10 Apr Qualitative versus Quantitative Analysis for Management Science
by Ralph H. Kilmann and Ian I. Mitroff
This article was originally published in Interfaces, Vol. 6, No. 2 (1976), pages 17-27.
This paper neither presents the kind of data nor the kinds of symbols that are found in the typical paper in Management Science. It deals with an important class of variables that have been slighted in the literature of OR/MS: qualitative variables. This paper presents a methodology for gathering and for analyzing a special class of qualitative variables, i.e., the projective, symbolic images that managers have of their ideal organization. More specifically, this paper shows how personality variables impact on (1) the raw images that managers have of their ideal organization and (2) the kinds of information sources and methodologies that managers and scientists most typically prefer to use in complex decision-making situations.
This paper attempts to show that certain phenomena which are fundamental to the study of organization design and decision-making, and which may be currently thought of as inherently incapable of quantification can be partially quantified. More to the point, this paper attempts to argue that even if these phenomena could never be completely quantified, this does not lessen their importance. Even further, it does not prove that they cannot be studied systematically and with increasing degrees of both sophistication and precision. The general thesis of this paper is that in its relentless pursuit of the ideal of quantification, OR/MS have failed to develop their ability to treat important qualitative variables in a manner that would do adequate justice to them. In a word, OR/MS have failed to develop one of their most important aspects – their qualitative side – their ability not just to treat and to tolerate, but even more, to appreciate and to understand qualitative phenomena .
We have been investigating a conceptual framework that appears to highlight not only the important differences between qualitative and quantitative analysis, but the relationship between the two, and in fact, how the two even oppose one another at times. This framework is C. G. Jung’s personality typology , , , . The Jungian typology was chosen for two main reasons: (1) the dimensions of the Jungian typology can be directly related to different managerial and organizational styles; as a result, the system helps to shed light on a wide variety of organizational and managerial phenomena; (2) the Jungian typology does not prescribe one of the four major personality types as superior or better than any of the others but instead points out that each type has major strengths as well as weaknesses . For the purposes of our study, two particular dimensions of the Jungian typology were of special importance. The first dimension corresponds to the kind of “input-data” an individual characteristically prefers to take in from the outside world. The second dimension corresponds to an individual’s preference for the kind of “decision-making process” the individual characteristically brings to bear upon his preferred kind of input-data.
According to Jung, individuals can take in data from the outside world by either sensation or intuition but not by both simultaneously. As a result, individuals tend to develop a preference for one mode of input or the other. Sensation refers to those individuals who typically take in information via the senses, who are most comfortable when attending to the details of any situation, and who prefer concrete, specific facts. In contrast, intuition refers to those individuals who typically take in information by means of their imagination, by seeing the whole – the gestalt – of any situation. These individuals typically prefer the hypothetical possibilities in any situation to the “actual” facts. It should be stressed that all individuals perceive with both these functions at different times. But as Jung argues, individuals tend to develop a preferred way of perceiving, and in fact, cannot apply both types of perception or data-input at the same exact time.
According to Jung, there are two basic ways of reaching a decision: thinking and feeling. Thinking is the process of reaching a decision that is based on impersonal, analytical modes of reasoning. Feeling on the other hand is the process of a reaching of a decision that is based on personalistic, value judgments that may be highly unique to the particular individual. Thus, however one takes in data (either by intuition or sensation) an individual may come to some conclusion about the data either by a logical, impersonal analysis (thinking) or by a subjective, personal process (feeling).
Combining the two data input modes (sensation and intuition) with the two decision making modes (feeling and thinking) in all possible ways results in the following four Jungian personality types:
(1) sensation-thinking (ST),
(2) sensation-feeling (SF),
(3) intuition-thinking (NT), and
(4) intuition-feeling (NF) .
In essence, the four types describe different forms of organization design and decision making  as well as different forms of science . The ST type, one can easily argue, has become the epitome of the industrial revolution, bureaucracy, and rigorous scientific investigations. That is, the ST type emphasizes precision, control, specificity, impersonal (objective) analysis, logical and orderly reasoning, etc. Thus, the ST approach represents quantitative analysis in the extreme because one can clearly specify variables, their inter relationships, and their precise measurement under controlled conditions. In fact, some might argue that the objective of management science as well as other sciences is to eventually conceptualize and/or discover theories, methods and measures such that the phenomena of interest can be so quantified. For some, the extent to which the field has not been so quantified or subject to quantitative analysis is the extent that the field is underdeveloped or is not even to be considered a science !
The Jungian typology, however, specifies three other forms of management science, the SF, NT, and NF. Each of these serves as an alternative approach to investigating phenomena, in particular, management science (organization design and decision making). The NT type stresses conceptual analysis as opposed to precise quantification – because the nature of the phenomena may be such that variables simply cannot be specified precisely enough for quantification. The NT type is thus considered a form of qualitative analysis, striving towards impersonal objectivity and the specification of variables and their interrelationships if only in a conceptual-verbal manner. In the same vein, organization design and decision making can be approached largely as abstract concepts and be described via broad, general categories and typologies. It is interesting to note that at the micro level of analysis (i.e. individuals and small groups) there are more empirical studies than at the macro level of analysis (i.e. organizations and institutions). The latter focus receives more theoretical attention since variables are more difficult to define precisely, to measure, and to control. Also, at the more macro level (e.g. organization design and organization decision making) the greater complexity of the phenomena precludes the specification of straightforward mathematical relationships among variables – the relationships change as fast as they can be specified.
The SF and NF types define a different kind of qualitative analysis. Both of these types rely on strictly subjective and value laden criteria for analyses rather than impersonal rules, logic and principles. The feeling function, as was noted, derives judgments, evaluations, and conclusions by a highly personalistic and unique criteria – unique at times to a single individual. Rather than attempting to find the common theme or character of some set of phenomena, the feeling function strives to generate differences. The type of science that feeling types can be expected to advocate would involve the incorporation of ethics into scientific investigations and thus allowing values to determine what should be investigated (because it is deemed important via one’s values) and how it should be investigated (not violating ethical positions) regardless if the study in question is interesting, feasible, or fundable. Also, a scientist under this approach would not reject his hypothesis simply because a particular set of data tended to refute it because he may still believe in the hypothesis and may thus require more convincing evidence (i.e. his analysis is not detached and impersonal as the typical ST science approach is generally de scribed ). Similarly, the study of organization design and decision making can be approached in an SF or NF manner by designing structures and processes that are unique, personalistic and based on different values than precision, logic and control. Instead, organizational structures and processes can be geared to individual needs (versus organizational goals) and can be designed for flexibility, openness, freedom and to serve humanity (versus to optimize some revenue-cost equation). The major difference between SF and NF is that the former still strives for some precision while the latter can be as abstract and loosely defined as the NT.
Of particular interest is that the NF is the complete opposite of the ST approach and therefore the NF provides the greatest challenge to the approaches to management science which have been dominated by ST. In other words, the ST and NF are hot just different; they actually oppose one another since they are based upon conflicting preferences for information and decision making styles. In the same way, NT and SF are opposite one another. However the NT and SF type are two different forms of qualitativeness and therefore these do not highlight the distinction of quantitative versus qualitative as does the ST-NF conflict.
The remainder of this paper describes a study which was undertaken to provide some empirical support for the arguments just presented. In particular, the study explores whether individuals with different psychological types do in fact have different views of organization design and what constitutes an ideal organization. If this turns out to be the case, then we have a methodology for developing the different forms of qualitative versus quantitative management science – by systematically investigating how ST, NT, SF, and NF individuals engage in the act of inquiry . A forthcoming study will explicitly explore the prevalence of the ST (quantitative) norm of management science by providing managers a choice of four types of information to make a decision on a typical management problem. Also, we plan to investigate whether managers’ psychological types explain the decision choices they make. This would give us a further basis for developing a more comprehensive theory and methodology for studying different qualitative and quantitative approaches to management science.
Before proceeding to the two studies, we should mention that to consistent with the theme being developed in this paper, we have chosen to use an NF-NT form of investigation. The present study, the NF, utilizes story telling to examine managers’ view of their ideal organization – relying only on an “intuitive” content analysis to uncover some of the similarities and differences among the stories of the four types (i.e. not a rigorous statistical method). The forthcoming study will take more of the conventional ST approach by utilizing such analytical-quantitative measures as statistical correlations and analysis of variance. According to the theme of this paper, each of these methodological approaches is appropriate and necessary to investigate complex phenomena.
THE IDEAL ORGANIZATION
The question as to whether different individuals do indeed have fundamentally different qualitative and quantitative notions with respect to what constitutes an ideal organization is not merely an interesting theoretical question. It also has profound design implications for modern corporations. In an attempt to shed light on this question, a procedure was devised for exploring the kinds of words, symbols, and raw projective images that different kinds of managers used in describing their ideal organization.
A short personality test was administered to three different groups of managers for the purpose of determining the personality type of each individual manager. Immediately after the completion of this test, each individual was asked to then write a short story that expressed his or her concept of an ideal organization. After the completion of this task, each individual was placed into one of four different discussion groups depending upon the individual’s personality type; i.e., the particular personality test that was used sorted individuals into one of the four personality types; each discussion group contained those and only those individuals whose type was the same. Each group was asked to organize itself in any manner it so desired, to discuss the stories of each individual, and then to come up with a single group story that best expressed the entire group’s concept of an ideal organization.
Prior to the completion of all exercises, neither the personality test, the purpose of the stories, or the fact that the individuals were placed into the various groups on the basis of the similarity of their personality profiles was explained to the individuals. Since the exercises were a natural part of a class or a workshop session, the individuals were informed that their responses would form the basis of an extensive discussion on the nature of different personalities and organizational behavior. Explicit feedback was later given to each individual and to each discussion group as to what their personality scores and stories meant. As a general rule, interest was high for all the exercises and the rapport between the authors and the individuals was good.
The individuals and the groups were asked to express their concepts of their ideal organization and the form of stories for the deliberate reason that we wanted to tap their raw unconscious, projective images of what the concept of an organization meant to them , , . For this same reason the individuals and discussion groups were also asked to describe their concept of their ideal (as opposed to real) organization although one of the three groups of managers after having first described their ideal organization was asked to describe how their real organization differed from their ideal.
Each group of managers was tested separately. The first group of managers was composed of 25 middle to high level business executives, a number of them presidents of their own medium to large-sized companies in the Pittsburgh area. Each was currently enrolled in an executive MBA program at the University of Pittsburgh. The second and third groups of managers were also composed of about 25 members each. These last two groups, however, were composed of middle level supervisors in the Pennsylvania State Department of public assistance. Whereas there was only one woman in the first group of 25 executives, the ratio of women to men was about three to two for the last two groups.
Two findings in particular are immediately apparent and striking from a content analysis of the individual stories and the stories of the four Jungian discussion groups: (1) there is a remarkable and very strong similarity between the stories of those individuals who have the same personality type (e.g., ST); (2) there is a remarkable and very strong difference between the stories of the four personality types. That is, individuals of the same personality type tend to tell the same kind of story or have the same image of their ideal organization whereas different personality types tend to have very different images. Indeed, the differences between the stories of different types are so strong that one is tempted to say that the “ideal” of one type is the “hell” of the other, and vice versa.
The stories of ST individuals are characterized by an extreme emphasis and concentration on specifics, on factual details. ST types are extremely sensitive to the physical features of their work environment. For example, the stories of ST types display an extreme preoccupation with environments that are neither “too hot” or “too cold” but “just right”. The ideal organization of the ST’s is one that is characterized by complete control, certainty, and specificity. In their ideal organization, everybody knows exactly what his or her job is. There is not uncertainty as to what is expected in all circumstances. Further, ST organizations are impersonal. The emphasis is on work, and work roles, not on the particular individuals who fill the roles. It thus comes as no surprise that the ideal organization of ST’s is authoritarian. There is a single leader at the top and a well-defined hierarchical line of authority that extends from the very top down to all of the lower rungs of the organization. In an ST organization, the individuals exist to serve the goals of the organization, not the organization to serve the goals of the individuals. Finally, the goals of an ST organization are realistic, down-to-earth, limited, and more often than not, narrowly economic.
The stories of NT’s are marked by an extreme emphasis on broad, global issues. In describing their ideal organization, NT’s show an almost complete disdain for specific, detailed facts. NT’s neither specify the detailed work rules, roles, nor lines of authority but instead focus on general concepts and issues. To put it somewhat differently, if the organizational goals of ST’s are concerned with well-defined, precise micro economic issues, then the goals of NT’s are concerned with fuzzy, ill-defined, macro economic issues like “an equitable wage for all workers.” NT organizations are also impersonal like ST organizations. However, where ST’s focus on the details of a specific impersonal organization, NT’s focus on impersonal concepts and theories of organization. For example, they are concerned with concepts of efficiency in the abstract. Likewise, whereas in an ST organization individuals exist to serve the particular organization, in an NT organization individuals exist to serve the intellectual and theoretical concept of the organization in general. In a word, if ST organizations are impersonally realistic, then NT organizations are impersonally conceptual.
The stories of NF’s are also marked by an extreme preoccupation with broad, global themes and issues. NF’s also show an extreme disdain towards getting down to specifics. NF’s are similar to NT’s in that both take a broad view of organizations. However, NF’s differ from NT’s in that where the emphasis of NT’s is on the general theory or theoretical aspects of organizations, the emphasis of NF’s is on the most general personal and human goals of organizations. Thus, NF organizations are concerned with “serving humanity”, e.g., “with making a contribution to mankind.” NF’s differ from both ST’s and NT’s in that for both ST’s and NT’s the individual exists to serve the organization where for NF’s the organization exists to serve the personal and social needs of people. Since in Jungian personality theory the NF type is the extreme opposite of the ST type (as the SF type is the extreme opposite of the NT), it is not surprising to find that the ideal organization of NF’s is the exact opposite of ST’s. Thus, if an ST organization is authoritarian with well-defined rules of behavior, then an NF organization is completely decentralized with no clear lines of authority, with no central leader, and with no fixed prescribed rules of behavior. The stories of NF’s incessantly talk about “flexibility” and “decentralization.” As a matter of fact, many of the stories of NF’s contain diagrams of their ideal organization which show them to be circular or wheel-like in structure rather than hierarchical. NF organizations are also idealistic as opposed to realistic. In essence, NF organizations are the epitome of organic, adaptive institutions .
If the ideal organizations of ST’s and NF’s are extreme opposites, then the organizations of NT’s and SF’s are also extreme opposites. If NT’s are concerned with the general theory of all organizations but not with the details of any particular organization, then SF’s don’t care about theory at all or issues in general. SF’s are instead concerned with the detailed human relations in their particular organization. SF’s are like ST’s in that both are concerned with details and facts. However, SF’s differ from ST’s in that the latter are concerned with detailed work rules and roles whereas the former are concerned with the human qualities of the specific people who fill the roles. SF’s are in this sense similar to NF’s. Both SF’s and NF’s are concerned with the people in the organization. SF’s differ from NF’s in the sense that where NF’s are concerned with people in general, SF’s are concerned with individuals in particular. SF organizations are also realistic as opposed to idealistic. Like ST’s, SF’s are also concerned with the detailed work environment although where for ST’s the environment of concern is physical, for SF’s it is the interpersonal environment that is of concern.
While it is unfortunately beyond the scope of this paper to demonstrate the full implications of the above results, it can be shown that the marked differences in personality illustrated above have tremendous consequences for the design of modern large-scale organizations (cf. ). It certainly can said that the well-known, different types of communication networks introduced by Bavelas ,  are not mere abstract creations that bear no relation to some actual organizational concepts. The different types of networks are the results – the psychic creations if one prefers – of fundamentally different types of personalities. If the chain and wheel networks had never been created, then our results imply that not only would they have to have been invented, but even stronger, that they would have been created by ST and NF types respectively.
Finally, it is important to stress what we believe may be one of the most important outcomes of this kind of research, namely the direct investigation of the relationship between personality variables and different concepts of organization. We are certainly not contending that the relationship has not been studied at all (cf. ). Our contention is instead that it has been studied indirectly, for example, by having different managers tell stories to the Thematic Apperception Test . The relation of the TAT stories to different concepts of organization is indirect at best. In our approach we are interested in the direct relationship between personality variables and a manager’s concept of his or her ideal organization. In concluding this section, the reader is referred to the appendix for some examples of actual stories.
In this paper we have argued for the relevance and appreciation of alternative forms of science and organizations, particularly on the quantitative – qualitative dimension. The Jungian framework has been proposed as a basis for defining these different forms. We can further argue that if the different psychologies of individuals tend to reflect different forms of science and organizations, then management science has to seriously consider explicit investigations of these four forms for the variety of substantive issues in management science – that is, not just the design of ideal organizations, but decision making, leadership, conflict, management information systems, strategic planning, etc. There is also some support for these arguments in the physical sciences  and not just the social sciences or management science. Finally, there is a growing body of literature , , ,  which indicates that the higher one goes in an organization the more that what we have called SF and NF skills are called for. As Mintzberg has so aptly put it:
. . . The manager’s nerve center information is of a special kind. He appears to find it most important to get his information quickly and informally. As a result, he will not hesitate to bypass formal information channels to get it, and he is prepared to deal with a large amount of gossip, hearsay, and opinion which has not yet become substantiated fact [15, p. B-105].
If this is indeed the case, then as much as we have trained business students in ST and NT methods and techniques , we need to train them for SF and NF skills as well – perhaps even more.
In the end the biggest challenge may be whether management science is flexible and adaptive enough to learn how to handle the rich qualitative materials that are abundantly available in everyday management life. It is not clear at all that management practice needs to learn how to adapt itself to management science if it is to survive but rather whether management science needs to learn how to adapt itself to practice if it is to survive .
An ST Ideal Organization
My ideal work organization is one in which the structure is such that one person ultimately sets up the rules and regulations. He should have knowledgeable persons advising him. The information should flow downward and upward with all the persons involved being asked their opinion. The employees should all be judged on the basis of their ability to do their job and if they are not suited for their job then they should either be transferred to a more appropriate job or let go. The building should be such that it is conducive to good work. The equipment should be the best for each job and kept in repair. Each person should receive adequate training for the job. There must be rules and regulations that apply. There should be uniformity in their application. Yet we should be aware of the rare ^situation where we have to take the individual’s side. Suggestions for improvement should always be encouraged and given serious consideration.
An NF Ideal Organization
My idea of an ideal organization is one that has unlimited funds, rules and regulations that are flexible and one that would encompass all of the problems people could have.
This organization would take the place of all other social agencies, private and public. It would have sufficient staff to offer whatever help is necessary to the client including help with money management, hospital and other medical bills, social services, employment counseling and emotional problems. It would have to be a large staff but hopefully it would eliminate all of the duplicating of services that are now in effect. It should be free of political pressures.
It would be a tremendous cost but it should benefit the entire United States. I guess what I’m trying to say is we should have some form of socialized medicine. The term medicine would not strictly related to medical problems. If we could start helping people at the point they have problems and not shift them from one phase to another we would eliminate or contain most of the problems we are faced with today.
For instance a man who is unemployed, whose wife is pregnant, no way to pay doctor bills, other children in the home who come home crying be cause they don’t have what other children have, this man is likely to go out and do something dishonest to try to get what his family needs. In short every one should get what help they need when they need it.
An SF Story: “Utopia in the Business World”
The day had been a particularly harrowing one at the office with more than the normal amount of frustrations with the administration, the workers and even the public. I went home and fell exhausted into bed.
Suddenly I awoke and looked around. Where was I? What was this strange place? Who were these people? At that moment I was approached by a smiling person with hand extended who said, “Welcome to our organization. We are glad to have you with us. My name is…. I will take you around to meet the rest of the staff.”
Everyone I met was very friendly and in the days to come proved to be most helpful. My duties were explained to me quite clearly and thoroughly. The procedure with which I had to work was written in such a way that there was very little chance of misinterpretation.
All of the staff worked quite well with each other with a minimum of disagreements. The separate department heads would meet once a week with the Administrator who would keep them informed of new developments. The department heads would then keep the workers informed. Once a month the Administrator would address the entire staff. There was a free and easy exchange of ideas. There was no CIA atmosphere nor was there always a lot of rumors floating around. No one ever said “I hear by the grapevine.” There was no need to “hear by the grapevine.” Everyone was fully informed as to the opportunities available to him.
A door slammed and suddenly I was transported from the Ideal Organization back to the world from which I came.
An NT Ideal Organization
The ideal organization is one in which the goals are developed in response to the interrelation between environmental and member generated factors. In other words, the goals would be a reaction to a need of the environment as well as a need by the members to satisfy the environmental need. The structure itself will provide its own goals, controls, divisions of labor, motivation and reward structure. Also, free entry and exit by members should be present, providing constant feedback and fresh perceptions of environmental needs. Thus, if environmental needs and/or organizational needs change, the structure will naturally change with these changes.
1. Ackoff, Russell L., The Design of Social Research, Chicago, University of Chicago Press, 1953.
2. Barton, Allen H., and Lazarsfeld, Paul, “Some Functions of Qualitative Analysis in Social Research”, Bobbs-Merrill Reprint Series in the Social Sciences, S-336, pp. 321-361.
3. Bavelas, A., “A Mathematical Model for Group Structures”, Journal of Applied Anthropology, Vol. 7 (1948), pp. 16-30.
4. Bavelas, A., “Communication Patterns in Task-Oriented Groups”, Journal of Acoustical Society of America, Vol. 22 (1950), pp. 725-730.
5. Cyert, Richard M., and MacCrimmon, Kenneth R., “Organization”, in Handbook of Social Psychology, 2nd Edition, edited by Gardner Lindzey and Elliot Aronson, Vol. 1, Addison-Wesley, Reading, Mass., 1968, pp. 568-611.
6. Filstead, William J., Qualitative Methodology, Markham, Chicago, 1970.
7. Jung, C. G., Analytical Psychology, Its Theory and Practice, Vintage Books, New York, 1968.
8. Jung, C. G., Psychological Types, Rutledge and Kegan Paul Ltd., London, 1923.
9. Kaplan, Abraham, The Conduct of Inquiry, Chandler, San Francisco, 1964.
10. Kilmann, Ralph H., “An Organic-Adaptive Organization: The MAPS Method”, Personnel, Vol. 59, No. 3 (1974), pp. 35-47.
11. Kilmann, Ralph H., and Taylor, Vern, “A Contingency Approach to Laboratory Learning: Psychological Types Versus Experiential Norms,” Human Relations, Vol. 27 (1974), pp. 891-909.
12. Marshall, I. N., “The Four Functions: A Conceptual Analysis”, Journal of Analytical Psychology, Vol. 12 (1967), pp. 1-31.
13. Mason, Richard O., and Mitroff, Ian I., “A Program for Research on Management Information Systems”, Management Science, Vol. 19, No. 5 (January, 1973), pp. 475-487.
14. McClelland, David C, Atkinson, J. W., Clark, R. A., and Lowell, E. L., The Achievement Motive, Appleton-Century-Crofts, New York, 1953.
15. Mintzberg, Henry, “Managerial Work: Analysis from Observation”, Management Science, Vol. 18, No. 2 (October, 1971), pp. B-97 – B-110.
16. Mintzberg, Henry, The Nature of Managerial Work, Harper & Row, New York, 1973.
17. Mitroff, Ian I., The Subjective Side of Science: A Philosophical Inquiry into the Psychology of the Apollo Moon Scientist, Elsevier, Amsterdam, 1974.
18. Mitroff, Ian I., Nelson, John, and Mason, Richard O., “On Management Myth – Information Systems”, Management Science, Vol. 21, No. 4 (December, 1974), pp. 371-382.
19. Mogar, Robert E., “Toward a Psychological Theory of Education”, Journal of Humanistic Psychology, Vol. IX, No.1 (Spring, 1969), pp. 17-52.
20. Murray, H. A., Explorations in Personality, Oxford University Press, New York, 1938.
21. Myers, I. B., Myers-Briggs Type Indicator, Educational Testing Survey, Princeton, N.J., 1962.
22. Rosenthal, Robert, Experimenter Effects in Behavioral Research, Appleton, New York, 1966.
23. Grayson, C. J., “Management Science and Business Practice”, Harvard Business Review(July-August, 1973), pp. 41-48.