KNOWLEDGE MANAGEMENT SIGNIFICANCE AND COMMUNICATION COMPLEXITY IN THE CONTEXT OF INNOVATIVE ENTERPRISES: CASE OF POLISH NEWCONNECT MARKET

The main goal of this study is to show how the communication complexity influences the knowledge sharing in the organizational dimension, on the example of the NewConnect market in Poland. This paper presents and investigates the above mentioned topics of showing the importance of different forms of communication in sharing knowledge in innovative enterprises based on the questionnaire that has been conducted in NewConnect market in Poland. The empirical study proved that: (1) there is no correlation between the level of communication management in innovative enterprises, (2) the age of the enterprise level and the significance of knowledge management in innovative enterprises, (3) the age of the enterprise and the level of communication complexity in innovative enterprises. The findings of this research give varied and valuable arguments to managers from each sector in paying much attention to different forms of communication with stakeholders in sharing knowledge. This paper is valuable to academics and practitioners in search of reliable data on the influence of modern and traditional forms of communication on sharing knowledge in innovative enterprises, showing the fulfilment of the gap this type of study in the literature.


Introduction
Communicating with stakeholders and knowledge management processes are the basis for the functioning of today's businesses-in different industries and sectors. In addition, establishing lasting relationships with stakeholders (both internal and external) can determine the broadly understood quality of innovative processes. Therefore, it is possible-in some simplification-to assume that communication with stakeholders and sharing resources (in particular the knowledge) have an impact on the development of enterprises by delivering the thereby increase the solidity and effectiveness of communication processes. It is worth remembering that active and long-term communication with various, diversified classes of entities (both internal and external) is a source of costs for the enterprise. This is a kind of "an investment". Therefore, the complexity of communication should be properly planned, organized, implemented and controlled, as well as integrated with innovative processes.
At this point it is also worth emphasizing that the complexity of communication of contemporary enterprises-also operating in innovative industries and sectors-should not be determined only by modern forms/methods of communication. Traditional forms are still of great importance, e.g. direct conversations with employees or clients and business partners (Woźniak, Wereda 2018). Such activities, seemingly "outdated", give the opportunity to create trust between the enterprise and stakeholders (Wereda, Zaskórski 2018). It is worth remembering that trust is particularly important in the processes of creating, developing, implementing and commercializing innovations. This indicates the legitimacy of a specific diversification (by combining modern and traditional solutions) of activities and methods taken into account in shaping the complexity of communication processes of innovative enterprises with various groups of stakeholders.

Methodology of the research
The subject scope of the study concerns methods of communication (in a traditional or modern way) of innovative enterprises with various stakeholder groups (internal and external). The article also links the issue of the complexity of communicating with stakeholders and the importance of knowledge management in the development of enterprises. The subjective scope of the research is innovative enterprises operating on the NewConnect market in Poland. The study included 60 enterprises (15.7% of entities from the populationpopulation comprised 381 companies from Poland). Data are actual for December 2018 (New Connect Statistic Bulletin 2018). The subject structure of the activities of the surveyed entities is contained in Table 1. Source: own elaboration The study used a random systematic selection (taking into account the criterion of the leading profile of activity indicated for the purposes of the NewConnect registry) in the layers. Respondents were CEOs or managers (of the highest or middle level) responsible for the area of relations with the environment, IT or innovations, employed in enterprises listed on the NewConnect market. One respondent from each company was qualified for the study ( Table 2). The structure of the research sample-taking into account different criteria-is described in detail in Table 3.
The empirical study was carried out between November and December 2018 and covered the entire country (16 provinces in Poland). The largest number of surveyed enterprises is based in central Poland. In turn, the least studied entities are located in northern and eastern Poland (Figure 1). Such spatial distribution of the surveyed entities results mainly from the location of companies listed on the NewConnect market-the selection of entities for the research sample reflected the spatial distribution of entities in the entire population. In addition, such a ENTREPRENEURSHIP AND SUSTAINABILITY ISSUES ISSN 2345-0282 (online) http://jssidoi.org/jesi/ 2020 Volume 7 Number 3 (March) http://doi.org/10. 9770/jesi.2020.7.3(35) spatial distribution of the surveyed enterprises points to the fact that the so-called "innovative" entities predominate in western, southern and central Poland-and therefore in the areas that are best developed in terms of industry, services and trade. Northern and eastern Poland are mainly agricultural and recreational areas. The research tool was the Computer Assisted Self Interviewing questionnaire. Respondents on a 5-point scale assessed both the level of use of particular communication channels with stakeholders and the degree of influence (importance) of knowledge management on the development of the enterprise. The results of the evaluation of each question (factor) determined by the respondents were subjected to statistical analysis-factor analysis. The calculations were carried out using the IBM SPSS Statistics 24 software (PS IMAGO 4.0). The study also employed a method of critical analysis of the literature, as well as methods of analysis, synthesis and induction.
Five hypotheses were put forward to achieve the goal of the study: 1. Hypothesis No. 1: Knowledge management significance is at a high level in innovative enterprise. In order to answer the above problem, the article will present the methodology of constructing these indicators, referring to specific partial factors (Table 5 and Table 7). These factors ( Composite ratios were used in the study because: they give a chance to take into account a large number of factors, enable holistic analysis, and provide the basis for a complex, multi-faceted quantification and evaluation of the studied phenomena (Nardo et al. 2005).

ENTREPRENEURSHIP AND SUSTAINABILITY ISSUES
Factors included in the study (both for the purposes of the construction of KMSI and CCI) were designed to measure (on a 5-point scale) the approach of enterprises to the implementation of individual activities as part of communicating with stakeholders and the impact of these relationships on knowledge management. The value of ENTREPRENEURSHIP AND SUSTAINABILITY ISSUES ISSN 2345-0282 (online) http://jssidoi.org/jesi/ 2020 Volume 7 Number 3 (March) http://doi.org/10.9770/jesi.2020.7.3(35) "1" meant that the activity is very rarely implemented or its impact is very low, and the value of "5", that the action is implemented very often or its impact is very large. The reliability of the scale was analyzed using the Cronbach's alpha coefficient to verify the quality of the data. In order to increase the transparency of the analysis, all indicators will be described together. The KMSI indicator will be presented first. For a full list of 13 factors describing the impact of knowledge management on the development of an innovative enterprise, the value of Cronbach's alpha coefficient was 0.689 (Table 4). Taking into account the methodological recommendations, the obtained value could be considered sufficient. The conducted analysis also indicated that there is no possibility to increase the reliability and quality of the scale in case of removing further factors (Table 5). In contrast, for the four CCI indices, the following Cronbach alpha coefficient values were obtained: CCI_int (0.725), CCI_indcust (0.898), CCI_instcust (0.731) and CCI_buscoop (0.701) ( Table 6).

ENTREPRENEURSHIP AND SUSTAINABILITY ISSUES
ISSN 2345-0282 (online) http://jssidoi.org/jesi/ 2020 Volume 7 Number 3 (March) http://doi.org/10. 9770/jesi.2020.7.3(35) For the construction of KMSI, as well as up to four CCI-indicators, methodological recommendations regarding the development of composite indices developed by OECD (2008) were used. The adopted methodology for the construction of all indicators included the following stages (Nardo et al. 2005): 1. determining the scope of measurement and the legitimacy of using the composite indicator, 2. selection of partial factors, 3. evaluation of the quality of empirical data, 4. assessment of the relationship between partial factors, 5. giving weights to the partial factors and their aggregation to the composite indicator.
The results of the implementation of the first three stages for the KMSI indicator are included in Tables 4 and 5, and for the four CCI indices in Tables 6 and 7. In the assessment of relations between partial factors and their aggregation, the factor analysis method was used for the composite indicators KMSI and CCI (by means of the main component analysis-PCA) (Hudrliková 2013). The Kaiser-Mayer-Olkin coefficient and the Bartlett sphericity test were used to verify the correctness of the PCA analysis. The limit value of the KMO coefficient is commonly adopted at the level of 0.5 to 0.7 (Williams et al. 2012). In the case of the KMSI coefficient, the value of KMO statistics was 0.554 (Table 8), and for CCI, respectively: CCI_int (0.675), CCI_indcust (0.713), CCI_instcust (0.740) and CCI_buscoop (0.729) ( Table 9). Bartlett's sphericity test for all five indicators showed that the hypothesis of uncorrelated coefficients can be discarded-test statistics are at a significance level lower than 0.001. Further PCA analysis is justified and methodically correct (Table 8 and Table 9). In a further analysis for all five indicators, the method of distinguishing main component factors with Varimax rotation was applied. However, the selection of components was based on the Kaiser criterion, which assumes that the eigenvalues of factors will be greater than "1" (Table 10). In the case of the KMSI indicator, factor analysis gave the basis for qualifying 13 factors to 5 components (Table 11).  Assigning individual factors to constituents made it possible to name all components of the KMSI indicator and to give component weights. The weights have been normalized by the sums of the squares of charges that correspond to the part of the variance explained by the component. The construction of four CCI indices was carried out in the same way. The CCI_int indicator consists of only one component, CCI_indcust and CCI_buscoop ratios from two components, and CCI_instcust from three components. The rules of all five indicators are included in 12.

Source: own elaboration
The obtained formulas of KMSI and CCI indicators will be used to verify hypotheses, and the values adopted by these indicators will be described in further parts of the article.

Research results
The distribution of KMSI values is characterized by weak left-side skewness, which means that the majority of values were above the average value (Table 13). Considering the fact that each of the 13 factors included in the structure of the KMSI indicator was assessed on a 5-point scale ("1" means very rare/sporadic use of the action, and "5" very frequent use of the measure), the average value of the indicator at the level 3.4354 indicates that on average, the entirety of the importance of knowledge management in the development of innovative enterprises is at a moderate level. The "limit" (median) on a 5-grade scale is 3.00. Generally, it can be assumed that the low level of significance of knowledge management is for KMSI values in the range <1, 2.5, average level in the range <2.5, 3.5), and high in the range <3.5; 5>. However, this is a contractual and standardized division, because precise indication of the level of the importance of knowledge management requires the identification of the needs and capabilities of the company in this respect. The distribution of CCI_int values is characterized by strong left-side skewness, which means that the vast majority of values were above the average (Table 14). The CCI_instcust indicator also has left-sided skewness (Table 15). On the other hand, relatively low right-side obliquities are characterized by CCI_buscoop (Table 16) and CCI_indcust (Table 14)-which means that just over half of the values were below the average value.
Taking a similar assumption, as in the case of RBM, that the low level of communication with a given stakeholder class is CCI in the range <1; 2.5), average in the range <2.5; 3.5), and high in the range <3.5; 5>, it can be assumed that communication complexity:  with internal stakeholders (in an enterprise) is at a high level (average at the level 4.5625) (Table 14);  with individual customers is at a low level (average at the level 1.8690) (Table 14);  with institutional customers is at an intermediate level (average at the level 3.4067) (Table 15);  with business co-operators is at an intermediate level (average at the level 3.3146) (Table 16). Knowing the average level of importance of knowledge management for the development of innovative enterprises and the complexity of communication of these enterprises with various stakeholder groups, it is possible to make an in-depth analysis of this issue from the perspective of individual thematic areas (i.e. two components of KMSI and CCI indicators). For this purpose, the results of the factor analysis were used. This enabled the grouping of individual factors for KMSI characterizing the activities under knowledge management in five thematically coherent components (Table 13). Due to the fact that particular factors and components of the KMSI were assessed on a 5-point ordinal scale, the Friedman test was used to assess the degree of importance of knowledge management and the design of a uniform ranking of components (Table 17 and Table 18). The lowest level of significance of knowledge management in the surveyed enterprises concerned the informational consistency (the C5 component)-a result in the Friedman test with an average rank of 1.57. The highest complexity was noted for the component (C1) associated with the acquisition of information resources-the average rank at 4.95 (Table 17).
A detailed list of 13 partial factors assumed in the study for the construction of the KMSI indicator and those subjected to the Friedman test is presented in Table 17. The respondents relatively most often indicated that in the development of enterprises the activities related to: increasing demand for trainings (f-11), enhancing the competences (i.e. knowledge and skills) of employees (f-8), increasing the scope of obtained data, information and knowledge from the environment (f-4), and increasing the scope of obtained data, information and ENTREPRENEURSHIP AND SUSTAINABILITY ISSUES ISSN 2345-0282 (online) http://jssidoi.org/jesi/ 2020 Volume 7 Number 3 (March) http://doi.org/10. 9770/jesi.2020.7.3(35) knowledge from employees (f-5). On the other hand, in the enterprises surveyed, the least chance for development of enterprises (in the context of knowledge management) is seen in: increasing the reluctance of workers to changes in the enterprise (f-12) and the introduction of conscious restrictions to data, information and knowledge for the various positions and management levels (e.g. in the form of procedures of access to information resources) (f-2).  For a detailed analysis of the complexity of communication with stakeholders, the results of factor analysis were also used. This enabled the grouping of factors for individual CCI indices in (Table 19):  one component-for CCI_int;  2 components-for CCI_indcust;  3 components-for CCI_instcust;  2 components-for CCI_buscoop.
Also for all four CCI indices, the Friedman test was used to assess the complexity of communication with stakeholders and to create a uniform ranking of components (Table 19 and Table 21).   (Table 19).
For communication with internal stakeholders, a detailed list of 4 partial factors adopted in the study to construct the CCI_int index and subjects subjected to the Friedman test is presented in Table 20. Respondents relatively often indicated that in the formation of the complexity of communication with internal stakeholders the most important is the action related to email accounts (f-4 ), and the smallest of communication in the form of paper documentation (f-1). For communication with individual clients, a detailed list of 8 partial factors adopted in the study to construct the CCI_indcust index and subjects subjected to the Friedman test is also presented in Table 20.
Respondents relatively most often indicated that in shaping the complexity of communication with individual clients the most important activities are: communication in the form of direct meetings (f-3), email accounts (f-5)

ENTREPRENEURSHIP AND SUSTAINABILITY ISSUES
ISSN 2345-0282 (online) http://jssidoi.org/jesi/ 2020 Volume 7 Number 3 (March) http://doi.org/10. 9770/jesi.2020.7.3(35) and traditional telephone calls (f-4), and the smallest with the use of social networking sites/portals (f-8) and corporate portals (personalized user accounts) (f-6). For communication with institutional clients, a detailed list of 8 partial factors adopted in the study to construct the CCI_instcust index and subjects subjected to the Friedman test, is presented in Table 20. Respondents relatively often indicated that in the formation of the complexity of communication with institutional clients the most important are activities related to: email accounts (f-5), traditional telephone calls (f-4) and in the form of paper documentation (f-2), and the smallest of: corporate portals (personalized user accounts) (f-8), teleconferences (f-6), as well as social networking sites/portals (f-7). In turn, for communication with business partners, a detailed list of 6 partial factors adopted in the study to construct the CCI_buscoop indicator and those subjected to the Friedman test is presented in Table 20. Respondents relatively often indicated that in the formation of communication complexity with business partners the most important activities include: email accounts ( f-4), traditional telephone calls (f-3) and in the form of paper documentation (f-1), and the smallest of: teleconferences (f-5) and corporate portals (personalized user accounts) (f-6).  In order to verify hypotheses 3-4 the Spearman's rho correlation coefficient was used (Table 22 and Table 23).
On this basis, one can make a negative verification of the hypotheses:  No. 3, that says that the higher level of communication complexity (with internal and external stakeholders), the higher level of knowledge management significance in innovative enterprises; however, it should be noted that a weak, positive and statistically significant correlation only occurs in the complexity of communication with institutional clients and business partners (Table 22);  No. 4, that says that the older enterprise, the higher level of knowledge management significance in innovative enterprises;  No. 5, that says that the older enterprise, the higher level of communication complexity (with internal and external stakeholders) in innovative enterprises; however, it should be noted that a weak, positive and statistically significant correlation only occurs in the complexity of communication with internal stakeholders of innovative enterprises (Table 23) .

Conclusions
Communication with stakeholders and knowledge management are important processes related to the planned and structured development of innovative enterprises. It is worth noting that these processes do not have to always be correlated with each other. They can often overlap independently and affect themselves only seemingly. Furthermore, an enterprise, in order to be considered as the innovative unit, may not have both the complexity of communication with stakeholders and knowledge management development (knowledge significance) at a high level. It must be emphasized that the peculiarity of these both categories depends heavily on industry, market, regulations, customers, etc. The universal (global) dependencies and applications should not be adopted here.