‘ . , ,....;..::l.r .52. 7.. : .r V: .I .3va ‘Ir. 4......3. :5... NlVERS SITY LIBR ARI l2“: ll \Illlll llllllllHilllllll‘llll 3 1293 0102 This is to certify that the dissertation entitled DEVELOPMENT AND VALIDATION OF AN INSTRUMENT TO ASSESS PREFERENCES TO HOLD COMPETING VALUES presented by Cornelius VanderVeen has been accepted towards fulfillment of the requirements for - Ph D degree in Ed. Admin. 31%} profth‘) Date Nov. 44 1993 MS U is an Affirmative Action/Equal Opportunity Institution 0-12771 LIBRARY Michigan State University ' PLACE Ill RETURN BOX to remove this chookoutirom your record. TO AVOID FINES return on or baton duo duo. DATEDUE \ DATE DUE DATE DUE MSU in An Atflnnativo Action/Equal Opponunity institution Walls-9.1 DEVELOPMENT AND VALIDATION OF AN INSTRUMENT TO ASSESS PREFERENCES TO HOLD COMPETING VALUES BY Cornelius VanderVeen A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requrements for the degree of DOCTOR OF PHILOSOPHY Department of Educational Administration 1993 ABSTRACT DEVELOPMENT AND VALIDATION OF AN INSTRUMENT TO ASSESS PREFERENCES TO HOLD COMPETING VALUES BY Cornelius VanderVeen There is increasing recognition that many persistent problems require, not choosing between mutually exclusive alternatives but, managing interdependent alternatives that often appear to be opposites. Examples are Centralization / Decentralization and Individual / Teamwork. Choosing one to the exclusion of the other will lead to system.dysfunction. The purpose of this study was to develop an instrument to assess respondents’ preferences to simultaneously hold competing values. The survey sought to differentiate respondents who take polar positions from those who are able to see the benefit in both polar opposites. An initial survey was administered. Data analysis guided the process of refining the instrument for a second administration. The preference trait was found to be uni-dimensional rather than multi-dimensional. That is, factor analysis revealed no sub-clusters of related items. No age or gender effects were found. Split-half reliability was acceptable. The study concludes that the trait in question can be validly and reliable assessed and that such assessment has important implications for adult learners and for educators of adults. i ACKNOWLEDGEMENTS I am grateful for the long term support and encouragement provided by the chairman of my committee, Dr. S. Joseph Levine. Appreciation is also extended to the members of my doctoral committee, Dr. Mary Jim.Joseph, Dr. RObert Anderson, and especially Dr. Cass Heilman who provided guidance in the completion of this dissertation. ii TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES Chapter I. Chapter II. BACKGROUND AND DESCRIPTION OF THE PROBLEM INTRODUCTION POLARITY THEORY STATEMENT OF THE PROBLEM RESEARCH QUESTIONS SIGNIFICANCE OF THE STUDY LIMITATIONS OF THE STUDY DEFINITION OF TERMS SUMMARY REVIEW OF THE LITERATURE INTRODUCTION VALUES SURVEYS POLARITY THEORY POLARITIES IN GENERAL POLARITIES IN BUSINESS POLARITIES IN EDUCATION POLARITIES IN ADULT LEARNING STATISTICAL ANALYSES iii l4 19 19 22 23 23 25 26 28 30 33 34 36 39 Chapter III. METHODS AND PROCEDURES INTRODUCTION THE INITIAL SURVEY SECOND SURVEY Chapter IV. FINDINGS INITIAL SURVEY SECOND SURVEY SUMMARY OF RESULTS Chapter 40 41 45 47 57 63 V. CONCLUSIONS, RECOMMENDATIONS, AND IMPLICATIONS INTRODUCTION CONCLUSIONS RECOMMENDATIONS FOR FURTHER RESEARCH SIMPLICATIONS FOR ADULT LEARNING APPENDIX A. INITIAL SURVEY APPENDIX B. SECOND SURVEY BIBLIOGRAPHY iv 65 68 72 75 79 82 83 Table Table Table Table Table Table Table Table Table Table Table Table Table 13. 10. 11. 12. Age dis Occupat Comparison of scores between males and females LIST OF TABLES tribution of first sample ions of the respondents Variances associated with factors accounting for at least 1% of the variance Factor survey Factor survey Factor survey Factor survey Survey .50 or Occupations of respondents (second survey) Comparison of scores between males and female Correlations between individual items and total Total loadings and correlations for items in factor #2 loadings and correlations for items in factor #3 loadings and correlations for items in factor #4 loadings and correlations for items in factor #5 items that had a correlation of greater with total score S score score percentiles Page 49 49 52 53 54 55 56 58 63 71 Figure Figure‘ Figure Figure LIST OF FIGURES The six elements of the Precise-General Polarity Development and validation process Survey instructions and format Differentiation—Integration Polarity vi Page 12 41 43 77 CHAPTER I BACKGROUND AND DESCRIPTION OF THE PROBLEM INTRODUCTION The permanence of change and the importance of managing change has been amply documented during the past two decades (Toffler, A. 1985, Naisbitt, J. 1982, Kanter, R.M. 1983). Deregulation has brought about significant change in the banking and airline industries. A reduction in, but by no means elimination of, gender discrimination.in employment has brought about change in business, industry, and government. Women are employed in technical and managerial positions in greater numbers than ever before. Increased global competition has brought about massive change in the way manufacturing is accomplished in this country. Vertical integration, :hn manufacturing organizations, is on the decline. Many organizations that formerly purchased sub-assemblies from their own subsidiaries are selling the subsidiaries and purchasing from whichever supplier can produce the high quality, low cost product. There are widespread campaigns to design quality into both products and production processes. The drive for increased quality has replaced traditional structures with new forms of structure that promote organizational synchronicity and eliminate historical, internal turf barriers. Manufacturing organizations are discovering the need to maintain a strong focus on manufacturing issues and, at the same time, develop 2 a strong focus on service issues that meet customer satisfaction needs of both internal and external customers. Recent writers including Stalk and Hout (1990), Morton (1991), and Barker (1989) suggest that the rate of change is increasing faster than ever and that no end to this increase is in sight. Whether leaders are doing an adequate job of planning and implementing change is open to debate. It is clear however that leaders and managers, in both the public and private sectors, are faced with the challenges of change and are spending more of their working hours managing the change process than they have in past decades. Experienced.organizational development consultants, find themselves working with organizations to broaden the focus of employees from a plant focus to one of international scope. This change in focus is not only occurring at the executive levels of organizations but extends to engineering, production and financial staffs as well. At the same time, intra- departmental foci are being replaced with foci on products, systems and customers. A manager who used to be in charge of a function such as invoicing now may find herself still supervising the invoicing function but doing so as a part of a customer service team that deals with the total customer service delivery system. The focus is no longer on internal organizational functions. The ‘primary' focus is now' on customers and how to provide those customers with products and services of exemplary quality. 3 Supervisory and managerial ranks are being constricted with supervisory functions, in many cases, being drastically altered and even eliminated by the creation of self-directed, autonomous work groups. The role of supervisor is no longer one of controlling employees. It is instead one of serving employees so that the employee teams can be effective. The writer has worked with an automobile components manufacturing plant in the midwest that has recently changed from a traditional mode of supervision to one in which the supervisory level has been eliminated and replaced with a work group advisor pool concept. One-hundred-fifty hourly employees are working in five distinct teams that make all of the decisions regarding how the product is to be produced. If a problem arises that the group itself is unable to solve, the group can ask for an individual, by name, from the work group advisor pool to assist them is solving the problem. The important aspect of this system is that the group decides if and when it needs help and it decides which individual from the pool will be asked to supply the needed help. If a given member of the work group advisor pool is not perceived as competent, cooperative, or lacking in any other necessary attribute, that member will simply not be called on to provide service and will eventually be out of a job. This system truly provides for the work group advisor (formerly supervisor) to recognize that the customer on a day-to-day basis is the production staff and to recognize that job 4 security entails serving the needs of that customer. Supervisory and management development programs can no longer succeed by presenting methods and techniques predicated upon the idea that the job of the manager or supervisor is relatively static, circumscribed, and predictable. Since the new role of the supervisor or manager is to serve the internal customer, i.e. the work teams, the knowledge, skills, and abilities needed by the supervisor or manager will be dictated by the needs of the teams and those needs will be in constant flux. The broader foci at both managerial and worker levels requires decisions to be made more collaboratively than has historically been the case. Increased collaboration brings about increased opportunity for interpersonal and intergroup conflict. The training response has been to teach people how to make decisions in groups. Such interventions are helpful when choices are clear and when data illuminates the best pathway to organizational success. They are inadequate when groups or individuals are engaged in conflicts involving values. The interpersonal infrastructure of manufacturing organizations has long been anchored by the collective bargaining agreement. Management and labor have traditionally agreed to a wide variety of provisions in order to create organizational stability for a two or three year period at a time. This practice is waning as both labor and management S discover that market demands, technology, needed skills, and a host of other variables no longer remain static for two or three years at a time. The emerging practice is for the bargaining parties to develop what is referred to as Modern Operating Agreements (MOAs) . MOAs are flexible documents that allow the parties to develop and modify short term “Letters of Understanding" in order to respond to quickly changing market requirements and other internal and external environmental conditions. Rapid change is not restricted to our work lives. It applies to home life as well. The nuclear family consisting of mother, father, and children living in the same house for a matter of decades has become the exception rather than the norm. A nuclear family, in the modern era, may consist of gays or lesbians, with or without children, double income famdlies with no children, and various combinations of I'revolving step relationships" as children of divorced parents move back and forth in shared custody arrangements. Increased job mobility, career changes, and complex demands of two career families all add to the probability that a family will not remain in the same neighborhood for a generation. Some of the continual adjustments to change that are required at the family level are made easier by the collection of relevant data to guide family, decision making processes. Many of these changes require that clear, mutually exclusive choices be made. Others involve no apparent acceptable alternatives and clearly deal with values. Most of the change described in the popular literature is linear change. Linear change involves giving up one state or condition in favor of a different state or condition. The next change involves giving up the recently adopted state or condition for still another newer state or condition. One moves through change from condition A to condition B to condition C. For example, Glenn Varney (1989) cites extensive research indicating " . . . that collective decision making is a more productive process than individual decision making.‘ Varney provides a host of reasons why organizations should move from individual decision making to collective decision making. In this example, as with many others that could be cited, the present condition and the proposed, alternative condition are typically seen as mutually exclusive. One does not adopt the new condition while retaining the old. In linear change there is little consideration of backtracking, circularity, or the simultaneous pursuit of competing values. There is little consideration of achieving both the former condition and the new condition, in a dynamic flow, over an extended period of time. In fact, actions by organizations to pursue competing values are frequently described in pejorative terms. Such descriptions include wishy-washy, vacillating, unorganized, want their cake and eat it too, and the left hand doesn't know what the right hand is doing. There is an increasing awareness that effectively leading .7 change efforts and managing the implementation of change require the ability to recognize the validity of and to pursue competing values. One must simultaneously retain the old value and adopt the new value even though the two values may appear to be antithetical. However, the pursuit of competing values places leaders in the middle of a dilemma. Dedicating organizational resources to one value, has an adverse impact on the competing value. One operational example of such a dilemma is the need to promote and support effort and accountability by individuals in an organization and, at the same time, to promote and support effort and accountability on the part of teams. Organizational teamwork and individual effort are competing values. If an organization rewards individual effort, there is reduced incentive for putting one's energy into being a team player. If teamwork is rewarded by the organization, there is reduced incentive to compete with one's team members for individual recognition. Organizations that try to stress both are seen as indecisive, vacillating, and wishy-washy. Recent years have seen crusades for teamwork in numerous organizations. Such crusades frequently fail because teamwork is presented as replacing focus on the individual instead of existing simultaneously with focus on the individual. In such situations, individuals often lose their sense of personal potency because they lack the skills to engage in collaborative enterprise. In addition to the loss of personal 8 potency, elimination of recognition for individual effort and accomplishment can result.in.a loss of personal identity'which is only' partially offset by' the development of a team identity. The challenge, felt by leaders, to achieve both individual and group excellence derives from increasing global competition.in.the private sector and.decreased funding in the public sector. They must "have their cake and eat it too' in order to assure the survival of their organizations. In.the public school setting, leaders are aware that they must provide programming and delivery resources to challenge the most able students. This legitimate need requires the allocation of scarce resources to support students at the upper end of the achievement curve. At the same time, these leaders are aware that they must provide programming and delivery resources to provide equal educational opportunity for all students. This legitimate need requires the allocation of scarce resources to support students at the lower end of the achievement curve. Quality and equality are competing values and neither can be ignored. Hewever, the dedication of organizational resources to one value has an adverse impact on the competing value. On a less strategic level, Quinn (1988) has identified Innovating and Coordinating as competing values. Leaders that excel at innovation are not typically as strong at coordination. In addition, if the resources available to a leader are directed primarily at innovation, they cannot also 9 be directed at coordination. Quinn maintains that effective leaders mmst both innovate and coordinate and that both must be done well. In addition to Innovating-Coordinating, the Competing Values Framework developed by Quinn includes Brokering-Monitoring, Producing-Facilitating, and Directing- Mentoring. Quinn's Master Manager accomplishes all eight of these dimensions and does so with excellence. The traditional conception of managing competing values has been one of balance where one finds the appropriate mix of actions supporting each value and maintains that mix. This would suggest that effective leaders and managers need not excel at either innovation nor coordination but be able to meet minimum standards at each. Schools need not excel at quality nor equality as long as they don't fail at either. Johnson (1992) presents a theory which involves a much more dynamic process of managing competing values in which wide swings, instead of balance, are permissible and even necessary. He calls this l'Polarity Theory“. POLARITY THEORY Polarity Theory distinguishes between problems to be solved and.polarities or dilemmas to be managed. Problems to be solved tend to have a sense of finality to them. One can choose one alternative and exclude the other. Deciding whether The Evening College should offer a course on the history and geography of the Middle East next semester or an art appreciation course involves solving a problem, or perhaps 10 a set of problems. Once the problem is solved, the next problem can be addressed. Polarities, on the other hand, involve an ongoing tension between two apparent opposites which need each other. To continually choose one over the other will lead to disfunction. Polarities are, in reality, unsolvable problems because the competing values are mutually dependent instead of mutually exclusive. Johnson uses the example of breathing to illustrate a clear and obvious polarity. This polarity involves the apparent opposites of inhaling and exhaling. To inhale is a desirable and necessary activity. Although inhaling is a desirable and.necessary activity, it does not take one long to see that continually choosing inhaling over its opposite of exhaling will result in extreme system.disfunction. One must indeed place equal value on inhaling and exhaling and continually vacillate between the two in order to maintain the survival of the system. The Polarity Theory Model has a structure containing six elements. There is a vertical axis separating the opposite poles and a horizontal axis separating the positive and negative aspects of each of the opposing poles. The labels used for the opposing poles are value-neutral. That is, neither of the polar labels can be seen as either good nor bad. Referring back to the earlier example, neither inhaling nor exhaling are inherently good or bad. The labels 'energized and lazy' constitute an example of labels that are 11 not value-neutral. A. more appropriate set of labels, expressing the same polarity would be 'energized and relaxed. " While the term “lazy' carries a pejorative connotation, the term.'relaxed' does not. Once labels are established for the two polar extremes, advantages and disadvantages are listed for each pole. The reason the poles must be described in terms that are value-neutral is to facilitate the development of advantages and disadvantages. It is extremely difficult for people to develop a list of advantages for a term.that is seen as inherently undesirable. Likewise, it is difficult to consider disadvantages of terms that are seen. as being inherently good. The polarity labeled “precise and general' is described in Figure 1. Also included in this figure is a label for each of the four quadrants of that polarity. Figure 1 shows the six elements in the structure of a polarity to be managed. In this example, the polarity is Precise - General. Two of the elements, shown on the horizontal axis, are the labels of the polar opposites. The two upper quadrants show the positive aspects of the polar opposites while the two lower quadrants show the negative aspects of the opposites. When Precise is pursued to the exclusion of General, the organization will eventually experience rigidity. In order to solve the rigidity problem, the organization must move toward the General pole in order to achieve sufficient flexibility. Movement toward the General pole will, in fact, allow more flexibility and be seen as a 12 solution to the earlier problem" IMembers of the organization will generally rejoice and exhort each other to "never make that mistake again.‘ 'Precise just doesn't work for this organization.‘ + l + | Flexible I Clear I I I General ----------------- I ----------------- Precise I | I Ambiguous l Rigid I - I - Figure 1. The six elements of the Precise - General polarity Although the immediate organizational problem is solved, a new problem will invariably emerge. When General is pursued to the exclusion of Precise, the organization will 13 eventually experience ambiguity and must move toward the Precise pole in order to achieve sufficient clarity. When this happens, the members of the organization that resisted the move to General will have their day of celebration. They will be able to say “I told you so!“ As the organization moves toward Precise and abandons General, the problems associated with ambiguity will, in fact, be solved. However, after experiencing clarity for a while, the organization will eventually experience rigidity and find itself in the position of having to move toward the General pole in order to obtain some flexibility. According to Johnson, organizations that manage polarities well understand the cyclical nature of polarities and don't mind being on what he calls an “infinity loop“. The competing values of being precise and of being general need each other in a cyclical, dynamic relationship involving movement back and forth between the poles. Polarity Theory refers to the idea of static balance between the poles as the “Single Pole Myth“. The theory suggests that effective management of polarities requires regular changes in emphasis between the poles. The need for continual shifts in emphasis comes from continual change in organizations, people, and the environment in which people and organizations function. The balance point of today may place the system out of balance tomorrow. Shifts of emphasis may be large or small but Johnson maintains that they are continual in systems that manage polarities well. 14 The identification of a polarity as opposed to either a problem to be solved or a mutually exclusive choice, is not without considerable ambiguity. In the case of inhaling and exhaling, referred to earlier, it is clear that one cannot choose either to the exclusion of the other. When the choice becomes whether to purchase a series of single operation presses or a:mu1ti-station.press, it is clear that a choice of one over the other can be made. But what about a pair of terms sudh as socialism and capitalism which, when put in terms that are value neutral, would be expressed as public ownership and private ownership? Individuals who see only the positive aspects of private ownership and the negative aspects of public ownership of the means of production would be difficult to convince that the opposing poles are mutually dependent and need each other in order to exist. As with many areas of social inquiry, Polarity Theory encounters identification, classification, and measurement problems that are not easily overlooked or overcome. STATEMENT OF THE PROBLEM Although the literature reveals an increasing awareness of the need to simultaneously value polar opposites and although Johnson has developed a model to understand the interplay between polar opposites, there are no instruments available to assess one's preference to simultaneously value polar opposites as opposed to valuing only one pole or the other. 15 Robert Carkhuff (1988) writes “Insider trading and corporate raiding notwithstanding, financial capital is no longer the dominant ingredient in economic productivity growth. In its place are the two emerging ingredients: human capital, or people who think, and information capital, or operational information that can be translated into products and services.“ The validity of Carcuff’s assertion can be readily observed.in the business world of today; Productivity gains in the auto industry, for example, are resulting from better utilization of human.capital brought about by increased use of systems of worker collaboration. Gains brought about through the use of robotics also support Carcuff’s assertion as robots can essentially be viewed as information systems. The effective utilization of human capital and of information capital requires interpersonal and group interaction. Increased interpersonal and group interaction in turn increases the potential for values conflicts in the management of competing values. Robert Quinn (1989) has developed an instrument that assesses the extent to which leaders and.managers effectively pursue the eight competing values that make up his Competing Values Framework. Leaders and managers are thus able to discover the areas in which they may need further development. The instrument is descriptive so that leaders and managers, as self directed learners, can design and plan their own learning experiences or program planners can develop curriculum based 16 on an organization profile. The Polarity Theory developed by Johnson.is much.broader in its scope than Quinn’s Competing Values Framework. The Competing Values Framework limits its application.to the roles of manager and leader. Polarity Theory is intended to apply to managerial and leadership behavior, family systems, individual development, international politics, ethics, and religion. In fact, it is difficult to see competing values situations where the Johnson model cannot be applied. Participants in Polarity Theory workshops and seminars have expressed the need for a self-discovery instrument that would help them to more clearly see their own preferences for tolerating or embracing competing values. Additionally, such an instrument would enable organizations to assess the extent to which their managers, leaders, and other employees are able to function in an environment of competing values and ongoing change. Programs designed to increase leadership and management skills, whether they are designed for schools, industry, or government, must of necessity address values. Setting organizational priorities involves valuing some outcomes over other outcomes. Dealing with substance abuse problems in the ‘work. place, for example, requires facing“ up to certain personal and organizational values. Should those who are addicted to controlled substances be punished because they are evil or should they receive medical and psychological 17 assistance because they are suffering from an addictive illness? A major instructional difficulty associated with programs that focus on personal and organizational values is helping participants to become clear about their own values and the values held by their organizations. This difficulty is in part attributable to the fact that values are personally ingrained and an everyday part of the organizational culture and are therefore largely invisible to the members of the culture. The Polarity Theory Model raises the level of difficulty considerably. It does so by first of all asking participants to be clear about where they stand with respect to a number of values issues. It then goes on to demonstrate the need for accepting both.of the competing values in each of the conflicting values sets. For many participants, this experience requires a major paradigm shift. Quinn (1989, describes a simple agenda for self- improvement. “This agenda involves three general steps: Learn about yourself, develop a change strategy, and implement the strategy.“ He suggests that would—be Master Managers accomplish the first step by completing the Competing Values Instrument. Although, as noted earlier, Quinn has developed an instrument to assess program participant's preferences to recognize and to adopt competing values relevant to the Competing values Framework model, Johnson has developed no such instrument for his Polarity Theory model. 18 With the exception of the fairly recent emergence of Gestalt psychology and other holistic approaches, it seems that most of the history of psychology has been.more concerned with segmenting and categorizing human experience than with integrating jig. Major theorists have emphasized different aspects of behavior and categorized these as cognitive (Ellis, 1962, 1971), affective (Rogers, 1961), and psychomotor (Wolpe, 1969). With such a foundation in theoretical psychology, it should be no surprise that instruments designed to identify traits tend to categorize rather than to integrate. There are, of course, a large number of survey instruments that have been developed to examine attitudes, values, and beliefs. These instruments vary widely in the extent to which they have demonstrated validity and reliability. They also vary widely in the particular attitudes, values, and beliefs they purport to examine. What these surveys appear to'have in common is a bias to categorize respondents baseduupon.theiattitudes, values, and beliefs that are held. Construct validity studies conducted on these instruments are, for the most part aimed at validating the categories. Two such instruments are discussed in the literature review. No instruments were found that were designed to determine the degree to which respondents simultaneously value categories that are, on their face, mutually exclusive. 19 RESEARCH QUESTIONS The purpose of this study is to develop an instrument that will assess preferences to simultaneously hold competing values. Four research questions are formulated to accomplish this purpose. 1. What survey items and format will validly and reliably measure respondents’ preferences to simmltaneously value polar opposites as opposed to valuing one pole or the other? 2. .Are either' age: or sex related. to respondents' preferences to value polarities? 3. Given survey items that include a broad range of polarity topics, which items, if any, combine to form clusters indicating that all items in a cluster are measuring a common domain of polarities? 4. If distinct clusters emerge, what are the common elements within each cluster that will enable a clear description of the cluster domain? SIGNIFICANCE OF THE STUDY The Polarity Theory model has been presented to nearly 1,000 workshop participants by various members of the Human Resource Development Center staff at the University of Toledo. These participants, representing public school leaders as well as leaders from business and industry, have ranged from non- supervisory workers through the vice-presidents of Fortune 50 companies. The Theory has been well received by the participants, at all levels, as a useful tool for personal and professional development in programs which use self-discovery strategies for learning and change. Other parts of the programs make use of the Keirsey Temperament Sorter developed 20 by Keirsey, et. a1. (1984), the Learning Styles Inventory by Kolb (1981), and.Human.Synergistics' Stress Processing Report (1988) to facilitate self-discovery. Program participants have indicated that an instrument to help them see their propensities for recognizing and valuing polarities would be very helpful to their self-discovery process. One barrier to the development of effective work teams relates to the issue of management of group conflict. The movement from segmented organizations to synchronous organizations is proving to be a painful transition and the pain results from a long history of organizational turf barriers based on functional, departmental demarcation. In the heavy industry segment of the economy, for example, employees from production have traditionally been rewarded for “shipping iron“ while people from the quality department have been rewarded for making certain that the product is shipped only if it meets stringent quality standards. These two functional areas have been organizationally separate and their interests have been in opposition. In today’s synchronous organizations, representatives of these two functions find themselves on the same team and are expected to reach consensus with other team members in group decision making. The competing values of “Quantity“ and “Quality“ must be simultaneously embraced and neither the production representative nor the quality representative can. be an advocate solely for their narrow area of responsibility. 21 Although training in polarity theory may help both the production person and the quality person to see more clearly the nature of their conflict by carefully developing all four quadrants of the polarity, it can do little to help them see themselves more clearly. What is needed is a self-discovery instrument that helps these people gauge their own preferences for accepting both poles in competing values situations. Those who discover a personal preference to value both poles can make a decision to act against their normal inclination when they perceive a need to crusade for a cause and those who discover a personal preference to only value one pole or the other can :make a decision to act against their normal inclination when they see their crusading causing nonproductive conflict in the group. Recognizing one's overall preference for accepting competing values may thus prove useful to both the individual and the group. Of additional significant importance is the identification of clusters of related polarities that may be accepted more readily than others. It may be for example that work life polarities are accepted.more readily than home life polarities. Perhaps such related clusters differ as a function of age, education or other demographic attributes. Effective program design for training in Polarity Theory content and development of personal strategies for the effective management of polarities would benefit from further understanding of polarity clusters and differential effects of 22 various demographic attributes. Clusters and demographic effects can't be identified in the absence of an instrument to measure preference to recognize and value a broad range of polarities or competing values. This study proposes to develop and validate an instrument which will measure respondents' overall preference to value polarities. Analysis of the data will determine whether a preference to value polarities is a single dimension trait or a multi-dimension trait based on clusters of polarities. Finally, the study will examine whether a relationship exists between one's preference to value polarities and one's age, or SEX. LIMITATIONS OF THE STUDY Establishing the validity of a survey is a complex, ongoing process. An instrument is valid if it can be shown to measure that which it purports to measure. The present study develops an instrument and begins the validation process. However, complete validation is well beyond the scope of this study. Although the term “polarity“ is defined in the Definition of Terms section, the concept itself is not precisely developed nor operationally defined. This study attempts to develop a method of measuring a construct that lacks precise definition. Any measurement, based on an incompletely defined construct is bound to be shaky at best. The reason this study 23 is sound, despite the lack of a clearly defined construct, is that the survey resulting from. the study will help to operationalize the polarity construct. This study uses college students, with a limited range of age and educational level, as subjects. It could verwaell be that the polarities preferences in question vary markedly with reference to these and other demographic characteristics. Survey administration to a more diverse group is outside the scope of the present study. DEFINITION OF TERMS 1. Polarity - For the purposes of this study, the term polarity refers to a pair of words or phrases that are of neutral value, apparent polar opposites, and interdependent. 2. Preferences - This term is ascribed its customary meaning as 'ChOOSlng as more desirable“. 3. Value - This term is also ascribed its customary meaning of “a principle considered worthwhile“. 4. Competing Values — This study considers competing values to be pairs of values terms that are apparent opposites. 5. Clusters - These are sub-sets of survey items that factor analysis determines are closely related and account for major portions of the variance in total survey scores. 6. Polar Opposites - The pair of words or phrases used to label the horizontal axes in the Polarity Theory model are referred to as polar opposites. SUWARY Constant change is evident in virtually all aspects of work and family life. Problem solving models are effective aids in managing change when the task is to choose between 24 alternatives that are mutually exclusive. However, there is a growing realization that many recurrent problems are simply not amenable to resolution through the use of traditional problem solving methods. Successful management of these problems involves a continual shift in emphasis, back and forth, between mutually dependent and apparent opposite values rather than choosing between the values. Although our understanding of these dilemmas or polarities is assisted by newly developed models, assessment instruments to measure one's preference to simultaneously hold competing values have not been developed. The purpose of this study is to develop and perform preliminary validation of such an instrument . CHAPTER II REVIEW OP THE LITERATURE INTRODUCTION The purpose of this study is to develop and validate an instrument that will measure respondents' preferences to simultaneously support opposing or competing values. One with a greater preference would have a greater tolerance for ambiguity and would have less need to polarize issues. One with a lesser preference to simultaneously support competing values would have less tolerance for ambiguity and would be more likely to polarize issues. Although the construct of paradox, dilemma, or polarity is not new, acceptance of the idea that competing values situations are inherent in both personal and organizational systems, and must therefore be successfully managed, is relatively new. A theoretical framework for understand the dynamics of how polarities flow, over an extended period of time, is only now being developed. This study will extend the polarities literature by beginning to operationalize behaviors pertinent to valuing both of the poles of a given polarity. The literature review contained in this chapter is divided into seven sections. The first section examines two instruments, taken as examples, which measure and categorize values preferences but fail to assess the extent to which competing values can be simultaneously held. The second section describes the Polarity Theory developed by Johnson 25 26 (1992). The third section discusses polarities in general. The fourth section considers polarities in the management and leadership of business with particular attention to the work of Robert Quinn. The fifth section reviews polarities commonly found in public education. The sixth section examines polarities that are prevalent in the field of adult education and training. The seventh and final section provides support from the literature for the techniques of statistical analysis that are used in the study. The scope of this review includes identification and description of various polarities contained in the literature. What is not included is extensive discussion of models for managing polarities. Such models are simply not available. If they were available, the present study would perhaps not be necessary. Historically, writers have dealt with polarities by providing advice as to which of competing alternatives should be selected by the reader. A best choice has been advocated. More recently, there has emerged an increased tolerance for recognizing that polarities cannot be managed by treating them as problems to be solved. This more tolerant view, although helpful, has not provided solid direction regarding the management of polarities. At best, we have been exhorted to “balance“ the competing values. VALUES SURVEYS Oliver (1985) developed a self-discovery instrument which 27 he calls “the Personal Value Statement (PVS) . The PVS questionnaire contains twenty items, with each item in turn containing three words arranged in triads such as “Power, Style, People“ and “Influence, Elegance, Charity“. Respondents are instructed to prioritize each of the words in each triad based on the importance of the word to the respondent. Each word is assigned a value of one, two, or three. Each word is then assigned to one of five categories identified as Political, Aesthetic, Social, Theoretical, and Economic. Numerical values assigned by the respondents to each of the words are summed for all of the words in each of the five categories. Respondents can thus discover whether their personal values are predominantly political, aesthetic, social, theoretical, or economic. Ignoring, for the moment, whether or not it is an entirely appropriate juxtaposition, if one can posit that aesthetic and theoretical are polar opposites, the PVS is simply not structured to reveal the extent to which a respondent can simultaneously value both aesthetics and theory. Of perhaps even greater importance is the fact that the instrument provides no support for movement back and forth between the polar opposites, over time. Presumably one is not expected to be capable of excelling at being political, then social, then aesthetic all within a one- hour period. Coblentz (1990) has developed an instrument which is more narrowly focused than that of Oliver. His instrument is 28 designed to assist organizational change agents in identifying their personal styles in facilitating organizational change. Like Oliver, Coblentz displays words in triads. Respondents are asked to distribute twelve points among the words in each triad. The thirty words, from the ten triads are assigned to three distinct categories labeled master, saint, and prophet. With these three very general labels, it is difficult to even conceptualize competing values. Evidence of a reluctance on the part of Coblentz to allow for competing values is apparent when he explains how to score the three columns labeled master, saint, and prophet. He writes “If any column totals tie, you may prioritize them on the work sheet in the order in which you prefer to begin working on them. “ Foci that must be prioritized cannot, at the same time, be simultaneously valued. POLARITY THEORY Johnson (1992) has developed a model which describes both the structure and the dynamic of polarities over time. The model has the broadest possible application in that it is described as applicable to individuals, groups, and even nations. The content of a polarity may include issues relative to one's work, family, religion, politics, and almost any other area one can imagine. The author suggests that polarities are nanaged more or less well but that they are never completely resolved. He states “What distinguishes a 29 problem to be solved from a polarity to be managed is that one has a solution and the other doesn't.“ The structure of a polarity, according to Johnson, includes a horizontal axis with a polar opposite at each end. It also includes a vertical axis which bisects the horizontal axis to produce four quadrants. The quadrants above the horizontal axis contain the advantages of the respective polar opposites while the quadrants belOW' the ‘horizontal axis contain the disadvantages of the respective polar opposites. The dynamic of a polarity is that it is always in.motion with increased emphasis first on one pole then on the other. “Another assumption I question is that it is possible to merge the up sides of both poles so that you can simultaneously have the benefits of both. I don't think the shift in emphasis needs to be dramatic, but I think it does take place. When a tightrope walker has considerable experience, there will be very little movement in the long pole used for balance. But it does move ever so slightly shifting the emphasis from one side to the other.“ Johnson (1992) According to Johnson, all polarities can be structured according to his model and all polarities, when considered in their entirety, possess the dynamic characteristics presented by the model. It is frequently the case that different individuals see different quadrants of a polarity. They typically are aware of the advantages of their position and the disadvantages of the opposite pole. They' may not recognize that the issue is, in fact, a polarity; However, if 30 opposing sides share their incomplete visions, the entire polarity will become apparent. The author refers to those who are aware of only two diagonal quadrants of the polarity as either “crusaders“ or “tradition bearers.“ Crusaders are aware only of the negative aspects of the current situation and of the positive aspects of the opposite pole. Tradition bearers are aware only of the positive aspects of the current situation and they fear the negative aspects of the opposite pole. Both crusaders and tradition bearers are accurate in their perceptions. There is, in fact, a negative result of clinging to the nearby pole just as there is a positive result of such action. Likewise, there are both positive and negative results of moving toward the opposite pole. POLARITIES IN GENERAL What is perhaps the broadest application of Polarity Theory may be found in the following quotation from “A Brief History of Time“ by Steven W. Hawking (1988) . “Although light is made up of waves, Planck's quantum hypothesis tells us that in some ways it behaves as if it were composed of particles: it can be emitted or absorbed only in packets, or quanta. Equally, Heisenberg's uncertainty principle implies that particles behave in some respects like waves: they do not have a definite position but are ’smeared out' with a certain probability distribution. The theory of quantum mechanics is based on an entirely new type of mathematics that no longer describes the real world in terms of particles and waves; it is only the observations of the world that may be described in those terms. There is thus a duality between waves and particles in quantum mechanics: for some purposes it is helpful to think of particles as waves and for other purposes it is better to think of waves as particles.“ 31 At a less cosmic level, Cameron and Quinn, (1988) use the word “paradox“ to refer to the kinds of issues this study calls “polarities“ . They report that writers are increasingly recognizing the need for individuals and organizations to be sensitive to “the presence of simultaneous opposites or contradictions“. Despite the need, to be sensitive to simultaneous opposites, Maruyama, (1976) suggests that traditional methods of thinking about organizations are not appropriate for analyzing these complex phenomena. He uses term such as competitive, quantitative, classificational, and unidirectional to describe current modes of thinking about organizational behavior. What is needed according to Maruyama is thinking that has attributes that are, for example, symbiotic, qualitative, relational, and contextual. .Russo and Schoemaker, (1989) list four elements of the decision making process. These elements are framing, gathering intelligence, coming to conclusions, and learning from feedback. They also suggest a step that precedes the four-element process which they call the metadecision. It is suggested that this step remain separate from the framing step and that it consider such issues as should a decision be made at all, who should nake the decision, what's the primary issue, and will this decision affect other decisions. It would appear that Polarity Theory is most appropriately placed in the metadecision phase of the decision making process 32 described by Russo and Schoemaker. The literature regarding the importance of attending to paradoxes or polarities is broadly dispersed. The work of Quinn which has been.mentioned earlier is primarily addressed to an. audience of leaders, managers and supervisors in business, industry and government. The work of Russo and Schoemaker, also mentioned earlier, is focused on both manufacturing and service industries. Siporin and Gummer, (1988) assert that the ability to “perceive the oppositional elements of a paradox and to deal with them“ often precedes adaptive change in family therapy; They go on to suggest that paradox is normal, inherent, and prevalent in human nature. Bennis, (1989), addressing leaders in education as well as business, maintains that complexity is here to stay and that our inability to tolerate ambiguity is an important reason why leaders are unable to effectively lead. Finally, the Myers-Briggs Type Indicator, a psychological assessment instrument that reflects temperament, is widely used for individual therapy and for employee development in virtually every type of organization imaginable including the:military. Both the Myers-Briggs instrument and the similar Keirsey Temperament Sorter (Keirsey and Bates, 1984) display the four polarities of temperament as introversion and extraversion, sensing and intuiting, thinking and feeling, and judging and perceiving. 33 POLARITIES IN BUSINESS Leadership theories have advanced through a progression beginning with positing “one-best“ way approaches, advancing to various contingency theories which suggested that the “best way“ was to be determined within situational contexts, and finally to the present awareness that simultaneous “best ways“ may in fact be contradictory but still valid. Blake and Mouton (1964), wrote persuasively about the need to balance foci between tasks and relationships. Weick (1976), suggests that organizations need to simultaneousLy demonstrate both loose and tight properties. The value of attending to both individual and group dynamics is described by Smith and Berg (1987). Sims and Gioia (1986), Van de Ven (1983, and Quinn (1990), conceptualize leadership as a balancing of conflicting demands and values. The task is to identify the relevant polar opposites and determine the ways in which key polarities are balanced by effective leaders. Ouchi (1981), compares successful Japanese and American companies. He developed Theory Z to describe some of the successful American Companies. These companies avoid excessive formality, automatism, and individualism. and maintain a careful balance between intimacy and objectivity. Theory 2 companies do not reject individualisnn They recognize that American culture places strong value on this trait and they balance the trait with an ecological view that places individualism within the organizational context of 34 interdependence. Theory Z companies also value objective information for decision making while, at the same time, they apply subjective criteria to determine the “suitability“ of the decision. The relevant competing values identified by Quinn and his associates (Quinn, et. a1. 1990) are contained in four “contradictory models of organizational effectiveness: 1) the rational goal model, emphasizing achievement and goal attainment; 2) the internal process model, emphasizing bureaucracy, internal control and stability; 3) the human relations model, emphasizing the building of teamwork and commitment to organizational goals; and 4) the open systems model, emphasizing adaptation to the external environment.“ POLARITIES IN EDUCATION -The topic of global education, within the context of American public education, involves a number of polarities. The commonness/uniqueness paradox is discussed by Tye(1991). Any given American school is both like other schools and, at the same time, different than any other school. To the extent that schools are alike in that they are part of a society engaging in global markets, global education may be a good idea. To the extent that schools are under local control and responsive to local needs, local parochialism may see global education as antithetical to community interests. Lamy (1991), suggests that in order to minimize the controversy 35 over global education we ought to “clearly define the substantive focus and learning objectives of any interdisciplinary program.“ He also states “Controversy is inevitable, and it should. be welcomed. by educators as essential to the learning process.“ Duke (1987) states that the time of school administrators is often consumed by the need to react to problems. Although he doesn't suggest that problems ought not be addressed, he does suggest that more focus on reflection is needed. Thus the action/reflection polarity is an issue. He also addresses the self need/other need polarity by suggesting that school administrators are so busy attending to the needs of students and staff that they seldom pay attention to what they, personally need. The rigor/relevance polarity is alluded to by McPherson, et. a1. (1986). The authors believe that theory and practice in educational administration are related in an essential, even symbiotic, way. Despite the essential relationship, they suggest “Few textbooks in educational administration have considered theory and practice in meaningful juxtaposition.“ The centralized/decentralized polarity has been of interest to educators for some time. Advocates for site-based management recognize the legitimate need of neighborhoods to influence the operation of neighborhood schools. At the same time, since financial resources are acquired at the district level, some district level control is necessary. 36 POLARITIES IN ADULT LEARNING Perhaps the most striking example of a polarity in adult learning is the pedagogy/andragogy polarity. Knowles (1989) changed the subtitle of his book from “Modern Practice of Adult Education: Andragogy Versus Pedagogy“ to “Modern Practice of Adult Education: From Pedagogy to Andragogy“ . He acknowledges making this change because it had been brought to his attention that pedagogical assumptions sometimes worked best with adults who had to learn in totally new content areas and andragogical assumptions worked best with children in a number of situations. He now advocates a more situational approach to the choice of andragogical and pedagogical learning strategies. The rigor/relevance polarity, applied to social research, is especially pertinent to adult education in developing countries. Part of the dynamic in this dispute involves the polarity between the objective and the subjective. Freire (1982) argues forcefully for integration of the objective with the subjective in his social research. This integration is necessary, according to Freire, in order for the research to be valid. Such a view contrasts sharply with the traditional view that social research must be conducted through strictly objective means. Hall (1982) proposes action research using objective approaches and subjective approaches as each is appropriate. His objection to traditional social research which has stressed objectivity is that it oversimplifies, 37 alienates, is not conducive to action, and is not consistent with principles of adult education. The objective versus subjective polarity is applied to the environment of adult and continuing education by Evans (1983). The author discusses the difficulty of faculty in maintaining sufficient objectivity to perform learner evaluations in a one-to-one faculty student learning arrangement that values subjective interaction. Adams (1972) describes tension at the Highlander Folk School over the issue of individualism versus collective action. Highlander had a tradition of organizing mine workers when there was a suggestion that an employment agency be established to help a small segment of the community. The staff was divided in its commitment to helping individuals as opposed to helping large groups. Myles Horton, founder of the Highland Folk School was conmitted to valuing both individualism and collective action. The content centered versus learner centered polarity can be observed from kindergarten through adult learning. Overly (1979) observes that informal learning is powerful because selection of the learning activities rests with the learner. He also recognizes the validity of homogenizing the learning process in the formal education setting. It is suggested that non-formal learning environments are intermediary processes between formal learning and informal learning. The issue of self-directed learning versus other-directed 38 learning is closely related to the content centered versus learner centered.polarityx Manz andeanz (1991) suggest that organizations ought to foster self-directed learning to a greater extent than is currently the case. They also recognize the validity of persuasive influence from the instructor. It is suggested by the authors that one strategy for the successful management of this polarity is to have a clearly acknowledged recognition that the instructor’s point of view is advisory only. Tway, et. al. (1990) address the debate between content centered and learner centered competency based training models. At first glance, any competency based model would appear to be content centered. The authors have developed a process in which the learners specify the competencies as a way of managing this polarity. The polarity between individual action and group action, discussed in previous sections of this chapter, is extant in adult learning as well. Longstreet (1979) notes the tension between self-fulfillment and societal-fulfillment. She states “It must further be understood that human fulfillment involves not simply the balanced interaction of self—fulfillment and societal fulfillment, but the tension between these as well.“ This view seems consistent with the position taken by Johnson that successfully managed polarities do not involve finding a balance point and clinging to it but, managing movement between the polar opposites. 39 STATISTICAL ANALYSES The format used in the survey asked respondents to circle a single number on a line that separated the two polar opposites. An example is “Head 5 4 3 2 1 0 1 2 3 4 5 Heart“. Babbie (1973) suggests the use of ordinal response formats in the construction of indices that use multiple items to measure a given trait. Babbie also distinguishes between this type of format and the “Likert Scale“ which assesses agreement or approval of one value at a time rather than two at a time. The search for meaningful clusters was conducted through the use of factor analysis. .Dillon.and.Goldstein (1984) point out that the factor analytic procedure reduces the number of variables in.a data set by extracting factors in.a manner that maximizes the amount of variance in a total data set that can be accounted for by the extracted factors. Thorn (1989) provides rationale for the use of the t-test to assess the significance of differences between sample means. T-tests were used in the present study to test for gender differences. The Pearson Product Moment correlation coefficient was used.extensively in this study'toiassess relationships between various items and total score. Anderson et. a1. (1987) discuss the appropriateness of this procedure for assessing linear relationships between variables. CHAPTER III ITHODS AND PROCEDURES INTRODUCTION The purpose of this study is to develop an instrument that will assess preferences to simultaneously hold competing values. Four research questions are formulated to accomplish this purpose. 1. What survey items and format will validly and reliably’ measure respondents' overall preference to simultaneously value polar opposites as opposed to valuing one pole or the other? 2. Are either age or sex related to respondents' preference to value polarities? 3. Given survey items that include a broad range of polarity topics, which items, if any, combine to formiclusters indicating that all items in a cluster are measuring a common domain of polarities? 4. If distinct clusters emerge, what are the common elements within each cluster that will enable a clear description of the cluster domain? Development and validation of the survey instrument was accomplished with a multi-step process. That process is outlined in Figure 2. 40 41 Step 1. Assemble 49 polarity pairs Step 2. Administer 49 item survey to sample Step 3. Analyze data for clusters and item analysis Step 4. Construct second survey from items with ’ highest predictive value Step 5. Administer refined survey to new sample Step 6. Analyze data to confirm predictive value of individual items Figure 2. Development and validation process An instrument with forty-nine items was developed from a variety of sources described in detail in this chapter. The instrument was administered to a sample and the data were analyzed to determine whether the trait was uni-dimensional or multi—dimensional. The survey was refined and only the items that had a correlation of .30 or greater, with the total survey score, were retained and incorporated into the final survey. The final survey was then administered to a new sample. The data from the second administration were analyzed to assure that each individual item had a correlation of at least .30 with the total survey score. This chapter reviews the purpose of the study, the study subjects, the development and administration of the instrument, the experimental procedure, and statistical analyses. THE INITIAL SURVEY A pool of forty-nine items was assembled for the initial Esurvey. These items were selected to represent the broadest possible range of polarities from work life, home life, 42 politics, learning styles, personality' types, management styles, and other topical areas. The rationale for the broad range of topical areas was to enable testing for clusters of related polarities if such clusters existed. The operational definition used in selecting items for inclusion in the survey required that the items express pairs of values that are interdependent” 'The conceptual test for interdependence required that a pair was interdependent if focus on one value, to the total exclusion of the other value, would reasonably lead to system disfunction. The items used in the initial survey are included in Appendix I. Some of the items originated from. participants in Polarities workshops conducted at the University of Toledo. Participants, as part of the training, were asked to list polarities they encounter at home and in the work place. Dr. Barry Johnson, consultant and author, added to the list. Polarities identified from the writings of Dr. Robert Quinn were also included. Various assessment instruments, including the Learning Styles Indicator, developed by Dr. David Kolb, the Leadership Behavior Analysis, by Kenneth Blanchard, and the Stress Processing Report, by Dr. Clayton Lafferty, were examined for content that expressed a polarity. Dr. Barry Johnson.was consulted regarding an appropriate format for the presentation of survey items. The format and instructions shown in Figure 2 were adopted. 43 Instructions Circle one number on each line indicating which.termiyou value more and by how much. Circle only one number for each pair of terms. Example: Easy going 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 — 4 - 5 Serious If you value being easy going over being serious but you value them.almost the same, you would circle “1“ to the left of the “0“. If you thought that being easy going is just plain lazy and placed more value on being easy going, you would circle “5“ to the right of the “0“. If you place equal value on being easy going and on being serious you would circle the '0' . CIRCLE ONLY ONE NUMBER FOR EACH PAIR OF ITEMS. 1. Head Heart 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - 5 Figure 3. Survey instructions and format. An alternative format which would have respondents circle two numbers on each line indicating the extent to which each polar opposite was valued was rejected. This latter alternative format was rejected because it would require additional time to complete the survey and did not appear to add additional value for self-discovery learning. Dr. Johnson reviewed the forty-nine pairs included in the original survey to ascertain content validity. Each pair was 44 determined to consist of terms for which a case could be made that the words are mutually dependent, apparent opposites. A procedure was carried out to assist in establishing the construct validity of the instrument. A panel, each person being knowledgeable regarding the construct upon which Polarity Theory is based, should be able to determine from observed behaviors, whether individual respondents would score high, medium, or low on the survey. Such prediction should be possible only if the survey measures the same construct is that used by the panel members in making their judgements. This assumes, of course, that the trait in question is manifest in observable behavior. The survey, consisting of forty—nine pairs, was administered to three staff members in the Division of Continuing Education. A panel of three persons who are expertly familiar with Polarity Theory and who have personal knowledge of the behavioral traits of the three respondents were asked to rate the polarity propensities of the three respondents on a scale of high, medium, and low. Each expert rated each respondent independently from the other experts. The ratings of the experts were compared to the survey scores of the respondents. The initial survey was administered to sixty graduate students from the College of Education and Allied Professions and to thirty-eight graduate students from the College of Business at the University of Toledo. None of the respondents 45 had any knowledge of Polarity Theory. Survey administration was conducted by regular faculty, in their classrooms. Data from the administration of the initial survey were analyzed for split-half reliability using the Pearson product moment coefficient of correlation between the even numbered and odd numbered items. A t-test was administered to determine whether significant differences exist between the total scores for males and the total scores for females. Factor analysis was used to determine whether the polarity preference trait was multi-dimensional or uni-dimensional. This process was also used to determine which specific survey items combined to form related clusters. A Pearson product moment correlation was computed to determine whether age was significantly related to total survey scores. Finally, the score on each item was correlated with the total score. Individual items that had a correlation with the total score of less than .30 were discarded. Possible duplicate items that, on their face, were closely related and showed a high correlation with each other were carefully examined and only one of the items was retained. SECOND SURVEY The revised survey was administered to seventy-three students from the College of Business at The University of Toledo. As with the initial survey, none of the respondents had any knowledge of Polarity Theory and survey administration 46 was accomplished by persons other than the researcher. The second survey consisted of twenty items and is included as Appendix 2. Data from the administration of the second survey were analyzed for split-half reliability using the same procedures applied to the initial survey. A t-test ‘was administered.to determine whether significant differences existed between the total scores for males and the total scores for females. A Pearson.product moment correlation.was used to determine whether or not age was related to total scores. Correlation analysis was used to determine the strength of relationship between each individual item and total scores. CHAPTER IV FINDINGS INITIAL SURVEY The first analysis step consisted of administering the survey to three staff members in the Division of Continuing Education at the University of Toledo. These staff members were selected on the belief that they represented a broad range of preference to simultaneously value both poles of competing values. One of the staff members tends to gather considerable information before taking a position. He frequently encourages “devil's advocate“ positions on both sides of issues. It was predicted that he would score low on the survey, indicating a pmeference to simmltaneously hold competing values. A second staff member was selected.because he frequently argues a position very forcefully and is more concerned with persuading than with clarifying. It was predicted that he would score high on the survey, indicating a preference to hold one value to the exclusion of the other. The third staff person was predicted to score in the middle of the survey range because her behaviors seemed to fall in between those exhibited by the other two staff members. A three person panel was assembled. Panel members were both expertly familiar with polarity theory and also familiar with the behavioral traits of the three respondents. Panel members were shown blank surveys and asked to independently predict whether the scores of each respondent would be high, 47 48 medium, or low. The panel members were in perfect agreement as to which respondent was “high,“ which was “medium,“ and which was “low“ . The survey scores from the three respondents were combined with data from an additional ninety-eight respondents and percentile scores were computed. Panel ratings were compared to survey percentile scores for the three respondents. The respondent who was rated “high“ had a percentile score of .92, the Fmediumf respondent had. a percentile score of .44, and the “low“ respondent had a percentile score of .03. The perfect inter-rater agreement between panel members coupled with the agreement between panel member ratings and the percentile scores provides evidence of construct validity. The initial survey consisting of forty-nine items was administered to ninety-eight respondents. Survey administration occurred in regularly scheduled , credit classroom settings therefore survey return rate was not a problem as it may have been if surveys were mailed to respondents with a request that completed surveys be returned by mail. Respondent anonymity was promised to respondents and severely limited the amount of demographic data that could be requested. Demographic data on age, sex, and occupation was collected. Fifty-five of the respondents were females, forty- two were males, and one respondent failed to indicate gender. Table 1 presents the age distribution for the sample. 49 Table 1 Age distribution of first sample Age Frequency 20-24 43 25-29 15 30-34 8 35-39 11 40-44 13 45-50 8 Mean Age = 29.67 Standard Deviation = 9.1 Standardized Kurtosis Standardized Skewness -1.97 2.7 The skewness coefficient indicates the distribution of age to be significantly, positively skewed. That is, a preponderance of respondents were in the 20 to 29 age groups. The kurtosis coefficient indicates a flat distribution. These two measures suggest that age is not normally distributed in the sample and that generalization to populations where age is normally distributed may not be warranted. Table 2 shows the occupations of the survey respondents. This information is provided for replication purposes only and is not used in the analyses. Table 2 Occupations of the respondents Occupation Frequency Retail Sales 6 Health Care 10 Educational Administration 10 Full Time Student 36 Teacher 7 Law Enforcement 4 Self Employed 3 Business And Industry 18 No Occupation Given 4 50 Analysis of the actual survey data scores began with a comparison of the total scores from all odd numbered items and all even numbered items. A Pearson product moment correlation coefficient was calculated to determine the extent of agreement between the odd and even sub-scores. The correlation coefficient of .8882 indicates a high level of agreement between the two halves of the instrument and contributes to confidence in its reliability. A t test was performed to determine whether a significant difference could be found between the scores of males and females. No significant difference was found. The results of the t test are displayed in Table 3. Table 3 Comparison of scores between males and females Mean Standard Deviation Males 101.7 34.4 Females 104 36.1 t = 0.368 (p.=.713) A Pearson product moment correlation was performed to test for a relationship between age and total score on the survey instrument. The correlation coefficient of -.0663 (p.=.52) indicates no significant relationship between age and total score. The analysis of age and gender effects, described above, dealt with demographic aspects of the sample. The next analysis was performed to determine whether the polarity 51 preference trait was uni-dimensional or multi-dimensional. A factor analysis was used to ascertain whether individual items tended to contribute to one trait, uni-dimensional or to a number of discrete traits, multi-dimensional. The factor analysis procedure empirically combines various survey items to determine whether combinations of related items can account for significant amounts of variance in the total survey scores. This procedure assumes no a priori relationships among sub-sets of items. Instead, clusters of related items are identified only if they bear an empirical relationship to one another. Table 4 lists the twenty-eight factors that are able to account for at least one percent of the variance in total survey scores. 52 Table 4. Variance associated with factors accounting for at least 1% of the variance Item # % Var. Cum. % 1. 26.0 26.0 2. 6.4 32.4 3. 5.0 37.5 4. 4.4 41.9 5. 4.3 46.2 6. 3.4 49.6 7. 3.2 52.8 8. 2.9 55.7 9. 2.8 58.5 10. 2.7 61.1 11. 2.6 63.8 12. 2.5 66.2 13. 2.2 68.4 14. 2.1 70.5 15. 2.0 72.6 16. 1.9 74.5 17. 1.8 76.3 18. 1.7 78.0 19. 1.5 79.5 20. 1.5 81.0 21. 1.4 82.4 22. 1.3 83.7 23. 1.2 84.9 24. 1.2 86.2 25. 1.2 87.3 26. 1.1 88.4 27. 1.0 89.5 28. 1.0 90.4 The above table suggests that there are not a few distinct clusters that are capable of accounting for major blocks of variance*within the surveyu Rather, the variance is widely dispersed. This finding lends credence to a conclusion that the polarity preference trait is uni-dimensional rather than multi-dimensional. Despite this initial finding, a decision was made to go ahead and examine the items that had the highest loading on the first few factors to see if clear patterns emerged. 53 Interpretation of the factor analysis results ignored the first factor as containing common elements having to do with the survey format, general subject matter, and other unidentifiable issues. A total of six factors were extracted. The twenty-one items that loaded higher on factor #2 than on any other factor (with the exception of factor #1) are shown in Table 5. Also included in this table are the correlation coefficients between each item and respondents’ total scores. Table 5. Factor loadings and correlations for survey items in factor #2. Item Loading Corr. 1. Sensible/Imaginative -.48* .39 2. Organized/Adaptable -.63* .39 4. Principles/Emotions -.38* .44 8. Evidence/Hunch -.46* .33 11. Clear Reason/Strong Feeling -.48* .46 12. Value Judgements/Logical Judge. -.28 .59 25. Introvert/Extrovert .19 .39 26. Excitement/Tranquility .12 .56 27. Approval/Independence .29 .54 29. Quality/Quantity .35* .51 30. Thinking/Doing .25 .61 34. Planning/Producing .33* .67 35. Innovating/Coordinating . .34* .56 37. Talking/Listening .29 .48 39. Passive/Aggressive -.21 .49 40. Fiction/Non fiction -.25 .48 42. Government Control/Indiv. Rights .26 .43 43. Exercise/Relaxation .11 .52 44. Literal/Figurative .18 .67 48. Enthusiastic/Peaceful -.25 .47 49. Saving/Giving .23 .56 * : value greater than .30 54 The eight items in the above table that have a factor loading of .30 or greater are followed by an asterisk. Examination of these eight items suggests a common theme that may' be expressed along’ a continuum! of structure ‘versus flexibility. The structure pole contains the terms sensible, organized, principles, evidence clear reason, quality, planning, and coordinating. The flexibility pole contains the terms imaginative, adaptable, emotions, hunch, strong feeling, quantity, producing, and innovating. Items twenty-nine (quality/quantity) and.thirty-four (planning/producing) don’t seem to adhere to this conceptual scheme as clearly as the other items. The ten items that loaded.higher on factor #3 than on any other factor (with the exception of factor #1) are shown in Table 6. Also included in this table are the correlation coefficients between each itemiand respondent’s total scores. Table 6. Factor loadings and correlations for survey items in factor #3. Item Loading Corr. 5. Head/Heart —.37 .53 7. Actual/Possible .37 .43 13. Passion/Commitment —.35 .54 16. Individual Effort/Teamwork .33 .43 18. Firm/Gentle -.46 .59 19. Democrat/Republican -.46 .46 24. Results/Plans .38 .60 31. Union/Management -.54 .41 46. Challenge/Support .45 .66 47. Loyalty/Expertise .36 .62 All ten of the items in Table 6 have a factor loading 55 greater than .30. These items appear to be very diverse. There doesn’t appear to be a simple cluster description that fits most of the items as was the case in the previous cluster. The first item (Head/Heart) in this cluster appears, on its face, to be closely related to the structure/flexibility cluster discussed earlier. The nine items that loaded higher on factor #4 than on any other factor (with the exception of factor #1) are shown in Table 7. Also included in this table are the correlation coefficients between each itemland respondent’s total scores. Table 7. Factor loadings and correlations for survey items in factor #4. Item Loading Corr. 6. Planning/Impulse -.46* .40 10. Socialism/Capitalism .46* .34 14. Deliberate/Spontaneous -.45* .48 17. Cooperation/Competition _ _.22 .46 20. Art/Science .39* .51 28. Liberal/Conservative .28 .53 33. Clear Direction/Flex. Guidelines -.31* .58 41. Laws/Choices .50* .51 45. Absolute/Tentative -.24 .61 * = loading greater than .30 The six items with an asterisk after the factor loadings in the above table have a loading of greater than .30. A description for this cluster might be rules versus choices. The terms planning, deliberate, science, clear direction, and laws align with the rules end of the continuum. Impulse, spontaneous, art, flexible guidelines, and choices seem to describe the choices end of the continuum. It does not appear 56 that the socialism/capitalism pair fits this scheme. An additional difficulty is that the description “rules versus choices“ seems barely distinguishable from the description “structure versus flexibility“ used in factor #2. The nine items that loaded higher on factor #5 than on any other factor, with the exception of factor #1, are shown in Table 8. Also included in this table are the correlation coefficients between each itemiand respondent’s total scores. Table 8. Factor loadings and correlations for survey items in factor #5. Item. Loading Corr. 3. Speculative/Realistic -.29 .39 9. Punctual/Leisurely -.36* .44 15. Guiding Father/Nurturing Mother -.38* .48 21. Adventures/Traditions .35* .57 22. Abstract/Concrete .26 .46 23. Logic/Intuition .43* .47 32. High Quality/Low Cost .34* .48 36. Masculinity/Femininity -.34* .41 38. Intense/Relaxed .37* .52 * = loading greater than .30 The seven items with an asterisk after the factor loading in Table 8 have factor loadings greater than .30. A cluster scheme that may be used to describe some of the items in this factor is active versus passive. The terms related to the passive side are leisurely, nurturing mother, traditions, femininity, and relaxed. The terms punctual, guiding father, adventures, masculinity, and intense appear to fit the active side of the continuum. It does not appear that logic/intuition nor high quality/low cost fit such a scheme. 57 This cluster has the lowest factor loadings of the first five factors and two of the seven items don’t seem to fit the scheme. The weakness of the five factors examined, coupled with the absence of clear, unifying cluster themes leads to the conclusion that the trait is uni-dimensional rather than multi —dimensional . SECOND SURVEY Since the trait seems to be uni-dimensional, factors were judged to be no longer relevant. The decision to review the items and keep only those with the highest correlations between item scores and total survey scores was made. Additionally, duplicate items and items that appeared to be possibly ambiguous were deleted from the survey. Of the original forty-nine items, there were twenty-four that had a correlation of .50 or greater with the total score. Those items, with their respective correlation coefficients are listed in Table 9. 58 Table 9. Survey items that had a correlation of .50 or greater with total score. Item Correlation Head - Heart .53 Value Judgements - Logical Judge. .59 Passion - Commitment .54 Firm.- Gentle .59 Art - Science .51 Adventures - Traditions .57 Abstract - Concrete .56 Saving - Giving .56 Results - Plans .60 Excitement - Tranquility .56 Approval - Independence .54 Liberal - Conservative .53 Quality - Quantity .51 Thinking - Doing .61 Clear Direction - Flex. Guidelines .58 Planning - Producing .67 Innovating - Coordinating .56 Intense - Relaxed .52 Laws - Choices .51 Exercise - Relaxation .52 Literal - Figurative .67 Absolute - Tentative .61 Challenge - Support .66 Loyalty - Expertise .62 Respondents were trained in'polarity theory following administration of the instrument. After they received the training and fully understood the theory, some of the respondents made suggestions regarding improvement of the instrument. Those suggestions resulted in the removal of three items from the survey. Passion - Commitment was dropped for reasons of ambiguity. One interpretation of this item could be “Are you doing what you are doing because you feel passionate about it or are you doing it because you made a commitment to someone?“ Under that interpretation, the item would represent a polarity. An alternative interpretation of 59 commitment involves personal commitment in which one is personally committed to a course of action because one does, indeed, feel passionate about that course of action. Quality - Quantity was deleted because of fairly recent changes in the*way quality is viewed” Experts, such as Philip Crosby (1979), in the field of quality improvement would suggest that the best way to obtain sufficient quantities of acceptable product is to improve quality thereby eliminating the need for costly product inspection and repair activities. In this case, improved quality is a pathway to improved quantity and not an apparent opposite. Loyalty - Expertise was deleted because it can be argued that one can be both loyal and expert and that they are not opposites. The reason they were initially included was from the frame of reference that one could either reward loyalty (longevity) or expertise (productivity). Results - Plans, Planning - Producing, and Thinking - Doing were experienced by respondents as being very closely related. The last two were seen as particularly close in meaning. Since the Thinking - Doing pair had the lower correlation coefficient of the two, a decision was made to delete that item. Of the remaining twenty items, there are a number that are arguably close in meaning. The goal was to produce a survey that was valid, reliable, and short enough to be used efficiently in self-discovery learning situations. Whether 60 the survey ends up being twenty items long or fifteen items long is not important. At twenty items in length, it is certainly not unduly burdensome to complete. The revised survey of twenty items was administered to seventy-three respondents. Survey administration occurred, once again, in a classroom setting involving students taking credit classes in the Master of Business Administration program. Demographic data on age, sex, and occupation was collected. Forty-two of the respondents were females, twenty- nine were males, and two respondents failed to indicate their gender. The mean age of the respondents was 22.7 years and the standard deviation for age was 2.13. The average age for this second survey administration is significantly lower that for the earlier administration.because the earlier administration utilized subjects from the College of Education as sell as from the College of Business whereas the second administration utilized subjects from the College of Business only. The average age of graduate students in Business is lower than the average age for graduate students in Education at The University of Toledo. Table 10 describes the occupations of the survey respondents. 61 Table 10 Occupations of respondents (second survey) Occupation Frequency Retail Sales 11 Health Care 1 Full Time Student 33 Teacher 1 Law Enforcement 2 Self Employed 2 Business and Industry 16 No Occupation Given 7 Scores from the odd numbered items were totalled.as were scores from the even numbered items. A Pearson product moment correlation coefficient was calculated to determine the extent of agreement between the odd and even sub-scores. The correlation coefficient of .8460 indicates a high level of agreement between the two halves of the instrument as was the case in the initial survey. A t-test was performed to determine whether a significant difference could be found between the scores of males and females. No significant difference was found” The results of the t-test are displayed in Table 11. Table 11. Comparison of score between males and females Mean Standard Deviation Males 39.38 15.16 Females 44.33 16.17 t = 1.301 (p = 0.197) .A Pearson product moment correlation was performed to test for a relationship between age and total score on the 62 survey instrument. The correlation coefficient of .0805 (p = .5074) indicates no significant relationship between age and total score. The final analysis of the data collected in the second survey, as with the initial survey, involved examining the correlations between each individual item and the total score. This analysis was accomplished in order to further refine the instrument in terms of its ability to measure the polarity construct with minimal contamination from non-related variables. Table 12 lists the correlation coefficients between each of the individual survey items and the total score from all twenty items. The total scores contain numerous sources of variability, only one of which relates to respondent preferences to simultaneously value apparent opposites. The square of any individual item correlation represents the percent of total score variability that can be explained by that item. Therefore, items that correlate highly with the total score can be said to accurately predict the total score. For example, the item “Challenge - Support“ with a correlation of‘ .60 accounts for thirty-six percent (.60 times .60) of the total score variance. 63 Table 12 Correlations between individual items and total score Item. Correlation Head - Heard .52 Value Judge. - Logical Judge. .61 Firm - Gentle .47 Art - Science .32 Adventures - Traditions .50 Abstract - Concrete .56 Results - Plans .54 Excitement - Tranquility .43 Approval - Independence .63 Liberal - Conservative .57 Clear Direction - Flexible Guidelines .54 Planning - Producing .66 Innovating - Coordinating .57 Intense - Relaxed .50 Laws - Choices .63 Exercise - Relaxation .64 Literal - Figurative .55 Absolute - Tentative .51 Challenge - Support .60 Saving - Giving .59 It can be seen from the above table that all but three of the items have a correlation with total score of .50 or greater and none of the correlations are below .30. SUMMARY OF RESULTS The goal of the study was to develop a survey instrument that would validly and reliably measure subjects' preferences to simultaneously hold competing values. Additionally, it was hoped that the resulting survey would be simple enough to allOW' for self-administration and self-interpretation to maximize its application in self-discovery learning. The survey format used in the present study allows for self- administration and self-interpretation when accompanied by 64 normative data for comparison.by the learner; The findings of this study indicate that the survey instrument assesses, validly and reliably, the preferences in question and that further study is needed to continue the validation process. Content validity for the instrument was enhanced by using managers who had. participated in polarity workshops to generate potential survey items based on actual polarities they must manage in the work place. It was further enhanced through review by an expert in polarity theory. Preliminary reliability considerations were addressed through the use of a split-half procedure. Construct validity was established by comparing independent ratings of a panel of experts to scores obtained from.survey results. No significant age nor gender effects were found. Factor analysis suggests that the trait under examination is primarily uni-dimensional rather than multi-dimensional. No factor, other than the first, general factor, accounted for more than 6.4 percent of the total variance. Factor clusters lacked clear, unifying characteristics and, in some cases, an item in one factor appeared to be more closely conceptually related to a different factor than the one in which it was found. CHAPTER V CONCLUSIONS. RECOMMENDATIONS, AND IMPLICATIONS INTRODUCTION The purpose of this study was to develop an instrument that would validly and reliably assess the preferences of respondents to simultaneously hold competing values. The Polarity Theory developed by Dr. Barry Johnson suggests that problem solving approaches are often used unsuccessfully to guide decision making. He suggests that many of the situations to which problem solving techniques are applied are not problems to be solved in the traditional sense. He advocates making a distinction between problems to be solved and polarities or dilemmas to be managed. Traditional problem solving techniques are appropriately applied to situations containing mutually exclusive alternatives but such techniques are not appropriate for application in situations containing mutually dependent, and apparently opposite alternatives. The approaches advocated by Johnson for the management of polarities can be taught and applied. More than one thousand workshop participants have benefitted from.such training that was delivered. by The ‘University’ of Toledo, Division of Continuing Education. Participants have included supervisors, middle managers, and executives from Fortune 500 companies. It has been determined by workshop participants, and The University of Toledo Continuing Education staff concurs, that a survey instrument which would assess the extent to which 65 66 workshop participants are able to simultaneously recognize the legitimacy of competing values, would be a useful tool for the training and application of Polarity Theory. No such instrument was available prior to this study. A difficult problem in the development of this instrument was in the development of a working definition of a polarity. Johnson defines a polarity as a set of apparent opposites that need each other. After providing that brief definition, he gives some examples to help in understanding his theory. These examples include such polarities as inhaling/exhaling, and general/precise. Although these examples may be clear, extending the definition to a broad range of word-pairs results, sooner or later, in ambiguity. Does the word-pair masculine/feminine constitute a polarity or can one choose to exhibit, throughout one's entire life only feminine behaviors and exclude all masculine behaviors? Johnson would say that such a choice would lead to a poorly managed polarity because he would argue that masculine/feminine is indeed a polarity. He would suggest that there are positive and negative aspects to both masculinity and femininity. Further, choosing one to the total exclusion of the other would invariably lead to system dysfunction. It could be argued, on the other hand, that this is not a polarity at all and that one can indeed make a mutually exclusive choice. There is no way to objectively settle the argument. This study was undertaken, despite this difficulty, because the results of the study 67 contribute to the understanding of the underlying construct and thus helps to clarify and operationalize the definition of a polarity. 68 CONCLUSIONS A.major question in this study revolved around the issue of whether the preference trait in question was uni- dimensional or mmlti-dimensional. If the trait was multi- dimensional, various sub-sets of polarity preferences would be reflected in the total score and survey respondents would score differently on some clusters of items than on other clusters. Perhaps there would be a cluster of items relating to work and a separate, distinct cluster of items relating to family. Respondents, given a sizeable sample, would be expected to score somewhat differently on these two clusters of items. If the trait was uni-dimensional, there would.be no identifiable sub-sets of polarity preferences. Factor analysis failed to reveal distinct clusters. One cannot unequivocally conclude, based on this single study, that such clusters are non-existent. The present study made use of college students for subjects. Existence of clusters could become evident with a sample of different characteristics. A different response protocol might also yield a different picture with respect to clusters. What can be concluded from this study is that clusters were not found. The fact that a panel of individuals who are expert in polarity theory was able to predict, with perfect agreement, whether the scores of three respondents, known personally to the panel members, would.be in high, medium, or low ranges is important. This procedure, although admittedly carried out 69 with a small sample, gives strong evidence that the survey instrument does in fact measure the preference trait in question. Without this procedure, one could claim that the survey measures something but would have to rely strictly on content validity to make a claim that the “something“ was the trait in question. The procedure involving the panel added another important element in addition to construct validity. It could be argued that a preference to simultaneously value polar opposites has no identifiable behavioral manifestations. If this were the case, learning about one's preferences would perhaps satisfy one's curiosity but would lead to no useful outcomes because one would not be able to use the information for personal growth. This would be analogous to learning the precise weight of one’s brain. What does one do with such information? If, on the other hand, preferences to simultaneously value polar opposites has clear behavioral manifestations, knowledge of one's preferences and the corresponding behaviors can serve as the basis for self- directed learning and personal growth. Quinn (1989) suggests that self-improvement can be based on a three step process of learning about oneself, developing a change strategy, and implementing the strategy. This assumes that what is learned about oneself has behavioral manifestations that, unlike brain weight, can be changed. As far as polarity preference is concerned, one can 70 develop an appreciation for different sets of preferences and proceed with an action plan to change one's behaviors. In this study, panel members used their personal knowledge of the behaviors of each of the three respondents as a basis for predicting respondent scores on the instrument. The ability of panel members to predict survey scores based on observed behavior indicates that the preference trait has clear behavioral manifestations. It has been mentioned in earlier chapters that in order for a polarities preference instrument to have maximum utility for self-directed learning, it ought to be capable of self- administration and self-interpretation. The self- administration capability was demonstrated by directing respondents to read the instructions and complete the survey. Respondents reported no ambiguity nor misunderstanding of the survey instructions or format. The survey does not require expert assistance for administration. Self-interpretation refers to the idea that no expert assistance should be required for a respondent to score the instrument and to determine the meaning of the resulting score. The issue of scoring is a simple one. All that is required for scoring is to add the scores of each of the individual items to arrive at a total score. Self- interpretation of that score requires a comparison of the total score to the percentiles shown in Table 13. 71 Table 13. Total score percentiles Total Score Percentile 22 1 24 5 25 10 26 15 29 20 30 25 31 30 34 35 36 40 41 45 43.5 50 46 55 47 6O 51 65 52 70 57 75 S8 80 63 ' 85 65 90 68 95 91 99 The percentile associated with a given score locates that score in terms of the normative population. The instrument is relatively easy to self-interpret however, respondents need to have received training or become self-trained in polarity theory in order to understand what it is that is being measured. An additional interpretation requirement for valid population percentiles is a larger sample that is demonstrably representative of the general population. Such broad sampling to produce general population norms is beyond the scope of this study. 72 RECOMMENDATIONS FOR FURTHER RESEARCH This study indicates that the survey instrument has construct validity but additional studies are required to strengthen such a claim. One method of further defining the construct of polarity and, at the same time, gaining additional knowledge regarding construct validity’ of the instrument would be to compare survey scores with scores on the Myers-Briggs Type Indicator. It seems reasonable to speculate that a person who is able to simultaneously value polar opposites does not have a strong need for psychological closure. Such a person would be able to remain continually open to new experience that would help to illuminate issues. If speculation of this nature has a basis in fact, a person who scores low in the instrument developed in this study (high preference for valuing polar opposites) should score high on the Perceiving end of the Judging/Perceiving scale of the Myers-Briggs instrument. One might also speculate that a person who simultaneously values polar opposites relies more on intuition than on data. The basis for such speculation derives from the assumption that data is used to make choices and polarities, by their very' nature, are not amenable to clear choices. This speculation could be assessed by comparing scores on this instrument with scores for the Sensing/Intuiting scale of the Myers-Briggs instrument. Persons who score high on the Intuiting dimension should have low scores (high preference 73 for valuing polar opposites) on the instrument developed in this study. If the relationships between the Myers-Briggs scores and the scores from the polarities instrument supported the speculation, construct definitions for the two Myers-Briggs scales would help to further define and operationalize the construct upon. which. Johnson's Polarity Theory and. this instrument are based. Additional research would illuminate relationships between survey scores and demographic variables that were beyond the scope of the present study. Educational level would be of particular interest. If, as is often claimed, education is a broadening experience, persons with more education ought to be able to simultaneously value polar opposites more readily than those with less education. .Although age and gender were considered in the present study, additional work needs to be done on these demographic variables. The issue of whether the preference trait is uni- dimensional or multi-dimensional may be related to gender. An investigation of interaction between educational level and cluster analysis could also be carried out. These sorts of analyses ought to be considered with alternative response protocols. Such.protocols might involve interviews or simply asking respondents to judge whether they tend to see themselves as people who align with causes or who, seeing both sides of issues, are peace makers. 74 The final recommendation for further research suggests administration of the survey as a pre-test to training in Polarity Theory followed by a post-test administration. Given the apparent uni-dimensional nature of the preference trait, parallel forms of the instrument could be developed for use in such a design. Immediate post-test scores could also be compared with delayed post-test scores. If delayed post-test scores were significantly lower than pre-test scores, it would suggest that the preference trait could be affected through training. 75 IMPLICATIONS FOR ADULT LEARNING The polarity model, as a whole, has important implications for practitioners of adult learning. Andragogy and pedagogy both have their upsides and downsides. If a system moves away from pedagogy and toward andragogy, such movement should not be based upon the belief that pedagogy is a “problem! foerhich.andragogy'is the “solution“ in any final sense. If a system supports andragogy, to the exclusion of pedagogy, the system will eventually experience the downside of andragogy. It is currently fashionable to be crusaders for andragogy and the crusade may we warranted. If so, the crusade is warranted not because pedagogy was a mistake but because adult learning practitioners have clung for too long to the pedagogical pole of the polarity. Successful management of this polarity will require moving back and forth between pedagogical and andragogical approaches so that the upsides of both may be experienced and the downsides of both may be avoided. Management of this polarity will not be easy. Crusaders for andragogy fear the downside of pedagogy and want the benefits of andragogy. Those supporting the pedagogical tradition fear the downside of andragogy and don't want to lose the benefits of pedagogical approaches. Successful management of this polarity will require the recognition that each side is in a perfect position to give the other side early warnings regarding the need to move the system toward 76 the other pole. A manager of an adult learning enterprise must keep the two sides talking in order to assure that the enterprise manages the polarity effectively. Looking beyond the general applicability of Polarity Theory to the field of adult learning, the instrument developed in this study may illuminate a particular polarity of importance to the field. That polarity may be considered as integration versus differentiation. One who has a preference to simultaneously value polar opposites is able to integrate those opposites into a meaningful whole. On the other hand, one who has a preference to value only one of the polar opposites is able to clearly differentiate the opposites. Both integration and differentiation are useful. Without integration, it is impossible to manage diversity. Without differentiation, diversity cannot be valued. Figure 3 shows the upsides and downsides of the integration- differentiation poles. 77 Positive l Positive Clear focus on part I Clear focus on whole values uniqueness I Values connectedness Competitive challenge I Collaborative challenge Differentiation -------- I ---------- Integration Unclear about whole I Unclear about part Devalues connectednessi Devalues uniqueness Leads to win/lose l Devalues competition Negative I Negative Figure 4. Differentiation - Integration polarity This study has shown that learners have preferences for differentiation or for integration and that those preferences can be validly and reliably measured. Those that score low on the survey have a preference to simmltaneously value polar opposites and are integrators. Those that score high on the survey have a tendency, given polar opposites, to prefer one pole over the other and are differentiators. Given a classroom of learners, the adult educator will probably find that the room contains integrators and differentiators, for either pole, no matter what pair of competing values is being considered. For example, if the class is considering the relative merits of capitalism and socialism, the room will probably contain some members who strongly favor socialism, some who strongly favor capitalism, and some who are able to see some merit in each of the positions. If the adult educator is aware of the integrating and 78 differentiating forces in the room, that educator can use the differentiators for capitalism to describe the negatives of socialism. and the positives of capitalisml The differentiators for socialism can be used to describe the negatives of capitalism and the positives of socialism. The integrators can be used to facilitate discussion between the two opposing groups. This provides a :model for group empowerment in which the learners can direct their own learning process to achieve understanding instead of engaging in a debate to achieve a victory for one side over the other. Understanding polarity theory and having an instrument to assess preferences to hold competing values can help the educator of adults to maintain or enhance learner self-esteem in the learning environment. Without polarity theory, differentiators, those who tend to take extreme positions, are often described in pejorative terms such as rigid, radical, and unreasonable. Integrators, those who tend to simultaneously value opposites, are similarly described in pejorative terms such as vacillating, wishy-washy, and indecisive. The instrument allows learners to determine whether they tend to be differentiators or integrators and the theory provides a sound basis for valuing who they are irrespective of their style. Differentiators provide a valuable service in illuminating the quadrants of a polarity being examined and integrators can facilitate dialogue between those who differentiate each of the polar opposites. APPENDICES 79 Appendix A. Initial Survey Instructions Circle one number on each line indicating which termiyou.value more and.by'howrmucht Circle only one number for each.pair of terms. Example: Easy going 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - 5 Serious If you value being easy going over being serious but you value them almost the same, you would circle “1“ to the left of the “0“. If you thought that being easy going is just plain lazy and placed more value on being easy going, you would circle “5“ to the right of the “0“. If you place equal value on being easy going and on being serious you would circle the '0'. CIRCLE ONLY ONE NUMBER FOR EACH PAIR OF ITEMS. 1. Sensible Imaginative 5 - 4 - 3 - 2 — 1 - 0 - 1 - 2 - 3 - 4 - 5 2. Organized Adaptable 3. Speculative Realistic 4. Principles Emotions 5. Head Heart 6. Planning Impulse 7. Actual Possible 8. Evidence Hunch 9. Punctual Leisurely 10. Socialism Capitalism 11. Clear Reason Strong Feelings 12. value Judgements Logical Judgements 13. Passion Commitment 14. Deliberate Spontaneous 15. Guiding Father Nurturing Mother 16. Individual Effort Teamwork 17. Cooperation Competition 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 80 Firm Democrat Art Adventures Abstract Logic Results Introvert Excitement Approval Liberal Quality Thinking Union High Quality Clear Direction Planning Innovating Masculinity Talking Intense Passive Fiction Laws Government Control Exercise Gentle Republican Science Traditions Concrete Intuition Plans Extravert Tranquility Independence Conservative Quantity Doing Management Low Cost Flexible Guidelines Producing Coordinating Femininity Listening Relaxed Aggressive Non-fiction Choices Individual Rights Relaxation 44. 45. 46. 47. 48. 49. Literal Absolute Challenge Loyalty Enthusiastic Saving 81 Figurative Tentative Support Expertise Peaceful Giving 82 APPENDIX B. SECOND SURVEY AGE SEX __M _F OCCUPATION Circle one number on each line indicating which term you value more and by how much. CIRCLE ONLY ONE NUMBER FOR EACH PAIR OF TERMS. Example: Easy going 5*- 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - 5 Serious If you value being easy going over being serious but you value them almost the same, you would circle “1“ to the left of the “0“. If you thought that being easy going is just plain lazy and placed no value on being easy going; you would circle “5“ to the right of the “0“. If you place equal value on being easy going and on being serious you would circle the “0“. 1. Head Heart 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - S 2. Value Judgements Logical Judgements 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - S 3. Firm Gentle 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - 5 4. Art Science 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - 5 5. Adventures Traditions 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - 5 6. Abstract Concrete 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - 5 7. Results Plans 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - 5 8. Excitement Tranquility 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - 5 9. Approval Independence 5 - - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - 5 10. Liberal ' Conservative 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - 5 11. Clear Direction Flexible Guidelines 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - 5 12. Planning . Producing S - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - 5 13. Innovating Coordinating 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - S 14. Intense Relaxed 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 -,4 -'5 15. Laws Choices 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - 5 16. Exercise Relaxation 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - l 17. Literal Figurative S - 4 - 3 - 2 - 1 - 0 - l - 2 - 3 - 4 - 5 18. Absolute Tentative 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - 5 19. Challenge Support 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - 5 20 Saving Giving 5 - 4 - 3 - 2 - 1 - 0 - 1 - 2 - 3 - 4 - 5 BIBLIOGRAPHY 83 BIBLIOGRAPHY Adams, F2 1976. “Highlander Folk School: Getting Information, Going Back and Teaching It.“ In Education, Participation, and Power: Essays in Theory and Practice. Cambridge, MA: Harvard Reprint Series. Anderson, D. R. et. al. 1987. Statistics For Business and Economics. St. Paul, MN: West Publishing Co. Babbie, E. 1973. Survey Research Methods. Belmont, CA: Westwood Publishing Co, Inc. Barker, J. 1989. Discovering the Future: The Business of Paradigms. St. Paul, MN: ILI Press. Black, R. R. and Mouton, S. 1964. The Management Grid: Key Orientations for Achieving Productivity Through People. Houston, TX: Gulf Publishing. Bennis, W. 1989. Why Leaders Can't Lead. San Francisco, CA: Jossey-Bass Publishers. Carkhuff, R. R. 1988. The Age of the New Capitalism. Amherst, MA: Human Resource Development Press. Coblenz, J. J. 1990. “Beyond-Personality Inventory: Archetypes of change-agent Style.“ In The 1990 Annual: Developing Human Resources, Pfeiffer, J. (Ed.). San Diego, CA: University Associates. Crosby, P. 1979, Quality Is Free, New York,NY: McGraw Hill. Dillon, W; R. and Goldstein, M. 1984. Multivariate Analysis Methods and Applications. New York, NY: John Wiley and Sons. Duke, D. L. 1987. School Leadership and Institutional Improvement. New York, NY: Random House. Ellis, A. 1962. Reason and Emotion In Psychotherapy. New York, NY: Lyle Stuart and Citadel Press. Ellis, A. 1971/ Growth Through Reason: Verbatim Cases in Rational-emotive Therapy. Palo Alto, CA: Science and Behavior Books. 84 Evans, M. 1983. “Academic Evaluation In Individualized Education: Problems and Strategies“ (Continuum Vol. XLVII no. 4) National University Continuing Education Association, Washington D. C. Freire, In 1982. “Creating Alternative Research Methods: Learning to do it by Doing it: In Hall, Budd, Gillette, Arthur, and Tandon, Rajesh, (Eds.) Creating Knowledge: A. MOnopoly? New Delhi, India: Society for Participatory Research in Asia. Hall, B. 1982. “Breaking the Monopoly of Knowledge: Research Methods, Participation, and Development“ In Hall, Budd, Gillette, Arthur, and Tandon, Rajesh, (Eds.) Creating Knowledge: A Monopoly? New Delhi, India: Society for Participatory Research in Asia. Hawking, S. W. 1988. A Brief History of Time. New York, NY: Bantam Books. Johnson, B. 1992. Polarity Theory. Amherst, MA: Human Resource Development Press. Kanter, R. M. 1983. The Change Masters: Innovation for Productivity in the American Mode. New York, NY: Simon and Schuster. Keirsey, D. and Bates, M. 1984. Please Understand Me. Del Mar, CA: Gnosology Books. Knowles, M. S. 1989. The Making of an Adult Educator. San Francisco, CA: Jossey-Bass Kolb, D. 1985. Learning-Style Inventory. Boston, MA: McBer and Co. Lamy, S. L. 1991. “Global Education: A Conflict of Images“ In Tye, Kenneth A. (Ed.), Global Education: From. Trujt to Action. Alexandria, VA: Association for Supervision and Curriculum Development. Longstreet , W. S . 197 9 . “Open Education - A Coming to Terms with Uncertainty“ in Overley, N. v. (Ed.), Lifelong Learning: a Human Agenda . Alexandria , VA: Association for Supervision and Curriculum Development. Manz, C. C. and. Manz, K. P. 1991. Strategies for Facilitating Self—Directed Learning: A Process for Enhancing Human Resource Development. (Human Resource Development Quarterly, Vol. 2, No. 1.) San Francisco, CA: Jossey Bass Inc. 85 Maruyama, M. 1976. “Toward Cultural Symbiosis.“ In Evolution and Consciousness: Human Systems in Transition, Jantsch, E. and Waddington, C. H. (Eds.). Reading, MA: Addison-Wesley. McPherson, R. B., Crowson, R. L., and Pitner, N. J. 1985, Managing Uncertainty: Administrat ive Theory and Practice in Education. Columbus, OH: C. E. Merrill Publishing Co. Morton, S. C. (Ed.). 1990. The Corporation of the 19903. New York, NY: Oxford University Press. Naisbitt, J. 1982. Megatrends: Ten New Directions Transforming Our Lives. New York, NY: Warner Books. Oliver, J. E. 1985. “The Personal Value Statement: An experiential learning instrument.“ In the 1985 Annual: Developing Human Resources, L. Goodstein, and J. Pfeiffer, (Eds.). San Diego, CA: University Associates. Ouchi, W. G. 1981. Theory Z. New York, NY: Avon Books. Overly, N. V. (Ed.). 1979. Lifelong Learning: a Human Agenda. Alexandria, VA: Association for Supervision and Curriculum Development. Quinn, R. E. 1984. “Applying the Competing Values Approach to Leadership: Toward an integrative model.“ in J. G. -Hunt, R. Stewart, C. Schriesheim, and D. Hosking (1335.), Managers and Leaders: An International Perspective. New York, NY: Pergamon. Quinn, R. E. 1988. Beyond Rational Management. San Francisco, CA: Jossey-Bass Publishers. Quinn, R. E. and Cameron, K. S. (Eds.) 1988. Paradox and Transformation. Cambridge, MA: Ballinger Publishing. Quinn, R. E., Denison, D. R., and Hooijberg, R. 1990. Analyzing the Structure of Paradox: A test for the competing values model of leadership roles . Unpublished paper: University of Michigan, School of Business Administration. Rogers, C. 1961. On Becoming a Person. Boston, MA: Houghton Mifflin. Russo, J. E. and Schoemaker, P. J. 1989. Decision Traps. New York, NY: Doubleday / Currency. 86 Sims, H. P. and Gioia, D. A. 1986. The Thinking Organization. San Francisco, CA: Jossey-Bass. Siporin, M. and Gummer, B. 1988. “Lessons From Family Therapy. The Potential of Paradoxical Interventions in Therapy.“ in Paradox and Transformation, R. Quinn and K. Cameron, (Eds.). Cambridge, MA: Ballinger Publishing. Smith, K. and Berg, D. 1987. Paradoxes of Group Life. Understanding conflict, paralysis, and movement in group dynamics. San Francisco: Jossey-Bass. Stalk, G. and Hout, 12 FL 1991. Competing Against Time. New York, NY: Collier Macmillan Publishers. “Stress Processing Report“. 1988. Plymouth, MI: Human Synergistics. Thorn, M. B. 1984. Statistics For the Behavioral Sciences. Mountain View, CA: Mayfield Publishing Co. Toffler, A. 1970. Future Shock. New York, NY: Random House. Tway, D. C., Davis, L. N., and Mink, O. G. 1990. Reality Referenced Training: An Empirical Study of Self-Norming Needs Assessment. (Human Resource Development Quarterly Vol. 1, No. 4.) San Francisco, CA: Jossey- Bass Inc. Tye, B. B. 1991. “Schooling in America Today: Potential for Global Studies.“ in Tye, K. A. (Ed.), Global Education: From Thought to Action. Alexandria, VA: Association for Supervision and Curriculum Development. Varney, G. G. 1990. Building Productive Teams. San Francisco, CA: Jossey-Bass. Van de Ven, A. 1983. Review of “In Search of Excellence.“ Administrative Science Quarterly, 28: 621-24. Weick, K. 1976. Educational organizations as loosely coupled systems. Administrative Science Quarterly, 21: 1-19. Wolpe, J. 1969. The Practice of Behavior Therapy. New York, NY: Pergamon Press. MICHIGAN STATE UNIV. LIBRARIES WIHIIIIHHIII”IIIWIWWIHWilli“WWI 31293010298903