FACULTY OF BUSINESS AND ECONOMICS AN ANALYSIS OF DIFFERENT KINDS OF ORGANISATIONAL CULTURE M.C.Knowles, Prasuna Reddy& Kinga Konczey Working Paper 25/02 November 2002 ISSN 1327–5216 Abstract This study analysed the different types of organisational cultures in which a sample of 370 managers were working. The four types of organisational culture were classified as power, role, task and person. The most prevalent organisational cultures in which these managers believed they worked were the role, power and task cultures, with role culture being the most prominent. When asked what type of organisational culture they preferred to work in, respondents’ ratings favoured task culture over power, person and role cultures. These preferences particularly applied to respondents who were older in age, had higher levels of salary, and greater years of work experience. Gender differences were also found on ideal culture ratings with women showing greater preference for person cultures. Furthermore, when ratings on current and ideal cultures were compared, respondents rated their organisations as having greater power or role cultures than desired, and less task or person cultures than desired. The implications of these kinds of discrepancies are discussed with respect to leadership and management practice. This paper is a work in progress. Material in the paper cannot be used without permission of the author. DEPARTMENT OF MANAGEMENT AN ANALYSIS OF DIFFERENT KINDS OF ORGANISATIONAL CULTURE It has only been relatively recently that organisational culture has become a focus of contemporary research. This has been stimulated by two main factors. One is the realisation that leadership and culture are closely interrelated and thus two sides of the one coin (Schein, 1984; Schein, 1992). This is especially important in the area of leadership development for if leadership is to be enhanced from a practical point of view then attention also needs to be directed simultaneously at modifying organisational culture. The second factor stems from research into change management where again it has been realised that, if organisations are to introduce programs of planned change successfully, in one way or another there has to be a concomitant change in organisational culture (Lorsch, 1986; Ogbonna, 1993; Schneider, Arthur & Richard, 1996; Hupfield, 1997). One limitation of previous research, however, is that organisational culture has tended to be regarded in a general sense and too little attention has been directed at differentiating between different kinds of cultures. An exception to this lies in the work of Harrison (1972) who developed a questionnaire to distinguish between four types of organisational culture – power, role, task and person. Power cultures are those characterised by struggles for superiority of the strong over the weak. This tends to produce business strategies aimed at increasing size and influence and thus domination of the market. Role cultures are those in which behaviour is governed by rules, regulations and legitimacy. Priority is given to developing appropriate policies and procedures, and thus emphasis is placed upon means rather than ends. Task cultures are those that focus upon the mission of the organisation. This engenders a strong sense of purpose in its members which tends to over-ride all other considerations. Thus priority is given to ends rather than means. Person cultures are those that attract people who desire to work in congenial environments in which a premium is placed upon interest in the job itself as well as personal development. This tends to foster the creation of the cult of the individual. AIMS The purpose of the present study was to use the Harrison questionnaire in order to pursue three objectives. Since, as has already been stated, the literature to date deals with organisational culture in a generic sense, the primary objective of the current study was to determine what kinds of cultures managers in contemporary organisations work in, albeit power, role, task or person. In other words, which of these cultures are the most prevalent and which the least prevalent? In addition to identifying the kind of culture in which managers currently worked the Harrison questionnaire was also able to determine the kind of culture in which managers preferred to work. This was the second objective of the present study. The third objective was to ascertain the degree of correspondence between the managers’ current and preferred organisational cultures. A high degree of correspondence would indicate a close fit between the managers’ current and preferred cultures and thus a relatively satisfactory state of affairs. A low degree of correspondence would indicate a major gap between their current and preferred cultures, a manifestly unsatisfactory state of affairs. METHOD Participants The sample comprised 370 managers (135 women, 203 men, 32 unspecified) who were all undertaking a Master of Business Administration degree in an Australian university. Their ages ranged from 20 to 50 years (mean 28 years). Forty per cent of the sample were born in Australia or New Zealand; 38 per cent were born 2 in Asia, 8 per cent were born in Europe, and the remainder were born in North America or Africa (less than 2 per cent each), or did not specify their country of birth. Materials The questionnaire included demographic questions and a measure of organisational culture developed by Harrison (1972). This instrument contains 15 scales that represent different aspects of organisational culture, such as communication, coordination and control, decision making and conflict. Each of the scales has four statements and respondents are asked to rank the statements in degree of relevance, with a rank of 1 indicating most relevant. Respondents rank the statements first in terms of how the statements relate to the organisation in which they currently work, then in terms of the kind of organisation in which they would prefer to work. Each of the statements corresponds to one of the four primary cultures of power, role, task, and person, so that lower scores on the statements indicate greater relevance. RESULTS Ratings on the four cultural factors (power, role, task, person) were considered first in terms of organisations where respondents currently worked. Table 1 presents descriptive statistics for these ratings by type of work represented in their organisations. Table 1: Ratings on Current Organisational Culture by Type of Work Type of Work General Management/ Human Resources Marketing Accounting/ Finance Technical Other Total n 109 Power M SD 33.74 9.59 Role M SD 31.50 6.46 Task M SD 35.56 7.59 Person M SD 49.01 7.96 67 53 52 35 316 31.57 30.08 36.21 33.23 33.02 31.43 30.25 31.06 29.54 30.99 37.04 36.87 32.94 36.57 35.78 49.51 51.34 49.18 50.60 49.71 8.66 7.38 8.16 8.52 8.87 5.46 5.10 5.63 4.93 5.75 7.09 6.63 6.81 6.70 7.20 6.80 6.15 7.17 5.91 7.11 From Table 1, respondents believed that they worked mainly in role (mean range 29 to 31), power (mean range 30 to 36) or task cultures (mean range 33 to 37) rather than person cultures (mean range 49 to 51). A within-subjects analysis of variance test showed significant differences on mean ratings (Greenhouse-Geisser corrected) for the four cultures, F (2, 808) = 385.43, p<.001. Paired comparisons of means indicated role culture to be more prominent than power, task, and person cultures; power culture more prominent than task and person cultures; and task culture more prominent than person culture. Mean differences on ratings of organisational culture in descending order, with the more prominent culture shown first, were: role : person (18.22, t (398) = 30.30, p<.001); power : person (16.38, t (398) = 22.91, p<.001); task : person (13.58, t (398) = 31.30, p<.001); role : task (4.65, t (398) = 8.77, p<.001); power : task (2.81, t (398) = 3.78, p<.001); and role : power (1.87, t (399) = 3.91, p<.001). Multivariate analysis of variance tests found that ratings on the organisational cultural factors did not differ by respondents’ gender, F (4, 325) = 1.43, p=.225. However, ratings on power and task cultural factors did show differences according to the type of managerial work. Respondents engaged in Technical work tended to rate their organisations more highly on task culture, F (3, 276) = 3.79, p=.011, and lower on power culture, F (3,276) = 5.08, p=.002, than respondents engaged in Marketing or Accounting/Finance types of managerial work. Participants also ranked cultural factors in terms of ideal organisational cultures. Ideal ratings on the four different kinds of organisational culture are shown in Table 2. 3 Table 2: Ratings on Ideal Organisational Culture by Type of Work Type of Work General Management/ Human Resources Marketing Accounting/ Finance Technical Other Total N 105 Power M SD 49.24 7.62 Role M SD 39.80 5.85 Task Person M SD M SD 24.65 6.13 36.10 8.11 67 53 51 36 312 48.72 47.68 47.94 50.78 48.83 39.18 38.34 38.88 40.33 39.33 25.67 24.87 26.37 25.47 25.28 7.47 7.45 8.76 4.74 7.50 5.35 5.42 6.22 5.07 5.65 5.31 5.33 7.56 5.41 6.01 35.64 36.62 36.80 33.39 35.89 7.75 7.96 7.95 7.74 7.95 Respondents tended to rate ideal cultures in terms of task (mean range 25 to 26), person (mean range 33 to 37) or role (mean range 38 to 40) rather than power (mean range 48 to 51). A within-subjects analysis of variance test found significant differences on mean ratings (Greenhouse-Geisser corrected) for the four ideal cultures, F (2, 859) = 580.72, p<.001. Paired comparisons of means of the ideal cultures indicated task culture was preferred over power, role, and person cultures; person culture preferred over power and role cultures; and role culture preferred over power culture. Mean differences on ratings of ideal organisational culture in descending order, with the preferred culture shown first, were: task: power (23.11, t (396) = 36.64, p<.001); task : role (13.69, t (396) = 29.41, p<.001); person : power (12.73, t (396) = 18.49, p<.001); task : person (10.38, t (396) = 21.14, p<.001); role : power (9.43, t (396) = 23.16, p<.001); and person : role (3.30, t (396) = 5.33, p<.001). The results of multivariate analysis of variance tests showed no differences on mean scores for any of the ideal cultures by type of managerial work, F (4, 271) = 1.78, p=.133. However, ideal culture ratings did differ by respondents’ gender, F (4, 324) = 3.27, p<.01. Univariate tests showed significant differences between men and women on ideal ratings of person cultures, F (1, 327) = 8.76, p<.01, with women giving greater preference for person cultures (mean=34.40) compared to men’s ratings on person cultures (mean=37.05). Relationships between current and ideal cultural ratings and other characteristics of respondents were examined by correlational analysis. These results are summarised in Table 3. Table 3: Correlations Between Ratings on Current and Ideal Organisational Cultures and Characteristics of Respondents Respondent Characteristics Age Salary Years of Work Experience *p<.01 Power Current Ideal .08 .11 .10 .09 .07 .08 Role Current Ideal .10 .15 .06 .10 .11 .09 Task Current Ideal -.12 -.14 -.13 -.22* -.20* -.20* Person Current Ideal .02 -.04 .02 .06 .10 .06 As shown in Table 3, ratings of current organisational culture were not related to respondents’ age, salary levels or years of work experience. Apart from task culture, none of the ratings of ideal cultural factors were related to these respondent characteristics. However, greater preference was given to ideal task culture by respondents who were older in age, had higher levels of salary, and greater years of work experience. Paired t-tests were used to compare ratings of current organisational cultural factors and ideal or desired cultural factors. Mean scores for current and ideal ratings are shown in Table 4. 4 Table 4: Ratings on Current and Ideal Organisational Cultures Current Ideal N 356 356 Power M SD 32.86 8.73 48.54 7.72 Role M SD 31.10 5.75 39.20 5.66 Task Person M SD M SD 36.04 7.06 49.40 7.39 25.52 6.27 36.08 8.10 Significant differences were found between current and ideal ratings for all four types of cultures. Organisations were rated as having greater power culture than desired power culture, t (355) = -25.51, p<.001, and greater role culture than desired role culture, t (355) = -20.07, p<.001. In contrast, organisations were rated as having less task culture than desired task culture, t (355) = 21.49, p<.001, and less person culture than desired person culture, t (355) = 23.20, p<.001. DISCUSSION In keeping with the first objective of this research, the overall results of the study indicate that wide differences exist between the kinds of organisational cultures in which managers work. Given that much of the literature has tended to refer to organisational culture in a general or generic sense, these findings suggest a need to consider culture in a more detailed manner. As the present research clearly illustrates, people work in very specific organisational cultures. Of these, the most prevalent culture by far is role, followed next by power and task cultures. Few managers see themselves as working in person cultures. These findings may have important implications for that part of the literature that deals particularly with change management which is replete with examples of programs of planned change that have been unsuccessful. As the research so frequently reports, some 80 per cent of all change programs fail (Beer, Eisenstat & Spector, 1990; Kotter, 1995). One reason for this perhaps is that these programs tend to treat all organisational cultures in the same way and not cater for the considerable differences between them as found in the present research. It is also to be noted that managers engaged in technical work, tended to rate their organisations more highly on task culture and lower on power culture than managers engaged in other types of work, notably marketing and accounting/finance. These findings provide further evidence for challenging the notion of a generic culture within organisations. Thus, change programs need to consider not only the nature of the organisation but also the possibility of sub-cultures as a function of the type of work within the organisation. A second important result of the present study concerned the type of culture in which these managers preferred to work; in other words, their ideal culture. In order, these cultures were found to be task, person and role, with power last. Significant differences were also found in the preference order for these ideal cultures. Such strength of preference cut across all types of work including general management, human resources, marketing, accounting and finance, and technical, indicating the pervasive nature of these preferences. There were, however, some respondent characteristics that related to preferences for ideal cultures. The first of these included age, salary and previous work experience with task cultures being preferred by those who were older, had higher levels of salary, and a greater number of years of work experience. Presumably this reflects the yearning of these respondents to take on challenging tasks and prove their worth, consistent with their enrolling in their present MBA program. The second characteristic was gender, with women, compared to men, showing a clear preference for person cultures. This is a salient finding for it changes the focus from questions such as whether or not women and men in executive positions follow the same route in climbing the corporate ladder (Lyness and Thomson, 2000), or how women deal with broader gender issues such as the glass ceiling (Kotlis, 1993), to the more basic question of the type of culture characterising the organisation in which women choose to work. In other words there are vocational implications for the types of organisations in which women would want to pursue their careers as well as the way these organisations are presently managed or led. 5 Notwithstanding the importance of the results just described, perhaps the most remarkable finding of the present research was revealed when the ratings on current and ideal culture were compared and substantial differences were found on all aspects of organisational culture. The findings showed that both power and role cultures were rated as being more prevalent than desired, and task and person cultures were less prevalent than desired. In many respects the extent of the differences between current and desired cultures is surprising and the magnitude of such differences has profound implications for the organisations in which the present sample of managers work. One implication is that there may have been a poor employment match between these managers and their current organisations. If this were the case it would indicate that these organisations should put more emphasis upon improving their human resource management practices including selection, induction or training. This is something that could be immediately followed up and checked upon. It is difficult, however, to think that such limitations could exist on such a large scale. Thus the second explanation is perhaps more plausible and this is that leadership in these organisations could be behind the times. This proposition arises out of research that has been referred to previously that stresses the fact that leadership and culture are intimately interrelated (Schein, 1992). If managers desire to work in a different kind of organisational culture than the one they do at present, then what they want at the same time is for the style of leadership to be different to what currently exists in their organisations. In a similar vein, a considerable part of the literature over the past decade has constantly argued not only that there is all the world of difference between management and leadership but also that leadership is something over and above management (Zaleznik & Kets de Vries, 1975; Bass, 1985; Conger and Kanungo, 1998). As such it needs to be especially catered for and developed (Avolio, 1999). The results of the current research are also consistent with this view. 6 REFERENCES Avolio, B. (1999). Are leaders born or made? Psychology Today, 32, 5, 18. Bass, B. M. (1985). Leadership and performance beyond expectations. New York: Free Press. Beer, M., Eisenstat, R. A., & Spector, B. (1990). Why change programs don’t produce change. Harvard Business Review, Nov/Dec, 158-166. Conger, J. A., & Kanungo, R. N. (1998). Charismatic leadership in organizations. London: Sage. Harrison, R. (1972). Understanding your organization’s character. Harvard Business Review, May/June, 119-128. Hupfield, S. F. (1997). Restructuring and organisation: management’s role in defining organizational culture. Frontiers of Health Services Management, 13, 4, 40-43. Kotlis, A. P. (1993). Women in management: The ‘glass ceiling’ and how to break it. Women in Management Review, 8, 4, 9-15. Kotter, J. P. (1995). Leading change: why transformation efforts fail. Harvard Business Review, MarchApril, 59-67. Lorsch, J. W. (1986). Managing culture: the invisible barrier to strategic change. California Management Review, 28, 2, 95. Lyness, K. S., & Thompson, D. E. (2000). Climbing the corporate ladder: Do female and male executives follow the same route? Journal of Applied Psychology, 85, 1, 86-101. Ogbonna, E. (1993). Managing organization culture: fantasy or reality. Human Resource Management Journal, 3, 2, 42-54. Schein, E. H. (1984). Coming to a new awareness of organizational culture. Sloan Management Review, 25, 316. Schein, E. H. (1992). Organizational culture and leadership. San Francisco: Jossey-Bass. Schneider, B., Arthur, P. G., & Richard, A. (1996). Creating a climate and culture for sustainable organizational change. Organizational Dynamics, 24, 4, 6-19. Zaleznik, A., & Kets de Vries, M. (1975). Power and the corporate mind. Boston: Houghton Mifflin. 7