Social capital and entrepreneurial career resilience : The role of entrepreneurial career commitment

Article history: Received: September 19, 2018 Received in revised format: September 29, 2018 Accepted: October 14, 2018 Available online: October 14, 2018 Resilience is one of the constructs which received less attention in the contemporary entrepreneurial literature. This paper explored the relationships among social capital, commitment and career resilience of entrepreneurs in Northern Nigeria. 576 active entrepreneurs were selected to participate in this study using cluster sampling approach and were served with questionnaires of which 390, representing 68% response, were retrieved. The data were analyzed using Smart-PLS 3 software of Partial Lease Square Structural Equation Modelling (PLS-SEM). The findings suggest that bonding social capital influenced both career commitment as well as career resilience, but bridging did not. In addition, career commitment mediated the relationship between bonding social capital and career resilience, but did not mediate the relationship between bridging social capital and career resilience. Theoretical, practical and methodological implications were also provided. © 2019 by the authors; licensee Growing Science, Canada


Introduction
Entrepreneurship plays a significant role in the world economy and contributes sizably to job creation, productivity, and income.In many countries, entrepreneurial activities are the main driving forces behind poverty reduction, wealth and job creation, export earnings, income redistribution and reduction in income inequality.In Sub-Saharan African economy, Nigeria, has the highest Total Early-Stage Entrepreneurial Activity (TEA) rates, with 39% of the adult population involved in early-stage entrepreneurial activity either directly or indirectly (Amorós & Bosma, 2014) Yet, in the Northern part of the country, entrepreneurs have always been faced with different degrees of shocks and disruptions as a result of the environmental uncertainties (Imran et al., 2017b;Hasan, 2016).Religious crisis such as Boko Haram and other forms of social crises have been upsetting the region since independence.Consequently, majority of the entrepreneurs are constantly winding up their businesses.It has been estimated that 70-80% of these entrepreneurs quit within 3-5 years of their startup.However, the surviving entrepreneurs have devised a lot of strategies to cope with, recuperate from and prevent the future occurrences as well as the consequences.Central to these but underexplored, is the role played by social capital (Aldrich & Meyer, 2015;Bernier & Meinzen-Dick, 2014) and commitment (Yang & Danes, 2015) of these entrepreneurs in building resilience.
Social capital refers to "features of social organization, such as trust, norms, and networks, that can improve the efficiency of society by facilitating coordinated actions."(Putnam, 1993, p.167).Although, community as well as individuals' connections serve as good avenues for accessing different resources such as information, financial as well as emotional and psychological encouragements in critical and unforeseen situations both prior, during and after disaster or crisis; and in spite of series of disaster, crisis, political insecurity, and growing loss of lives and properties, traditional researchers and policy makers concentrated heavily on physical infrastructures mainly from government and other exogenous agencies in response to such events neglecting the very important role played by social capital in driving resilience (Aldrich & Meyer, 2015;Weichselgartner & Kelman, 2015), specifically, "Despite the evidence about its efficacy, resilience research and disaster management practice have yet to fully embrace social capital as a critical component" (Aldrich & Meyer, 2015, p. 256).In addition, another neglected issue in building resilience is the commitment defined as a "volitional psychological bond reflecting dedication to and responsibility for a particular target" (Klein et al., 2012, p137).Entrepreneurs with high level of commitment have all what it takes to continue no matter the hardship and disruptions.This paper therefore, argues that entrepreneurs who suffer trauma are more likely to be resilient when they possess large chunk of social resources and are committed to their career.
Although it is crucial for entrepreneurial outcomes, resilience has received negligible attention in the entrepreneurship literature (Bullough & Renko, 2013;Hussels et al., 2014;Kordnaeij & Ebrahimi, 2015) and in developing economies (Abiola & Udofia, 2011;Imran, et al., 2018a;Imran et al., 2018b), it basically remains equivocal what entrepreneurial resilience essentially entails and which possible elements such as situational factors (e.g.social capital) and specific behaviours (e.g.commitment) aid to develop and increase this individual capacity (Duchek, 2018).In this paper, we aim at responding to this research gap.We aim to construct a thoughtful understanding of the complex nature of this construct and important influencing elements (social capital and commitment).
The paper is organized as follows.First, we focus on the review of related literature of entrepreneurial resilience, social capital and commitment as a preamble to consider how they support or facilitate the emergence of the former.This concludes with a proposed schema (Fig. 1) showing some links between the two facets of social capital (bonding and bridging) and the resilience as well as the role of commitment.Second, we explain our methodological approach.Third, we present the analysis, result and discussions and finally we present our findings and discussion before finally outlining the conclusions and suggestions for future study.

Literature Review
Successful entrepreneurs set their targets to meet despite the flexible and ambiguous nature of environment.Entrepreneurs who exhibit high level of tolerance for risk, ambiguity and quick adaption and adjustment to change, may be ready to perform superbly (Ayala & Manzano, 2014).Additionally, in this contemporary interconnected universe both environmentally, technologically and socially, no entrepreneur is self-sustained.Thus, none can subsist disruption and retain their advantage (Bhamra et al., 2011).Entrepreneurs who possess resilience attributes are willing and ever ready to work tirelessly to meet their targets.They quickly respond to environmental changes in order to explore new opportunities that accrue and are ready to learn and improve from their previous mistakes (Cooper et al., 2004(Cooper et al., , 2013(Cooper et al., , 2014)).Defined as individuals' ability to adapt to, and recover from disturbing events (Cheshire et al., 2015), recently, resilience has garnered attention in the literature and policy making as a result of several catastrophes and disasters occurrences.In an attempt to mitigate the consequences of disasters and traumas, many empirical studies have explored different possible resilience attributes in this paper, we examine the role of social capital and commitment in building resilience of entrepreneurs in northwestern Nigeria.
It has been argued that, the more individual possesses social capital the greater the chance of achieving higher desired outcome.Therefore, the extent review of literature revealed that the construct acknowledged great attention from both scholars and policy makers, cutting across majority of academic disciplines.For its wider acceptability and applicability, Van Deth, (2003) argued it "has become a minor industry in the social sciences" (p.79).The concept is rooted in sociology and was first studied by Durkheim in 1897 when examining how social influence relates to suicide (Durkheim, 1951), but Hanifan was the first to introduce the term social capital in academic arena (Robison & Ritchie, 2016).Woolcock, (1998) highlighted a number of other concepts employed by scholars as synonymous to social capital including ties/network, moral resources, intangible assets, sociability, social capability, social energy and etc.
Based on Granovetter (1985) strong and weak ties argument, Putnam (2000) distinguished social capital into two components, i.e. bonding and bridging.So many scholars adopted this categorization and used it to predict different outcomes (Archuleta, 2015;Chen et al., 2015) Putnam, (2000, p. 23) argued that these dimensions "are not 'either-or' categories into which social networks can be neatly divided, but 'more or less' dimensions along which we can compare different forms of social capital".On one hand, bonding social capital is the horizontal norms of reciprocity, trust and social relations that is based on informality, similarity and closeness, developing relations and strong ties in the group, normally among intimate friends, intimate business partners and families or within community.It connotes ties among individuals with similar ethnicity, class and other sources of social characteristics which typically involves rather strong and regular interaction, built based on mutual attraction and common interest.It also includes the resources and emotional support available within the type of interaction.This type of social capital can be found among people with intense feeling of sympathy.It is characterized by strong feeling of love, caring and concern (Chen et al., 2009;Robison & Ritchie, 2016).It is often synonymously referred to as intra-community social capital (Sørensen, 2016), strong ties (Kivijarvi, 2015), homogeneous network (Reeskens & van Oorschot, 2013), close knit solidary ties (Anthias, 2007) localized social capital (Rabiul, 2015).According to Putnam, (2000, p. 22), bonding social capital is an "inward looking [networks that] tend to reinforce exclusive identities and homogeneous groups" Bridging social capital on the other hand, is horizontal trust, norms of reciprocity and social relations that occur in irregular relationships.It can be thought as a bridge that link different plot of land or countries that are dissimilar in terms of their populations, size and resources (Robison & Ritchie, 2016).It emphasized on the connections with different groups of individuals who do not have similar social identity but usually keep weak ties that contain better prospects (Kawachi et al., 2008).It is tie that connect individuals from different social setting, groups and organizations outside their immediate community based on mutuality and respect, usually through organizations and social group which offer access to potential external resources (Chen et al., 2009;Hampton, 2011;Linnenluecke, 2017).Thus, network connections that are shallow but wide and heterogeneous such as among different ethnic groups (Jun & Ha, 2015).According to Putnam (2000, p. 22), bridging social capital is an "Outward looking and encompass people across diverse social cleavages".It is often synonymously referred to as extra-community social capital (Sørensen, 2016), weak ties (Kivijarvi, 2015), heterogeneous network (Reeskens & van Oorschot, 2013) loose knit solidary ties (Anthias, 2007).
Additionally, literature in organizational behavior acknowledged an increased attention on the concept of commitment.The construct is very broad with studies from various fields, with different objectives devoted to explore its predictors as well as its outcomes (Cooper et al., 2014;Meyer et al., 2013;Raineri & Paillé, 2016;Tang, 2008).In fact, it has been one of the most researched areas in the field of organizational psychology for many decades.Commitment is apparently conspicuous in the work environment and have been found to have outcomes relevant to individuals and organizations.Commitment is considered as primary justification why some individuals want to continue being employed and others do not.
Several theories emerged on the reason behind individuals' commitment, underscoring commitment as a moral bond, affective attachment and/or calculative cost-benefit analysis (Solinger et al., 2015).Due to inherent shortcomings of the prior conceptualizations of commitment, and couple with the fact that commitment scholars have combined different forms of psychological bond (e.g.instrumental, acquiescence) under the umbrella of commitment and have distinguished among targets, as well as the foci to which individual can commit (organization, goals, supervisor), they consider it necessary to recast the concept as a specific type of bond, one that has similar meaning and works correspondingly across all targets (Klein et al. 2014;Klein et al., 2012) and has general applicability across different environment and more importantly, to remain relevant and unique concept for managing and understanding behavior in organization.Klein et al. (2012, p. 137) therefore define commitment as a "volitional psychological bond reflecting dedication to and responsibility for a particular target"
The principal element in resilience is having supportive and caring relationships within and outside the family (bonding and bridging).Such relationships that promote love and trust and offer reassurance and encouragement help boost an individual's resilience.Social capital is one of the constructs that received attentions in the resilience research as it is found to contribute greatly to individuals and community resilience (Goulden et al., 2013).The wider prominence of the concept and its relation to resilience might not be unconnected to the global need to react to natural disaster, diseases, crisis and continuous insecurity.Consequently, more researches are emanating delineating the salience role of social capital.Majority of the studies documented a significant positive relationship (e.g.Aldrich & Meyer, 2015).In fact, according to social capital theory, network resources are as critical as tangible resources in promoting resilience (Aldrich, 2017).Aldrich (2012) provided a good background for the study of social capital-resilience nexus.The work highlighted the prominent role of social capital as a crucial element that aid individuals to prepare for, adjust and recover from unforeseen events.It confirmed that composition of bonding, bridging and linking are recovery driver.It also delineates the way social resources functions after disasters.Survived victims with robust social network have upper hand than those that do not because they can easily acquire needed resources that would aid in fostering recovery.More importantly it argued, since disasters, war and other social crises are, and will continue to be, a challenge for developing countries, it is imperative that individuals build social capital in advance.When preparing for disaster, focusing only on physical infrastructural solutions, is not sufficient to avoid the negative impact.
The critical conceptualization of social capital into bonding (within) and bridging (between) explored the critical importance it has in building resilience.For instance, strong ties and connections allows individual to receive signals and make adequate preparations on impending problems (Hawkins & Maurer, 2010).Hence, bonding is the most readily available resources an individual can acquire in building resilience (Aldrich & Meyer, 2015).Bridging assists in providing critical resources and information that aid in quick and sustained recovery, it creates employment and other daily-life opportunities more than the strong ties (Hawkins & Maurer, 2010).Further, Goulden et al. (2013) argued that, in an attempt to hedge against or adequately prepare for uncertainties, individuals and communities rely on bonding and/or bridging social capital.They, discovered that bridging is unique form of social capital than bonding in building resilience.
Bernier and Meinzen-Dick (2014) investigated bonding, bridging and linking effect in building and maintaining resilience, the study found them to be very crucial predictors of resilience.Chiesi (2014) used name generator to study immigrants' social capital and its influences on the resilience of their business during crisis in Italy.Their study discovered that immigrant's social capital influences their business resilience more than cultural capital or ethnic belonging and that bridging social capital is more important to them than bonding.But Beekman et al. (2009) argued that social capital is relative in predicting resilience, it could either add to resilience or lessen it depending on the components of the social capital and the environment type (rural or urban).Torres and Marshall (2015) used small business disaster recovery framework to study the influence of social relations on entrepreneurs.The study interviewed 450 entrepreneurs in Mississippi who were victims of Katrina.Their finding indicated that, after economic shock, entrepreneurs with higher constellation of social capital reported higher level of resilience.But, they argued that bonding social capital is more important than bridging in building resilience.Entrepreneurs who are well connected with community and other institutions are presumably well equipped to immediately prevent and or react to any business disaster and build resilience.Finally, in order to identify and explain the antecedents of resilience during economic crises of 1990-1993and 2007-2009, Pal et al. (2014) ) use two phases (survey and interviews) exploratory study on eight entrepreneurs in textile sector in Sweden.One of these antecedents identified was social capital.They found that resources accessed through network to be an important enabler of resilience.
It has been observed that, generally, entrepreneurs are constrained by lack of resources and consequently unable to absorb shock or bounce back after major setback (Van Gils, 2005).Hence, having and exploring their contacts, entrepreneurs will have access to various resources to facilitate their resilience ability.We therefore propose that:

H1 -There is a significance positive relationship between bonding social capital of entrepreneurs and their career resilience.
H2 -There is a significance positive relationship between bridging social capital of entrepreneurs and their and career resilience.

Relationships of Social Capital and Career Commitment
As a physical, intellectual and emotional energy that an individual exerts in a venture, commitment has enjoyed enormous interest from both academic, practitioners and policy makers both as an antecedent as well as outcome of a relationships at different levels of analysis (individuals and organizational).Yet, previous studies focus more on the organizational level of analysis, very few concentrate on the individual level (Nangoli et al., 2013).Undoubtedly, commitment as an attitudinal outcome has been found to be affected by different variables such as social capital.
Critical literature searches and review have indicated that social capital has been predominantly one of the major factors which influence individual commitment (Aküzüm & Tan, 2014;Brien et al., 2015).For instance, Brien et al. (2015) performed a survey on the relationship between social capital (communication, influence and trust) and commitment in hotel industry in New Zealand and found that communication and influence has strong relationship with commitment while trust weakly related to commitment.Likewise, Esmeili et al. (2014) found relational and cognitive social capital as significant predictors of commitment.Also, Bozionelos (2008) studied the intra-organizational network resources both instrumental and expressive and their relationship with subjective/objective career success and organizational commitment.The findings suggest that the constructs are highly related.Furthermore Nangoli et al., (2013) documented that elements of social network (network degree and network transitivity) have significantly predicted commitment with the latter effects strongly than the former.Based on these studies we also proposed that: H3 -There is a significance positive relationship between bonding social capital and career commitment.H4 -There is a significance positive relationship between bridging social capital and career commitment.

Relationship between Career Commitment and Career Resilience
Scholars have argued that commitment is a vital element of the resilience of entrepreneurs.Cooper et al. (2013) and Yang and Danes, (2015) maintained that, people with greater level of commitment are the most resilient.As such, commitment is a vital element of the resilience of entrepreneurs.Yet, majority of the studies that related these constructs used resilience to predict commitment (Lee & Cha, 2015).Hence, this study discovers a paucity of research relating these constructs contrariwise.In other words, studies on the relationship between commitment and resilience are very few (Jokštaitė & Pociūtė, 2014;McCormick, 2000;Negru-Subtirica et al., 2015;Salisu et al., 2017;Tiet et al., 2010;Yang & Danes, 2015).For instance, Negru-Subtirica et al. ( 2015) conducted a longitudinal study on the relationship between vocational commitment and career adaptability.They found significant relationship between these two constructs.Jaaron and Backhouse (2014) conducted two case studies in two British service organizations to examine the link of service thinking (affective commitment and structured organization) and organizational resilience.Their results show that it is only through committed people who possess better resources that organizations can succeed in recovering from difficulties.The study also identified that affectively committed people typically trigger their latent accommodative competencies to face stressors, and that they are more capable of thinking "outside of the box" to generate creative thoughts which will aid faster recover.Therefore, we propose that: H5-There is a significant positive relationship between commitment of entrepreneurs and their career resilience.
In another scenario, commitment is found to be an intervening variable between some antecedents and we suspected that it could as well mediate the relationships in this paper.We therefore pose these propositions:

Methodology
This paper gathered data from entrepreneurs in north-western Nigeria through self-administered questionnaire.576 respondents were selected using cluster sampling technique in three states of the region -Kaduna, Kano and Katsina, The paper adopts cross-sectional approach where the data is collected at once (Sekaran, 2013.)Before distribution, questionnaires were pretested to assess the appropriateness of the wordings and layouts, and the extent to which measures embodied all the aspects of the constructs.The questionnaires were sent to five experts, (two professors, two Ph.D. holders and one industrialist).Based on the pre-test result, necessary modifications were later integrated into the final questionnaire.Out of 576 questionnaires distributed, 390 or 68% were retrieved and used.Most of the participants were males (90%) and married (72%).Averagely, the participants were young, i.e. 36 years old, had 10 years of experience in current industry and 10 years in start-up, and majority (72%) possesses at least B.Sc./HND.

Social Capital
 Bonding  Bridging

Measures
Although there are various scales used to measure the constructs of this study in the previous researches, this paper selected the best, based on their psychometric properties and their applicability and/or currency.In addition, all items were rated in 5-point Likert scale.First, ECR was measured by unidimensional 10 items of Connor and Davidson (2003) Resilience Scale (CD-RISC) validated by Campbell-Sills & Stein (2007) from the original 25 items, three factor model CD-RISC developed by (Connor & Davidson, 2003).Sample of the item include, "I tend to bounce back after illness or hardship".Second, ECC was measured by Klein et al. (2012), Unidimensional Target-free (KUT) developed by (Klein et al., 2014) based on Klein et al., (2012) conceptualization.To suit the context, the scale was modified and reworded.Example of the questions are "how committed are you to your career".Third, Social capital was measured using 58 items (30 measuring bonding and 28 measuring bridging) personal social capital scale (PSCS) developed by Chen et al. (2009).Sample of the item includes for bonding "among your family members how many can you trust?" and for bridging "among governmental groups/organizations, how many will help you upon your request?"

Result and Analysis
In this study, PLS-SEM (Smart-PLS) is used for data analysis.It is a variance based structural equation modelling method that can assess both inner and outer models and it is the most widely used tools in social and behavioural science researches in examining quantitative data.It exhibits more statistical power then most of the statistical tools available.Finally, it handles complex models such as assessment of second or high order constructs (Hair et al., 2014;Henseler et al., 2009).

Descriptive Analysis of the Latent Constructs
The latent variables' descriptive statistics in this study was computed using their means and standard deviation.From Table 1, commitment has the highest mean of 4.1313 and bridging social capital has the least mean of 2.4332.

Common Method Variance (CMV)
To check the possibility of CMV, Harman's One Factor Test (Podsakoff et al., 2012) was conducted.The eigenvalues show no single factor accounted for > 50% of the variance and the first factor accounted for only 17.7% of the total variances indicating that CMV may not be a worrying case in this study.

Measurement (outer) Model
We assess outer reflective items by determining individual item and internal consistency reliability, then convergent and discriminant validity to ensure they possess the required reliability and validity.First we determine the internal consistency reliability.As seen in Table 2, composite reliability (CR) ranged between 0.853 and 0.926, rho_A ranged between 733 and 902 and Cronbach's alpha (CA) ranged from 803-901.These thresholds exceed the minimum standard level of 0.70, hence internal consistency reliability is achieved.Second, the convergent validity was assessed by examining the loadings, average variance extracted (AVE) as well as CR.As seen in Table 2 and Fig. 2, the loadings were all beyond the standard value of 0.6 (Hair et al., 2012).The values of CR, are also greater than 0.7.The AVE values exceed the threshold of 0.5 (Hair et al., 2014).
We assess the discriminant validity using two criteria, AVE-SV (Fornell & Larcker, 1981) and heterotrait-monotrait (HTMT) matrix.Using the AVE-SV technique in Table 3, the constructs passed discriminant validity test as the diagonal values were greater than the horizontal and vertical values.We assess (HTMT) based on the two thresholds of 0.85 (Kline, 2016) or 0.90 (Gold et al., 2001).The value less than this threshold indicates discriminant validity.Also, we also assess HTMTinference, where a confidence interval (CI) having the value of <1 shows discriminant validity.In this paper, as shown in Table 5, when both thresholds are considered, none of the values in the matrix is greater or equal to any of the two benchmarks.Also, the highest upper limit confidence interval value is <1 and therefore, discriminant validity is achieved.

Assessment of the Formative Hierarchical Component Model
Outer weights in formative measurement models was analysed for their relevance by first checking if the collinearity is not at acute level (Hair et al., 2016).Consequently, collinearity issues for the second-order (bonding and bridging) are determined.As presented in the Table 6, two arrays of constructs were checked for collinearity in the inner model, i.e.PNS, FCM, PTM, MWH and MPR as predictors of bonding social capital and PGA, PSG, RRI, RHG and RPG as predictors bridging social capital.The variance inflation factor (VIF) values for each of the constructs are less than the cut-off value of <5 (Hair et al., 2016), indicating that these constructs are distinct and are measuring different aspects of social capital.The suitability of the higher-order constructs was evaluated on the basis of their conceptual properties.Since this study used formative higher-order components, the internal consistency reliability and convergent and discriminant validity assessment are not necessary because the items for formative constructs need not to be highly correlated (Henseler et al., 2009).Following Becker et al. (2012), the evaluation of the second-order constructs was performed from the link between second-order and first-order constructs.Table 7 shows the formative second order construct assessment.The bootstrapping results using sub-samples of 5000 cases indicate the weights and path coefficients for each of the formative second order constructs.The results show that PNS, FCM, PTM, MWH and MPR & PGA, PSG, RRI, RHG and RPG are significantly related to bonding and bridging respectively.

Testing of Direct Effect
As depicted in Fig. 3 and Table 8, we assess the direct effects of bonding and bridging on the ECR and ECC.We have found significant positive relationship between bonding and ECR (β = 0.206,t=4.048,p=0.000 and CI= 0.121,0.293),between bonding and ECC (β = 0.134, t=2.029, p=0.021 and CI= 0.024, 0.244), as such, H1 and H3 were supported.But relationship between bridging and ECR & ECC {(β = -0.053,t=1.106, p=0.135 and CI= -0.135, 0.023) (β = 0.042, t=0.652, p=0.257 and CI= -0.064, 0.148) respectively} were not significant, and H2 and H4 were not supported.We also found significant positive relationship between ECC on ECR (β = 0.450, t=10.171,p=0.000 and CI= 0.377, 0.524) therefore, H5 was supported.Overall, this study tested five direct relationships and found three significant while two were not significant.This paper used bootstrapping procedure in Smart-PLS (Hair et al., 2016) to test for mediation.A mediation effect is considered statistically significant when p-value is less than 0.05 or if confidence intervals do not contain zero value.The results in Table 9 show the mediating ability of ECC in the relationship between bonding and ECR but not between bridging and ECR and therefore hypotheses H6 was accepted, but and H7 was not accepted.

Assessment of Structural Model
To assess the outer model, we have examined the explanatory and predictive power.The former is determined using two criteria, coefficient of determination (R 2 ) and effect size (f 2 ).The latter is also determined using two criteria, path coefficient (β) and predictive relevance (Q 2 ).From the results, the bonding and bridging explained 22.2% of total variance in ECC.While bonding, bridging and ECC explained 36.4% of total variance in ECR.For the effect size (f 2 ), as seen in Table 10, the ECC contributes 0.466 to R 2 when predicting ECR.But, when prediction ECC, bonding contributes more to R 2 (0.313) than bridging (0.101).Finally, Q 2 was assessed using blindfolding procedure in PLS-SEM.Q 2 values >0 means that the predictor variable possesses the predictive relevance for the criterion variable (Hair et al., 2011).The Q 2 values of endogenous variables signify that the research model has good predictive relevance (see Table 11).

Discussions
Although, there are many studies that relate the constructs of this study in various ways, there are still gaps that remain unexplored.This paper explored the link of these constructs in the context of entrepreneurship in the northern region of Nigeria.As presented above, results provide support for the prepositions that bonding is positively related to both ECR and ECC, but bridging has no impact on both the ECR and ECC.The paper also found that ECC indirectly play a role in the relationships of bonding and ECR, but bridging-ECR nexus was not mediated by ECC.The findings corroborate and adds to previous research in a number of ways.To the best of our knowledge, we found no prior research that combined, measured and empirically tested the relationships among these variables in the entrepreneurial context and in a developing country.As such, this is the significant effort to pool these variables into a single framework and assess their connectedness.First, the paper has found bonding social capital to be a significant predictor of ECR as well as ECC.This finding suggests that entrepreneurs who amass relationship with their close contact can benefit more by getting the required resources and consequently become committed and resilient.In the case of resilience, so many studies (e.g.Bernier & Meinzen-Dick, 2014;Goulden et al., 2013) have argued in favour of the finding of this current study and according to (Aldrich & Meyer, 2015) bonding social capital is the most readily available resources an individual can acquire in building resilience.Further, in the case of commitment, the findings of this paper corroborate with the previous findings (e.g.Aküzüm & Tan, 2014;Brien et al., 2015).Bridging social capital on the other hand predicts neither ECR nor ECC and this is quiet at odd with the literature.One of the possible reasons could be the respondents' nature of smallness and/or newness.Most of the entrepreneurs in the region are new and small business owners with very small investment.Second, the result provided clear evidence that resilience often depends on how highly committed these entrepreneurs are.The more they are committed, the more they absorb shocks and bounce back.This paper is among the first to explore the relationship between ECC and ECR.Scholars found positive relationship between resilience and commitment (Lee & Cha, 2015) but, this study found positive relationship between ECC and ECR (in reverse way).Some of the very few researchers that documented the relationship between these variable (e.g.Negru-Subtirica et al., 2015;Salisu et al., 2017;Tiet et al., 2010;Yang & Danes, 2015) and therefore this paper corroborates with their findings.Finally, the main theoretical implication of this paper is to show how commitment of these entrepreneurs could mediate the relationships of social capital and career resilience.The finding suggested that the relationship between bonding and ECR was mediated by ECC, but it did not mediate the relationship between bridging and ECR.This paper therefore argued that although bonding predicted ECR, it can also influence the ECR through ECC.In other words, because these entrepreneurs have immediate contacts which guide them to important resources, that is why they are committed and consequently led them to be resilient.

Limitations and Future Directions
This paper has a number of limitations which other studies can tackle.The nature of this study is crosssectional and hence no causal conclusions can be drawn from the results.Future research should adopt a longitudinal research design.In this paper, no demographic variable was used and hence, we were could not identify their effects in the model.Therefore, future study may consider incorporating them to predict ECS.We also discover that ECC explained ECR and although there were few studies that relate these constructs clockwise, there is a need for more studies on how they are related.

Conclusion
This paper shed light into career resilience in the entrepreneurial context and the significance of social capital and ECC.In general, the paper presented seven relationships, five direct and two mediations.Three direct hypotheses were supported, but two were not supported.In addition, one mediation hypothesis was supported and one was not.However, this paper can only deliver limited understanding outside the context in which the data were collected.It is one of rare pieces of studies on predictors of resilience in the context of entrepreneurship.But the context of our sample means that more studies of this nature could be definitely required in other countries.These may shape entrepreneurial career development through different nationalities.For its significance, and to broaden the findings presented in this paper, career resilience of entrepreneurs requires more investigation most especially in developing countries.

Table 1
Descriptive Statistics for all Research Constructs of the Study

Table 2
Loadings, CR and AVE

Table 6
Collinearity Assessment

Table 7
Path Coefficient/Weight Assessment

Table 8
Structural Model Assessment (Direct Effect)

Table 9
Result of Mediation Effect (Hypotheses testing)