MANAGERIAL ROLES IN SMEs AND THEIR EFFECT ON PERCEIVED MANAGERIAL EFFECTIVENESS IN LITHUANIA

: The aim of this study was to analyze the managerial work of Lithuanian managers in terms of managerial roles and their effect on perceived effectiveness of managers’ work in small and medium – sized companies in Lithuania. Literature analysis reveals the concept of managerial work in terms of roles played by managers and discusses the issue of managerial effectiveness. Both concepts are seen as highly complex phenomena where role accomplishing is recognized as an important influential factor in effective managerial work performance. A survey was used as the most appropriate tool for data collection. The respondents at managerial positions were randomly selected in SMEs in Lithuania. Exploratory factor analysis was used to single out managerial roles, and multiple regression analysis was conducted to identify the relationship between managerial roles and perceived managerial effectiveness. The study revealed significant results in terms of performance of managerial roles and questioned the universalistic model of managerial roles by pointing out to their sensitivity to organizational context. The results demonstrate that a part of the activities of managers still fall within the framework of the traditional managerial roles. Yet, findings also suggest that there is a substantial part which can be subsumed under the new separate roles as Analyzer, Representor, and Networker. Regrouped managerial roles revealed a strong positive correlation with perceived managerial effectiveness.

The latest approach presented by Mintzberg (1973) is one of the most influential studies focused on the concept of manager's work through managerial roles. The created model involves ten managerial roles classified as interpersonal, informational and decisional, and provides a clear understanding about managerial work. Despite the criticism concerning structured observation methodology, Mintzberg's framework has been frequently used in studies of managerial work (Kurke & Aldrich, 1983;Shapira & Dunbar, 1980;Allan, 1981;Pavett & Lau, 1983, Paolillo, 1987Tengblad, 2006, etc.).
Investigation of what managers do in different perspectives revealed certain limitations of these studies. According to several scholars (Stewart, 1989;Hales 1986;Chapman, 2001), attention should be paid not only to the content of the management work, but also to its effectiveness and managers' contribution to the achievement of organizational goals. The development of this idea has been reflected in a number of studies in the field of managerial effectiveness, although there is so far no consensus on this concept. Managerial effectiveness has been investigated from different perspectives as managerial roles and work behaviours (Martinko & Gardner, 1990;Analoui, 1999;Willcocks, 2002;Rastogi et al., 2004;Metts, 2007;Wang, 2011;Hamlin & Patel, 2012;Bamel et al., 2015), personal characteristics, skills and competencies (Boyatzis, 1982;Shipper et al., 2003;Allen et al., 2006;Narayan & Rangnekar, 2011), person, process and product approach (Campbell et al., 1970). Though there is an agreement that managerial effectiveness depends on a wide range of personal, organizational and environmental factors (Analoui, 2010), nevertheless difficulties occur trying to explain managerial effectiveness concerning its measurability and comparability.
It should be noted that most research in managerial work in terms of activities and roles was conducted in Western countries, while research in emerging economies is rather fragmentary (Pearson & Chatterjee, 2003;Bao, 2009;Analoui et al., 2010;Pathak et al., 2010;Bamel et al., 2011;Bamel et al., 2015, etc.) and seeking to explore the relevance of traditional managerial roles in terms of cultural disparities of the organizations in Eastern Asian, African, Middle Eastern countries. The findings of these studies revealed that Mintzberg's model has some utility, but economic, social and cultural specifics in these countries should be taken into account.
However, the analysis of numerous scientific literature related to the managerial work research has shown almost a lack of studies concerned with SMEs managers work in different terms (Paolillo,1984;Muir & Langford, 1994;O'Gorman et al., 2005;Florén, 2006;Andersson & Florén, 2008). Overall, managerial work investigations are usually associated with large companies, but in a small company the spectrum of managerial work differs qualitatively and quantitatively from the work of managers of larger organizations. Therefore the issue of managers' work roles and their effectiveness still needs more studies to gain a better understanding of these concepts in the context of SMEs.
In recent years Lithuania has become one of the rapidly growing and innovative economies. However, managers' work and the effectiveness of its performance still remains the important issue both in Lithuanian large companies and in SMEs. Thus the purpose of this study is to investigate the managerial roles performed by managers in SMEs in Lithuania and how they are related to perceived managerial effectiveness. The study reflects whether managerial roles of managers in the context of SMEs are bound to exhibit distinct features other than suggested in Mintzberg's concept. Study results contribute to the research of managerial work in terms of roles and perceived managerial effectiveness in emerging economies.

Conceptual framework of managerial work and effectiveness
This section presents a literature review on the main approaches to managerial work, with particular focus on roles and managerial effectiveness.
The early studies (1951)(1952)(1953)(1954)(1955)(1956)(1957)(1958)(1959)(1960)(1961)(1962)(1963)(1964)(1965)(1966)(1967)(1968)(1969) of managerial work, as pointed by Tengblad &Vie (2012), built a strong foundation for further investigations revealing common patterns of managers' work and its high fragmentation because of a large range among management positions. Later, other approaches to managerial work became popular: managerial work as activities and roles, as power and control, as performance and labour process (Korica et al., 2017). The managerial roles approach was popularised by Mintzberg (1973), who aimed to address a gap between too theoretical attitudes to the managers' work and practical realities (Korica et al., 2017). He diverted the analysis on the investigation of the real content of managerial work, concentrating on the manager's activities which could be characterized by brevity, fragmentation and variety, high level of interaction and a preference for verbal communication (Sancino & Turrini, 2009). According to Mintzberg (1973), a manager can be treated as a head of organization that faces challenges every day in a very complex environment. Based on an observational study of top executives, Mintzberg concluded that a manager's work could be described in terms of ten job roles. Managers perform managerial roles in a different matter depending on their level and functions, environment, personality and context. According to the framework suggested by Mintzberg (1973), managerial roles are divided into three categories: interpersonal roles (as figurehead, leader, and liaison), where managers are formally in charge because of their special status and formal authority; informational roles (as monitor, disseminator, and spokesperson) place managers in a unique position to receive, store and send information; decisional roles (as entrepreneur, disturbance handler, negotiator, and resource allocator) involve managers in significant decision-making about organizational activities. Mintzberg asserts that the ten managerial roles are common in all managerial jobs despite the functional areas or hierarchical levels of management (Mintzberg, 1990).
There are differences in managers' work with respect to the relative importance of roles according to the functional areas and levels in the hierarchies (Mount & Bartlett, 1999). However, the organization would be the one who determines the need for a particular role, and its complementation depends exactly on the manager's skills and capabilities (Kumar, 2015). As managerial roles form an integrated whole and are intensely interacting with each other, taking one of them out can harm managers' activities (de Oliveira et al., 2015).
Mintzberg's managerial roles framework has been criticized by other scholars for a limited sample, for the premise that top managers can represent typical managers, regarding the focus on what managers do without linking it with their achievements, and overall, questioning the purpose and emphasis of the roles and their universality (Snyder & Glueck, 1980;Snyder & Wheelen, 1981;Stewart, 1982;Martinko & Gardner, 1985;Tengblad, 2006).
In spite of the criticism, Mintzberg's approach raised a tide of further investigations. Kurke & Aldrich (1983) managed to prove Mintzberg's ideas by developing a replication study using the same methodology. Their study "confirms the soundness of his results, reinforcing the image of managers as operating in a work setting characterized by fragmetation, brevity, concentration on live media, and dependence on others for initiating contacts" (Kurke & Aldrich, 1983, p. 983). On the ground of controversial discussions, Tengblad (2006) replicated Mintzberg's (1973) study and found more differences than the study of Kurke & Aldrich (1983). Tengblad (2006) concluded that managerial work is neither stable as Mintzberg (1973) claimed, nor very changeable as described in the post-bureaucratic theory, and that "claims of the emergence of radically different managerial work are much exaggerated" (Tengblad, 2006(Tengblad, , p. 1437. Although Mintzberg managerial roles are often generalized, it has to be taken into account that different managerial levels require distinct job content, activities, functions, and performance capabilities. A wide diversity of studies were exploring the link between managerial roles and different aspects of hierarchical levels and functional areas (Paolillo, 1987;Pavett & Lau, 1983;Kotter, 1982), as well as gender (Smith & Schellenberger,1991), the relationship between managerial roles and context (Dierdorff et al., 2009), personality, experience and education. It has been noticed that the manager's level in the hierarchy and functional area have a strong effect on managerial work, and the importance of role-related behaviour increases depending on the manager's place in hierarchy. Paolillo (1987) found that six of the ten roles are influenced by the functional area, thus confirming the meaning of the functional aspect. Pavett & Lau (1983) concluded that roles were more important to top managers than to lower level managers.
A review of the literature reveals other classifications of managerial roles (Morse & Wagner, 1978;Luthans et al., 1985;Quinn, 1990;Mintzberg, 1994;Leslie et al., 2002;Dierdorff et al., 2009;DuBrin, 2012). Mintzberg (1994) reduced his previously determined managerial roles from ten to six. Morse & Wagner (1978) proposed nine managerial roles (strategic problem solving, resource managing, conflict handling, organizing, information handling, motivating, providing for growth and development, coordinating, and managing the organization's environment) that are different from Mintzberg's (1973) framework though are built on it (Morse & Wagner, 1978, p. 24). Quinn (1990) focused on effective manager's performance and determined eight roles such as director, producer, monitor, coordinator, facilitator, mentor, innovator, and broker, which are significant for manager's activities. Leslie et al. (2002) extended the list of roles by adding the role of a manager as an innovator and pushed it back into Mintzberg's (1994) roles classification, because an ability to innovate is considered as a manager's competitive advantage in business. DuBrin (2012) extended Mintzberg's findings and delineated 17 roles (strategic planner, motivator and coach, team builder and team player, technical problem solver, etc.) which reveal the complexity of managerial work. According to DuBrin (2012), the technological progress, knowledge of work importance, changes in organizational context develop significant shifts in the nature of managerial work. As we can see from different classifications, "the nature of managerial roles has changed over the past 50 years migrating from command and control models to contemporary roles that emphasize worker support, coaching, motivating, and facilitating" (Laud et al., 2016, p. 441).
The significance of managerial roles in different contexts steadily attracts a great deal of interest, and Mintzberg's framework despite its controversy is still valid and broadly used to describe managerial work.
As in this study we analyze the conceptualization and appliance of managerial roles as perceived by managers representing small and medium -sized enterprises (SMEs), addressing this specific question regarding their work content has raised some issues.
First of all, taking into account the previous theoretical and empirical studies, it should be emphasized that the investigation of managerial work is usually associated with large companies. However, different scholars (Muir & Langford, 1994;Florén & Tell, 2004;O'Gorman et al., 2005) state that the work of managers of SMEs differs from the managerial work in large companies, in particular because managers from SMEs need to have a variety of skills due to the constant requirement to assume different roles. According to Florén (2006), seeking to ensure the long-term success of an organization, there is a great importance for managers of small and medium-sized enterprises to have conceptual, communicative and technical skills and the abilities to combine them. For this reason, it is often difficult to clearly define the content of the managerial work of SMEs. A comparative study performed by Paolillo (1984) concerning the roles played by the manager on the basis of Mintzberg classification in SMEs and large companies revealed that the role of the Representor -the disseminator of information to external stakeholders -is central to small business managers. Likewise, Entrepreneur, Figurehead and Leader roles in small businesses are more important than in big ones,which is determined by the specifics of small and medium-sized enterprises, where the manager often deals with the development of the company's strategy and ensuring its implementation.

Managerial effectiveness
The issue of managerial effectiveness has been studied over the years in parallel with managerial work. Investigations of this concept raised a great deal of misunderstanding and ambiguity due to its complexity. Scholars focused on different aspects of managerial effectiveness seeking to reveal its essence, characterize an effective manager and its personality, and determine ways to measure managerial effectiveness (Stewart, 1991;Gupta, 1996;Analoui, 1999;Braithwaite, 2004;Bao, 2009;Bamel et al., 2011;Hamlin et al., 2011;Ruiz et al., 2014, and others). But still there is no clear agreement among scholars concerning the concept of managerial effectiveness (Bamel et al., 2015).
Managerial effectiveness is conceptualized and measured in different ways depending on the approaches emphasized by researchers: behavioural, personal effectiveness, person -process -product, effectiveness areas and objectives (Farahbakhsh, 2007). Gupta (1996) envisioned managerial effectiveness as the "ability of a manager to carry out the activities required of his/her position while achieving the results both current and in terms of developing further potential" (Gupta, 1996, p. 399). Rastogi & Dave (2004) acknowledged that managerial effectiveness is not only related with the manager's personality characteristics but also with performance and output.
Behavioural focus is clearly noticed in the analysis of managerial effectiveness. Balarman (1989) defined managerial effectiveness in behavioural terms providing a few job oriented criteria ( communication, delegation of work, planning and scheduling, etc.) for evaluation. The effectiveness of managerial behaviour was also highlighted in the studies (Hamlin & Serventi, 2008;Hamlin et al., 2011;Hamlin & Patel, 2012). The identified management performance criteria: effective planning and proactive control; active supportive management; delegation and empowerment; genuine care for staff; fight for interests of the staff; active learning and staff development; open and trusting managers approach; staff involvement; communication, coaching and well flow of information (Hamlin, 2011) may serve as predictors of the manager's effectiveness. Analoui (1999; proposed a system of parameters like managers' perception, skills and knowledge, choices and opportunities, inter-organizational relationships, etc. that helps to understand how managers can dispense their work in an effective way. Hales (1986) stated that effective managers only do what is related to their direct duty, and the attention should be focused on how they contribute to the achievement of organizational goals.
According to Mintzberg (1973), managerial effectiveness is an ability of managers to perform various managerial roles as a whole. This approach was taken into account in Leslie et al. (2002) study of managerial effectiveness in a global context. Researchers introduced a concept of managerial effectiveness being dependent on four different pillars of overall effectiveness. Manager's personality, capabilities, managerial roles and experience are the main cornerstones to describe effectiveness. In turn managerial effectiveness has been conveyed in five dimensions: managing and leading represents traditional roles of manager and leader; interpersonal relationships describe manager's ability to build and maintain relationships inside the organization with senior managers and peers; knowledge and initiative dimension combines personal knowledge and managerial competences related to confidence, initiative and independence; success orientation stands for goal achievements and attainments of desired organizational outcomes, and contextually adept focus on management of external relationships (Leslie et al., 2002).
Generally speaking, it means how managers' work is related to perceptions of their effectiveness. As there is a lack of unequivocal direct criteria of managerial effectiveness against which managerial work could be measured, managerial work performance will depend on managers' own perception. Taking this into account in our study, effectiveness centers upon the ability to perform multiple managerial roles.
Based on the literature review it could be indicated that there is a comprehensive knowledge accumulated regarding different aspects of managerial roles and effectiveness. However, research done in the context of emerging economies might bring additional contribution to this area.

Sample and data collection tool
Survey of managers of small-medium sized business enterprises was applied to get data on managerial roles and perceived managerial effectiveness. Convenience sampling was applied asking managers of SMEs to fill in the questionnaire. Micro companies with a number of employees less than 10 were not included into the survey, due to specific nature of such small firms. Companies employing more than 250 employees were also considered as big companies and were excluded from the research. The final sample included 96 mainly top and middle-level managers from different SMEs in Lithuania. Companies represented very diverse sectors of economy, the majority of answers came from service providing companies, others were involved in manufacturing or trade activities.
The original questionnaire was translated into a local language to avoid misinterpretations of the statements. To measure performance of managerial roles and managerial effectiveness, the respondents had to evaluate every question on the scale from 1 to 6, where 1 meant complete disagreement, while 6 represented complete agreement.
The majority of respondents (58%) were male, against 42% of female. The dominant group of respondents (53%) consisted of middle-aged managers between 35 and 54 years of age, while younger managers aged from 18 to 34 years made up 40% of the sample. 67% of the respondents had 7 and more years of experience in management job, while only 4% were novice in the field having less than one year management experience. 90 % of managers had university degree -46 % of them held master's and 44 % bachelor's degree, 10 % had lower level education. 47 % of the respondents defined their management positions as being middle level managers, 42 % as top management, and 11 % as bottom-line managers.
The questionnaire included two main constructs. The first construct was meant to define the managerial roles of the respondents, while the second one measured the self-perceived effectiveness of performing those roles. The first construct was developed for exploratory factor analysis (EFA). So it is original for this study, but the initial idea was mainly based on Mintzberg's approach developed in the whole set of publications (Mintzberg, 1973;1994). All questions of this construct were built around three main topics: managing relationships, managing information and managing actions. Such structuring stems from Mintzberg's works, but it was referred to in Leslie and his team's survey (2002) and it was borrowed for constructing this questionnaire as well. Another important source of inspiration were publications by Hamlin and his colleagues (Hamlin & Cooper, 2005;Hamlin & Sawyer, 2007;Hamlin & Serventi, 2008;Hamlin et al., 2011;Hamlin & Patel, 2012), where he analysed effective managerial and leadership behaviour and what it implies. The final construct included 57 questions, 37 of which were taken from the survey of Leslie et al. (2002), others represent input of the authors of this publication in interpreting ideas presented by Mintzberg, Hamlin and other researchers in the area.
The second construct aimed to measure self-perceived effectiveness of a manager's job. This is an original construct developed for this study. To measure effectiveness, Mintzberg's concept of management tasks (Mintzberg, 1973) was taken as a starting point. The questions formulated reflect different tasks paramount to managerial work.
The assumption is made that achieving those tasks would mean higher managerial effectiveness, while disregarding leads to low effectiveness. This construct consists of nine questions, where two questions were taken from Leslie et al. (2002) study and others were constructed by the authors of this survey.

Exploratory factor analysis and reliability of constructs
Exploratory factor analysis was carried out to define the management roles of SME managers. As it was mentioned before, 57 questions related to different areas of managerial work were formulated. The initial factor analysis solution gave ambiguous results. Eight factors were detected, but some variables had very similar correlations with more than one factor. The process continued by eliminating those ambiguous variables. The final solution included only 38 variables, but those variables helped to create a clear picture of managerial roles as they are seen by the managers of SMEs.
A principal component factor analysis was conducted on 38 items with Verimax rotation. The Kaiser -Meyer -Olkin measure verified the sampling adequacy for the analysis. The KMO value was equal to 0.924, which is 'marvellous' and well above the acceptable limit of 0.5 (Field, 2013). Bartlett's test of Sphericity had p -value lower than 0.01, which indicates that correlation matrix of variables was significantly different from an identity matrix, therefore, some correlation between variables exists, and data can be used for factor analysis. An initial analysis was run to obtain Eigenvalues for each factor in the data. Six factors had Eigenvalues over Kaiser's criterion of 1. In combination those six factors explained 73.6 % of the variance.
The items that cluster on the same factor suggest that factor 1 describes the role of Analyser, factor 2 -the role of Representor of the organization, factor 3 -Leader for employees, factor 4 -creative Innovator in problem solving, factor 5 -Decision maker, and factor 6 -Networker inside and outside the organizations. The first factor explains about 18 % of data variance, the second one about 17%, the third about 13%, the forth -10%, the fifth -8%, and the sixth -7%.  The reliability of constructs received was checked using Cronbach's Alpha reliability criterion. All six constructs have high internal consistency with the lowest Alpha of .755 for the role of Networker and the highest of .956 for the role of Representor. Construct's reliability was checked for the self-perceived effectiveness as well, and it was found that the reliability of measurement scale is high, with Cronbach's Alpha equal to .94. An inspirational (charismatic) leader Skilled and competent representative of organization Capable to maintain stability in organization Flexible to adapt to changing environment Effective at managing information and disseminating its flows between organization and its environment Extremely effective in managing conflict to enhance the quality of the decision Capable to evaluate risks and take necessary actions to follow the strategy of meeting company's goals and expectations Consistently drives for better outcomes

Multiple regression analysis
Hierarchical regression was intended to apply to find out the relationship between different characteristics of managers and their perception of the effectiveness of a manager's job. The first model included demographic characteristics of managers as predictors -age, gender, education, managerial experience and management level. The second one included six managerial roles as found using factor analysis. Self-perceived managerial effectiveness was taken as a dependent variable.
Regression analysis revealed that demographic characteristics cannot be used to explain the way managers understand the effectiveness of their job. ANOVA F-test p-value for the first model was greatly above the level of significance and the model could explain less than 4% of variance of data. Therefore, regression analysis was rerun with only manager's roles as predictors.
Since neither factor analysis per se, nor previous literature analysis could suggest the priority of roles making effect on perception of effectiveness of managerial work, it was decided to apply a stepwise regression model, which includes predictors to the model based on pure mathematical relation between the variables. It is the most suited for exploratory model building (Field, 2013), which was the case in the present study.
A five-step model building was carried out in regression analysis. Final regression model contains five out of six variables included in the primary model. Only decision making role did not show important effect on self-evaluated effectiveness (see Table 3). Evaluations of managerial roles can firmly predict how managers will see the effective-ness of their work. The final model explains 82.2 % variations of the dependent variable. But the model strongly suggests that it is not necessary to analyse all managerial roles to know the effect on perceived effectiveness. The first regression model with just one predictor -evaluation of the role of Representor, can explain about 71% of variance (see Table 2). Adding a second predictor increases the value of forecast only by about 5%, and the last two predictors add just about 1% of precision to the forecast each. The role of Representor is the best predictor due to the fact of a very high correlation between this variable and the variable of effectiveness (r=0.843). Other predictors also have strong correlations with the dependent variable (Analyzer r=0.788; Innovator r=0.77; Networker r= 0.702; Leader r=0.652). A strong correlation exists with the Decision maker role as well (r=0.729), but it has a more random character and cannot serve as a predictor. All correlations are statistically significant at the 0.01 level.
The analysis shows a good model fit, which means that the model can be used for prediction of the dependent variable. Durbin-Watson statistic (2.19) is very close to 2, therefore, it confirms the assumption of independence of errors. ANOVA tests for intermediate models and for the final regression model (F=83.198, p=0.00) suggest that models are significantly better at predicting the outcomes than using the means as "a best guess" (Field, 2013).
Having in mind theoretical explanations, there is no surprise that all five predictors have positive relationship with self-perceived effectiveness. It means that higher evaluation of roles included into the model leads to higher perception of effectiveness of the work performed. In the final model with 5 predictors, the role of Representor provides the largest impact on perceiving effectiveness. The increase of the evaluation of the Representor role by 1 point on the scale from 1 to 6 leads to an increase in evaluating effectiveness of 0.296 points. Other roles have less exposed effects, but still quite significant (see Table 4). The smallest effect is achieved by the networking role where the change in the scale by 1 point in evaluating the networking role still leads to 0.121point change in evaluating effectiveness. None of confidence intervals cross zero, which indicates that parameters are significant in representing population. But they are quite large, particularly when we include more predictors. So, using models with less predictors increases the representativeness of the model for the whole population. Multicollinearity is not a problem for the models with VIF statistics below 10 and tolerance statistics above 0.2 (Field, 2013). The data were also checked for meeting the assumption of homoscedasticity and linearity. Partial plots analysis did not reveal issues with meeting those assumptions. No obvious outliers representing cases that might have undue influence a. Dependent Variable: Self-perceived effectiveness on predictor's regression coefficient were detected. To test the normality of residuals, the histogram and normal probability plot were analyzed. The histogram was symmetrical and approximately bell-shaped, but probability plot was a little bit S-shaped, which indicated some problems of skewness. This refers to some violation of the assumption of normality. But this issue was considered as minor. General reliability of the model should be satisfactory to analyze its implications.

Discussion and conclusions
The purpose of this research is to determine the managerial roles of SMEs managers as measured by the developed survey instrument and how they are related to perceived managerial effectiveness. Based on our research analysis, six managerial roles were distinguished using EFA. The study revealed how SMEs managers evaluate the importance of their performed managerial roles. Roles profile according to the priorities can be presented as Leader (28.1%), Decision maker (21.9%), Innovator (14.6%), Networker (8.3%), Representor (6.3%), Analyzer (4.2%). 16.7% respondents indicated the equal importance of several roles at once. It is also noteworthy that in our study no significant differences in roles profiles with respect to managerial positions in the organizations were found. These results differ from some studies (Alexander, 1979;Paolillo, 1981), which concluded that the manager's level in the hierarchy has a strong effect on the extent managerial roles are required. This inconsistency occurred due to the different context of hierarchical structure of the researched companies as they were mostly large. In our case we have another segment -small and medium -sized companies. In small companies hierarchical structure depends greatly on their managers and owner's will because they are the ones who decide when and what can be done. The roles presented in our framework only partly coincide with the roles described by Mintzberg (1973;1994), Morse & Wagner (1978), Leslie et al. (2002). The Leader and Innovator roles do not deviate from the interpretation of mentioned scholars or at least they are very close to their interpretations. The role of the Leader implies being with employees, trusting them and involving them in the decision making process, rewarding them for their achievements, communicating intensively on all issues of their interest. The role of the Innovator overlaps with the role of the Entrepreneur as described in Mintzberg's (1973) classification. It implies creative attitude to new opportunities and thinking "out of the box", encouraging others to think in the same way, readiness to instigate and accept changes.
In the same study Mintzberg (1973) stressed the importance of involvement of managers in decision making activities. But, according to Mintzberg, there is no single decision making role. Decision making includes a whole set of different activities, and these sets can be seen as separate managerial roles. Therefore, he describes Entrepreneur, Disturbance handler, Resource allocator and Negotiator roles as separate ones, falling into the category of decisional roles. The current study indicates that managers of SMEs distinguish decision making as a separate area of their responsibility. But in their mind decision making is not just problem solving and coming up with solutions, it is more about speed, toughness and rationality. Making decisions means acting without procrastination, in a planned and organized way, with the confidence in what has been decided. Such interpretation of the Decision maker role can be attributed to our study and is somewhat more specific and wider than in Mintzberg (1973) or Leslie et al. (2002) studies. Indeed, interpretation of decision making as being a tough process where speed and self-confidence is paramount can have several reasons. We assumed that the peculiarities of SMEs affect the process of decision making. Differently from Tengblad's (2006) replication study of Mintzberg's research, where he noticed that top managers better delegate the subordinates to make decisions than do it themselves, managers of small companies cannot rely on analytical departments and extensive assistance while making decisions. The majority of analytical work should be done by themselves. Lack of time and sometimes skills and competences (due to the fact of diversity of issues) creates permanent pressure on managers, which makes the decision making process quite tough.
Another assumption might be linked to the specificity of the region. For small open economies, including Lithuania, that are characterized by limited home markets, both the establishment of SMEs and their internationalization are important (Diskienė et al., 2015). However, most SMEs are national companies born in the national market and serving first of all national consumers. Operating on an emerging market and being a part of a big European market, local managers face very tough international competition from the part of bigger, financially more robust, and more experienced competitors. This creates pressure for managers to react speedy and "in the right way". Indeed, the study on Lithuanian SMEs internationalization (Diskienė et al., 2015) has revealed that SMEs engaged in international activities beside the other factors stress the significance of personal attitudes, managerial experience, skills and competences.
In this study three roles (Analyzer, Representor and Networker) were found as more specific to SMEs. This is in line with Paolillo (1984) research on managerial roles in small and large companies who concluded that the roles of the Leader and the Representor are central to small business managers, as well as the roles related with external communication.
The role of the Analyzer means finding out important relevant information, analyzing it and applying for the prospects of organization. It requires dealing with organization's internal and external information. Noticing changes in the environment and making sense of them, dealing with internal risk issues or analyzing different processes are particularly important for the managers of SMEs. This role implies using information in a rational, logical way. So managers should be involved in dealing with information in an analytical way, they should be analysts. This role partly reflects the role of the monitor in Mintzberg's classification. Both roles require working with information. But the major difference is that monitoring is separated from the application of received infor-mation. Receiving of information becomes the aim by itself. The role of the Analyzer in this study refers not only to obtaining information but also to utilizing it. It is important not only to forecast future and track changes, but to define threats and opportunities and to influence processes. This finding can be referred to the specifics of SMEs, where finding necessary information, analysis, decision making and implementation are parts of the job of the same manager. Small scale activities do not let specialize in separate activities so respondents find hard to think about getting and using information as two separate stages in dealing with information.
Representor's role implies dealing with outsiders for the benefit of the organization. It includes negotiating on behalf of the organization, interacting with private and public sector representatives, communicating and advocating the interest of the organization in different milieu, exercising influence to promote the company's interest. With reference to Mintzberg's classification, this role partly covers the role of Negotiator, partly the Spokesperson role, but it also has something to do with decisional roles. Appearance of the Representor role can easily be linked to the specifics of management in small organizations. Lack of specialized human resources alongside complexity of task require from managers to see the environment as "a whole". Negotiating with, explaining to, persuading different stakeholders seem as parts of the same job for those managers. Therefore, they do not make difference between activity of negotiation and activity of a spokesperson. Both of them can be seen as activities of promoting the company's interests to outside stakeholders.
The role of the Networker means creating effective alliances and networks inside (building teams) and outside (looking for contacts and creating and maintaining important relations) of organizations. This role is new to this study and represents mainly a mix of Liaison and Leader roles as defined by previous studies (Mintzberg, 1973;Leslie et al., 2002;DuBrin, 2012). The importance of new information technologies and social networks can be assessed by considering why these roles fall into one. Crossing organizational boundaries social networks become part of routine environment for every manager. Managers themselves become members of different networks -social and professional, moreover, their employees make part of different networks as well. Managers feel that they cannot stand apart from these trends and need to use them as an important tool to achieve organizational goals.
Finally, it must be recognized that the work of managers of SMEs, as Muir & Langford (1994) stated, is like a "multi-role behaviour", and they take on "a myriad of roles" (Floren, 2006, 279).
The results of regression analysis show the direct link between evaluation of the accomplishment of managerial roles and perceiving of managerial effectiveness by the manager. It is not surprising due to implied process -outcome relationship. Good performance of managerial roles should lead to better achievement of goals, to higher effectiveness of managerial work. At least such relationship exists in the mind of respondents. So, the findings of this research are in line with the conclusions made in Leslie et al. (2002) and Hamlin & Patel (2012) studies, where positive effect of the roles on managerial effectiveness has also been identified.
The study conducted suggests that the strongest link of perceived managerial effectiveness is with the role of the Representor of the organization. Managers who perceive themselves as being good negotiators, good organizations' agents dealing with outside entities in promoting organizational cause tend to directly link these qualities to higher managerial effectiveness. The roles of Innovator, Analyzer, Leader and Networker also affect evaluations of effectiveness, but their impact is more miniscule than the assessment of the role of Representor.
Our study yields an important conclusion that managerial roles can be better explained when specific settings of managers are taken into account. Universalistic view allows describing the work of a manager, but it lacks precision and can discard important work priorities. The findings do not assume that Mintzberg's framework cannot be applied in investigation of managerial work of managers in SMEs. Generally, there is no disagreement between managers in what kind of activities they participate in. The issue is how they understand the link between different sets of activities. The current research suggests that managers of SMEs perceive managerial roles as much more integrative than it is suggested by Mintzberg's (1973) model: information gathering cannot be taken apart from analysis and utilization of the information collected; public relations, negotiations and pursuing of the company's interest should also be taken as one managerial role and so on.
It is also noteworthy that Mintzberg's managerial roles model is recognized as more universal and neutral in the light of national culture, it still needs to be adapted in different contextual conditions. Our study was conducted according to certain contextual criteria: specific sample characteristics (local SMEs managers), monocultural business environment, different managerial level managers, etc. In terms of cultural differences more studies need to be conducted in less researched regions with emerging economies in order to evaluate the influence of the contextual and cultural aspects and to provide more evidence if managerial roles adjust in terms of distinct culture and determine other criteria. Another important criterion to check against might be the organization's size defining specializations within the organization and, therefore, affecting the scope of integration of different management activities, and the content of separate roles. New informational technologies changing the way people interact could also be chosen to check against. Developing of understanding of the managerial role playing in different contexts using those and other criteria could substantially improve our understanding of more effective managerial work.