Abstract: In this study, an attempt is made to demonstrate the relationship between style and organizational performance. With data from randomly selected companies quoted on the 1st tier market of the Nigerian Stock Exchange (NSE), this study empirically established, using the ordinary least square, a relationship between style and organizational performance. The practical implication of this finding is that the employees of Nigerian quoted companies are sensitive to the leadership behaviour of their chief executives. Consequently, the research calls for employee centred leadership behaviour.
INTRODUCTION
Style is the leadership style of the top management. Executives of organizations are usually interested on how they can make their organizations more effective and result oriented. Consequently, they are often interested on how they can lead their organizations successfully.
This study examines, empirically, the relationship between the style and organizational performance. The objective of the study is to specifically determine the relationship between leadership behaviour of the Chief Executive Officers and organizational performance of companies quoted on the 1st Tier of the Nigerian Stock Exchange.
Leadership is a subject that has been viewed from various perspectives: Folklore, conceptual framework and even empirical studies and yet there is no unanimity to its definition and content (Galbraith, 1977; Schein, 1992; Luthans, 1992; Kellerman, 2001; Tiger et al., 2001). Little wonder Sorcher and Brant (2002) are of the opinion that leadership is a complex multifaceted capability, with myriad nuances and subtleties.... However, in spite of these complexities, we examined some leadership styles and presented our perspective. The behaviour of managers is a symbolic communication to the employees down the line of what matters in the organization (Pascale and Athos, 1981; Ohiwerei, 1990). The management of each organization would need to be conscious of the type of leadership style and the behaviour they exhibit because these are seen, interpreted and emulated by the staff. Hannagan (2002) indicates that it can be argued that management is largely concerned with leadership because managers need to establish a sense of direction and to motivate people to move in that direction.
Pascale and Athos (1981) and Peters and Waterman (1982) examine management style in their presentations. Pascale and Athos (1981) specifically present the personal management styles of the Chief Executive Officers of Matsushita of Japan and International Telephone and Telegraph of USA. Ohiwerei (1990) is however, of the opinion that since to lead means the action of guiding or giving an example, direction given by going in front ... the leadership we are talking about here is not that of a group but that of an individual. He concludes by indicating that an individual can guide, give example and direct more easily than a group (Ohiwerei, 1990). Therefore, in this study we are concerned with leadership style and not management style for several reasons. First, effective management is largely concerned with leadership (Hannagan, 2002). Second, management is concerned with building competencies, control and the appropriate balance of power (Zaleznik, 2004), management is about coping with complexity (Kotter, 2001), organizing and staffing (Peters and Austin, 1985; Ndiokho, 1994; Kotter, 2001) and controlling (Peters and Austin, 1985; Umoru, 1993; Ndiokho, 1994; Kotter, 2001). Leadership, by contrast, is concerned with coping with change (Umoru, 1993; Kotter, 2001). This it does by setting a direction (Kotter, 2001; Zaleznik, 2004), aligning people (Kotter, 2001), motivating (Kotter, 2001), inspiring (Kotter, 2001; Zaleznik, 2004) and by unleashing energy (Peters and Austin, 1985; Umoru, 1993). Third, we believe that leadership style unlocks and frees latent energy in employees to accomplish organizational goals. In our opinion, management style is the macro view and leadership style is the micro view that mirrors what superior officers do. Leadership and management styles are two distinct but complementary systems of action (Kotter, 2001). In the view of Zaleznik (2004), a managerial culture emphasizes rationality and control...leadership is simply practical effort to direct affairs.
The style of leadership depends on the manager`s perception of and attitude to people and his assumption about the nature of human nature (Mullins, 1996; Livingstone, 2003) and as Hannagan (2002) puts it one of the greatest contributions to the understanding of leadership has been the work of McGregor (1960) in developing Theories X and Y. Theory X sees man as inherently lazy and dislikes work while theory Y sees man as regarding work as a natural phenomenon and does not only accept work but seeks it. These two theories are the two ends of a continuum. At the negative or harsh end is the theory X and the more benevolent end is theory Y. Leadership styles that have been suggested include those by Likert (1961), Vroom and Yetton (1973), Bass (1985) and Goleman (2000).
Likert (1961) identifies four models of leadership styles or systems: System 1: Exploitative authoritative, System 2: Benevolent authoritative, System 3: Consultative and System 4: Participative group. These models of leadership are a continuum. At one end of the continuum is the Exploitative authoritative and at the other end is the Participative group. The most authoritative is system 1 and the most democratic is system 4. Likert consistently found that most effective organizations exhibit system 4 characteristics and the least effective organizations portray system 2 characteristics. According to Dowling (1975) system 4 leadership improves performance and profit.
Vroom and Yetton (1973) identify five leadership styles in terms of the participation of subordinates in decision making process. These are autocratic 1 (A1), autocratic 11 (A11), consultative 1 (C1), consultative 11 (C11) and group participation (G). These are again a continuum with autocratic 1 being the most autocratic and group participation being the most democratic. In group participation leadership style, managers act as coordinators in order to enable the group reach a consensus which is accepted and implemented. Of course, once a consensus is reached all group members are committed to the implementation which impact positively on performance.
Bass (1985) contracts two styles of leadership; the transactional and the transformational. The transactional leader determines what the subordinates need to achieve (personal and organisational goals), classifies the requirements and assists subordinates to have the confidence that they can reach the determined goals. Transformational leader motivates subordinates to do better than they would have by raising motivation and the importance and value of their tasks within the organisational goals. This leadership style emphasises the importance of the subordinates and their tasks and therefore motivates employees to greater performance.
Goleman (2000) found that leaders who used styles that positively affect climate in an organization make better financial results than those who did not. The six styles identified are coercive, authoritative, affiliative, democratic, pacesetting and coaching. The coercive leader requires immediate compliance in a drive for achievements. Subordinates are required to do what they are told. Initiatives are stifled and employees` sense of responsibility is eroded. The authoritative leader mobilises subordinates towards a direction by having a clear vision and making subordinates know how their work fit into the overall organisational strategy. The affiliative leader creates strong harmony and keeps subordinates happy. He creates strong loyalty, improves communication through sharing ideas and initiatives. The democratic leader forges consensus through participation. The pacesetting leader sets high performance standards for himself/herself and everyone else. This style does not tolerate poor performance from all and is obsessed with doing things better and faster. In the coaching style, the leader (who is the coach) helps subordinates identify their strengths and weaknesses and encourages them to establish long-term development goals. The coach relies on delegation and agrees with the subordinates their roles and responsibilities.
In the light of the above, it is evident that all the authorities are agreed that leadership styles range from autocratic to the very democratic. We are of the opinion that the leadership style adopted depends on the superior`s managerial frame (such as background, experience, confidence, perception of the nature of human nature and leadership inclination), the willingness of the subordinates to accept responsibility and take decisions and the situations manager find himself/herself (such as the task to be performed, work procedure and the shared value of the organisation). Leadership style used by a given leader is not static but dynamic based on what is to be done (Goleman, 2000; Schaeffer, 2002).
Having identified several leadership styles the next step is to examine what leaders do. In the view of Prentice (2004) effective leaders take personal interest in the long-term development of their employees and they use tact and other social skills to encourage employees to achieve their best. Leadership does not only exercise expertise but it also empathizes with the subordinates (Pagonis, 2001; Ohiwerei, 1990; Goleman, 2004), communicates vision (Ohiwerei, 1990; Tiger et al., 2001), understands the changing needs of customers and employees (Tiger et al., 2001), have a sense of direction (Omolayole, 1983; Heifetz and Laurie, 2001; Tiger et al., 2001), energizes and motivates followers (Ndiokho, 1994; Tiger et al., 2001; Goleman et al., 2001; Livingstone, 2003; Goleman, 2004; Maccoby, 2004), impacts a strong sense of value to followers (Tiger et al., 2001; Peters, 2001; Bennis and Thomas, 2002), provides emotional balance (Goleman et al., 2001; Heifetz and Laurie, 2001; Goleman, 2004) and maximizes followers well being not their comfort (Heifetz and Laurie, 2001).
Leadership, with the activities highlighted above, galvanizes organizational efforts to achieve intended and emerging organizational goals.
MATERIALS AND METHODS
The research population comprised companies quoted on the Nigerian Stock Exchange. There were twenty six industrial categories in the 1st Tier Securities as at December 31st, 2005. Each of them consisted a listing of between one company as in commercial/service industry and thirty-eight as in banking industry. There were two hundred and three (203) listed companies as at that date.
A sample of the research population was taken since time and costs were constraints. Great care was exercised to get a fair representation of the population as sample. Consequently, proportional stratified sampling method was utilized in selecting participating companies. Six industrial sectors of banking, insurance, food/beverages and tobacco, healthcare, industrial/domestic products and packaging were sampled. Fifty three companies were selected from the six industrial sectors of banking 19 (38), insurance 11 (23), food/beverages and tobacco 7 (13), healthcare 6 (11), industrial/domestic products 6 (12) and packaging 4 (8). The number sampled is immediately beside the name of the industries while the number of the listed companies in each industry is in parenthesis. These industrial categories and the organizations sampled were selected by the lottery method. The Chief Executive Officers (CEO) of each of the companies or their designated representatives were the respondents. These categories of respondents were selected because the organization was our unit of analysis and the persons occupying these positions were not only expected to be knowledgeable about the variables we were interested in; they were also the driving force for each of the organizations.
Present study was both ex-post facto and cross-sectional survey research methods. The reasons for the choice of these methods were first that we investigated events in which the interactions between the dependent and independent variables had already occurred and were not manipulated. Second, the questionnaire method of gathering data was adopted. Third, data for the dependent and independent variables were gathered as at a particular point in time.
Both primary and secondary data were used in the conduct of this research. The primary data were from the responses to questionnaire administered to the selected sample as indicated above. The questionnaire elicited information from the companies on the research variables. Secondary data were the financial statements of the selected companies for 2005 from which we calculated accounting based organizational performance.
Reliability (internal consistency) was tested using the Cronbach`s coefficient. The Cronbach`s alpha of each of the variables should not be less than 0.7 (Nunnally, 1978; Fron et al., 1992). The Cronbach`s alpha value of leadership behaviour of the chief executive officer (S) was 0.85. The high value Cronbach`s alpha test indicated that the stability, dependability and predictability of the measuring instrument were very high (Asika, 1991).
Organizational performance was defined using return on total assets. Style was defined as the characterization of the leadership style of the chief executive officer of each organization adopted to achieve organizational goals. This consisted of concern for employees. Six Likert type questions obtained from Blake and Mouton (1985) were used to measure style (Appendix 1).
The Ordinary Least Square (OLS) regression models were used in the analyses of data collected. The simple regression model was used to identify the relationship of the independent and the dependent variables.
The model tested, in the study, was
|
P = a0+a1S+u |
| Where: |
| P | = | Return on Total Asset (ROTA) |
| S | = | Style (leadership behaviour of CEO concern for employees) |
| u | = | Error term |
| a1 a1 > 0 |
= | Coefficient of the research variable |
Research hypothesis: The research hypothesis tested was: Leadership behaviour of the Chief Executive officers is positively related to organizational performance.
Limitations of the study: The first limitation of this study is that the majority of the limited liability companies in Nigeria are not quoted on the Stock Exchange. Therefore, the size of the sample was small in relation to all companies in Nigeria.
Second, the study was restricted to only Style as independent variable and Return on Total Assets (ROTA) as the measure of performance. It is however expected that further research would examine the relationships of other independent variable and ROTA and other performance measure with style.
RESULTS
Here, we present the analyses of the data acquired. Table 1 shows the spread of the administered questionnaires, the responses of each industry and the response rates.
Table 1 shows that the total number of quoted companies of the six industries selected were 105. The proportional stratified sample of the population at 51% was 53 companies. The response rate of each industry ranged from 63% in banking industry to 100% in packaging. At the end, 40 questionnaires or 75% of the sample size were returned and usable.
Testing for the assumption of regression: We tested below the acquired data to check if it qualified to be used for regressions based on the assumptions of regression model.
Measurement error: That the dependent and independent variables are observed without measurement errors. We have indicated that style (leadership behaviour of the CEO) was reliably measured with Cronbach`s coefficient 0.85.
Normality: That the population distribution of the variable should be randomly distributed. To test the normality assumption, the skewness and kurtosis of the data were determined. The results are shown in Table 2.
In Table 2, the variable had critical value indicating that the skewness is different from zero at 5% level of analysis. However, the actual value of skewness in the table was -0.020 which was not close to the problem value of 2 as indicated by Hoyle (1995). It will be seen from the table that the mean and the median of the data coincide indicating normality of the data. We also tested for the difference between the mean and the median and we found that there was no statistically significant difference between the two at 5% level of analysis. Moreover, the value of kurtosis of the variable was -0.903 which was not close to the problem value of 4 recommended by Hoyle (1995). These results indicate that the distribution of the variable was normal.
Randomness: That the dependent variable is an unvaried random variable for each specific combination of independent variables. According to Oyesiku (1995) this is satisfied by random sampling choice. As indicated earlier, we used proportional stratified random sampling method in which the industries and organizations were selected by lottery method. This assured randomness.
Autocorrelation: That there be no autocorrelation. Autocorrelation refers to a situation where the stochastic error term is correlated with itself over time. The generally accepted autocorrelation test is the Durbin-Watson test, which calculates the d-statistics. In Table 2, Durbin-Watson value was 2.25 for style. This was within the acceptable range of one (which is the lower limit) and four (which is the upper limit) (Iyoha, 2004). Therefore, autocorrelation was not a threat to the results of the regression.
| Table 1: | Spread of administered questionnaires and responses |
| Source: Author`s fieldwork | |
| Table 2: | Skewness and Kurtosis of the entire survey |
| S: Style | |
Homoscedasticity: That the various arrays of the dependent variable have the same variance. According to Iyoha (2004) large R will suggest the presence of heteroscedascity. Our R in this study was 0.52. Therefore, homoscedasticity was assumed.
Linearity: That the regression of Y (the dependent variable) and X (the independent variable) is linear. The F-tests suggest linearity.
Same number of cases and variables: That the number of cases (observations) must be larger than the number of independent variables. In this study, we had 40 observations and one independent variable.
The descriptive statistics for the entire survey indicated that the mean and the standard deviation of the independent variable, style (leadership behaviour of CEO concern for employees) were 3.9 and 0.70, respectively. The regression results of the entire survey were:
ROTA = -0.13 + 0.52 S (-1.974) (3.797)* |
(1) |
| R2(%) | = | 27.5 |
| F | = | 14.412* |
Equation 1 means that Style (S) had a positive and significant relationship with ROTA at p<0.05. Style (S) accounted for 27.5% of the variation in ROTA. The F-value indicates that Style (S) explained a significant amount of the variation in ROTA at p<0.05.
In order to better appreciate the relationship under investigation, we divided the entire sample into two groups: finance and non-finance and ran separate regressions for them. Finance group included banking and insurance industries, while non-finance group included industrial/domestic products, food/beverages and tobacco, healthcare and packaging industries. The two groups consisted of 20 companies each.
Finance group regression results: The descriptive statistics of the finance group were mean 3.73 and standard deviation 0.45. Considering the 5 point scale used the mean value was moderate. We present below the regression results of finance group.
ROTA = -0.02 + 0.21 S (-0.169) (0.890) |
(2) |
| R2(%) | = | 4.2 |
| F | = | 0.792 |
Equation 2 means that Style (S) had a positive but not significant relationship with ROTA at p<0.05. Style (S) accounted for 4.2% of the variation in ROTA. The F-statistic denotes that Style (S) did not explain a significant amount of the variation in ROTA at p<0.05.
Non-finance group regression results: The descriptive statistics of the non-finance group were mean 4.08 and standard deviation 0.87. Considering the 5 point scale used the mean value was high.
We present below regression results of non-finance group.
ROTA = -0.11 + 0.58S (-1305) (3.047)* |
(3) |
| R2(%) | = | 34 |
| F | = | 9.287* |
Equation 3 means that Style (S) had a positive and significant relationship with ROTA at p<0.05. Style (S) accounted for 34% of the variation in ROTA. The F-statistic suggests that Style (S) explained a significant amount of the variation in ROTA at p<0.05.
DISCUSSION
It was hypothesized that leadership behaviour of the Chief Executive officers is positively related to organizational performance.
Leadership behaviour of the Chief Executive Officers (S) was found to be positively related to organizational performance. It had positive relationship at the group levels and at the entire survey level. While the variable was found significant at p<0.05 in non-finance group and at the entire survey level, it was not significant in the finance group.
The positive relationship between leadership behaviour of the Chief Executive Officers (S) and organizational performance was as expected by literature (Pascale and Athos, 1981; Peters and Waterman, 1982).
The mean values for S ranged from 3.73 to 4.08. These are indeed strong values which indicate strong affirmation of the components of leadership behaviour of the Chief Executive Officers (S). Consequently, leadership behaviour of the Chief Executive Officers (S) had more concern for employees than task. Leadership allowed employees to participate (Likert, 1961; Dowling, 1975), valued and accepted the opinions of others, avoided conflicts, kept people together, maintained friendly relationships and assisted when required.
The positive but not significant relationship of leadership behaviour of the Chief Executive Officers (S) with organizational performance in the finance group was not entirely unexpected. Iyayi et al. (2005) found that employees who moved from one bank to another considered their former banks as having too much concern for task. The mean score of 3.73 indicates that the finance group merely crossed the border line into concern for employees. This again was confirmed by Iyayi et al. (2005) who found that employees, who moved, perceived their new organizations as having less work related concern and more employee related concerns than their former organizations.
In the non-finance group, leadership behaviour of the Chief Executive Officers (S) had positive and significant relationship with organizational performance. The mean of leadership behaviour of the Chief Executive Officers (S) was 4.08. This means that this group had more concern for employees than the finance group.
POLICY IMPLICATION
Leadership behaviour of the chief executive and indeed of all cadres of management, should emphasize and enhance employee concern. When employees are certain that their leaders empathize with them, they are more ready and willing to develop organizational citizenship behaviour that enhances organizational effectiveness.
CONCLUSION
This study examined the relationship between style and organizational performance. It established, empirically, that leadership behaviour of the Chief Executive Officers is positively related to organizational performance in Nigerian quoted companies.
Consequently, it was recommended that leadership behaviour of chief executive officers and indeed the entire management should be employee concerned.
APPENDIX 1
Style questionnaire
Kindly tick on one of the multiple choice answers that agrees most
with how you would likely act as the leader of a work group. The response
categories are: Never (1) Seldom (2) occasionally (3) Frequently (4) Always
(5).