ALIGNING ENTREPRENEURIAL ORIENTATION AND OPEN INNOVATION FOR BETTER EXPLANATION OF INDUSTRIAL TUNISIAN SMES PERFORMANCE

The main objective of this research is to examine the effect of the alignment as gestalt of Open Innovation and Entrepreneurial Orientation on the Industrial Tunisian SMEs performance. To achieve this aim, we carry out a quantitative survey with a questionnaire distributed to 110 Industrial Tunisian SMEs and mobilize the perspective of alignment as gestalt of the configurational approach. Our results reveal an empirical taxonomy which distinguishes four Tunisian SMEs configurations named, The Conservative Enclosed, The Committed Exceptional, The Vigilant Receptive and the Innovative Involved. Our findings show that The Committed Exceptional and Innovative Involved configurations register the same high level of performance. However, the Conservative Enclosed is the least performed unlike the Vigilant Receptive configuration which is characterized by an average level of performance. Overall, theoretical and empirical contributions are heralded to advance the relevance of Open Innovation approach to the entrepreneurship theory.


Introduction
The unpredictable changes in the business environment, the economic crisis and the increased influence of digital culture are the starting point for the rapid expansion of the movement of search for best managerial and strategic practices that guarantee high levels of performance. Examples include Open Innovation (OI) and Entrepreneurial Orientation (EO).Thus, trapped in an unstable economic climate characterised by shortened innovation cycles and unpredictable global competition, small and medium enterprises (SMEs) find difficulties to innovate alone because of their small size and resources scarcity.This is why, there is a call to stress the importance adoption of Open Innovation approach (Bigliardi et al., 2020;Chesbrough, 2003) particularly in SMEs (Gassman et al., 2010). The open innovation model should be seen as a strategic imperative to overcome obstacles related to innovation processes. It exist different research on the influence of OI on performance (Federico and Biancardi, 2020). Despite the relevance of this relationship, several framework and empirical research reveal some controversial results (Moretti and Biancardi, 2018). Sometimes this relationship is positive (Nitzsche, 2016) sometimes is negative and nonlinear.To overcome this problem, research argue for the integration of contextual factors which can be studied with OI.

ISSN: 2320-5407
Int. J. Adv. Res. 9 (11), 245-255 246 Based on literature review and according to Kherrazi and Said (2020), the dynamic of innovation promotes the coherence of strategic and management modes. In addition, it appears that since the work of Schumpeter (1934), researchers have explored the relationship between innovation and entrepreneurship (Al Quadah, 2018) as well as innovation and entrepreneurial orientation (Majdouline et al., 2020). However as pointed out by West et al., (2014) the relationship between OI and the disciplines of management and entrepreneurshipsuch as EO is not sufficiently studied. In particular, few studies have examined the combined interaction between EO and OI. So, this insufficiency prompts us to study this relationship in the Tunisian context with the aim of a better explanation of performance levels. In this sense, Covin and Slevin (1989) argued that EO is an essential strategic orientation that encourages companies to innovate by partnering with external collaborators. In the same vein, Yun (2015)  Several studies have been carried out to identify the drivers of OI or to explain its effects on performance. But, those research deal with linear and direct relationships between individual factors. The results are therefore incomplete and don't reflect the total reality of a complex and multidimensional phenomenon like performance. Nevertheless, we will not study the only impact of OI but a simultaneous impact of OI and EO on performance. This choice is in relation with research which call for the evaluation of performance through the combined influence of EO and OI Mobilizing the theoretical framework of strategic management, organisational entrepreneurship (Lumpkin and Dess, 1996;Covin and Slevin, 1989), OI literature and alignment as "gestalt" (Venkatraman, 1989), this research aimed to give better explanation of how the combined interaction of OI and EO affect performance. Hence, the necessary contribution of a holistic approach such as configurational approach (Heredia et al., 2019;Majdouline et al., 2020;Wiklund and Shepherd, 2005) which suggests that explanation of performance is achieved through configurations combining EO and OI (Yun et al., 2016). In addition to these theoretical motivations, our choice for this subject is supported by considerations that relate to the Tunisian context. We cite the considerable effort devoted in recent years by the Tunisian Government to encourage innovation (Ben Miled-M'rabet, 2012) and to put in place the mechanisms necessary to establish a culture of innovation in Tunisian companies. The empirical exploration of this article is therefore based on the study of industrial Tunisian SMEs with the objective of providing a better explanation of their performance by the constitution of an empirical taxonomy identified by a combined interaction of EO and OI.
In light of alignment as gestalt, we provide some insights on the impact of alignment of OI and EO on performance. We therefore ask the following research question: To what extent does the alignment as 'gestalt' of OI and EO has an impact on performance? In response to this question, we make four parts. The first one is devoted to the conceptual framework. The relationship between OI, EO and performance involving the configurational approach is presented in the second part. Then, we present the methodological choices and the questionnaire survey in the third part. Finally, the fourth part is devoted to the discussion of results, conclusion and further research.

Conceptual Framework Open Innovation (OI)
In an environment marked by economic change and the advent of digital, innovation has become source of competitive advantage.According to Lopes et al., (2017), companies and especially SMEs have been engaged in innovation by investing in internal R&D and protecting their intellectual property. They therefore adopted a 'closed innovation' model that required strict control of internal knowledge to prevent their flight out of the company. Nevertheless, to guarantee the continuity of innovation, some SMEs can no longer rely solely on their internal resources or R&D. Indeed, as admitted by Ibarra et al., (2020), SMEs are constrained to continually generate financial resources. In terms of literature, Chesbrough (2003) proposes the adoption of OI as an appropriate solution for these challenges.
According to Martinez et al., (2014) 'open innovation supports the creation of competitive advantage that can generated profit through investment of innovative design that performance in terms of survival can be maintained'.

247
Thus, Chesbrough (2003) defines open innovation as 'the respective use of intentional knowledge inputs and outputs to accelerate internal innovation and expand markets for the external use of innovation'. According to Jinhyo et al., (2020) open innovation encompasses the opportunity to leverage external sources of knowledge to thrive the internal innovation and the use of external channels to reach the internal knowledge market. In the literature, we find the 'Inside-out', 'Outside-In' and 'coupled-process' models. We interest in our research to inbound and outbound OI.

Inbound Innovation
It consists in renewing the company's stock of knowledge by integrating the knowledge from external agents (customers, suppliers, universities, associations, business network). According to Chesbrough and Crowther (2006), companies generally follow this path to maximize their profitability, to make the R&D function internally flexible to improve innovation capacity and to open up to the international market.
The theoretical framework of Laursen and Salter (2006) dealing with the 'breadth' and 'depth' concepts of OI is the basis of several research. Breadth is defined as 'the number of external sources of knowledge used by the firm in its innovation process'. It reflects the number of partners with which a company collaborates Lazzarotti and Manzini, (2009). Depth refers to 'the intensity with which the firm draws knowledge from these external sources (Laursen and Salter, 2006). It reflects the phase's variety.
According to Laursen and Salter (2006), these two concepts of inbound OI represent the openness of firms' external search processes.
In the present research, we mobilize the theoretical framework of Laursen and Salter, (2006) to study inbound innovation.

Outbound Innovation
It consists in capturing technology, ideas and innovations and commercializing them through external distribution channels. According to Gassman et al., (2010) it is 'a process that refers to the realization of profits, commercializing ideas, selling intellectual property and multiplying technology while transferring ideas to the outer environment'.

Entrepreneurial Orientation (EO)
EO has been one of the key concepts of entrepreneurship studies that foster competitiveness (Alarcon, 2020; Covin and Wales, 2019) and has received substantial conceptual and empirical attention (Campos et al., 2012). At the company level, EO manifests itself through its philosophy, decision-making practices and strategic behaviours (Wales, 2016).
In the present research, we consider EO as a multidimensional construct (Lumpkin and Dess, 1996) made up of three independent dimensions which are innovativeness, proactivity and risk taking.

Innovativeness
It is defined by Lumpkin and Dess (1996), as the tendency to engage and support new ideas, novelty, experimentation and creative processes that may result in new products, services or technological processes. Proactivity Proactivity is the ability to beat competitors and being the first on the market to introduce new products and services or new technologies (Zahra and Covin, 1995). It is the ability to anticipate and react to the external environment change and the future events (Covin and Wales, 2019).

Risk-Taking
According to Miller and Friesen (1978), risk taking means 'the degree to which managers are willing to make large and risky resource commitments -i.e., those which have a reasonable chance of costly failures'.

Conceptual model and development of research hypotheses
Many researchers were interested to the relationship between OI, EO and performance such as Yun (2016), and Ibarra et al., (2020). Also, Parkman et al., (2012) have studied the EO' mediating role for the relationship between OI and performance. In addition, Cheng and Huizingh (2014) investigated the impact of different strategic orientations on OI and they found empirical evidence of the dependence of OI to the context and particularly found that EO is the most interacted with OI. We refer also to research conclusion of Yun  In the literature review, researchers affirm that the configurational approach integrates the principle of equifinality which assumes the existence of several configurations that can be efficient (Drazinand Van de Ven, 1985; Doty et al., 1993). For our research, this principle implies the existence of multiple combinations of variables which are the dimensions of EO and OI that can perform well. In light of the above, we state the second hypothesis: H2: It exist one or many gestalts that achieve high levels of performance.

Methodological Framework
We present successively the methodological framework chosen for the present research, the measure of variables and the data collection.

Methodological Choice
The configurational approach highlights the existence of possible combinations between a set of mutually theoretical variables. In the present research, we choose the empirical taxonomic approach based on the quantitative method to generate different configurations. Indeed, in order to establish taxonomy of industrial Tunisian SMEs, we pursue a quantitative approach of collecting and analysing empirical data. Thus, we proceed in two stages.The first one is descriptive and consists in generating an empirical taxonomy through cluster analysis method. Thus, Jolibert and Jourdan (2006) recommended this method to operationalizing alignment as 'gestalt' and to ensure internal coherence.
The second stage is explanatory and consists in testing research hypotheses. This involves assessing the predictive validity of the generated taxonomy. So, we carry out variance analysis in terms of performance. After the validity of the taxonomy, we carry out post-hoc tests of multiple comparisons of means, particularly the Scheffe test to identifiy the performing configuration(s).

Operationalization of variables EO measurement
We opt for measurement scale of Hughes and Morgan (2007), which takes into account the multi-dimensional criterion of EO. Taken separately, Innovativeness, proactivity and risk-taking are measured each one with 3 items. Respondents are required to indicate their level of agreement or disagreement with each of these three variables over the past three years. The answers are ordered according to Likert's five-point scale ranging from 1: totally disagree to 5: totally agree.

Open Innovation measurement
To measure inbound OI, we ask questions about breadth OI and depth OI proposed by Laursen and Salter (2006). The breadth consists of the number of external sources of information used by the firm. (2016) is expressed in terms of the intensity of use of sources of information or knowledge that lie outside the boundaries of the company.The depth OI and the breadth were constructed using the same 16 sources of knowledge. As part of the study, respondents were asked to rate the 249 degree of use (partner variety) and importance (phase variety) of the 16 external sources in the development or improving their products or manufacturing processes. The variables are measured on a 5-point Likert scale ranging from 1: Not at all used to 5: Extremely used.

The depth measurement of Laursen and Salter
Outbound OI was measured with 4 items on a five-point Likert scale ranging from totally disagree to totally agree which is derived from the work of Cheng and Huizingh (2014).

Performance measurement
We have used subjective measures which are based on managers' perceptions. These measures are proven to be reliable and generate consistent results compared to objective performance measures that are generally difficult to access.Through the measurment scale of Arend (2003), firm's performance is measured by indicators such as sales growth, market share growth and profitability. The respondents are asked to assess the evolution of these indicators in comparison with those of competitors over the three past years. This is on a five-point Likert scale of measurement, ranging from 1: Strongly decrease to 5: Strongly increase.
Using the SPSS 21 software, we have purified the measurement scales for the different variables through validity tests by principal component analyse (PCA) which we have supplemented with reliability analyses (Cronbach's alpha).

Sample and data collection
We conducted our survey using a questionnaire administrated face to face to managers of Tunisian industrial SMEs located in Great Tunisia.
The statistical processing of the field survey data through SPSS 21 shows that our sample is made up of small companies at 70 per cent and medium-sized companies at 30 per cent.

Results and Discussion
We present below the result of taxonomic structure and the result of hypotheses tests.

Taxonomic Structure
As recommended by Venkatraman (1989), the appropriate method to generate taxonomy is 'cluster analysis' which consists of grouping SMEs into configurations. So, the companies belonging to the same cluster tend to be internally similar and the most different from other. To test the taxonomy' statistical validity and reliable results, a hierarchical classification followed by non-hierarchical classification was performed using SPSS 21.
The hierarchical classification was carried out according to the method of Ward (1963), which is an agglomerative algorithm that performs a series of mergers between observations. This method which maximizes the homogeneity of the groups is most commonly used to ensure a better assessment of the level of similarities or differences between the observations. The results obtained are interpreted from the dendogram which is a schematic visualization of data structure. This classification was completed by a non-hierarchical classification based on the "K-means" method which consists in fixing a posteriori the optimal number of configurations.
The dendrogram and non-hierarchical classification provides a breakdown of the empirical observations into four configurations. In addition, we performed on SPSS 21 an analysis of variance in order to test the statistical validity of the empirically obtained taxonomy. In adition, as affirmed by Jolibert and Jourdan (2006), we carried out the quality assessment of taxonomy using the variance analysis of the classification variables (EO' dimensions, depth OI, breadth OI and outbound OI).
250 The results of the variance analysis (see Table 1) show significant discrimination among the 110 industrial Tunisian SMEs (significance level, p< .05).
The result of the hierarchical analysis shows an optimal number of four SME configurations resulting in a balanced distribution of observations: {configuration 1: n = 16}, {configuration 2: n = 22}, {configuration 3: n = 34}, {configuration 4: n = 38}. The non-hierarchical classification was then carried out using the K-Means method using information from the hierarchical analysis (number and composition of groups).
From this analysis, we label configurations as following: We note that the number of SMEs in each configuration is greater than the statistical threshold of 10%. This justifies the validity of the obtained taxonomy.

Description of The taxonomy of industrial Tunisian SMEs
We present bellow a description of the empirical configurations generated from Tunisian SMEs.
251 We can interpret the profiles of each configuration presented in table 2 as following:

Configuration 1: The Conservative Enclosed
This configuration is the least representative with 16 SMEs. They are characterised by an average orientation of innovativeness and low level of proactivity and risk-taking. Innovations are minor and they are usually in the form of process-related improvements or imitations in product design. These SMEs are followers and passive without any orientation to be the first in the market. They express a little willingness to detect opportunities without being translated into real innovations. They are risk averse. These SMEs are similar to the model of conservative firms and they don't express any orientation towards open innovation. This is clear with low level of breadth, depth and outbound innovation. This profile is similar to the'closed innovation' model.

Configuration 2: The Exceptional Committed
Their entire innovation process is open to several different partners and they are engaged in deep relationship with external collaborators. They try to find a potential market through different channels to better drain and improve the fluidity of their under-expoited innovation (Cheng and Huizingh, 2014). Their commitment in inbound innovation allows them to take advantage of the richness of this external contribution through deep collaboration with multiple partners. Indeed, companies with an entrepreneurial orientation can better manage risk-taking by partnering with third parties. They accept high risks in terms of acquiring new knowledge that may be different from their basic knowledge and technologies. In addition, these companies engage in risk-taking to ensure an adequate supply of their products and services in potential new markets (Wales, 2016). They are also engaged in outbound OI with an average level. In this way, these entrepreneurial SMEs with high levels of innovativeness can benefit more from permeability of their borders to succeed in setting up activities of exploration of innovations and exploitation of technologies. Companies of this configuration are therefore ambidextrous and similar to conclusion of Nobakht et al., (2021) about the importance of organisation ambidextry that is conducive to simultaneous pursuit of exploitation and exploration activities.

Configuration 3: The Vigilant Receptive
The companies of this configuration are committed to open innovation. But they are vigilant towards the development of new products. They are distinguished by their proactive capacity which translates into a systematic search for opportunities and new markets. These companies are ready for all types of collaboration across a wide range of partners (high level of breadth). These SMEs don't make a preliminary selection and aren't engaged in deep relationships (low level of depth). So, it seems that the ambition of this collaboration is to save time in the market. This is why they try to explore the knowledge and technologies that are already provided by other companies. The result we have obtained is therefore consistent with the idea of Wales et al., (2013) who believe that companies with a strong proactivity are more open to external collaboration because they are constantly striving to be the first in the 252 market. However, outbound innovation is almost absent in these companies that can be justified by their average attitude towards risk-taking. These SMEs therefore behave within a degree of vigilance.
We conclude a better fit between the high level of inbound OI and average level of risk taking. We rely on conclusion of Schroll and Mild (2011), who affirm that inbound OI do not include a great risk taking in contrast to outbound OI.

Configuration 4: The Innovative Involved
SMEs of this configuration have high level of innovativeness and proactivity that leads to better monitoring the external environment in order to seize opportunities and respond in time to the future needs of customers. Therefore, these companies are able to identify external knowledge and select the appropriate partners with their internal needs and thus achieve bring added value in terms of strengthening their current capacities and exploring research and development.
The SMEs of this configuration are distinguished by high level of openness to inbound innovation particularly with an implication and commitment in deep relationship and make selection of partners. Their orientation towards inbound innovation is similar to that proposed by Lazzarotti and Manzini (2009), named 'integrated collaborators'.
In addition, this configuration is characterised by an average commitment to outbound innovation in the form of outsourcing of certain activities. Risk-taking behaviour among these SMEs is moderate because firms oriented towards outbound innovation should be more attentive. We cite also the conclusions of Carvalho and Sugano (2016), who stated that outbound innovation activities could be more risky and that the company may not capture the added-values from commercializing its innovation. For this reason, SMEs of this configuration refuse a total engagement in outbound innovation and show a certain precaution.

Testing the Impact of alignment as 'Gestalt' on performance
We verify from this test the effect of OI-EO's alignment on performance.  The interpretation of the performance's means score shows that the 'Committed Exceptional' (C2) and the 'Innovative Involved' (C4) are the most efficient. We note therefore, that with these two different profiles, it is possible to achieve the same result of a high level of performance. Also, our results reveal that the 'Conservative Enclosed' (C1) has the lowest level of performance and finally, the 'Vigilant Receptive' (C3) recorded an average level of performance.

Conclusion
The alignment as gestalt is validated by an empirical taxonomy of the characteristics of Industrial Tunisian SMEs which distinguishes four configurations that are associable with different levels of performance. This taxonomy provides a general view.
We have tested the predictive capacity of this taxonomy on performance. We found that 'Exceptional Committed' (C2) and 'Innovative Involved' (C4) have different profiles in terms of EO and OI and achieve the same high level of performance. This result is consistent with the principle of equifinality as developed by Drazin and Van De Ven (1985) and Dotyet al., (1993) which is specific to the configurational approach.
Our results can also be discussed with regard on the profile of each configuration.We admit the existence of three configuration clans. The first clan represents 54.54% which contains the 'Exceptional Committed' (C2) and the 'Innovative Involved' (C4) configurations. It is the performing clan which is characterized by an entrepreneurial orientation attitude and commitment towards the OImodel. In this sense, we are in line with the conclusion of Ibarra et al., (2020) who argue in favour the adoption of EO and OI as dynamic capabilities useful in times of crisis. Industrial Tunisian SMEs should also be increasingly involved in the adoption of these two strategies, particularly in an environment marked by health and economic crisis.
We find a second clan that represents 14.56% of our sample and includes the 'Conservative Enclosed' (C1) which is marked by conservative attitude and an orientation towards the closed innovation model. It achieves the least level of performance. So, this can be explained by the mentality of these companies characterised by a desire to protect their ideas and innovations, or even because OI is a new strategy for them and they need more time to become familiar with it and to be able to apply it and to copy with other performing configurations (C2 and C4).
The third clan represents 30.9% and contains 'the Vigilant Receptive' (C3) which ischaracterised by an average level of proactivity and innovativeness with an adoption of the OI model oriented more towards large network of partners with moderately deep engagement with them.

Contributions
The conclusion that can be drawn from the present research revolves around some theoretical and managerial contributions. From a theoretical point of view, the result of this research revealed the importance of the alignment perspective as a 'gestalt' integrated into the configurational approach. This holistic approach allowed us to identify combinations between EO and OI and thus go beyond linear and binary relationships. This enabled us to generate a taxonomy which contains four configurations of industrial Tunisian SMEs. This ensures a better explanation of the different levels of performance achieved by these SMEs.
Thus, we respond to the call of Wiengarten et al., (2013) cited in Majdouline et al., (2020) to conduct future research on the integration of EO and OI. This is because entrepreneurship and innovation are considered as two complex concepts that require more elucidation and empirical research to establish conceptual models and explanatory theories. Still at the theoretical level, we learned that the specificity of the configurational approach is that there is no single way to succeed and perform but it may exist more than one way to be performing. It is the specificity of equifinality principle.
On the managerial level, the taxonomy generated by our work is a concrete tool stemming from the Tunisian SMEs reality that could help companies that plan to engage in the strategy of open innovation.
Based on this research, industrial Tunisian SMEs could align themselves with the efficient configuration that suits them to meet the performance challenge. Indeed, this study calls on the managers of Tunisian industrial companies to encourage them to pay more attention to the appropriate conditions for a better implementation of open 254 innovation. By this contribution, the taxonomy formulated could therefore ultimately help SMEs in their strategic entrepreneurial positioning by making them aware of the adoption of OI and its implementation by facilitating their task and allowing them to save time.

Limitations and further research
This research work cannot escape some limits which can obviously become future perspectives of research. Thus, the data collected are specific to the industrial Tunisian PMEs, which may limit the generalisation of the results to other sectors of activity. So, it would be appropriate to look at other areas of activity. In the same vein, our taxonomy depends on the context in which the study is carried out. Nevertheless, the configurations of Industrial Tunisian SMEs are subject to comparison with other configurations. Thus, researchers and practitioners in any country can compare their configurations to those generated in this research. Thus, the specific characteristics of Tunisian SMEs and the conclusion that emerge are likely to be added and enriched the existing literature of EO and OI. So, it is interesting to carry out future research to be able to judge the possibility of applying our taxonomy generated to other countries.
Furthermore, the mobilization of cluster analysis is characterized by a certain stability of the configurations that seems to have a static character and so, can reduce the understanding of their evolutionary trajectories. To design a dynamic taxonomy, it would be advisable to carry out research based on a longitudinal analysis in order to know their future trend. The continuation of this research could be considered by mobilizing a combined data analysis method. So, with the alignment as 'gestalt', research can also be continued by adopting the alignment perspective as 'co-variation' through the mobilization of the Structural Equation Modelling.