https://orcid.org/0000-0002-5143-2773
Figures
Abstract
Supply chain resilience has garnered significant attention from both scholars and practitioners. However, the complex nature of the topic has resulted in a dearth of research on its key elements and formation mechanisms. To bridge this knowledge gap, we implemented grounded theory and conducted semi-structured interviews with 23 interviewees, which led to the identification of six key elements of supply chain resilience through open coding, axial coding, selective coding, and theoretical model saturation testing. These elements are product supply resilience, resource resilience, partner resilience, information response resilience, capital resilience, and knowledge resilience. Drawing from the key elements and the three phases of supply chain resilience (readiness, response, and recovery), we illustrated its formation mechanism and constructed a theoretical model of the influencing factors and pathways of supply chain resilience. We devised a questionnaire based on the coding results and confirmed its reasonableness and validity with a small sample of 109 questionnaires. Subsequently, a large sample of 409 questionnaires was used to test and validate the theoretical model using structural equation modeling, demonstrating that the identified key elements positively impact supply chain resilience. In sum, our paper enriches the comprehension of supply chain resilience by identifying its key elements and elaborating on its formation mechanism.
Citation: Ke Y, Lu L, Luo X (2023) Identification and formation mechanism of key elements of supply chain resilience: Exploration based on grounded theory and verification of SEM. PLoS ONE 18(11): e0293741. https://doi.org/10.1371/journal.pone.0293741
Editor: Mohamed Rafik N. Qureshi, King Khalid University, SAUDI ARABIA
Received: August 17, 2023; Accepted: October 18, 2023; Published: November 2, 2023
Copyright: © 2023 Ke et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper.
Funding: This work was supported by Guangxi Key R&D Plan (2022AB34029) and by the Research Fund Project of Development Institute of Zhujiang-Xijiang Economic Zone, Key Research Base of Humanities and Social Sciences in Guangxi Universities (ZX2023051) and by the Innovation Project of Guangxi Graduate Education (JG2023018). The Guangxi Key R&D Plan (2022AB34029) awarded to Lin Lu. The Research Fund Project of Development Institute of Zhujiang-Xijiang Economic Zone, Key Research Base of Humanities and Social Sciences in Guangxi Universities (ZX2023051) awarded to Xiaochun Luo. The Innovation Project of Guangxi Graduate Education (JG2023018) awarded to Youan Ke. Role of Funder: Lin Lu plays a role in study design, data collection and analysis, project management, publication decisions and review of manuscript. Xiaochun Luo plays a role in study design, data analysis, supervision and editing of manuscript. Youan Ke plays a role in investigation, data curation and analysis, manuscript writing and revision.
Competing interests: The authors have declared that no competing interests exist.
1. Introduction
In the context of economic globalization, labor specialization, digital intelligence, and the upgrading of personalized customer needs, supply chain networks are gradually evolving into supply chain ecosystems, resulting in a more complex structure that increases the likelihood that it will be affected by unexpected events [1]. Unfortunately, statistics indicate that nearly 75% of organizations experience supply chain disruptions annually [2], and these disruptions caused by risk can be expensive for organizations. Two distinct types of risks frequently impact supply chains: operational risk, which is related to systematic supply chain business risks and is due to the disturbance caused by the uncertainty of day-to-day operations, such as supply delays, demand fluctuations, equipment failures, and product quality. The other type of risk is disruption risk, caused by unexpected events that significantly impact the supply chain, such as natural disasters, wars, and policy changes. This type of risk is difficult to predict, has a low likelihood but a high impact, and may have irreversible negative consequences [3, 4]. Both types of risk, which are essentially the result of unpredictable internal and uncontrollable external factors, have the potential to disrupt the supply chain, and the latter type is more likely to do so.
Supply chain disruption is characterized by urgency, uncertainty, unpredictability, destructiveness, and diversity, which impede the receipt and delivery of expected goods and limit the ability of node companies to provide services to customers [5]. These impact the operational efficiency, market share, and profitability of supply chain nodes, ultimately weakening the performance of the supply chain through decreased revenue, customer complaints, lower productivity, and higher production costs. A more resilient supply chain can reduce the destructiveness and duration of supply chain disruption more effectively than its competitors and recover from disruption rapidly, thereby mitigating the impact of disruption on supply chain performance. Identifying the key elements of supply chain resilience, and comprehending and enhancing supply chain resilience, have significant implications for the future of nations and organizations [6].
Scholars have conducted extensive research on the elements of supply chain resilience, and some have worked on research projects related to the key elements of supply chain resilience [7–11]. However, few have focused their efforts on identifying the key elements. Understanding and enhancing supply chain resilience focuses on identifying its key elements and formation mechanism, enabling companies to invest efficiently with limited resources and effectively enhance supply chain resilience. Given the importance of supply chain resilience, the numerous elements affecting supply chain resilience, and the paucity of research on the identification of key resilience elements, this study employs grounded theory to conduct an exploratory study of key elements and the formation mechanism of supply chain resilience to answer two questions:
- RQ1: Among the various elements of supply chain resilience, what are the key elements?
- RQ2: At what phases of supply chain resilience do these key elements become operational, and what is the formation mechanism of supply chain resilience?
The following are the prospective innovations of this study: First, it enriches the study of key elements of supply chain resilience based on the dearth of scholarly research, and then it constructs and evaluates a scale for key elements of supply chain resilience. Second, this study innovatively integrates the key elements into the capabilities and three phases of supply chain resilience and uses a combination of qualitative and quantitative methods to construct a theoretical analysis framework of supply chain resilience based on the key elements, thereby enhancing the academic community’s understanding of the key elements and formation mechanism of supply chain resilience.
2. Literature review
2.1. Concepts of supply chain resilience
The demand for supply chain resilience is based on the fundamental premise that none of the risks can be eliminated entirely [12]. Organizations may lessen the disruption risks to the continued functioning of the supply chain by enhancing resilience, which is the prerequisite of the demand for resilience [13]. At the same time, improving supply chain resilience will surely be accompanied by a large amount of resource investment, which will inevitably increase the resilience shaping cost, and the excessive cost is bound to erode the profits of enterprises [14, 15]. In an environment that is dynamic and unpredictable, the ideal and actual states often diverge. Any supply chain disruption can have a ripple effect, with local disruptions propagating along the supply chain network and causing disruptions in the entire supply chain [16, 17]. In order to avoid disruptive effects, the supply chain must function even when disruptions occur. An emphasis on supply chain resilience has arisen due to the significant losses resulting from supply chain disruptions and their increasing frequency. Supply chain resilience is critical for preventing and responding to supply chain disruptions and ensuring continuity of supply chain operations. Therefore, supply chains need to be resilient to cope with the effects of environmental change.
According to Christopher & Peck [18], supply chain resilience refers to the capability of a system to bounce back to its original or more desirable state after the disturbance. Williams et al. [19] consider supply chain resilience as the ability of a supply chain to respond to unexpected disruptions and resume normal supply chain network operations. Dubey et al. [20] define supply chain resilience as the capability of a supply system to recover to its original state within an acceptable period after a disruption. It is clear from these definitions above that supply chain resilience is only defined in terms of post-response, which is reactive. Supply chain resilience can introduce the concept of disruption prevention to reduce the likelihood of disruptions and avoid their effects. According to Falasca et al. [21], supply chain resilience is the capability of a supply chain to reduce the likelihood of disruption, mitigate the consequences of disruption, and shorten the time it takes for the system to resume regular operation. Ponomarov and Holcomb [22] consider supply chain resilience as the adaptive capacity to cope with unforeseen events, respond to disruptions and recover from them. Hohenstein et al. [7] argue that supply chain resilience is the supply chain’s capability to be prepared for unexpected risk events, responding and recovering quickly to potential disruptions to return to its original situation or grow by moving to a new, more desirable state, which somehow affirms the supply chain’s capability to learn after disruptions and to turn threats into opportunities. Supply chain resilience is the capability of a supply chain to respond to disruptive events through both proactive and reactive forms [23].
A review of definitions reveals no consensus regarding the definition of supply chain resilience but that academics have consistently developed a comprehensive knowledge of the concept. It has evolved from reactive responses after the disruption to include both proactive preventions before disruption and reactive responses after the disruption. The capability to learn allows the supply chain to recover and function at a higher level, which is another aspect of supply chain resilience. In addition, supply chain resilience is confused with terms such as disruption, disturbance, disruptive event, and unexpected event, and there is no consensus. Particularly, disruption, disruptive events, and unexpected events narrow the set of possible events [24]. Disturbances are foreseen or unforeseen events that directly affect the regular operation and stability of an organization or supply chain [25] and are the pre-process and the manifestation of supply chain disruption. The supply chain will function normally if resilience successfully absorbs the disturbances. Otherwise, disruptions will occur, and supply chain disruption is the sequential result of the disturbance [26]. In this study, therefore, the term disturbance is used, which encompasses a broader range of events and more accurately depicts the reality of supply chain conditions prior to disruptions. In addition to rapid and efficient responses to disruptions, supply chain resilience requires monitoring and identifying risks to prevent disruptions. Many academicians view supply chain resilience as a capability that encompasses the capability to absorb, adapt, and recover [27–29]. In conclusion, this study defines supply chain resilience as a capability that includes the capability to absorb disturbance events before a supply chain disruption, the capability to adapt to the environment by responding rapidly after the disruption, and the capability to recover the supply chain to its original level or even higher.
2.2. Characteristics of supply chain resilience
2.2.1. Dynamic evolutionary.
The supply chain system has a close relationship with the ecosystem. According to the contingency theory, for organizations to achieve a higher performance level, their elements and behaviors must be compatible with the environment. Organizations can maintain resilience in dynamic and complex environments by employing a diverse and dynamic multitude of practices [30]. Supply chain resilience is a dynamic concept [31]; the nature of supply chain resilience is that it changes with the environment, and environmental changes create new resilience requirements [32]. In the process of continuous environmental change, the supply chain requires a dynamic capability to match it so that it has the appropriate anti-risk capability in the three phases of resilience (readiness, response, and recovery); otherwise, supply chain disruptions will occur. In the context of environmental uncertainty, supply chain resilience can continuously integrate, construct, and reconfigure supply chain resources to match the specific environment and accomplish dynamic evolution in response to dynamic changes and the condition of the supply chain. Supply chain resilience is a dynamic evolutionary capability of the supply chain [33].
2.2.2. Complexity.
The supply chain is becoming more complex due to increased market competition, demand volatility and product portfolios, shorter product life cycles, and faster innovation rates [34, 35]. In addition, as each level of the supply chain expands, it progressively develops into a supply chain ecosystem with a growing number of nodes. The same supply chain node companies may also exist in other chains, which may interact. Given the necessity to deal with the increasing complexity of supply systems, academicians and practitioners have focused on supply chain resilience [36]. The dynamic evolution characterizes the supply chain’s adaptability to a dynamic environment by exhibiting diverse resilience conditions and initiatives in response to changing environments and at different phases of the supply chain resilience [37]. Supply chain resilience is complicated by the interaction of the two dimensions of supply chain management complexity and management environment complexity, which are related to the supply chain’s internal complexity and disrupted by the external environment’s dynamic complexity [38]. Consequently, complexity is a characteristic of supply chain resilience.
2.2.3. Ambidexterity.
According to prior research, supply chain resilience requires proactive and reactive measures to effectively address disruptions [39–41]. At the same time, the supply chain has been continuously extended at all levels, and is gradually transforming into a supply chain ecosystem with a gradual increase in the number of nodes involved. The node enterprises in the same supply chain may also be in different supply chains, and each supply chain interacts with each other. The supply chain is increasingly characterized as a complex social system, with deepening complexity of association with the social environment, dynamics and non-linearity. The need to address escalating supply chain complexity has generated interest in supply chain resilience among scholars and supply chain practitioners [42]. Both supply chains and supply chain resilience have their unique characteristics and are evolutionary responses to specific environments. The dynamic evolutionary nature of supply chain resilience characterizes the ability of supply chains to adapt to dynamic environments, exhibiting different resilience profiles and initiatives in the face of different environments as well as at different stages of the supply chain [43], and is the ability to adapt to the dynamic complexity of the environment. Supply chain resilience is both related to the internal complexity of the supply chain and disturbed by the dynamic complexity of the external environment, i.e., the two dimensions of supply chain management complexity and management environment complexity together contribute to the complexity of supply chain resilience [39]. Therefore, complexity characterizes supply chain resilience.
2.2.4. Phase.
Supply chain resilience has been classified into different phases, each reflecting the peculiarities of the phase. The phases of supply chain resilience are readiness, response, and recovery [22, 23, 44], but Hohenstein et al. [7] argue that supply chain resilience consists of four phases: readiness, response, recovery, and growth, and the growth phase indicates that the supply chain can learn from disruptions and drive supply chain performance to a higher level. A more advanced manifestation of supply chain resilience is learning from disruptions and growing [45]. Knowledge management influences the recovery phase of supply chain resilience, which implies the supply chain’s capability to continuously learn from the feedback of disruptions to develop better plans and solutions for the future. Therefore, this study concurs with Ponomarov and Holcomb [22] that supply chain resilience includes three phases: readiness, response, and recovery, with the recovery phase also accounting for the growth implications. The three phases of supply chain resilience are depicted in Fig 1.
It is essential to recognize the possibility of disruption from any source and adequately prepare [46]. The readiness phase includes the forecasting, identification, and evaluation of environmental disturbances and the preparation for absorbing the disturbances to reduce the likelihood of disruption. The level of readiness is determined by how well the supply chain absorbs the disturbance. Following the failure of supply chain absorption disturbance, the response phase demonstrates a rapid response to supply chain disruptions as well as the dynamic evolutionary capability of resilience to continuously integrate, construct, and reconfigure the supply chain’s resources and capabilities in order to minimize disruption losses and prepare for subsequent recovery. The objective of the recovery phase is to quickly return the supply chain to its original state or even develop to a higher state while recovering and learning from the disruption.
2.3. Elements of supply chain resilience
According to academic research, recognizing the key elements of supply chain resilience requires clarifying the elements of supply chain resilience. Scholars have utilized inconsistent terminology when referring to supply chain resilience elements, these are referred to as enhancers [37], competencies [47], principles [25], elements [48], capabilities [28], antecedents [49], and enablers [50]. This study employs the term element, contributing to resilience development [51].
2.3.1. Classical literature review sorting.
This subsection reviews some of the scholars’ studies that identify the elements of supply chain resilience based on the systematic literature review approach, as shown in Table 1. In dynamic and complex environments, firms need to have the ability to perceive environmental uncertainty, and the possibility of supply chain disruptions can be reduced through proactive strategies such as increased awareness of environmental threats. In practice, when supply chain disruptions do not affect production, the probability is that they cannot attract management’s attention, and the lag in response strategies can cause irreparable harm to the supply chain [52]. From the results of literature review combing, scholars focus on certain elements of active resilience, such as supply chain risk management culture, situational awareness, and redundancy; at the same time, flexibility, redundancy, synergy, and agility are the focus of researchers’ attention, which deserves to be focused on; researchers usually add or adjust the elements that they think are needed on the basis of one another’s research; in addition to this, the elements of supply chain resilience, although becoming more and more comprehensive, but are somewhat cumbersome.
2.3.2. Stage division combing.
Numerous elements of supply chain resilience have been identified, and scholars have begun to explore which elements play a role in which stage of supply chain resilience, integrating the influencing elements of resilience into the stages of resilience. This is critical from a supply chain management perspective [11]. Table 2 summarizes the division of elements into different stages of supply chain resilience by some scholars, from which it can be observed that different scholars have different perspectives and identify different elements of resilience based on the differences in the division of resilience stages. It is worth noting that although the reference value is provided, the stage division of the elements is mostly summarized in the previous theoretical studies and lacks the support of empirical studies, which is insufficient in terms of scientificity and rigor. Therefore, there is a need to further clarify the stage division of supply chain resilience and the roles of numerous elements in each stage.
2.4. The need to identify key elements of supply chain resilience
Based on the resource base theory, firms possess different tangible and intangible resources, and by integrating and optimizing these resources, they gain and sustain firm performance and competitive advantage [54]. For supply chains, firms improve supply chain resilience based on the resources they possess through initiatives such as redundancy, optimizing supply chain network design, and investing in IT infrastructure, which in turn positively affects the level of supply chain performance. However, it is important to note that firms have limited resources and it is unrealistic to invest in all aspects of resilience. Given the complexity of the elements of resilience, enterprises need to focus on the key elements of resilience, transform the key elements into the resources they possess, improve output efficiency and the value of resources, and build supply chain resilience effectively and efficiently.
Based on the dynamic capability theory, dynamic capability is an ability to continuously integrate, construct and reconfigure internal and external resources and capabilities to adapt to environmental changes and gain competitive advantages [55]. As mentioned earlier, supply chain resilience is a dynamic capability with dynamic evolutionary and stage characteristics, and its evolutionary nature is closely linked to environmental changes [31]. In the preparation, response and recovery phases of supply chain resilience, the enterprise’s own state and the environment in which it operates change, and enterprise decision-making is always confronted with a lot of uncertainty and complexity, and there is variability in the resources and capabilities required to build absorption, adaptation and recovery capabilities. Ideally, firms would be able to continuously adjust their resources and capabilities to maintain a dynamic match with the needs of the environment. However, within the constraints of limited enterprise resources and capabilities, it is more realistic for enterprises to develop key dynamic resilience capabilities with the resources and capabilities they have available.
3. Research design
3.1. Research methods
Grounded theory is a qualitative research method that combines empirical and theoretical research, proposed by American scholars Glaser and Strauss in 1967. It bridges the gap between traditional qualitative and quantitative research to a certain extent and is one of the more scientific qualitative research methods. The three primary schools of grounded theory are classical grounded theory, represented by Glaser; proceduralised grounded theory, represented by Strauss; and constructivist grounded theory, represented by Charmaz [56]. The three primary schools of grounded theory embody positivist, interpretivism, and constructivist epistemologies. The differences in epistemologies are reflected in their methodological differences, which lead to differences in their approaches, concentrated in the coding sessions.
Different research questions should be matched with different research methods, and this paper’s grounded theory research method has its rationale. Firstly, grounded theory is appropriate for studies with insufficient explanations or imperfect theory construction. Given the dearth of research on the key elements of supply chain resilience, it is reasonable to use grounded theory to explore the research. Secondly, it is based on research scenarios. Theoretical questions in management research should be derived from specific management situations [57], and grounded theory is essentially a bottom-up approach to theory construction through a set of systematic operations [58].
This study employs proceduralised grounded theory to identify key elements of supply chain resilience, which are then coded using the qualitative data analysis software NVivo20. The proceduralised grounded theory divides the coding procedure into three steps: open, axial, and selective. Open coding extracts initial concepts from the data and categorizes them; axial coding integrates the categories and refines the main categories by delving deeper into the true meaning of the initial concepts and categories and the logical relationships between them; and selective coding converges the core categories, explores the relationships between the main categories and the core categories, and creates a narrative between the main categories and the core categories. After coding, a theoretical saturation test is required to ensure the scientific and rigorous character of the theoretical model construction.
3.2. Data collection
3.2.1. Interviewees.
The data required for the grounded theory method can be derived from first-hand data such as research interviews or second-hand data such as research literature, policy documents, and news reports. In this investigation, semi-structured interviews are used to collect first-hand data. In order to enhance the quality of the interviews, it is also necessary to establish the criteria for selecting interviewees. The basic requirements for the selection of interviewees are as follows: (1) select employees with a bachelor’s degree or higher as interviewees to ensure a good performance in logical analysis, understanding and judgment, and presentation skills; (2) require interviewees to have more than one year experience in work related to the supply chain and be familiar with the operational processes of the company, so that the interviewees can understand the purpose of the study and the content of the interview and give more professional and valuable information. In this study, 23 interviewees were selected, and all of them satisfied the basic requirements, and the basic information of the interviewees is shown in Table 3.
3.2.2. Interview data collection.
Before conducting the formal interviews, a more detailed semi-structured interview outline was developed, as shown in Table 4, based on this study’s research questions and objectives, in conjunction with the relevant literature and expert counsel. This study employed semi-structured interviews with 23 interviewees, each lasting at least 60 minutes, and interviews were conducted from 5 January to 29 January 2023. The interviews centered on three topics of supply chain resilience: comprehension, state, and elements. At the same time, the interview questions are flexible according to the actual situation in the interview process, and the interviewees are not bound to the interview outline. They are allowed to play freely and ask their questions. In addition to this, it is also necessary to use the STAR rule to extract information from the four aspects of the situation, task, action, and result.
4. Data analysis
The interview recordings are transcribed verbatim into textual materials, and the qualitative analysis software NVivo20 is utilized to guarantee a standard, regular, and systematic coding procedure. In this study, 19 of the 23 profiles are chosen randomly for formal coding, and four profiles are set aside for saturation testing as a basis for the re-validation of the identified and summarised results.
4.1. Open coding
Open coding is the process of conceptualizing and categorizing original data. Firstly, the interview text is conceptualized sentence by sentence without subjective bias or preconceptions, and the original meaning is retained as far as possible. Secondly, the initial concepts repeated less than three times are eliminated, and the categories are refined. Finally, 53 concepts are obtained, and 21 categories are abstracted from the concepts, as shown in Table 5.
4.2. Axial coding
The objective of axial coding is to analyze the logical relationships between various categories, to further summarise the categories derived from open coding, and to develop the main categories. Based on open coding, the logical relationships and evolutionary patterns between the 21 categories are compared. Six main categories are identified, namely product supply resilience, resource resilience, partner resilience, information response resilience, capital resilience, and knowledge resilience, as shown in Table 6.
Product supply resilience reflects the characteristics of supply chain output results and delivery, reflecting the company’s capability to respond to changes in market demand and provide services to customers and meet their needs, which is directly related to customer satisfaction. Resource resilience indicates that the company has a certain level of flexibility and redundancy in its human resources, procurement, production equipment, and logistics, which can absorb environmental disturbances and reflect its capability to deal with internal and external environmental uncertainties. Partner resilience implies that the company can change and modify its partners and operate collaboratively in response to environmental changes and demands to better adapt to the environment. Information response resilience reflects the capability of supply chain members to receive, process, and provide feedback on information to provide information support for timely action, allowing the company to respond quickly to changes in the external environment. Capital resilience reflects the economic strength of supply chain members to provide capital for supply chain recovery and growth from disruptions. Knowledge resilience is the capability of the company to learn from disruptions to improve operational efficiency while updating technology and developing new products to satisfy customer demand more efficiently and effectively, thereby enabling the supply chain to recover and grow.
4.3. Selective coding
Selective coding is the process of identifying core categories through an exhaustive analysis of all categories, analyzing the relationship paths between core categories and other categories, and developing a theoretical model. This study identified the core category of "key elements of supply chain resilience" through an in-depth analysis of 53 concepts, 21 categories, and six main categories, and the typical relationship structure is depicted in Table 7.
As shown in Fig 2, combined with the previous concepts, characteristics of supply chain resilience, and the results of coding, the formation mechanism of supply chain resilience can be explained as follows: the formation of supply chain resilience is a reflection of the integration of the capability of supply chain resilience (absorption, adaptation and recovery capability), the characteristics of dynamic evolution, ambidexterity (active resilience and reactive resilience) and phases (readiness, response and recovery phases). Before the supply chain disruption, that is, the readiness phase of supply chain resilience, the company primarily improves the absorption capability of supply chain resilience through product supply resilience and resource resilience, such as backup suppliers, inventory redundancy, multiple transportation channels, and other measures to absorb external disturbances, which exemplifies proactive resilience. As external disturbances intensify and the absorption capability fails, the supply chain is disrupted and enters the response phase. The company primarily increases the adaptability of supply chain resilience through partner resilience and information response resilience, which makes the supply chain respond quickly to the disruption and minimize the loss of disruption. Upon completion of the response phase, the supply chain gradually enters the recovery phase. With the capital support, the company collaborates with supply chain members to complete the recovery plan, improve the recovery capability of supply chain resilience, and progressively restore the supply chain. Concurrently, reflect on the causes of disruption, learn from disruption, and achieve growth. The capability to adapt and recover is a reflection of reactive resilience. In conclusion, product supply resilience, resource resilience, partner resilience, information response resilience, capital resilience, and knowledge resilience are the key elements that contribute to the formation of supply chain resilience, and they have different effects on different capabilities and different phases of supply chain resilience.
4.4. Theoretical saturation test
The theoretical saturation test is a coding analysis of the set-aside data to test for the emergence of new concepts and categories in the data. If new concepts and categories emerge, the data collection must be expanded for coding until no new concepts and categories emerge. This study randomly selects 19 of the 23 profiles for formal coding, while the remaining four are reserved for saturation testing. The absence of new concepts and categories after the test demonstrates that the theory developed in this paper is essentially close to saturation and that no further sample expansion is required.
5. Model verification
5.1. Research hypothesis
According to the grounded theory analysis, the key elements of supply chain resilience are product supply resilience, resource resilience, partner resilience, information response resilience, capital resilience, and knowledge resilience. The theoretical model is further validated and analyzed using a structural equation model to assess its rationality and reliability. The structural equation model is constructed with product supply resilience, resource resilience, partner resilience, information response resilience, capital resilience, and knowledge resilience as exogenous latent variables and supply chain resilience as endogenous latent variables. The following hypotheses are put forward:
- H1: Product supply resilience positively influences supply chain resilience;
- H2: Resource resilience positively influences supply chain resilience;
- H3: Partner resilience positively influencing supply chain resilience;
- H4: Information response resilience positively influences supply chain resilience;
- H5: Capital resilience positively influences supply chain resilience;
- H6: Knowledge resilience positively affects supply chain resilience.
5.2. Questionnaire design
Regarding the design of the measurement items, each exogenous latent variable mainly referred to the results of the previous spindle coding, as shown in Table 8. Regarding reliability and validity measures, the Likert 7-point scale is superior to the Likert 5-point scale [59, 60]. Therefore, the Likert 7 subscale is used with a scale of 1–7, ranging from strongly disagree to strongly agree.
5.3. Questionnaire pre-survey
Since the design of the scale is based on axial coding, the scale needs to be validated. The minimum sample size for the pretest should equal three times the number of measurement items on the scale [61]. The number of measurement items in this study’s pretest questionnaire is 33, so the minimum sample size for the pretest should be 99. The questionnaires were collected from 6 March to 27 March 2023, and the pre-survey of this study yields 143 questionnaires, of which 109 are valid, for an efficacy rate of 76.22%; the valid questionnaires collected satisfy the minimum sample size requirement of 99 for the pretest. This study uses SPSS 26 statistical analysis software for data processing and exploratory factor analysis.
5.3.1. Reliability test.
In this research, Cronbach’s α coefficient is used to measure reliability. Reliability is an indicator used to evaluate the consistency and stability of measurement results. In general, Cronbach’s α values between 0.7 and 0.8 have high reliability, and values greater than 0.8 are considered to have excellent reliability. As shown in Table 9, the overall reliability of the questionnaire is 0.930, and the reliability of each dimension ranged from 0.845 to 0.892, which is greater than the 0.7 standards for judging reliability [62]. The test results indicate that the scale’s reliability is high, and the variables have high internal consistency.
5.3.2. Initial scale purification.
In this study, the quality of the items is measured by the Corrected-Item Total Correlation (CITC) and Cronbach’s α coefficient after the item is deleted (CAID), and the combination of the two could be used to determine the appropriateness of an item. The CITC value can be used to characterize the overall correlation between a item and its dimension, and the threshold value for the CITC coefficient is typically 0.4. The items with CITC below 0.4 and CAID are more excellent than Cronbach’s α coefficients of the dimension to which they belonged [63–65]. Table 9 displays the results of the scale’s CAID and CITC. According to the calculated results, the CITC values of item RER2 of the resource resilience dimension and item KNR2 of the knowledge resilience dimension are less than 0.4, and Cronbach’s α coefficient of the respective dimension would increase after the deletion of the item. Consequently, the elimination of RER2 and KNR2 is considered.
5.3.3. Exploratory factor analysis.
Before conducting exploratory factor analysis, it is necessary to conduct the Kaiser-Meyer-Olkin (KMO) test and Bartlett’s sphericity test to determine whether the bias correlation between variables is sufficiently significant and whether the correlation coefficient matrix is a unit matrix, respectively. The closer the KMO value is to 1, the stronger the correlation between variables, and it is generally accepted that if the KMO statistic is more significant than 0.70 and Bartlett’s sphericity test reaches a significance level of 0.05 or higher, the model is more appropriate for factor analysis. As shown in Table 10, the KMO measure for this study is 0.844, and the P-value of Barth’s spherical test is less than 0.001, indicating some structure and correlation between the original variables. Combining these two factors, the data from this investigation are therefore suitable for factor analysis.
This study uses principal component factor analysis, the remaining 31 items are orthogonally rotated, and six factors with eigenvalues greater than one are extracted, yielding a cumulative interpretation variance of 75.241%, which is a high degree of explanation, and the analysis results are shown in Table 11. The first factor loadings are RER1~ RER8 items (RER2 has been removed), which have higher loading coefficients on the resource resilience factor; the second-factor loadings are PSR1~ PSR6 items, which have higher loading coefficients on the product supply resilience factor; The third-factor loadings are items KNR1~ KNR6 (KNR2 has been eliminated). These five items have high loading coefficients on the knowledge resilience factor; the fourth-factor loadings are CAR1~ CAR5 items, which have high loading coefficients on the capital resilience factor; the fifth-factor loadings are IRR1~ IRR4 items, which have high loading coefficients on the information response resilience factor; and the sixth-factor loadings are PAR1~ PAR4 items, and these four indicators have high loading coefficients on the knowledge resilience factor.
5.4. Formal questionnaire
5.4.1. Questionnaire design, distribution, and collection.
The scale of six key elements has been explored previously, and the supply chain resilience (SCR) variable is presented to validate the influence of six key elements on it. Supply chain resilience is measured concerning the scale of Ambulkar et al. [66] as We can cope with changes brought by the supply chain disruption (SCR1); We can adapt to the supply chain disruption easily (SCR2); We can provide a quick response to the supply chain disruption (SCR3); We can maintain high situational awareness at all times (SCR4). The rest of the scale design is the same as the pre-survey.
From 13 April to 15 May 2023, 543 questionnaires were collected, 409 of which are valid, for a valid recall rate of 75.32 percent. For credible parameter estimates, the ratio of a sample size to scale items must be at least 5:1, and a ratio of 10:1 or close to 10:1 is generally required to assure the validity of the significance test [67]. There are 35 items on the scale, and the ratio of valid questionnaires to items is 11.69:1, greater than the minimum of 10:1.
5.4.2. Reliability test.
Cronbach’s α coefficient can be used to test reliability. As shown in Table 12, the study results indicated that Cronbach’s α coefficient values for each scale dimension were above 0.7, higher than the 0.7 standards for judging reliability [62].
5.4.3. Exploratory factor analysis.
The KMO measure of the formal survey is 0.942, and the P-value of Barth’s spherical test is less than 0.001; the data from this investigation are therefore suitable for factor analysis. Using principal component factor analysis, orthogonal rotation is performed on 35 items, and seven factors are extracted with a cumulative interpretation variance of 65.987%, indicating a high degree of explanation; the analysis results are presented in Table 13.
5.4.4. Validity test.
The validity test includes content validity and construction validity, and construction validity can be divided into convergent validity and discriminant validity. In this paper, the six key elements of supply chain resilience are coded from the results of axial coding and validated by exploratory factor analysis, and the items of supply chain resilience are taken from the mature scales of existing literature. The items are pretested and validated by experts to ensure the quality of the scale content and therefore have high content validity.
Convergent validity can be examined utilizing item loading, composite reliability (CR), and average variance extracted (AVE). When the item loading is more significant than 0.6, and the CR and AVE values for each variable are more significant than 0.7 and 0.5, respectively, the scale’s convergent validity is high [68, 69]. As shown in Table 14, the minimum item loading is 0.611, which is greater than 0.6; the minimum CR value of each variable is 0.821, which is greater than 0.7; and the minimum AVE value is 0.518, which is greater than 0.5, indicating that the scale have good convergent validity.
The Discriminant validity criterion is that the square root of the AVE value of each variable (the main diagonal part) must be greater than the correlation coefficient between the variable and other variables [68]. As shown in Table 15, the square root of the AVE value of each variable is greater than the correlation coefficient between that variable and the other variables. Consequently, there is a high degree of discriminant validity between the items.
5.4.5. Model fit and hypothesis testing.
In this research, the Amos 28 software is used for model fitting and hypothesis testing, and Table 16 displays the fitting results of the theoretical model. Referring to the criteria of authoritative scholars, Hayduk [70], Bagozzi & Yi [71], Hu & Bentler [72], and Wu [73], all the fit indicators of this model satisfy the reference criteria, indicating that the overall fit between the data and the model is good. The results of the model’s hypothesis testing are shown in Table 17 and Fig 3. Product supply resilience, resource resilience, partner resilience, information response resilience, and capital resilience significantly positively affect supply chain resilience with path coefficients of 0.170, 0.200, 0.209, 0.229, and 0.218, respectively, and knowledge resilience positively affects supply chain resilience with a path coefficient of 0.103. In conclusion, all the six research hypotheses proposed in this paper are tested.
6. Summary
6.1. Conclusion
Based on grounded theory, this study systematizes and explores the key elements and formation mechanisms of supply chain resilience. Open coding, axial coding, selective coding, and a theoretical saturation test on the semi-structured interview data reveal that product supply resilience, resource resilience, partner resilience, information response resilience, capital resilience, and knowledge resilience are the six key elements of supply chain resilience. Among them, product supply resilience and resource resilience help to improve the absorption capability of supply chain resilience and act in the readiness phase of supply chain; partner resilience and information response resilience help to improve the adaptation capability of supply chain resilience and act in the response phase of supply chain; capital resilience and knowledge resilience help to improve the recovery capability of supply chain resilience and act in the recovery phase of supply chain; and capital resilience and knowledge resilience help to improve the recovery capacity of supply chain resilience and act in the recovery phase of supply chain. Based on these findings, a theoretical model is constructed, and corresponding hypotheses are proposed. Then, a questionnaire is designed based on the results of axial coding, and a small sample pre-survey verifies the reasonableness of the questionnaire. Finally, the structural equation modeling method is used to evaluate the theoretical model based on questionnaire data from the formal survey. The validation results indicate that product supply resilience, resource resilience, partner resilience, information response resilience, capital resilience, and knowledge resilience positively affect supply chain resilience, and the proposed hypotheses are all supported.
6.2. Implications
This study compensates for the lack of research in the identification of key elements of supply chain resilience, expands the study of key elements and formation mechanism of supply chain resilience, enriches the theoretical research perspective and empirical research scenarios of supply chain resilience, and is a valuable addition to the existing theoretical system of supply chain resilience, so it has some theoretical value. In the meantime, the scale of six key elements designed by this study through axial coding has been validated, providing a benchmark for designing associated questionnaires and developing the scale. In addition, the six identified key elements of supply chain resilience are the top priorities for enhancing supply chain resilience, indicating the management’s primary concerns. The categories and concepts illustrated in the coding process of the six key elements can serve as a detailed reference for practitioners seeking to enhance supply chain resiliency and have practical significance.
6.3. Research limitations and perspectives
This study mainly has the following limitations: (1) The six key elements of supply chain resilience identified come from the grounded theory, and their empirical data inevitably have some subjectivity, which may influence the coding results. (2) The division of the six key elements into the three phases of supply chain resilience based solely on qualitative interview data lacks a certain level of scientific validity. (3) The conclusions of this study are somewhat universal and lack industry-specific targeted research.
In the future, scholars can classify the phases of action of the six key elements from a quantitative perspective. Moreover, the key elements of supply chain resilience can be extracted based on different industries in a targeted manner. In addition, effective and efficient paths to build supply chain resilience can be proposed based on the six key elements of supply chain resilience in a targeted manner.
References
- 1. Chen J, Liu Y. Operations management innovation enabled by digitalization and intellectualization: From supply chain to supply chain ecosystem. Journal of Management World. 2021; 37(11), 227–240.
- 2. Scholten K, Stevenson M, van Donk DP. Dealing with the unpredictable: Supply chain resilience. International Journal of Operations & Production Management. 2019; 40(1), 1–10.
- 3. Kinra A, Ivanov D, Das A, Dolgui A. Ripple effect quantification by supplier risk exposure assessment. International Journal of Production Research. 2020; 58(18), 5559–5578.
- 4. Baz JE, Ruel S. Can supply chain risk management practices mitigate the disruption impacts on supply chains’ resilience and robustness? Evidence from an empirical survey in a COVID-19 outbreak era. International Journal of Production Economics. 2021; 233, 107972. pmid:36567758
- 5. Ivanov D, Dolgui A, Sokolov B, Ivanova M. Literature review on disruption recovery in the supply chain. International Journal of Production Research. 2017; 55(20), 6158–6174.
- 6. Sawyerr E, Harrison C. Developing resilient supply chains: Lessons from high-reliability organisations. Supply Chain Management: An International Journal. 2019; 25(1), 77–100.
- 7. Hohenstein NO, Feisel E, Hartmann E, Giunipero L. Research on the phenomenon of supply chain resilience: A systematic review and paths for further investigation. International Journal of Physical Distribution & Logistics Management. 2015; 45(1/2), 90–117.
- 8. Naimi MA, Faisal MN, Sobh R, Sabir LB. A systematic mapping review exploring 10 years of research on supply chain resilience and reconfiguration. International Journal of Logistics Research and Applications. 2022; 25(8), 1191–1218.
- 9. Karl AA, Micheluzzi J, Leite LR, Pereira CR. Supply chain resilience and key performance indicators: A systematic literature review. Production. 2018; 28, e20180020.
- 10. Kumar M, Raut RD, Sharma M, Choubey VK, Paul SK. Enablers for resilience and pandemic preparedness in food supply chain. Operations Management Research. 2022; 15(3), 1198–1223.
- 11. Jüttner U, Maklan S. Supply chain resilience in the global financial crisis: an empirical study. Supply Chain Management: An International Journal. 2011; 16(4): 246–259.
- 12. Liu JG, Jiang XH, Zhao JL. Resilience of the supply chain system based on interpretative structural modeling. Journal of Systems & Management. 2015; 24(4), 617–623.
- 13. Aslam H, Khan AQ, Rashid K, Rehman S. Achieving supply chain resilience: The role of supply chain ambidexterity and supply chain agility. Journal of Manufacturing Technology Management. 2020; 31(6), 1185–1204.
- 14. Wang Y, Gao Y, Teng C. Literature review and research prospects of supply chain resilience under disruption. Management Review. 2017; 29(12): 204–216.
- 15. Katsaliaki K, Galetsi P, Kumar S. Supply chain disruptions and resilience: a major review and future research agenda. Annals of Operations Research. 2022; 319(1): 965–1002. pmid:33437110
- 16. Ivanov D, Sokolov B, Dolgui A. The Ripple effect in supply chains: Trade-off ‘efficiency-flexibility-resilience’ in disruption management. International Journal of Production Research. 2014; 52(7), 2154–2172.
- 17. Dolgui A, Ivanov D, Sokolov B. Ripple effect in the supply chain: An analysis and recent literature. International Journal of Production Research. 2018; 56(1–2), 414–430.
- 18.
Christopher M, Peck H. Building the resilient supply chain. 2004. https://doi.org/10.1108/09574090410700275.
- 19. Williams Z, Ponder N, Autry CW. Supply chain security culture: measure development and validation. The International Journal of Logistics Management. 2009; 20(2): 243–260.
- 20. Dubey R, Gunasekaran A, Childe SJ, Papadopoulos T, Blome C, Luo Z. Antecedents of resilient supply chains: An empirical study. IEEE Transactions on Engineering Management. 2019; 66(1), 8–19.
- 21.
Falasca M, Zobel CW, Cook D. A decision support framework to assess supply chain resilience. In Proceedings of the 5th International ISCRAM Conference. 2008; (pp. 596–605).
- 22. Ponomarov SY, Holcomb MC. Understanding the concept of supply chain resilience. The International Journal of Logistics Management. 2009; 20(1): 124–143.
- 23. Chowdhury MMH, Quaddus M. Supply chain resilience: Conceptualization and scale development using dynamic capability theory. International Journal of Production Economics. 2017; 188, 185–204.
- 24.
Liu X. Research on the formation mechanism and dynamic simulation of cross border E-commerce supply chain resilience. Yunnan University of Finance and Economics, Yunnan. 2021. https://doi.org/10.27455/d.cnki.gycmc.2021.000796.
- 25. Kamalahmadi M, Parast MM. A review of the literature on the principles of enterprise and supply chain resilience: Major findings and directions for future research. International Journal of Production Economics. 2016; 171, 116–133.
- 26. Liu J. Supply chain disruption: Causes, consequences and countermeasures: Literature review from the perspective of supply chain governance. Journal of Zhongnan University of Economics and Law. 2022; (4), 130–144.
- 27. Hosseini S, Ivanov D, Dolgui A. Review of quantitative methods for supply chain resilience analysis. Transportation Research Part E: Logistics and Transportation Review. 2019; 125, 285–307.
- 28. Kochan CG, Nowicki DR. Supply chain resilience: A systematic literature review and typological framework. International Journal of Physical Distribution & Logistics Management. 2018; 48(8), 842–865.
- 29.
Shishodia A, Sharma R, Rajesh R, Munim ZH. Supply chain resilience: A review, conceptual framework and future research. The International Journal of Logistics Management, ahead-of-print(ahead-of-print). 2021. https://doi.org/10.1108/IJLM-03-2021-0169.
- 30. Dabhilkar M, Birkie SE, Kaulio M. Supply-side resilience as practice bundles: A critical incident study. International Journal of Operations & Production Management. 2016; 36(8), 948–970.
- 31. Ali I, Golgeci I, Arslan A. Achieving resilience through knowledge management practices and risk management culture in agri-food supply chains. Supply Chain Management: An International Journal. 2021; 28(2), 284–299.
- 32. Adobor H, McMullen RS. Supply chain resilience: A dynamic and multidimensional approach. The International Journal of Logistics Management. 2018; 29(4), 1451–1471.
- 33.
Bai Y. A study on the factors influencing supply chain resilience based on a dynamic capability perspective[D]. Nanjing: Nanjing Normal University, 2018.
- 34. Gupta S, Modgil S, Meissonier R, Dwivedi YK. Artificial intelligence and information system resilience to cope with supply chain disruption. IEEE Transactions on Engineering Management. 2021; 1–11.
- 35. Negri M, Cagno E, Colicchia C, Sarkis J. Integrating sustainability and resilience in the supply chain: A systematic literature review and a research agenda. Business Strategy and the Environment. 2021; 30(7), 2858–2886.
- 36. Hossain NUI, Fazio SA, Lawrence JM, Gonzalez EDRS, Jaradat R, Alvarado MS. Role of systems engineering attributes in enhancing supply chain resilience: Healthcare in context of COVID-19 pandemic. Heliyon. 2022; 8(6), e09592. pmid:35721689
- 37. Blackhurst J, Dunn KS, Craighead CW. An empirically derived framework of global supply resiliency: Framework of global supply resiliency. Journal of Business Logistics. 2011; 32(4), 374–391.
- 38. Sheng Z, Wang H, Hu Z. Supply chain resilience: Adapting to complexity—based on the perspective of complex system thinking. Chinese Journal of Management Science. 2022; 30(11), 1–7.
- 39. Cheng J, Lu K. Enhancing effects of supply chain resilience: Insights from trajectory and resource-based perspectives. Supply Chain Management: An International Journal. 2017; 22(4), 329–340.
- 40. Maharjan R, Kato H. Resilient supply chain network design: A systematic literature review. Transport Reviews. 2022; 42(6), 739–761.
- 41. Ozdemir D, Sharma M, Dhir A, Daim T. Supply chain resilience during the COVID-19 pandemic. Technology in Society. 2022; 68, 101847. pmid:35075312
- 42. DuHadway S, Carnovale S, Hazen B. Understanding risk management for intentional supply chain disruptions: risk detection, risk mitigation, and risk recovery. Annals of Operations Research. 2019; 283(1): 179–198.
- 43. Li X, Chung C, Goldsby TJ, Holsapple CW. A unified model of supply chain agility: The work‐design perspective. The International Journal of Logistics Management. 2008; 19(3): 408–435.
- 44. Rahman T, Paul SK, Shukla N, Agarwal R, Taghikhah F. Supply chain resilience initiatives and strategies: A systematic review. Computers & Industrial Engineering. 2022; 170, 108317.
- 45. Han Y, Chong WK, Li D. A systematic literature review of the capabilities and performance metrics of supply chain resilience. International Journal of Production Research,. 2020; 58(15), 4541–4566.
- 46. Blos MF, Wee HM, Yang J. Analysing the external supply chain risk driver competitiveness: A risk mitigation framework and business continuity plan. Journal of Business Continuity & Emergency Planning. 2010; 4, 368–374. pmid:21177223.
- 47. Wieland A, Wallenburg CM. The influence of relational competencies on supply chain resilience: A relational view. International Journal of Physical Distribution & Logistics Management. 2013; 43(4): 300–320.
- 48. Ali A, Mahfouz A, Arisha A. Analysing supply chain resilience: Integrating the constructs in a concept mapping framework via a systematic literature review. Supply Chain Management: An International Journal. 2017; 22(1): 16–39.
- 49. Ali I, Gölgeci I. Where is supply chain resilience research heading? A systematic and co-occurrence analysis. International Journal of Physical Distribution & Logistics Management. 2019; 49(8): 793–815.
- 50. Tortorella GL, Fogliatto FS, Gao S, Chan TK. Contributions of industry 4.0 to supply chain resilience. The International Journal of Logistics Management. 2021; 33(2): 547–566.
- 51. Scala B, Lindsay CF. Supply chain resilience during pandemic disruption: Evidence from healthcare. Supply Chain Management: An International Journal. 2021; 26(6): 672–688.
- 52. Zhang Y, Long J, Chen Y. Contingency policy and operation simulation of supply chain disruption with changing demand. Journal of Systems &Management. 2019; 28(6): 1202–1210.
- 53. Stone J, Rahimifard S. Resilience in agri-food supply chains: A critical analysis of the literature and synthesis of a novel framework. Supply Chain Management: An International Journal. 2018; 23(3): 207–238.
- 54. Ray G, Barney JB, Muhanna WA. Capabilities, business processes, and competitive advantage: Choosing the dependent variable in empirical tests of the resource-based view. Strategic Management Journal. 2004; 25(1): 23–37.
- 55. Teece DJ, Pisano G, Shuen A. Dynamic capabilities and strategic management. Strategic Management Journal. 1997; 18(7): 509–533.
- 56. Jia X, Heng L. The "jungle", history, and approach road of the grounded theory. Science Research Management. 2020; 41(5), 151–163.
- 57. Jia XD, Tan XH. The actual value of the classical grounded theory and its spirit to china management research. Chinese Journal of Management. 2010; 7(5), 656–665.
- 58. Martin PY, Turner BA. Grounded theory and organizational research. The Journal of Applied Behavioral Science. 1986; 22(2), 141–157.
- 59. Brown JD. Likert items and scales of measurement. Statistics. 2011; 15(1), 10–14.
- 60. Dawes J. Five point vs. eleven point scales: Does it make a difference to data characteristics? Australasian Journal of Market Research. 2002; 10(1). https://www.researchgate.net/publication/285731517.
- 61.
Wu M. SPSS statistical application practice. Beijing: China Railway Press. 2000.
- 62. Chin WW. Commentary: Issues and opinion on structural equation modeling. MIS Quarterly. 1998; 22(1), 7–16.
- 63. Yoo B, Donthu N. Developing and validating a multidimensional consumer-based brand equity scale. Journal of Business Research. 2001; 52(1), 1–14.
- 64. Song XK, Zhao YX, Zhang XH. Developing a fear of missing out (FoMO) measurement scale in the mobile social media environment. Library and Information Service. 2017; 61(11): 96–105.
- 65. Zhang L, Song XK, Sun XL, Zhu QH. A study on measurement of users’ information literacy under social search context. Journal of Modern Information. 2021; 41(10), 57–68.
- 66. Ambulkar S, Blackhurst J, Grawe S. Firm’s resilience to supply chain disruptions: Scale development and empirical examination. Journal of Operations Management. 2015; 33–34(1), 111–122.
- 67. Bentler PM, Chou CP. Practical issues in structural modeling. Sociological Methods & Research. 1987; 16(1), 78–117.
- 68. Fornell C, Larcker DF. Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research. 1981; 18(1), 39–50.
- 69. Hair JF, Ringle CM, Sarstedt M. PLS-SEM: Indeed a silver bullet. Journal of Marketing Theory and Practice. 2011; 19(2), 139–152.
- 70.
Hayduk LA. Structural equation modeling with LISREL: Essentials and advances. Baltimore: Johns Hopkins University Press. 1987. https://sc.panda321.com/extdomains/books.google.com/books/about/Structural_Equation_Modeling_with_LISREL.html?hl=zh-CN&id=AuGYVSo64qoC.
- 71. Bagozzi RP, Yi Y. On the evaluation of structural equation models. Journal of the Academy of Marketing Science. 1988; 16(1), 74–94.
- 72. Hu L, Bentler PM. Fit indices in covariance structure modeling: Sensitivity to underparameterized model misspecification. Psychological Methods. 1998; 3, 424–453.
- 73.
Wu M. Structural equation modeling: Operation and application of AMOS. Chongqing: Chongqing University Press. 2010.