Multilevel Analysis of the Effect of Corporate Entrepreneurship and Learning Capabilities on the Attitudes of Key Personnel in Small and Medium Enterprises in Korea: The Moderating Effect of Social Capital

This study investigates the multilevel impact of corporate entrepreneurship and learning capability on job satisfaction and organizational commitment and reveals the moderating effect of social capital. Using a quantitative approach based on secondary data from the Human Capital Corporation Panel (HCCP) provided by the Korea Research Institute for Vocational Education & Training (KRIVET). A total of 344 people working for 67 SMEs were processed to infer the results through the hierarchical linear model (HLM). We found that corporate entrepreneurship and learning ability had significantly positive effects on organizational commitment and job satisfaction. Further, social capital had a moderating effect only on the relationship between corporate entrepreneurship and organizational commitment, and there was no moderating effect on the relationship between corporate entrepreneurship and job satisfaction as well as the relationship between learning capability, organizational commitment, and job satisfaction. To inspire corporate entrepreneurship, individuals or teams should work with the company to create new organizations or incite innovation within the organization. Through social interaction, we must build and expand the consensus of our members and share values, attitudes, and objectives. Drawing on these findings, we suggest directions for future research.


Introduction
Since the turn of the 21st century, both government and private institutions have recognized the need for security and management of advanced technologies.Consequently, they have been expanding the collection of information in various industries.In an era where the securing, nurturing, and managing of talent determines success, companies are employing various strategies to recruit and retain key talents.Faced with changing and increasingly competitive business realities, the value and competitiveness of an entity are determined by Human Resource Management (HRM), an intangible resource rather than a tangible one (S.L. Shin, 2007).Successful recruitment, nurturing, and management of key personnel, and the reduction of the attrition rate are important defining aspects for any firm's core competitiveness.
Owing to fierce competition and rapid technological changes that have accompanied the globalization of the management environment, companies are increasingly interested in differentiated strategies for securing and retaining key talent (Ma et al., 2018).In particular, the leakage of key personnel leads to the leakage of key technologies.If key personnel are lost to foreign competitors, it can be a serious national problem (D. S. Shin, 2020).Since key personnel are a source of corporate competitiveness and greatly affect the survival of a company as a driving force for sustainable growth, companies are devising various strategies to secure and retain key employees (Choudhary, 2016).However, it is more difficult to find a solution to the management problem of key personnel in Small and Medium Enterprises (SMEs), which often have poor environments and limited internal resources compared to large enterprises (Josefy et al., 2015;Lai et al., 2016).According to the Transaction Cost Theory (TCT) as well as the Resource-Based Theory (RBT), opportunism (i.e., the intention of key talent to change or switch jobs) increases the cost of maintaining core talent in SMEs.Consequently, for SMEs, retaining core talent is a way to improve corporate performance.Therefore, a company's core workforce is an important factor in determining its competitiveness (Alvarez & Barney, 2017;Kauppila, 2015).When key personnel with high corporate strategic value leave the company, both indirect costs (e.g., loss of tacit knowledge, reduced labor productivity) and direct costs (e.g., recruitment, adjustment period and training for new employees, general administration) are incurred, resulting in large losses (Coetzer et al., 2019).Thus, SMEs struggle to develop measures and strategies to secure and retain key personnel.
The existing scholarship on HRM has studied the direct effect of personnel management on organization performance (Dyer & Reeves, 1995;Wright et al., 2001).The results of these studies have been verified empirically as well (Delery & Doty, 1996;Huselid, 1995;Macduffie, 1995).Based on a meta-analysis, these studies empirically support the importance of high-performance work practices toward organizational performance (Combs et al., 2006).Although many previous studies have identified humane aspects as a key factor in a company's competitive advantage, there has been an ongoing debate on the lack of research on the key factors for workers' positive attitudes, which are critical drivers for organizational performance (Batt, 2002;Lepak et al., 2006;Takeuchi et al., 2007;Wright & Boswell, 2002).In addition, previous studies on entrepreneurship heavily focused on the strategic growth of entrepreneurial organizations (Covin & Slevin, 1991;Hoskisson et al., 2011;Phan et al., 2009;Zahra et al., 2009).However, it might be the ideal time to refocus on both strategic and humane elements within entrepreneurial firms so that we can understand the various multilevel mechanisms that promote entrepreneurial actions (Santos et al., 2021).
Given these research gaps, our study utilizes the ambidexterity theory to investigate the multilevel impact on employee attitudes.Ducan (1976), who proposed the concept of ambidexterity theory, argues that successful organizations are ambidextrous organizations.Further, O'Reilly Ca and Tushman (2004) argue that ambidextrous organizations are more aggressive than other organizational types, particularly when developing new innovative products.Both He and Wong (2004) and Gibson and Birkinshaw (2004) argue that utilization and exploration are not mutually exclusive concepts and that the balance between utilization and exploration can overcome the inertia of organizations that are immersed in existing businesses and products, preventing blind immersion in search activities that are difficult to generate revenue from in the short term (Jansen et al., 2009).The ambidextrous concept has been explained by researchers in a number of different ways.Ducan (1976) focuses on the ability to easily handle two different structural processes required for organizational innovation.On the other hand, Adler et al. (1999) describe it as the ability of an organization to simultaneously pursue two different activities, such as efficiency and flexibility.Further, O' Reilly and Tushman (2008) describe it as the ability of an organization to simultaneously perform two opposing functions: exploration and utilization.What has gained ascendency within ambidexterity theory is that the balance between exploration and utilization has become an important foundation for defining ambidextrous concepts, and ambidextrous organizations are defined as organizations with the ability to explore new opportunities while actively utilizing existing capabilities and resources.
Studies on utilization and exploration have been actively conducted in the fields of management strategy and organizational learning.Exploration can be represented by words such as find, change, activity, and innovation, whereas utilization can be represented by words such as implementation, selection, production, and improvement (I.G. Lee, 2019).The proper balance between utilization and exploration is an important factor in the maintenance and expansion of organizations and systems (March, 1991).Studies on ambidextrous organizations have compared and analyzed both utilization-based and exploratory businesses (O'Reilly Ca & Tushman, 2004) and investigated the effects of exploration and utilization strategies on corporate performance (Ha, 2010).Research on ambidextrousness as an issue between organizations has implications for strategic alliances as well as performance (Lin et al., 2007;Tiwana, 2008) and investments in corporate venture capital (Wadhwa & Kotha, 2006).Hence, utilization and exploration research, which began in the strategic management field, is now being conducted in various management fields.
The path to becoming an ambidextrous organization is largely explained in two ways: structural and contextual ambidexterity.Structurally ambidextrous organizations are those in which each individual or group is in charge of exploration and utilization according to the organizational plan, whereas in situationally or contextually ambidextrous firms, these tasks are performed separately at the individual level (Y.D. Kim, 2020).Moreover, situational ambidextrous handles have more flexible properties than the former, wherein individual abilities and manager roles are important.Therefore, structurally ambidextrous firms are clearly engaged in multilevel activities.Using the ambidextrous theory dealing with two different processes, corporate entrepreneurship can be viewed as exploration (innovation, organizational change, etc.) and individual-level learning capability can be viewed as utilization (execution, implementation, production, etc.).Corporate entrepreneurship and learning capability are evolving concepts that focus on nurturing an environment that supports and motivates people to work with an entrepreneurial mindset, indulging in creativity and innovation within the existing organization and thus resulting in higher business performance prospects.Notably, entrepreneurship literature has underscored employee attitudes (e.g., organizational commitment and job satisfaction) to be highly significant in enhancing performance perspectives (Kompaso & Sridevi, 2010;Saks, 2006;Soomro & Shah, 2019).Therefore, this study deviates from the conventional general business-oriented measurement method and attempts to integrate the individual and organizational levels.This can be achieved by investigating the multilevel-level ambidextrous impact on individual-level attitudinal outcomes.
More importantly, we will investigate the moderating role of social capital, which can be defined as trust among organizational members, organic interaction activities, and vision sharing based on mutual understanding (K.K. Nam, 2007;Seol et al., 2018).Most scholars agree that social capital is formed via social relationships between members of a group and that it brings visible benefits to the whole group as well as individual members through the network of relationships.However, as a concept that can be generally applied in the field of social science, which has different conceptual definitions depending on the research purpose and area.Bourdieu (1986) described it as continuous networks, interactions, or recognition as an institutionalized relationship-the sum of actual or potential resources acquired by being a member of a particular group.Coleman (1988) understood it by its function-as not a single entity but a variety of entities that have two things in common: aspects of a certain social structure and a pattern of individual behavior deriving from such a structure.In other words, it is an entity that influences certain patterns of behavior in individuals belonging to a certain social structure.In addition, social capital, such as physical and human capital, is not completely replaceable.It is only replaceable depending upon specific activities.Putnam (2001) refers to the characteristics of social organizations such as trust, norms, and networks, and defines them as the ability to spread trust in a particular society.Fukuyama (1995) described it as the ability of people to work together for a common goal in a group or organization-a set of informal values or norms shared among members of a group seeking cooperation.Nahapiet and Ghoshal (1998) state that social capital constitutes all assets that are displayed through a network as the sum of substantial and potential resources that are inherent, useful, and can be obtained within the network of relationships owned by individuals or social units.Taken together, social capital can be conceptualized as trust among organizational members, organic interaction activities, and vision sharing based on mutual understanding, and it has the following characteristics (K.K. Nam, 2007;Seol et al., 2018).First, social capital is not the capital held by individual members but the capital inherent in relational attributes among members of a specific group, unlike economic, human, and cultural capital.Second, it requires constant effort for members who are in a relationship to maintain social capital continuously (S.I. Lee, 2017).Third, the benefits of economic, human, or cultural capital go to the person who owns the capital exclusively, but the benefits of social capital are shared (Putnam, 2001).Fourth, social capital has the characteristic that the more its members use it, the more its value accumulates and the more it increases.Therefore, economic, human, and cultural capital become increasingly less valuable as they are used, whereas social capital has the characteristic of becoming less valuable if not used (Adler & Kwon, 2002).Thus, we investigate how social capital moderates the multilevel effects of corporate entrepreneurship and learning capability on employees' organizational commitment and job satisfaction.As shown in Figure 1, we develop the conceptual model underlying our research and propose several hypotheses.

Corporate Entrepreneurship and Employees' Attitudes
SMEs are facing huge challenges in terms of employee satisfaction, organizational commitment, and enhancement of organizational performance globally, particularly in developing economies (Farooq & Sajid, 2015).However, few studies have focused on this issue in Korea, even though SMEs are regarded as an engine of growth for developing economies (Soomro & Shah, 2019).With this as its background, the present study attempts to explore the influence of corporate entrepreneurship on job satisfaction and organizational commitment in the SME sector of Korea.Recently, humane aspects of entrepreneurship are gaining huge attention from the entrepreneurial community and are based on the idea that ''entrepreneurs should extend their priorities beyond their profit margin, toward their employees, people, environment, and society'' (Parente et al., 2018).Previous studies have focused on the strategic growth of entrepreneurial organizations (Covin & Slevin, 1991;Hoskisson et al., 2011;Phan et al., 2009;Zahra et al., 2009).However, it might be time to refocus on the humane elements within entrepreneurial firms so that we can understand the employee attitudes that foster entrepreneurial behavior (Santos et al., 2021).Furthermore, organizational members' entrepreneurial competence is important for career development, individual life, and future preparations for the labor market.
Corporate entrepreneurship involves decision-making and implementation in a top-down manner with the goal of creating new businesses or fueling innovations.It is an organizational process that is combined with the redistribution of resources.Additionally, corporate entrepreneurship can be seen as a strategic innovation mechanism to secure an organization's competitive advantage in the industry (Lumpkin & Dess, 1996).Chua et al. (1999) define corporate entrepreneurship as the entrepreneurship that occurs in the process of individuals or groups belonging to an existing company creating new organizations or businesses or achieving innovations impacting the entire organization.Zahra (1991) describes a corporate entrepreneurship entity as an activity perspective, either formally or unofficially within an existing organization, creating new businesses and developing markets through innovation in the product or manufacturing process.In other words, the process of strategic innovation by members of the organization to improve the performance of the organization (Chua et al., 1999).Phan et al. (2009) employ this concept to explain the process in which established companies, not startups, act innovatively and take risks.It is a concept that seeks to continuously pursue organizational innovation.Kuratko, Ireland et al. (2005) depict it as actions that improve an entity's performance through innovation, risk-taking, and initiative to overcome the limitations or constraints of its resources.A few of corporate entrepreneurship's goals consist of converting market-oriented ideas into opportunities, pursuing change and innovation to secure a competitive edge, and taking risks and pursuing business ahead of competitors.Consequently, it is very similar to entrepreneurship, except that corporate entrepreneurship occurs within the corporate organization.Hornsby et al. (2002) mention managerial support, work autonomy, compensation and promotion, time availability, and organizational boundaries as internal factors influencing the operation of corporate entrepreneurship.Hodgetts and Kuratko (2007) find that, from the perspective of the organization, the value of entrepreneurship lies in the recognition of innovation and challenges at the organizational level.The constituent concepts of corporate entrepreneurship were innovativeness, initiative, and risk-taking (Covin & Slevin, 1989, 1991).Lumpkin and Dess (1996) expand the idea into five categories: innovation, initiative, risk- taking, autonomy, and competitive aggression.While various concepts in the spirit of corporate entrepreneurship are established, there are studies that present innovation, risk-taking, and initiative as important concepts for creating new businesses (Covin & Slevin, 1991;Kuratko, 2016;Zahra, 1995).
First, we argue that there is a positive relationship between corporate entrepreneurship and organizational commitment.Organizational commitment has been a major topic for management research since the 1980s and can be understood as an employee's intensive attribution to the organization, a sense of belonging, and identification (Neininger et al., 2010).Organizational commitment is receiving attention as an important management practice in the study of organizational behavior.This is primarily because organizational commitment has a positive relationship with various performance outcomes.For example, it is a variable that plays an important role in improving organizational performance measures such as a low turnover rate, absenteeism rate, and prevention of personal burnout (Angle & Perry, 1981;Buchanan, 1974;Park & Rainey, 2007;Stazyk et al., 2011;K. Yang & Pandey, 2008).Corporate entrepreneurship is considered a strategic means of boosting organizational vitality, generating competitive advantage, and improving performance (Bierwerth et al., 2015;Zahra & Covin, 1995).Furthermore, several studies show that corporate entrepreneurship improves the overall performance of a firm and brings significant competitive advantages (B.Antoncic & Hisrich, 2001;Covin & Slevin, 1989;Zahra & Covin, 1995).Whereas existing scholarship has focused on the impact of corporate entrepreneurship on financial performance (Zehir & Eren, 2007), there is a lack of research on corporate entrepreneurship's role in non-financial performance, such as employee attitudes and behaviors.Nevertheless, it is noteworthy that Lumpkin and Dess (1996) suggested a link between corporate entrepreneurship and nonfinancial outcomes such as organizational commitment.
Second, we argue that there is a positive relationship between corporate entrepreneurship and job satisfaction.Job satisfaction is used as a determinant of productivity, in sociology it acts as a measurement of alienation, and in economics as a proxy for utility as well as a predictor of various labor market phenomena.Lichtenstein (1984) emphasizes the subjective conceptual aspect of an individual rather than an objective fact because it is a complex combination of job cognition, emotion, and behavior (Jin & Song, 2012).Studies related to job satisfaction generally regard job satisfaction as a key employee attitude, which is a relatively stable emotional and evaluative tendency toward a specific person, situation, or thing.Thus, job satisfaction is an individual's attitude related to the job, including cognitive, evaluating, emotional, and behavioral factors.Kuratko, Hornsby, and Bishop (2005) found that recognition of corporate entrepreneurship factors, such as discretion and reward/enhancement, increases job satisfaction.Many studies have identified a positive relationship between corporate entrepreneurship and job satisfaction and its importance as a factor in corporate growth (Akehurst et al., 2009;J. A. Antoncic & Antoncic, 2011;Comeche & Loras, 2010;Kuratko, Hornsby, & Bishop, 2005;Van Wyk & Adonisi, 2011).
Therefore, the following hypotheses were established to examine the effect of corporate entrepreneurship at the organizational level on employees' attitudes at the individual level: Hypothesis 1.Corporate entrepreneurship has a positive effect on employees' attitudes.Hypothesis 1-1.Corporate entrepreneurship has a positive effect on organizational commitment.Hypothesis 1-2.Corporate entrepreneurship has a positive effect on job satisfaction.

Learning Capability and Employees' Attitudes
Learning capability refers to the ability related to learning methods and attitudes.For organizations, having members with learning capabilities is the key to gaining a competitive advantage.Learning capability is a basic condition for a human being, and one can be a lifelong learner.In the Secretary's Commission on Achieving Necessary Skills (SCANS) report (SCANS, 1991), a worker's learning ability is divided into foundation skills and workplace competencies.Moreover, learning and problem-solving abilities are presented as important factors of thinking power among basic functions.The National Institute of Adult Continuing Education (NIACE), a major organization for establishing and promoting lifelong learning policies in the UK, classifies the learning capabilities required for adults into basic literacy skills, vocational skills, and special job skills.Core abilities include communication; mathematical application; information utilization; self-learning; interpersonal relationships; and problem-solving abilities.Self-directed learning and problem-solving learning abilities, which are the key learning abilities that this study focuses on, are presented as important factors in learning abilities.Ulrich et al. (1993) described learning capability in an organization as the ability of individual members within the organization to generate and generalize influential ideas.Moreover, B. Yang et al. (2004) verified that there are critical relationships among individual-level learning activities (e.g., continuous learning, exploration, collaboration, and empowerment) and organizational-level learning activities (e.g., knowledge sharing, connection with the environment, and strategic leadership in learning).
According to Pedler et al. (1991), a learning organization is defined as an organization that continuously changes to promote the learning of all members and achieve strategic goals.For the learning process to be implemented in the organization, conditions must be well established to promote the dynamic interactions of individual knowledge within the organization.B. Yang et al. (2004) state that individual-level learning should be prioritized for organizational-level learning to be implemented.Dodgson (1993) emphasizes that individual members are subjects of knowledge creation and cannot create organizational knowledge without individuals; thus, organizational learning cannot be realized without individual learning.Therefore, organizations need to provide various channels and resources for active support of individual learning activities to create valuable knowledge for the organization.At an organizational level, learning capability can be understood as the organization's ability to learn effectively.It refers to collective behavioral patterns that can process knowledge and experience, generate new knowledge based on existing knowledge and experience, and accumulate knowledge for use when necessary (Garbi, 1998).An organization builds learning capabilities through iterative and continuous learning processes (Teo et al., 2006;Yeung et al., 1999).Goh and Richards (1997) describe individual learning capabilities as factors that facilitate organizational learning processes.Hult et al. (2001) conceptualize learning capability in an organization as preparation for the creation of new organizational knowledge, which consists of four directions: team, system, learning, and memory orientation.In other words, learning capability can be found by measuring the participation of individual members in creating the vision of the organization (team orientation), understanding the relationship between the activities of individuals and the overall vision of the organization (system orientation), the organization's interest in learning (learning orientation), and the atmosphere in which individual members can access knowledge and information (memory action orientation).More importantly, learning capability relates to the context of making employees committed within an organization (Sayeed, 2001).In addition, learning capability has been found to improve employees' work satisfaction as well as improve skills and abilities (Harel & Tzafrir, 1999;Kalleberg & Moody, 1994;McEvoy, 1997;Paul & Anantharaman, 2004).Therefore, in this study, the following hypothesis was established to verify the impact of learning capability on members' attitudes at an individual level.
Hypothesis 2. Learning capability has a positive effect on employees' attitudes.
Hypothesis 2-1.Learning capability has a positive effect on organizational commitment.Hypothesis 2-2.Learning capability has a positive effect on job satisfaction.

The Moderating Role of Social Capital
We propose that social capital moderates the relationship between corporate entrepreneurship, learning capability, and employees' attitudes.Social capital, which is also at the core of organizational management, increases employees' happiness and social transaction costs due to mutually trustful relationships among organizational members.This eventually contributes positively to organizational development and performance (J.H. Lee, 2014).In an organizational environment that is wellequipped with social capital, leaders can become highly efficient and contribute to the effective achievement of organizational goals.Social capital creates a sense of mutual trust through frequent contact and communication between organizational members, thereby saving individual or interdepartmental transaction costs and time (H.T. Lee, 2017).
Organizations where social capital exists would intensify the positive impacts of corporate entrepreneurship and learning capability on employees' attitudes due to mutual trust, smooth exchange of information over networks, compliance with respectful manners, and active participation among members (B.H. Kim, 2017).In addition, high social capital has positive intervention effects on different attitudinal aspects in organizations, such as organizational commitment and job satisfaction (Y.T. Lee, 2013; H. T. Lee, 2017;Li et al., 2016).This is because social capital can be shared among organizational members and utilize knowledge and ideas through frequent communication and common goal-setting based on mutual trust and faith among members.In particular, as competition deepens, it activates and promotes innovation (Alguezaui & Filieri, 2010).
First, we argue the moderating role of social capital between corporate entrepreneurship and employees' attitudes.Looking at previous studies, social capital has a positive effect on the career remuneration of organizational members and promotes the exchange of resources, product innovation, and corporate entrepreneurship activities within the company (Coleman, 2009;Hertati, 2015).Ashourizadeh et al. (2014) verified the positive effects of social capital on SME growth and corporate performance, and Nuryanto et al. (2020) confirmed that social capital promotes competitive advantage, which in turn has a positive effect on corporate performance.Social capital is divided into social capital outside the company that forms interactions with individuals or organizations outside the company and social capital inside the company centered on interactions between organizational members and departments.This study focuses on social capital within the company because SMEs that lack resources and capabilities often use their internal knowledge and corporate resources to acquire, utilize, and create new opportunities and rely heavily on internal members' knowledge, information, and competence (Boohene et al., 2019).Empirical studies have shown that social capital positively influences business (financial and non-financial) performance among SMEs.In the study of Chen et al. (2007), social capital predicts firm outcomes with regard to the profit or loss derived from the return on investment (ROI).Other studies, such as Fornoni et al. (2012) and Franco et al. (2016), demonstrate the favorable gains from social capital on some aspects of the performance of small businesses by utilizing the social network and mutually trustful interactions among organizational members.Thus, social capital will enhance the positive relationship between corporate entrepreneurship and employees' attitudes (e.g., organizational commitment and job satisfaction), such that the relationship is stronger when social capital is high.
Second, we argue the moderating role of social capital between learning capability and employees' attitudes.In the process of promoting organizational effectiveness, companies must link and combine various distributed knowledge, which is achieved through active and trustful interactions between the holders of related knowledge (Floyd & Lane, 2000).In particular, in the process of realizing entrepreneurial objectives, companies face various problems, and employees' learning capacity within the company and active interactions based on social capital play a critical role in promoting positive employees' attitudes by applying the knowledge held by individual members and departments during the problem-solving process (Hansen, 2002).In addition, even if the search for entrepreneurial opportunities is based on the entrepreneurial orientation of senior executives or strategic groups, it is the individual members of the company that implement organizational change to realize the entrepreneurial goals.Therefore, if members do not voluntarily and actively participate in knowledge exchange and problem solving processes, it is difficult to achieve the desirable organizational performance even if the senior executives try to increase entrepreneurial orientation.Therefore, the positive relationship between learning ability and employees' attitudes can be moderated by the amount and quality of knowledge exchange and mutual cooperation occurring among organizational members (De Clercq et al., 2010).In particular, small and mediumsized ventures have limited resources and capabilities, so the interaction and participation of internal members, such as sharing knowledge and information and effective use of knowledge-based retention and continuous learning, are more important in generating organizational effectiveness (Lumpkin & Dess, 1996).Here, social capital can be understood as the sum of intangible resources that an individual or organization can access by maintaining sustainable relationships and social networks (Bourdieu & Wacquant, 1992), and is one of the most important sources for information acquisition and knowledge management (Yli-Renko et al., 2002).Thus, social capital enhances the positive relationship between learning capability and employees' attitudes (e.g., organizational commitment and job satisfaction), such that the relationship is stronger when social capital is high.
Therefore, this study set the following hypotheses to verify the moderating effect of social capital in studying the effect of corporate entrepreneurship and learning capability on employees' attitudes.
Hypothesis 3. Social capital moderates the relationship between corporate entrepreneurship and employee attitudes.Hypothesis 3-1.Social capital moderates the relationship between corporate entrepreneurship and organizational commitment.Hypothesis 3-2.Social capital moderates the relationship between corporate entrepreneurship and job satisfaction.Hypothesis 4. Social capital moderates the relationship between learning capability and employee attitudes.Hypothesis 4-1.Social capital moderates the relationship between learning capability and organizational commitment.Hypothesis 4.2.Social capital moderates the relationship between learning capability and job satisfaction.

Research Design and Sample
To conduct this study, secondary data from the Human Capital Corporation Panel (HCCP) provided by the Korea Research Institute for Vocational Education & Training (KRIVET) were used.Head data were used for corporate variables, and work data were used for worker variables.The analysis was conducted only for manufacturing industries in South Korea with less than 300 employees, non-financial industries (research and development and IT industry), and research and development workers who could have high technical knowledge leakage problems.The analysis was conducted on workers in manufacturing, non-financial (R&D and IT industries), research, and development industries working in organizations with fewer than 300 workers, who may experience high technology leakage problems.Finally, the hypotheses were verified for a total of 344 people working for 67 SMEs (51 manufacturing, 16 non-financial).

Measures
The constructs used in this study are corporate entrepreneurship, learning capability, organizational commitment, job satisfaction, and social capital.To ensure measurement validity, all the items were developed utilizing measures that had been applied in previous studies.
Corporate Entrepreneurship (Average Ratings of the Employees).The key variable in this study is corporate entrepreneurship, which occurs when individuals or groups create new businesses internally within the current organization, innovating or transforming the entire organizational practices as members of an entity under a policy of taking a certain level of risk to achieve its management goals (Chua et al., 1999).Based on prior research, survey questions measured initiative, risktaking and innovation, which are representative dimensions of corporate entrepreneurship: ''Our organization encourages change and new attempts,'' ''Our organization rewards innovation properly'' and ''Our organization favors creative over sincere people.''The corporate entrepreneurship scale had a reliability of 0.811.
Learning Capability (Individual Employee Responses).Goh and Richards (1997) defined elements that promote the organization's learning process as organizational learning capabilities and suggested goal clarity, leadership and empowerment, experiment and reward, knowledge transfer, and teamwork as components of learning capability.Learning capability was measured by two items: ''Degree of effect of learning through interaction between seniors, juniors, and colleagues on job competency,'' and ''The degree to which self-learning affects job competency through work.''Cronbach's a for the two-item learning capability was .725.
Organizational Commitment (OC, Individual Employee Responses).Allen and Meyer (1990) classified organizational commitment as the psychological attachment to the organization, such as loyalty, warmth, goodwill, and a sense of belonging among organizational members.To measure it, three questions were asked, ''I feel like the problem of this company is my problem,'' ''If I leave this company, my life will be lost too,'' and ''This company deserves my loyalty.''Cronbach's a for the three-item organizational commitment was .701.
Job Satisfaction (Individual Employee Responses).Job satisfaction refers to the satisfaction of needs in a broad sense, it refers to individual favorable attitudes toward jobs.In general, the concept of job satisfaction is explained by considering both the factors of desire and attitude (Smith et al., 1969).Job satisfaction was measured by three items: ''I am satisfied with the content of my current work,'' ''I am satisfied with my current salary,'' and ''I am satisfied with the relationship at my current workplace.''Cronbach's a for the three-item job satisfaction was .668.
Social Capital (Average Ratings of the Employee).Social interactions between members of an organization play an important role in promoting the cognitive dimension of social capital, that is, the sharing of vision and code.Through social interaction, members form and expand a consensus, thereby sharing values, attitudes, and goals (Gulati & Sytch, 2008).Social interaction allows members of an organization to form common goals and values and plays a key role in sharing these goals and values (Tsai & Ghoshal, 1998).Social capital was measured by two items: ''Training and providing advice to the field department head on personnel-related issues,'' and ''Explaining the contents of the personnel system to employees from time to time.''Cronbach's a for the twoitem social capital was .801.
Control Variables.To improve the accuracy of the research results, we used individual-level control variables such as gender, age, educational background, working period, and marital status.Also, the corporatelevel controlled variables such as industrial and corporate age.

Methodological Approach
To conduct a study on the attitude of employees, which is a key variable affecting employee's behaviors considering both organizational and individual characteristics, an attempt was made to integrate the organizational and individual levels.Using the ambidextrous theory dealing with two different structural processes required for organizational innovation, organizational-level corporate entrepreneurship was viewed as exploration (innovation, organizational change, etc.) and individual-level learning capability was viewed as utilization (execution, implementation, production, etc.).Statistical significance was verified using SPSS 23.0 and HLM 8.0 as analytical tools.First, to verify the reliability and validity of the items, a reliability analysis and a factor analysis were performed at the individual and organizational levels.Next, to ensure the validity of whether corporate entrepreneurship and social capital are sublimated at organizational levels by adding them to organizational concepts, the level of consent within the organization was measured by r wg verification.A Pearson's correlation analysis, which is commonly used to investigate the relationship between factors, was also conducted.Finally, the basic information from the multi-level analysis was obtained, and the sizes of the two error terms that constitute the total variance were grasped through the null unconditional model, which is the basis of our analysis.The results of the study were revealed through a hierarchical linear model (HLM).
The research data in this study consisted of multilayered data at individual and organizational levels.An important point in multilevel analysis is the selection of targets and the size of samples.In multilevel research between individuals and organizations, it is important to select the subjects to be surveyed in a manner that ensures homogeneity and diversity among organizations.In addition, an important factor is the size of the sample, and although scholars suggest different opinions on the size of the sample, it is generally suggested that at least 30 samples should be analyzed (Kozlowski & Klein, 2000; J. M. Nam & Chun, 2012).This study's sample was selected from the manufacturing and non-financial industries among SMEs in Korea, and the sample was extracted from the members who carry out R&D work.The statistical analysis results for each demographic variable used in this study are as follows: The individual level was male (n = 296)/female

Reliability and Feasibility Analysis
In this study, Exploratory Factor Analysis (EFA) and Cronbach's a coefficient were used to verify the feasibility and reliability of the variables.Reliability was determined based on internal consistency, and internal consistency was analyzed using Cronbach's a value.All measurement variables were analyzed using a principal component analysis to extract the components, following which a varimax rotation method was used.As a result of exploratory factor analysis of the three items of job satisfaction, the three items of organizational commitment, and the two items of learning capability, the KMO (Kaiser-Meyer-Olkin) value measured the appropriateness of the sample was close to 1 with 0.757, and the sphere formation verification statistic value was 6611.92 (df = 28, p \ .000),which is significant, so the correlation matrix was suitable for factor analysis.In addition, the eigenvalues are 1.928, 1.819, and 1.661, which were greater than 1.The validity of the factors was demonstrated by explaining 24.079%, 22.732%, and 20.765% of the total variance, respectively.All factor loadings were 0.5 or higher, ensuring the validity of the concept of composition.As a result of the individual level reliability analysis, the Cronbach's a value was .668for job satisfaction, .701for organizational commitment, and .725for learning capability.As a result of exploratory factor analysis of the three items of corporate entrepreneurship and the two items of social capital, the KMO value measuring the adequacy of the sample was 0.731, close to 1, and the sphericity verification statistic was 631.527 (df = 10, p \ .000)indicating that the correlation matrix is suitable for factor analysis because the significance level is significant.The eigenvalues were 2.150 and 1.702 respectively, which are more than 1, and the factors demonstrate validity by explaining 43.009% and 34.048% of the total variance, respectively.In addition, all factor loadings were 0.5 or higher, which ensured the validity of the concept of composition.As a result of the reliability analysis of the organizational level, Cronbach's a value was .811for corporate entrepreneurship (three questions) and .801for social capital (two questions).
The r wg verification is a method for securing validity in multilevel analysis (HLM) and measure the level of intra-organizational agreement to corporate entrepreneurship and social capital (Bliese, 2000).This agreement shows how much consent exists within each organization in terms of organizational trust and confidence (James et al., 1993).Generally, it is desirable that the r wg value is greater than 0.7.
ICC (1) and ICC (2) are the two most frequently used methods for verifying reliability in multilevel analysis (James et al., 1993), where reliability is evaluated by the degree of consistency in the response of respondents (Kozlowski & Hattrup, 1992).ICC (1) was obtained from the analysis of variance (ANOVA) table, and the specific method of analysis was an F-test.The ICC (1) derived from the ANOVA model has a value of 21 to + 1, and the value is negative (2) when the variance in the group is smaller than the variance between groups; that is, when the mean square between (MSB) is smaller than the mean square within (MSW).If the ICC (1) value is negative (2), it means that the research model is likely to have a theoretical frog-pond effect (Spurlin et al., 1984).
The results of data aggregation analysis revealed that the r wg values for corporate entrepreneurship and social capital were 0.87 and 0.84, respectively, and therefore above 0.7.This meant that the characteristics of organizations could be well represented.The value of ICC (1) for corporate entrepreneurship and social capital was 0.06 each, which meant that more than 6% of the variance between individuals of these variables was explained by organizational characteristics.The value of ICC (1) was 0.76 and the value of ICC (2) was over 0.50, which is considered to be stable when analyzing it as an organizational level variable.

Correlation Analysis
Corporate entrepreneurship, social capital, industry, and age measured through the individual level, were analyzed by aggregating them to the organizational level.The final variables used in the study of social capital are hierarchical; hence, they are divided into individual and organizational levels.This is represented in Tables 1 and 2. At the organizational level, the correlation coefficient with individual capital ( + ) was significant, and the independent learning capability at the individual level was significantly correlated with organizational commitment ( + ) and job satisfaction ( + ) in the same direction as the research hypotheses.

Null Model (Unconditional Model)
The multilevel analysis starts from a null model (unconditional model) and is a common analysis step in all multilevel analyses.The basic model does not include independent variables but merely reflects the dependent variables and levels.Through this basic model analysis, the sizes of the two error terms that constitute the total variance can be determined.In other words, since the components of variance with random effects are estimated by dividing them into one and two levels, the ICC (Intra Class Correlation) value can be analyzed through the basic model.Generally, the dispersion ratio is 5%-25%.
The analysis results of the basic model of individual variables were analyzed as shown in Table 3, and the difference between individual and organizational levels on organizational commitment was 0.335 for the individual level and 0.070 for the organizational level.Thus, the dispersion ratio that accounts for the differences between organizations was 17.28% (0.070/(0.070 + 0.335)).Hence, about 17.28% of the differences in organizational   commitment are caused by differences between organizations.The individual-level variance for job satisfaction was 0.343, and the organizational-level variance was 0.025.Therefore, the dispersion ratio was 6.79% (0.025/ (0.025 + 0.343)).Hence, 6.79% of the differences in job satisfaction are caused by differences between organizations.

Relationship Between
The result of the analysis derived, using a hierarchical linear model for the verification of hypothesis 1-1 (the effect of corporate entrepreneurship on organizational commitment), is as follows.As shown in Table 4, we found that the higher the level of corporate entrepreneurship, the more statistically significant the positive influence on employees' perceived organizational commitment (b = .462,t = 0.096, p \ .001).This supports hypothesis 1-1 that ''corporate entrepreneurship will have a positive effect on organizational commitment.''Although these findings were not hypothesized, women have a higher organizational commitment than men, and the higher the educational level, the greater is the organizational commitment.We also found that the longer the working period, the greater is the organizational commitment.Note.Moderating effect (corporate entrepreneurship 3 social capital).Education 1 dummy processing (college and university graduation = 1).Education 2 dummy processing (Master's and Ph.D. graduation = 1).Gender dummy treatment (female = 1).Marriage status (single = 1). *p\.05.***p\.001.
The result of the analysis using a hierarchical linear model for verification of hypothesis 1-2 (the effect of corporate entrepreneurship on job satisfaction) is as follows.As shown in Table 4, we found that the higher the level of corporate entrepreneurship, the more statistically significant the positive influence on employees' perceived job satisfaction (b = .481,t = 0.092, p \ .001).This supports hypothesis 1-2 that ''corporate entrepreneurship will have a positive effect on job satisfaction.'' The result of the analysis using a hierarchical linear model for verification of hypothesis 3-1 (the moderating effect of social capital on the relationship between corporate entrepreneurship and organizational commitment) is as follows.As shown in the Table 4 and the Figure 2, we found that social capital exhibits moderating effects between corporate entrepreneurship and organizational commitment (b = .420,t = 0.204, p \ .05).This supports hypothesis 3-1 that ''social capital will moderate the relationship between corporate entrepreneurship and organizational commitment.'' The result of the analysis using a hierarchical linear model for verification of hypothesis 3-2 (the moderating effect of social capital on the relationship between corporate entrepreneurship and job satisfaction) is as follows.
As shown in Table 4, we found that social capital does not show moderating effects between corporate entrepreneurship and job satisfaction (b = .250,p = ns).Hence, hypothesis 3-2, ''Social capital will moderate the relationship between corporate entrepreneurship and job satisfaction,'' was not supported.
The Relationship Between Learning Capability, Organizational Commitment, Job Satisfaction, and the Moderating Effect of Social Capital In the basic model, there was a difference between companies in the variables at the individual level.Hence, multilevel analysis was conducted to investigate the effects of learning capability on organizational commitment, job satisfaction, and the moderating effects of social capital.The organizational level variables were industry (manufacturing and non-financial), corporate age, and social capital, and the individual level variables were gender, age, educational background, marital status, working period, and learning capability.A statistical model that expresses this in a formula is shown below.
Research model: condition model Individual level (1): The result of the analysis using a hierarchical linear model for verification of hypothesis 2-1 (the effect of learning capability on organizational commitment) is as follows.As shown in Table 5, we found that the higher the level of learning capability, the more statistically significant is the positive influence on employees perceived organizational commitment (b = .148,t = 0.048, p \ .01).This supports hypothesis 2-1 that ''learning capability will have a positive effect on organizational commitment.''Although these findings were not hypothesized, women are more committed to organizational commitment than men, and the higher the academic background, the greater the organizational commitment.
The result of the analysis using a hierarchical linear model for verification of hypothesis 2-2 (the effect of learning capability on job satisfaction) is as follows.As shown in Table 5, we found that the higher the level of learning capability, the more statistically significant the positive influence on employees' perceived job satisfaction (b = .213,t = 0.047, p \ .001).This supports hypothesis 2-2 that ''learning capability will have a positive effect on job satisfaction.'' The result of the analysis using a hierarchical linear model for verification of hypothesis 4-1 (the moderating effect of social capital on the relationship between learning capability and organizational commitment) is as follows.As shown in Table 5, we found that social capital does not show moderating effects between learning capability and organizational commitment (b = 2.014, p = ns).Hence, hypothesis 4-1, ''social capital will moderate the relationship between learning capability and organizational commitment,'' was not supported.
The result of the analysis using a hierarchical linear model for verification of hypothesis 4-2 (the moderating effect of social capital on the relationship between learning capability and job satisfaction) is as follows.As shown in Table 5, we found that social capital does not show moderating effects between learning capability and job satisfaction (b = 2.009, p = ns).Hence, hypothesis 4-2, ''social capital will moderate the relationship between learning capability and job satisfaction,'' was not supported.

Discussion
This study analyzed the effect of corporate entrepreneurship and learning capability on the attitude of the members of SMEs in South Korea with fewer than 300 employees and R&D workers who may have high technical leakage problems.In addition, using the ambidextrous theory, the effects of corporate entrepreneurship and learning ability on employees' attitudes and social capital as a moderating effect were verified.We found that both corporate entrepreneurship and learning capability have positive effects on employees' attitudes.We also found that social capital moderates the relationship between corporate entrepreneurship and organizational commitment.However, there was no moderating effect of social capital on the relationship between corporate entrepreneurship and job satisfaction.In addition, we failed to find the moderating effect of social capital on the relationship between learning capability and employee attitudes (e.g., organizational commitment and job satisfaction).These findings are insightful and have both theoretical and practical implications.

Theoretical Implications
First, it was verified that corporate entrepreneurship and learning ability had positive significant effects on organizational commitment and job satisfaction.In general, research results have shown that corporate entrepreneurship improves overall company performance and brings significant economic benefits (B.Antoncic & Hisrich, 2001;Covin & Slevin, 1989;Zahra & Covin, 1995).In addition, learning capabilities have been found to improve job satisfaction and organizational commitment, as well as develop key personnel and improve their knowledge and skills (Harel & Tzafrir, 1999;Kalleberg & Moody, 1994;McEvoy, 1997;Paul & Anantharaman, 2004;Sayeed, 2001).In line with previous studies, corporate entrepreneurship and learning capability confirmed the positive impacts on organizational commitment and non-financial outcomes such as job satisfaction, and the importance of these results as antecedent factors for corporate growth was also confirmed.Second, social capital had a moderating effect only on the relationship between corporate entrepreneurship and organizational commitment, but there was no moderating effect on the relationship between corporate entrepreneurship and job satisfaction as well as the relationship between learning capability, organizational commitment, and job satisfaction.Social capital is not the capital held by an individual but the capital inherent in the relational attributes among members of a particular group.It requires constant effort to keep social capital intact (Bourdieu, 1986).As competition intensifies, it activates and promotes innovative behavior (Alguezaui & Filieri, 2010).Hence, the interaction effect with corporate entrepreneurship was shown because it serves as a catalyst to increase the loyalty among employees (i.e., organizational commitment), which in turn may promote organizational innovation.
To inspire the corporate entrepreneurship that emerges in the process of creating new corporate ventures or transforming the entire organization innovatively, individuals and teams should work with the company to create new organizations or incite innovation and entrepreneurship within the organization (J.Nam et al., 2022).Through social interaction, we must build and expand consensus among the organizational members and share values, visions, and objectives.To facilitate the organization's learning process, we should clearly present our goals and strive to create an atmosphere of positive teamwork through sufficient experimentation, rewards, and transfer of knowledge (Goh & Richards, 1997).According to entrepreneurship perspectives, corporate entrepreneurship should motivate employees to demonstrate internal performance at work (Tseng & Tseng, 2019).The present study has provided further empirical support for this argument.

Practical Implications
The management of key personnel is more difficult for small and medium-sized enterprises with poor environments and limited internal resources than for large enterprises.Therefore, it is necessary to devise measures to secure talent and impact turnover that can be practically carried out in the environment of SMEs, which are at a disadvantage compared to large enterprises.The following specific measures are proposed: First, career development opportunities should be clearly presented.If the number of job seekers is small, such as in the Japanese recruitment market, an increasing number of job seekers consider non-financial factors such as career development opportunities.Therefore, it is important to arrange regular meetings with employees and clearly present individual goals or future career development plans.Employees may consider changing jobs if their career development opportunities are uncertain.In addition, regular feedback for individual employee promotion and career development is also a good way to avoid staffing leaks.
Second, if additional staffing is required, recruitment should proceed as soon as possible.In the case of a job market with a steep economic population decline rate, it is desirable not to burden existing employees with excessive work.There is a difference between improving the skills of employees through education and training and giving excessive work to employees.Moreover, it is important to know which employees want to take on more work.Leaving a lot of work to employees who do not want additional work can worsen the talent outflow.If a talent leak occurs, companies will struggle to meet the needs of existing customers.If additional staffing is required, recruitment should proceed as soon as possible since it takes a lot of time to find the right talent.
Third, we need to focus on developing the capabilities of talented people.Focusing on developing the skills of existing employees is also an effective way to prevent talent leakage.Providing opportunities for additional training and new roles to employees who want to develop a good career is a very important recruitment strategy that can prevent staff shortages that may arise if other employees leave the company.This method can be expected to increase employee loyalty and organizational commitment.
Fourth, we need to create opportunities for communication.It is important for management to have regular conversations about opportunities with employees to understand the problems or difficulties they have.It is also a good idea to ask employees to survey other employees.Such surveys can offer ways to initiate discussion about business problems or difficulties that employees are experiencing.The management can understand the overall atmosphere in the company from the survey results.Finding and solving the causes of difficulties that employees are experiencing is also an important way to prevent employees from leaving the company.

Limitations and Future Research Directions
Despite the theoretical and practical contributions of this study, it is not without limitations.One of the limitations of this study is that the Human Capital Enterprise Panel (HCCP), which has a large number of respondents from various companies, was designed to collect extensive information related to human resources.Therefore, it lacked the sophistication to replace variables for empirical research; that is, different design intentions did not measure the different aspects of each conceptual composition used in the study.It was also unable to distinguish characteristics from industry to industry by segmenting the SMEs.Therefore, thorough future research needs to be carried out by segmenting the industries of SMEs.
This study measured employees' attitudes, such as organizational commitment and job satisfaction, rather than their actual behaviors.Although the attitudes are good predictors of employee behavior and performance outcomes, future researchers may want to investigate if our independent variables would predict employees' behavior and/or performance outcomes such as individual performance, innovative behavior, entrepreneurial behaviors, prosocial behaviors, and actual turnover, etc.In addition, this study is based on a correlational research design, which makes it difficult to establish a causal relationship between the variables.However, it could be possible that the employees' positive attitudes may change over time.Thus, future researchers may consider the time effect of positive employee attitudes and use longitudinal analysis to explore how employee perceptions of corporate entrepreneurship and learning capacity change over time as the change unfolds.
In addition, based on the multilevel analysis method of this study, we tried to solve the problem of excessively simplifying the relationship between the institutional cause variables at the organizational level and the attitudes of workers.As a positive work based psychological state, employee engagement played a mediating role between corporate entrepreneurship and business performance (Ahmed et al., 2020).Given this, future researchers may explore the possibility of multilevel mediating roles among corporate entrepreneurship, employees' attitudes (e.g., organizational commitment, job satisfaction, employee engagement etc.), and their behavior or performance outcomes (e.g., individual performance, team performance, organizational performance, organizational citizenship behavior, innovative behavior, proactive behavior, etc.).
Finally, corporate entrepreneurship was studied in more depth to clarify the processes that affect workers' attitudes.There have been few studies focused on this issue in Korea, even though SMEs are regarded as an engine of growth for developing economies.Given the lack of existing studies, this study is meaningful in revealing the multilevel effects of corporate entrepreneurship on job satisfaction and organizational commitment in the Korean SME sector.Although our proposed relationships have been made to study the effects of organizational aspects on the attitudes of individual organizational members in Korean organizations, future researchers may want to investigate if similar effects can be found in other countries' organizations, such as the United States, Asian, Latin American, and European countries.

Conclusion
It should be noted that the key takeaway from this study is that corporate entrepreneurship and learning capacity should coexist based on the ambidexterity theory.It is important for the top management within SMEs to understand the critical roles of corporate entrepreneurship and learning capacity that promote positive employee attitudes.Using a multilevel model, this study also identifies the moderating role of social capital within entrepreneurial firms and elucidates the intervening role in the relationship between corporate entrepreneurship and learning capacity and employees' attitudes.We demonstrated the theoretical and practical implications of examining the multilevel mechanism.

Figure 1 .
Figure 1.Proposed research model: This figure is drawn based on the author's conception; the relationships between the variables are drawn based on relevant literature.

Figure 2 .
Figure 2. Social capital moderating effect between corporate entrepreneurship and organizational commitment.

Table 3 .
Individual and Organizational Level Dispersion Analysis Result.

Table 4 .
Relationship Between Corporate Entrepreneurship, Organizational Commitment, Job Satisfaction, and the Moderating Effect of Social Capital Result.

Table 5 .
The Relationship Between Learning Capability, Organizational Commitment, Job Satisfaction, and the Moderating Effect of Social Capital Result.