RESEARCH AND PRACTICE
IN HUMAN RESOURCE MANAGEMENT

Highlight, copy & paste to cite:

Shakeela, M. (2004). A Measure of Leadership Behaviour: Does the Age – Old Measure Require Redefining?, Research and Practice in Human Resource Management, 12(2), 140-151.

A Measure of Leadership Behaviour: Does the Age – Old Measure Require Redefining?

Mariyam Shakeela

Abstract

Despite the acknowledged importance of managerial leadership, and the continual interest in the concept, the phenomenon continues to attract controversy about conceptual and methodological issues. This research note adds to the debate by assessing independent dimensions of leadership behaviour with an adapted version of an instrument that was designed to capture two dimensions of leadership: (1) initiating structure or task, and (2) consideration or relationship. The results of the study reported in this paper, which were obtained from a sample of 409 managers from the Maldives and Mauritius, reveal that the respondents identified five dimensions of managerial leadership behaviour instead of the two that were expected. These findings are discussed in terms of the consequences and implications for human resource management and development.

Introduction

The influence of globalisation has created a world that faces challenges far beyond the comprehension of yesterday’s leaders. Consequently, contemporary organisations are characterised by such constantly changing dynamics as:

These trends and influences demand institutional flexibility, quality service, flatter organisations, interpersonal skills and innovation. Indeed, “the growing democratisation of Central and Eastern European and Asian countries, the unification of the European common market, and the rapid mobility of labour and capital around the globe” (Cascio 1992: 1) has forced organisations to trim their pyramidical frameworks, and to root out cumbersome and bureaucratic structures (Helgesen 1990). As a result of these transformations, corporations are being compelled to adopt strategies and cultures that demand new mindsets, as well as incumbents with skills and ability to operate in fast traumatic work settings of discontinuity (Limerick 1992). In fact, over the years organisational cultures of collaborative individualism have replaced property and capital as the source of power and profit (Phillips 1993).

Contemporary organisations are thus taking steps to adapt their ways and are calling for leadership styles that would manage these challenges and yet achieve the holistic organisational goals. Consequently, the strategic management of human resources is being recognised increasingly as the ultimate source of corporate adaptability and competitive power. Now, more than ever, responsive leadership is seen as a critical factor in organisational success. Therefore, to be effective, leaders need to adapt their styles to fit a broad range of individual and team situations.

This raises the need to study leadership behaviour as the leaders face these challenges and cope with the emphasis on people rather than products. Recognising this need, especially in some parts of the world such as the Maldives and Mauritius where there is no evidence that contemporary leadership behaviours have been analysed, a study was conducted on two leadership dimensions: initiating structure or task, and consideration or relationship, by using an instrument that can be traced to Fleishman (1953). The findings obtained in the study and reported in this paper when managerial responses to a questionnaire were factor analysed, were interesting and thought provoking. These study findings have the potential to contribute towards human resource management policies.

Method

Sample

Respondents for this study were 409 managers and leaders from various organisations, different backgrounds and managerial levels working in the Maldives and Mauritius. The reason for choosing these two countries was both for convenience (because the researcher has direct contact with the chosen countries and hence, found it comparatively easy for data collection), but more importantly because little, if at all is known about the leadership style of these countries. Indeed there is scant evidence that any research of this nature has been conducted in either of the two chosen countries.

In spite of the sample from Maldives containing a higher proportion of female managers, the overall managerial participation rate indicates the changing work roles of women in these contemporary societies. An important feature of the Mauritius sample was the age attribute which can capture greater personal experience and background knowledge. Consequently, nearly 80 per cent of the Mauritius sample was employed at the executive managerial level. Clearly, education is a valued dimension of managerial contexts, an aspect which is demonstrated by the number of managers who held university qualifications. Expatriate managers are not a majority of the examined cadres, but they are an important proportion of the study sample, which exposes the mobility of this cohort. Indeed over 50 per cent of the expatriate managers have been employed in organisations of the study site for over five years, and consequently, they strongly underpin the managerial cadres. Table 1 presents some interesting insights into the profile of the study sample.

Table 1
Demographic Profile Percentages
MALDIVES
(n=208)
MAURITIUS
(n=201)
TOTAL
(n=409)
Gender
Males 56.7 74.1 65.3
Females 43.3 25.9 34.7
Age
Under 20 1.5 2.0 1.7
20-30 32.2 6.5 19.6
30-40 35.6 24.9 30.3
40-50 26.4 32.3 29.3
Over 50 4.3 34.3 19.1
Managerial level
Supervisor 17.3 5.5 11.5
Middle 38.5 16.4 27.6
Executive 44.2 78.1 60.9
Education
High School 19.7 10.0 14.9
Trade/Vocational Certificate 34.6 17.4 26.2
University 45.7 72.6 58.9
Expatriate
Yes 26.0 35.8 30.8
No 74.0 64.2 69.2
Duration
Under 5 years 50.0 35.3 41.8
Over 5 years 50.0 64.7 58.2

Note: Values in parentheses are the number of respondents.

Procedure

The research strategy had three objectives and was designed to follow a pluralist approach. Nevertheless, in this paper there is a greater emphasis on quantitative methods rather than the qualitative dimensions. The first objective was to obtain data with a good balance across respondent backgrounds. For example, gender, age, managerial level and formal education were examined from data where there were adequate sample numbers in partitioned categories. A second pertinent feature of the study is that it was conducted on two island nations in the Indian Ocean; Maldives and Mauritius. This is a significant aspect of the study as this is the first time it has been reported that a comprehensive managerial related analysis has ever been undertaken in these countries. The third feature of the study design is that it was conducted as a cross-cultural analysis. Utmost care was taken to provide clear definitions of the requests and to keep the questionnaire as comprehensible, yet as succinct as possible while not compromising the quality and the validity of the questions asked. An overriding consideration in designing the research strategy and the instruments was also to enhance the determination of a robust data set that had relevance for the disciplines being examined and to ensure that the chosen sample was representative of the two nations.

Common techniques of quantitative research uses questionnaires, structured interviews and tests (Palomba & Banta 1999). The main data collection for this project was carried out through questionnaires. The questionnaire was attached with a cover letter which explained the purpose of the study and also carried full instructions emphasising anonymity and voluntary participation of the participants. All the questionnaires were administered with the support of sponsors who approved the conducting of the study in their region.

Prior to actual data collection, a pilot study was conducted using 50 questionnaires. They were distributed to the chosen participants of the pilot study which was administered over a period of one week. Since it is important to use carefully constructed instruments that have been tested for validity and reliability (Patton 1990), the initial data from the pilot study were thoroughly tested. On completion of the administration of the questionnaire, discussions were then held and interviews were carried out with the respondents, to identify any nomenclature problems that the participants may have encountered while answering the questionnaire. These interviews determined their views regarding understanding of questionnaire items in terms of clarity, sensitivity to cultural nuances and relevance to the industry. This was a vital element of the qualitative component of the study design. After careful assessment (quantitative), and particularly of the verbal responses offered, the questionnaire was modified to enhance better understanding and to accommodate the cultural sensitivities of the study participants.

Final collection of the data was carried out by sending 250 questionnaires to each of the sponsors of the respective countries. After the questionnaires were analysed, discussions were held with the focus groups. This was carried out to gain qualitative data which is not available on surveys alone. The resulting discussions with the focus groups elicited ideas, opinions and experiences which prompted them to respond to the questions. Hence, these discussions enhanced better understanding of the practical aspects of the answers, especially in cases where the answers differed from the expected response.

The instrument that was used to gather the data for leadership dimensions of initiating structure or task and consideration or relationship, and the explanation of the instrument is as follows.

Measure

The degree of initiating structure or task oriented style of leadership, and consideration or relationship oriented style leadership was assessed with a 15 item scale that was developed using American and Western concepts, and values. This instrument was adapted from a sub scale of the Leadership Opinion Questionnaire (LOQ) derived from the same research program as the Leadership Behaviour Description Questionnaire (LBDQ X 11, Stogdill 1963, Cook, Hepworth, Wall & Warr 1981). This instrument (LOQ) measures how leaders believe they behave in leadership roles. The origin of the instrument can be traced to Fleishman (1953). This instrument was used because of its popularity in assessing leadership behaviour. Frequent use in recent empirical research has accumulated a wealth of information about the scale psychometric properties and correlates (Cook et al. 1981). Respondents addressed each of the 15 items in terms of a five-point scale (1 = never, 2 = seldom, 3 = occasionally, 4 = frequently, 5 = always). Questions, 1, 2, 3, 4, 5, 6, and 7 measured relationship style and questions 8, 9, 10, 11, 12, 13, 14 and 15 measured task style.

The factor analysis produced unexpected and interesting results, highlighting that this instrument has unexplored dimensions.

Results

The study data from the 409 Maldives and Mauritius managers were assessed using SPSS and employing component analysis and the Varimax option with Kaiser Normalisation. Multivariate factor analysis was employed to determine construct validity, and Cronbach’s alpha coefficient was used to examine construct reliability.

The leadership scale was examined for its dimensionality. Table 2 shows the factor scores, eigenvalues, percentage of variance and the cumulative percentage of variance explained when the study data were evaluated by employing principal components method of data reduction. Eigenvalues greater than one were retained. In addition, reliabilities of the leadership constructs are also presented at the bottom of Table 2.

The reliabilities for each study construct were estimated using the Cronbach alpha coefficient (Cronbach 1951). Seminal work by Nunnally (1978) has often been the foundation of estimating the reliability of multi item constructs (scales). Indeed, on his guidance coefficient alpha scores of 0.7 are sought, although lesser values have been accepted. In this study, only two items (i.e., A24 and A16) had factor scores less than 0.50, but they were retained as they contributed positively to the construct reliability assessments. Overall, the reliability assessments are acceptable; given this is the first reported study of this type in these two island nations. There is every likelihood that the ‘novelty and the uniqueness’ of employing self reported survey may have contributed to reliabilities less than 0.70 values.

Two factors were expected, but five constructs were obtained. Forcing the data into a two-factor solution did not result in a meaningful configuration of the questionnaire items, and consequently, exceptionally low reliabilities were reported. Clearly, the respondents did not perceive two independent leadership behaviours in their workplaces.

Table 2
Principle Components Factor Analysis for Leadership
Item Descriptions Factors
1 2 3 4 5
Eigenvalues 2.841 1.996 1.474 1.120 1.043
Percentage of variance explained 18.941 13.308 9.829 7.468 6.951
Cumulative percentage of variance explained 18.941 32.249 42.078 49.545 56.496
Reliabilities 0.673 0.544 0.594 0.472 0.463
Task
A23 I push the staff for greater effort 0.750 - 0.06 0.268 0.104 0.011
A26 I emphasise meeting deadlines 0.804 0.083 0.064 - 0.015 - 0.049
A28 I see to it that the staff are working up to capacity 0.663 0.006 - 0.11 0.277 - 0.208
Supot
A15 I help my staff with their personal problems - 0.074 0.693 - 0.104 0.095 0.072
A18 I stand up for my staff even though it makes me unpopular with others 0.033 0.760 0.098 - 0.131 0.076
A19 I criticise a specific act rather than a particular member of the staff 0.06 0.629 - 0.078 - 0.07 - 0.026
Autho
A20 I ask for more than the staff can get done 0.214 - 0.051 0.777 0.034 - 0.156
A21 I rule with an iron hand 0.008 - 0.04 0.762 0.153 0.027
Cntrl
A14 I speak in a manner not to be questioned 0.086 0.035 0.247 0.653 - 0.178
A17 I refuse to explain my actions - 0.285 - 0.224 0.417 0.527 0.012
A24 I put the welfare of the department/section above the welfare of the staff 0.132 - 0.185 0.268 0.416 0.140
A27 I decide in detail what shall be done and how it shall be done by my staff 0.333 0.028 - 0.26 0.696 - 0.079
Rela
A16 I get the approval of the staff on important matters before going ahead 0.069 0.434 - 0.218 - 0.074 0.469
A22 I wait for my staff to push new ideas - 0.047 - 0.108 - 0.054 0.036 0.799
A25 I let others do their work the way they think is the best - 0.212 0.235 0.055 - 0.154 0.653

Notes: a. Task = task style behaviour; Supot = support, Autho = Authority, Cntrl = Control and Rela = Relationship.
b. Sample Size N = 409.
c. Extraction method: Principal Component Analysis, Rotation method: Varimax with Kaiser Normalisation a rotation converged in 3 iterations.

Discussion

Leadership styles of initiating structure or task, and consideration or relationship were studied using an instrument that was adapted from the Leadership Opinion Questionnaire (LOQ). It was administered to 409 managers from various service industries in Maldives and Mauritius. A majority of the managers held tertiary qualifications and there was a reasonable gender balance that is representative of the study countries. This instrument was developed in the 1950s and its appropriateness in these two island nations was a feature of the study findings. The applicability of Western theories to other cultures have been questioned by researchers (James 1997, Clark 1998, Gopinath 1998), and in fact, this was an actual concern in this study.

The findings from this research provided enlightenment on the used measure. First, this study establishes that the leadership instrument used for the study of espoused leadership behaviour, is applicable even in some non-Western environments. It also establishes that the instrument has the potential to measure more than two dimensions of leadership. In fact, it appears that the employed scale measures five independent dimensions of managerial leadership behaviour. This unexpected finding may have many explanations. One possibility is that this instrument was developed in the 1950s and managers of that time were comparatively less educated and knowledgeable than the managers of today. Another possible explanation is that today’s managers are empowered and are in a world that is fluid and in constant turmoil surrounded by demands far beyond what managers faced in the 1950s. The reality faced then was different, and hence what may have been conceived by the educated managers of today (i.e., in the period of this study) may be significantly different from yesterday’s managers’ comprehension of the instrument. It is also possible that the instrument was developed for a different level of managers and for a narrower leadership scope. This raises the possibility that this instrument may have produced a different set of results, if limited to lower level managers.

Conclusion

One of the most salient findings was that the leadership scale was extremely versatile as it was able to capture more than the two dimensions of leadership for which it had been designed. This is despite the study respondents being mostly educated senior and executive managers, who are expected to be ‘thinkers’ and not just administrative managers. Focus group discussions revealed that in the study contemporary work settings, managers/leaders are compelled to employ a variety of behaviours that go far beyond the traditional, widely acknowledged activities of initiating structure or task, and consideration or relationship. The context of leadership today is very different from what it was when this instrument was designed for measuring initiating structure or task, and consideration or relationship. Consequently, the perception of the study respondents was that leadership by educated and learned executives is not just the two dimensions that were examined. But keeping in line with conflicting demands of their jobs, their style is a mix of many dimensions that is not necessarily fixed or limited to just two types, but a bundle of behaviours that are dynamic, versatile, and in some situations need to be extremely sensitive.

In this context, it is debatable that this instrument developed in a Western context (and in the 1950s) is applicable in contemporary work settings. However, the fact that it measured five independent dimensions may imply that this instrument is still applicable, but that the scope needs widening, thus further refining the scale to incorporate the trends and influences that drive managerial/leadership behaviours. Nevertheless, finding five independent leadership behaviours has a salient implication in the development and management of human resources. It is presumed that this information is beneficial to staff development and training within organisations in terms of raising awareness and building corporate human resources that will effectively contribute to the success of the business. The findings will thus encourage consensus, cooperation and collaboration rather than competition and conflict and assist overall skill development for human resources. It is likely that these findings may also help shape human resource policies and contribute towards building a more positive and responsive organisational culture to meet the needs of learning organisations and help develop leadership for the new millennium.

Author

Mariyam Shakeela earned her Masters degree in Management at Curtin University of Technology. As an executive for over 10 years, in one of the leading and most diversified companies in Maldives, she has been exposed to many managerial and leadership roles both within and outside of Maldives. Her experience and observations as a professional in today’s dynamic, workplace prompted her interest in research on managerial and leadership behaviours.

E-mail: Shakeela@simdi.com

References

Cascio, W.F. (1992). International human resource management issues for the 1990s. Asia Pacific Journal of Human Resources, 30(4), 1-18.

Clarke, C. (1998). Reconciling Western management ideas with far Eastern realities. Long Range Planning, 31(4), 523-529.

Cook, J.D., Hepworth, S.J., Wall,T.D., & Warr, P.B. (1981). The experience of work: a compendium and review of 249 measures and their use. London: Academic Press, Harcourt Brace Jovanovich Publishers.

Cronbach, L.J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16, 297-334.

Fleishman, E.A. (1953). Leadership climate and human relations training. Personnel Psychology, 6, 205-222.

Gopinath, C. (1998). Alternative approaches to indigenous management in India. Management International Review, 38(3), 257-276.

Helgesen, S. (1990). The female advantage: Women’s ways of leadership. New York, N.Y.: Currency and Doubleday.

James, M. (1997). A conceptual framework for aligning managerial behaviors with cultural work values. International Journal of Commerce & Management, 7, 81-102.

Limerick, D. (1992). The shape of the new organisation: Implications for human resource management. Asia Pacific Journal of Human Resources, 30(1), 38-52.

Nunnally, J.C. (1978). Psychometric theory. (2nd ed.). New York, N.Y.: McGraw-Hill.

Palomba, C. A., & Banta, T. W. (1999). Assessment essential. San Francisco: Jossey-Bass.

Patton, M.Q. (1990). Qualitative research and evaluation methods. Thousand Oaks, California: Sage.

Phillips, N. (1993). Innovative management. London: Pitman.

Stogdill, R.M. (1963). Manual for the leader behaviour description questionnaire-form X11. Ohio: Bureau of Business Research, Ohio University.