's resources can work cooperatively to address concerns or take opportunities in relation to CSR practice and nickel mining activities. (3) The level of community adaptation is when respondents are asked about their perception of the manner in which the community handles CSR practice and nickel mining activities. Table 4 presents the indicators that measure COM-R. We used a 5-point Likert scale, from 1 = strongly disagree to 5 = strongly agree for constructs measurement [97]. The survey questionnaire was tested for internal reliability on 76 family leaders and found that the Cronbach's alpha (CA) (α) for CSR was 0.90, classified as excellent, and COM-R was 0.77 indicating acceptable internal consistency [98][99][100]. With PLS-SEM analysis (Hair et al. [100]), the composite reliability for CSR constructs between 0.81 and 0.83 and COM-R constructs between 0.81 and 0.82 was considered good [98].

Structural Equation Models
We empirically studied the effect of CSR practice on COM-R in the second part of this study by using SEM. We used the PLS statistical methods for taking advantage of first-hand data [99,103]. As noted in Figure 1, the conceptual framework of CSR was formed by three first-order subconstructs, namely: legal and ethical, economic and philanthropic dimensions, and measured as a second-order reflective construct. The present research framework (Figure 1) was formed using composite indices (i.e., special constructs). PLS is a proper technique; as suggested by Hair et al. [103] and Sarstedt et al. [104], we utilised Smart-PLS professional statistical analysis software.

Structural Equation Models
We empirically studied the effect of CSR practice on COM-R in the second part of this study by using SEM. We used the PLS statistical methods for taking advantage of first-hand data [99,103]. As noted in Figure 1, the conceptual framework of CSR was formed by three first-order subconstructs, namely: legal and ethical, economic and philanthropic dimensions, and measured as a second-order reflective construct. The present research framework (Figure 1) was formed using composite indices (i.e., special constructs). PLS is a proper technique; as suggested by Hair et al. [103] and Sarstedt et al. [104], we utilised Smart-PLS professional statistical analysis software.

Respondent Profile
This section defines the profile of respondent, consisting of gender, age, academic qualification, occupation, and average monthly earnings. This study indicated that respondents were relatively gender-neutral from 57.5% male-and 42.4% female-headed households. The distribution age of less than 25 years was 11.8%, aged 25 to 35 was 31.4%, aged 36 to 45 was 33.5% and over 45 years was 23.3%. Most of them are 36 to 45 years old and are considered productive community members. The distribution of educational qualification is different. The highest is high school graduate (39.8%), followed by junior high school (24.5%), diploma holders (7.1%) and undergraduates (9.4%). One respondent has a doctorate in philosophy. Respondents who worked at the mining company are only 2.4%. Most of them are self-employed and involved in informal business (41.4%) and farmers/fisherman (15.9%). The rest (22.4%) do not have permanent employment. The majority of the

Respondent Profile
This section defines the profile of respondent, consisting of gender, age, academic qualification, occupation, and average monthly earnings. This study indicated that respondents were relatively gender-neutral from 57.5% male-and 42.4% female-headed households. The distribution age of less than 25 years was 11.8%, aged 25 to 35 was 31.4%, aged 36 to 45 was 33.5% and over 45 years was 23.3%. Most of them are 36 to 45 years old and are considered productive community members. The distribution of educational qualification is different. The highest is high school graduate (39.8%), followed by junior high school (24.5%), diploma holders (7.1%) and undergraduates (9.4%). One respondent has a doctorate in philosophy. Respondents who worked at the mining company are only 2.4%. Most of them are self-employed and involved in informal business (41.4%) and farmers/fisherman (15.9%). The rest (22.4%) do not have permanent employment. The majority of the respondents (36.1%) earn IDR 1,000,000 to IDR 2,000,000. Approximately 34.5% have a monthly income less than IDR 1,000,000 per month, and 29.4% earn more than IDR 2,000,000 a month.

Measurement Model Verification
A method for testing the proposed theoretical framework is illustrated in Figure 1, which displays the structural model for reflective measurement using the PLS-SEM technique. Individual item reliability, convergent and discriminant validities were observed to assess the acceptability of the measurement model [103,105]. CA shows the verification for convergent validity. Each factor loading on the dependent variable must be significant and greater than 0.70 [106], whereas the average variance extracted (AVE) from each variable should be greater than 0.50. We evaluated discriminant validity by comparing the correlation between each pair of constructs and the root of AVE amongst those constructs [107]. We evaluated the reflective measurement model in terms of building credibility, reliability of individual items, legitimacy of discrimination, and convergence validity. We considered the reliability of individual items sufficient when the factor load is greater than 0.50 in each construct. We could use these items for reflective variable instructions (Table 5). The measurement for convergent validity and construct reliability signifies measures of internal consistency. Table 5 provide the reliable investigation of reflective constructs. The Cronbach alpha (CA) coefficient and composite reliability were greater than 0.70. AVE was investigated to assess convergent validity. The AVE values must be larger than 0.50. All our constructs exceeded this condition. The variance shared between the construct and others should be less than AVE to assess discriminant validity (Table 6).

Structural Model Evaluation
We used Smart PLS 3.0 to test the structural modelling and hypotheses [106]. We gauged the predictive power of the structural equation model by calculating the value of R 2 , which is the amount of variance described by the exogenous construct [108,109]. The model fit assesses a significant relationship between the variable and explained variance in the endogenous latent variables [110]. The overall model was fit with R 2 = 0.428. t = 10.842 was significant at p < 0.001. The results of the structural model analysis showed that CSR and COM-R have a direct and positive significant relationship (β = 0.654, p < 0.001) as shown in the Figure 2 below. Thus, hypothesis 1 (H1) was supported. The hypothesis testing showed a significant indirect and positive relationship between CSR with the three COM-R dimensions, consisting of hypothesis 2 (H2), the correlation of CSR practice to collective efficacy (β = 0.554, p < 0.001), hypothesis 3 (H3) CSR practice to community action (β = 0.521, p < 0.001) and hypothesis 4 (H4) CSR practice to community adaptiveness (β = 0.410, p < 0.001). Moreover, Table 7 indicates a positive and significant relationship between the CSR and COM-R dimensions. Therefore, the structural model has a significant predictive power. We used Smart PLS 3.0 to test the structural modelling and hypotheses [106]. We gauged the predictive power of the structural equation model by calculating the value of R 2 , which is the amount of variance described by the exogenous construct [108,109]. The model fit assesses a significant relationship between the variable and explained variance in the endogenous latent variables [110]. The overall model was fit with R 2 = 0.428. t = 10.842 was significant at p < 0.001. The results of the structural model analysis showed that CSR and COM-R have a direct and positive significant relationship (β = 0.654, р < 0.001) as shown in the Figure 2 below. Thus, hypothesis 1 (H1) was supported. The hypothesis testing showed a significant indirect and positive relationship between CSR with the three COM-R dimensions, consisting of hypothesis 2 (H2), the correlation of CSR practice to collective efficacy (β = 0.554, р < 0.001), hypothesis 3 (H3) CSR practice to community action (β = 0.521, р < 0.001) and hypothesis 4 (H4) CSR practice to community adaptiveness (β = 0.410, p < 0.001). Moreover, Table 7 indicates a positive and significant relationship between the CSR and COM-R dimensions. Therefore, the structural model has a significant predictive power.

Conclusions and Discussion
CSR is a commitment and obligation of a company to leverage natural resources that can provide

Conclusions and Discussion
CSR is a commitment and obligation of a company to leverage natural resources that can provide quality of life, economic benefits, and well-being for workers, community, and nation [5,111]. Natural resource management, such as nickel mining operating and processing activities in Indonesia, is one of the government's projects and development policies that can bring changes to economic, social, and environmental capital or a public-led dimension [69]. Strengthening of the society, community action [22,33], and community capacity are needed in facing various challenges and changes at all times [112,113], including nickel mining and exploration. The CSR initiated by corporate nickel mining is aimed at sustaining the surrounding community's well-being as well as COM-R. We therefore assessed the impact on COM-R of CSR practices around nickel mining activities.
CSR implementation by nickel mining companies and its effect on COM-R showed a positive and significant impact. The results of this study have been empirically proven from previous findings by Rama et al. [19], that investment through CSR practices provides an individual capacity effect and develops collaboration among community members. Moreover, this finding is also supported by Magis's [44] statement that changes occur through company activities that leverage natural resources and improve society capacity to achieve a resilient society through community action. This means that CSR investments can develop community capacity and cooperation to achieve community endurance, as mentioned in previous evidences [19,52,53]. The empirical findings derived from this study further strengthened Walton et al.'s [22] and McCrea et al.'s [36] results. Similarly, Wasylycia-Leis et al. [114] explained mining in relation to the change of endurance, such as social endurance, economy, and environmental impact. Developing the resilience and well-being of future sustainable community depends on stakeholders' commitment, especially to the company and government. CSR practices, such as economic, ethical, and philanthropy aspects undertaken by the company, have a positive impact on the community capacity. According to Case [17], building community capacity supports the resilience and well-being of the community in the face of environmental changes, difficulties, and risks in life. Soetanto et al. [20] stated that the role of corporate social responsibility is important to protect the built environment. The community began to realise that they could not endure the effects of extreme changes. Therefore, cooperation among the company, community leaders, and community is a step for making better decisions on the enhancement of community resilience, as stated by Magis [44]. The cooperation actions and collectiveness among community members are important aspects. Bandura [115] justified that collective effectiveness and community action are important to establish continuously the well-being of the community. Moreover, CSR practice has a significant indirect and positive impact on community adaptation, which has been reinforced by Rahmawati [52].
In the context of this research, CSR is able to foster community adaptation, which fits the assertion by Collins et al. [73] on the community's ability to absorb, react to, and regain from interference. Buikstra et al. [47] stated that community adaptation can be done by individuals and groups through changes in social networking and support, life experiences, environment, lifestyle, infrastructure, and support services and economy. We discovered a significant positive and indirect relationship between CSR practice and community action. CSR activities lead to good working relationships among groups in different communities and improve the community connection, as reported by Boadi et al. [53]. The community can also find relevant information and solutions when dealing with changes. Shucksmith [116] shared that community action, such as networks relation, is important. Local community partnerships with various partners, including government agencies and private companies, contribute to the social and environmental endurance and enhance the quality of life of the community [37]. Thus, partnerships between company and community through CSR programmes are an effective intervention in nurturing COM-R.

Limitations and Future Research
The results of this study are valuable and lead to the advancement of COM-R by strategic development model of CSR practices. This study analyses the role of CSR in improving society in the face of social, economic, and environmental changes, especially in communities surrounding the mining activity. However, this study requires further analysis for a thorough understanding of CSR practices in relationships with community resistance. The study will provide interesting evidence if the sample can be taken from different perceptions of stakeholders, such as local government and nongovernment organizations and not only the perception of the surrounding communities. Finally, the obtained model of CSR to improve the resilience of the community will be more holistic. The mixture of qualitative and quantitative approaches should be used to find their credible causal interactions best understood. In-depth explorative study could also be interesting, or even necessary, to investigate specifically how CSR awakens COM-R.

Conflicts of Interest:
The authors declare no conflict of interest.