Linking strategic improvisation and entrepreneurial self-efficacy to corporate entrepreneurship in Nigerian higher education institutions (HEIs)

Article history: Received June 5, 2016 Received in revised format October 2, 2016 Accepted October 26, 2016 Available online October 27, 2016 The role of leaders toward the development of entrepreneurship has been fully acknowledged. However, Leaders’ characteristics such as strategic improvisation and entrepreneurial selfefficacy were mainly examined in the private sector. Hence, it is imperative to extend empirical studies to public sector organizations. The present study, therefore, proposed and validated a model linking leaders’ strategic improvisation and entrepreneurial self-efficacy to corporate entrepreneurship in Nigerian Higher Education Institutions (HEIs). Using a structured questionnaire, 220 responses were generated from large HEIs in Kano State, Nigeria. The data of the study was analyzed using SmartPLS 3.0 to ascertain both measurement and structural model validity of the model. The results showed that both leaders’ strategic improvisation and entrepreneurial self-efficacy were significantly related to corporate entrepreneurship in HEIs. Implication and limitation of the study; and suggestions for future study are also provided. © 2016 Growing Science Ltd. All rights reserved.


Introduction
The pressure for change in public sector organizations has cut across almost all continents and sectors (Blackmore & Sawers, 2015;Yadav & Dabhade, 2013), including Higher Education Institutions (HEIs) (Bakar, 2014).The HEIs in Nigeria are facing similar pressure to increase efficiency, effectiveness and performance.These institutions have been accused of neglecting the changes that occur in the environment, which resulted in producing unskilled graduates that find it difficult to get jobs.While HEIs have been attributed to economic growth and development in other countries, HEIs in Nigeria are out-dated and not in line with the need of the society.These problems have been linked with the old traditional method of bureaucracy, rigidity, centralization and ambiguity in terms of power and responsibility (Nayyar & Mahmood, 2014), not to mention the rapid change that occurs in today's environment.Moreover, Dike (2014) reported that poor funding, obsolete facilities and dearth of skilled personnel have also played a role in the situation HEIs found themselves.However, all these problems associated with HEIs have been attributed to poor leadership that has the ability and skills to prioritize the limited available resources found at their disposal.Hence the need to identify relevant changes, that are capable of turning things around faster and better than their competitors or rivals (Cunha et al., 2012).
The significance of organizational changes or renewal for Nigerian HEIs, currently demonstrated a greater impact for their development.These changes provided an avenue for rejuvenating these institutions for increase performance and sustainability.One of the methods through which this renewal can be achieved is through corporate entrepreneurship (CE).Corporate entrepreneurship has been defined as those behaviours and activities display by individuals in established organizations (Holt et al., 2007).Important to realize is that several studies have established the link between corporate entrepreneurship and performance (Bakar & Mahmood, 2014;Simsek et al., 2011).However, less attention has been given to factors leading to corporate entrepreneurship especially in HEIs.Specifically, Kiani Mavi et al. (2016) stated that studies on corporate entrepreneurship often neglect interrelationship that exists between individual, environment and organizational factors.In addition, internal factors such leaders' characteristics have been neglected despite their role been acknowledged as one of the key factors that influence corporate entrepreneurship (Heavey & Simsek, 2013;Mustafa et al., 2016).The present study, therefore, investigated the relationship between leaders' strategic improvisational behaviour, entrepreneurial self-efficacy and corporate entrepreneurship.

Leaders' Strategic improvisation and corporate entrepreneurship
Strategic improvisation is one of the new concepts that enhance competitive advantage for organizations struggling to survive today's turbulent environment because it allows flexibility and adaptability (Bakar et al., 2015;Vera & Crossan, 2005).Strategic improvisation has been defined as the ability to spontaneously recombine knowledge, processes and structure in real time, resulting in creative problem solving that is grounded in the realities of the moment (McKnight & Bontis, 2002).In this study, strategic improvisation is defined as leaders' ability to respond to unforeseen circumstances intelligently and effectively in order to solve a problem or utilize an opportunity.In essence, strategic improvisation is processes were planning and action is strategically exhibited at the emergent of behaviour.
Several studies have justified the link between strategic improvisation and corporate entrepreneurship or its elements.For example, Brown and Eisenhardt (1997) have established that improvisation plays an important role in successful innovation and strategic renewal.Similarly, Vera and Crossan (2005) also revealed that improvisation is positively related to organizational innovativeness, especially if combined with some external factors.In essence, studies have linked improvisation with starting a new business (Baker, Miner, & Eesley, 2001;Hmieleski & Corbett, 2006) and new venture (Hmieleski & Corbett, 2008), which are all part and parcel of corporate entrepreneurship.Hence, the study postulates that: H1: Leaders' strategic improvisation is positively related to corporate entrepreneurship in Nigerian higher education institutions.

Entrepreneurial self-efficacy and corporate entrepreneurship
Self-efficacy is one of the most dominant factors explaining the reason people pursue specific given tasks with the zeal to succeed at all cost (Bandura, 1997).Hence, the conclusion that the main motivating factor lies not in the objective assessment but rather the subjective one, which is the inner perceived individual self-belief (Markman et al., 2002).However, ESE has been defined as the inner strength of an individual that pushes someone to successfully engage in entrepreneurial activities and tasks (Boyd & Vozikis, 1994).Leaders with high entrepreneurial self-efficacy are more likely to identify and utilize entrepreneurial opportunities as well as outcome such as corporate entrepreneurship (Chen et al., 1998;Fiet, 2002).
The relationship between ESE and individual entrepreneurial intention is well documented (Hallak et al., 2012;Naktiyok et al. 2010), likewise, the linked between ESE and new venture creation, renewal and growth (Jung et al. 2001;Marta et al., 2016).For example, Naktiyok et al., (2010) conducted a study in turkey where they established the linked between ESE and intention to start a new business.In a similar study Jung et al. (2001) also established that ESE has a significant and positive relationship with new business intention and action.Hence, the study postulates that: H2: Entrepreneurial self-efficacy is positively related to corporate entrepreneurship in Nigerian higher education institutions.

Research methodology
The study adopted a quota sampling, which is considered as a form of proportionate stratified sampling technique (Sekaran & Bougie, 2013).The use of the technique is premised on the fact that the sample are from different sub-groups with different numbers.The study in determining the population of the study that cut across ten higher institutions in Kano state, adopted Dillman (2007) method by selecting 370 out of 756 samples.The selection was based on strata in which all the ten tertiary institutions were included.The questionnaire was adapted from previous studies, SI from Vera and Crossan (2005), ESE from Wilson et al. (2007) and CE from Covin and Slevin (1989) respectively.The 100% questionnaires were distributed within a period of three months, out of which only 62% response was received.The questionnaire is divided into four sections namely: corporate entrepreneurship, strategic improvisation, entrepreneurial self-efficacy and lastly the demographic section.Questions (statements) were set on a five-point Likert scale ranging from strongly agree (5) to strongly disagree (1) aimed to determine their level of acceptance on the construct of the study.

Result and discussions
The data generated was analysed using Partial Least Square (PLS) version 3.0, which was divided into two basic sections.The measurement model and the structural model are the two steps of analysing data using structural equation modelling (SEM) as suggested by Hair et al. (2013) and Henseler et al. (2009).Consequently, a bootstrapping techinque was applied to determine the level of significance for the loadings, weights, and path coefficients as suggested by Chin (1998).Hence, the establishment and validation of the measurement model.Table 1, 2, and 3 provided values that are higher than the threshold suggested in the literature.Specifically, internal consistency values were all above 0.50 threshold (Anderson & Gerbing, 1988;Bagozzi & Yi., 1998;Gefen & Straub, 2005).Similarly, convergent validity values of both composite reliability (CR) and average variance extracted (AVE) were all above 0.70 and 0.50 as suggested by Henseler et al. (2009) and Vinzi et al. (2010).
The final validity of the measurement model for the discriminant validity was evaluated using three different methods.The most common method of validating the discriminant validity as suggested by Hair et al. (2010) and Fornell and Larcker (1981) has been cross loadings criteria.The cross loading is a process were the construct items loads better in the mother construct than in another construct of the study, while, the Fornell and Larcker (1981) criteria expect the square root of the AVE to be higher than when interrelating to the other construct.In essence, the only difference between cross-loadings and the AVE method, the later validate at the construct level while the former validate at the item level.All the values presented in Table 1 have achieved cross-loadings and AVE discriminant validity.Nevertheless, these methods was recently criticized by Henseler et al. (2015), that in most cases the above methods fails to fully scrutinize and validate the issue of discriminant validity.They, however, suggested the multitrait-multimethod matrix, to assess discriminant validity, the heterotrait-monotrait (HTMT) ratio of correlations as a more rigorous method of achieving discriminant validity.Henseler et al. (2015) stated that using HTMT.85 values most not be greater than .85 (Kline, 2011).Hence, discriminant validity may not a problem looking at the values stated in Table 3.The structural model in Figure 2 provided a justification for the postulated hypotheses for this present study.Precisely, the structural model aimed at assessing hypotheses, R-square, effect size and the predictive relevance of the model.The present study has R-square value of 21%, which states how well the exogenous constructs have explained the endogenous construct.The value of 0.208 (21%) can be categorized as moderate according to Chin (1998), by classifying R2 into three namely 19%, 33% and 67% to represent small, medium and large respectively.

The structural model
The hypotheses result of the study in Table 4 presented an evaluation of the path model of the structural model in fig. 2 above.The result provided empirical justification for all the hypotheses.Specifically, the study hypothesized positive relationship between strategic improvisation and corporate entrepreneurship in HEIs.The result was accepted with a beta value of 0.248, t-value 3.589 at significant level of 1% (p-value=0.000).In the same vein entrepreneurial self-efficacy also has a positive and significant relationship with corporate entrepreneurship in HEIs.In this case, the findings showed a beta value of 0.296, t value 4.295 at significant level of 1% (p-value= 0.000).Therefore, all the two hypotheses of the study cannot be rejected.To assess effect size (f 2 ) Hair et al. (2013) submission was adopted, they argued that result (p-value) only indicates significance of the relationship and not the effect of individual construct.Hence, creating serious concern on data and result interpretation, as such, the need to examine change in R 2 after deleting a construct to understands its effect.According to Cohen (1988) effect size are classified into 0.35 large, 0.15 medium and 0.02 small.The f 2 value of the study is 0.093 for SI and 0.066 for ESE, falling into the smaller effect category.This means, the variables are within the accepted limit.Finally, to fully meet the requirement of the structural model a blindfolding procedure was carried out to evaluate the predictive relevance (Q 2 ) of the whole model.The main reason for conducting blindfolding test was to determine how well the model and its estimate were structured in the study.Hence, a model could be reported to have a predictive relevance if the result of the blindfolding is greater than zero (Chin, 1998).The present study has a value of 0.094 which is greater than zero.Conclusively, it can be said that the model has a predictive relevance.

Discussion and conclusion
We argued that leaders' characteristics, namely SI and ESE are important factor that will help in revitalizing HEIs in Nigeria.Specifically, we argued that individual plays a significant role in opportunity recognition and utilization for the benefits of the organisation if given the opportunity, hence, the birth of CE.The CE has been regarded as the driving factor that determines the future of any organization (see., Elia et al., 2016;Guth & Ginsberg, 1990;Ireland et al., 2009), because in CE organization finds adaptability through innovativeness and strategic renewal.The HEIs are important sector of the economy, not to mention their role in determining the type of work force of an organization and the Nation at large.Studies on CE mainly focuses on external factors (Zahra, 1991), hence, ignoring the impact of middle line managers who are responsible for organizational strategy implementation.In fact contemporary researchers (see., Kakapour et al., 2016;Bakar et al., 2016) have identified the need for studies on individual factors as antecedence of corporate entrepreneurship.
Consequently, the study provided empirical support for the effect of leaders' strategic improvisation, entrepreneurial self-efficacy on corporate entrepreneurship.The study as one of the pioneering studies attempted to test the relationship between strategic improvisation, entrepreneurial self-efficacy and corporate entrepreneurship in public sector, precisely the HEIs.The results are similar to findings obtained in the private sector, in which this relationship was found to be positive and significant to either entrepreneurial intention or strategic renewal and firm performance.Thus, provides an additional literature on antecedence of CE, especially in the public domain.Based on the findings from this study, we recommended that HEIs management should double their effort in recognizing and appreciating leaders' with these characteristics, as it will not only result to CE but also improve performance of the whole institutions.In addition, programs should be design were these characteristics can be developed and maintained for the use of the institutions.However, future studies can extend this result in other public sector domain and can also look at the individual effects as antecedence of corporate entrepreneurship.In addition, a comparative analysis can also be looked at where both individual and organisational factors are identified as antecedence of CE in Nigeria HEIs.

Fig. 1 .
Fig. 1.Measurement modelThe measurement model presented in Fig.1above provided confirmatory factor analysis (CFA) of the model.It includes individual item reliability, convergent and discriminant validity of the measures.Consequently, a bootstrapping techinque was applied to determine the level of significance for the loadings, weights, and path coefficients as suggested byChin (1998).Hence, the establishment and validation of the measurement model.Table1, 2, and 3 provided values that are higher than the threshold suggested in the literature.Specifically, internal consistency values were all above 0.50 threshold(Anderson & Gerbing, 1988;Bagozzi & Yi., 1998;Gefen & Straub, 2005).Similarly, convergent validity values of both composite reliability (CR) and average variance extracted (AVE) were all above 0.70 and 0.50 as suggested byHenseler et al. (2009) andVinzi et al. (2010).

Table 1
Construct validity and reliability

Table 2
Discriminant validity