This chapter presents students and early career executives with a sound understanding of theory. Theory is explored in terms of both anatomy (parts of the whole) and physiology (relationships with each other) to better explore the complexity of theory itself before applying it to the study of leadership. It is difficult for students of leadership to fully embrace all the nuances of leadership study if they first do not understand what a theory is and how the study of leadership theory fits in with the general dynamics of applying specific leadership theories to the practical world. LEARNING OBJECTIVES Describe why the study of theory is important in the study of leadership in health organizations. Define and distinguish the basic elements and relationships of a theory. Demonstrate the utility of theory in the study of leadership, leadership principles, and leadership applications. Describe and compare two or more conceptual models, and discuss how the models relate to theory and support the discussion of leadership. Design a simple model of leadership principles or applications (from constructs and concepts in this chapter or other literature), and summarize the relationships between the model's theoretical elements and the application of leadership principles. Justify and defend the constructs in a simple model of leadership. WHY STUDY THEORY? Although the study of theory may seem nonessential and without practicality in a leadership course, you should ask yourself two simple questions: First, what is a theory? Second, what is leadership? If you can answer the second question without answering the first, there may be a gap in your understanding of the complexities of the art and science of leadershipand how leadership theory supports the development processes of early career executives. The authors of this text have more than 50 years of combined experience in health leadership positions in civilian and military health organizations and in academia. We have been educated academically and trained practically in the understanding, application, synthesis, and evaluation of leadership. As a result, we feel that it may not be possible to fully embrace all of the fundamentals of leadership and leadership theories without first understanding the complex relationships in theory structure and processes. Failing to understand theory may result in students of leadership not understanding what they are reading. More importantly, lack of theoretical methods knowledge could lead to misinterpretation, resulting in an inability to apply leadership knowledge, skills, and abilities in the organizational world. Consider one of the early leadership theories posited by scholars in the mid-1800ssomething called the "great man" theory. In essence, this theory suggested that to be a "great man," one had to emulate the specific "traits" of the acknowledged leaders of the time. There is overlap between great man and trait theories of leadership. What is missing is an understanding of what great means and how the specific traits define the word great for us to study. In reality, the great man theories of the early 1800s offered us no benchmark for success and no road map for others to follow. Essentially, the words great and trait were not universally understood or applied. As a result, the theory offered no utility for education or application. Simply put, the behavior of great leaders and their traits were observed and emulated without any attempt to understand the foundations of behaviors or influence of traits. Another example is the transformational leadership theory, which was introduced in 1978.1 This theory suggests that leaders can inspire followership by raising subordinate goals toward higher levels of motivation through developing a spirit of trust, respect, and loyalty. For example, John F. Kennedy's presidency is often characterized as inspiring the civil rights movement and a manifest destiny toward U.S. exploration of space. In support of this proposition, a janitor at NASA was once asked what he was doing cleaning a building after hours when he was not authorized to receive overtime pay for the work. His reply: "I am helping to put a man on the moon." President Kennedy was successful in transforming the perceptions of the janitor to view his job not only as a task, but also as one part of a greater visionthe U.S. quest for space exploration. Kennedy was an astute student of leadership. As a result, he was able to recognize the core values necessary to change individual behavior and form followership among the masses.2 Scholars of leadership theory can deconstruct Kennedy's natural abilities and provide tools, skills, and a road map for young scholars and early careerists to follow. Deconstructing "transformational theory" provides developmental insights that young executives can build on until their natural abilities associated with maturity and experience develop over time. Deconstructing theory and methods of theory building can greatly assist in this effort. Understanding the elements of a theory, models, and the components of theories and models is critical for analyzing, synthesizing, and evaluating which elements and components of a personal leadership model would be best for you in your career, and allow you to determine how to put them into practice. As former University of Alabama football coach Paul "Bear" Bryant said, "We beat most teams because we have a plan." You need a leadership plan (your model) as well to be the best you can be in your career. The rest of this chapter starts to build the foundation for you to develop your leadership plan, your model. WHAT IS A THEORY? Theory is the primordial soup from which complex questions can be modeled and discussed in bounded rationality where like-minded executives can agree on issues and causality. It is important to know that there is no one accepted definition of theory within the organization behavior literature. In layperson's terms, a theory is an advanced form of an idea or an opinion that has some basis in the empirical world. Regardless of the definition selected, a theory must be capable of support by qualitative measures or quantitative data. If a theory is incapable of initial development based on qualitative or quantitative properties, the burgeoning theory may not have evolved past the opinion or idea phase, and it may not be valuable to the profession or the advancement of knowledge. The term theory is derived from an ancient Greek word meaning to contemplate, to contemplate the divine, or to speculate. Simply put, a theory is a way to capture and represent a set of ideas, constructs, variables, and observations within a context to demonstrate how a part of the world works or could work better. A theory intends to evaluate variables that are operationally defined and measured in a dynamic world. Theories are analytical structures that seek to illustrate how linked ideas, domains, constructs, and variables perform under various conditions by using quantitative, qualitative, or combined methods. "A particular feature of science (including social science) is that it is continually evolving as a result of the scientific method which calls for a constant testing of ideas and observations of scientific facts, theories, and models."3 Theories start out as models that have been developed using empirical thinking. Qualitative information and analysis (also known as theory building methods) such as observation and literature review start this empirical process. Quantitative analysis (also known as theory testing methodology) tests models as hypotheses to determine if the model represents the world better than what was known before. This empirical hypothesis testing can be performed as qualitative research alone in certain cases, but may also be carried out in combination with quantitative methods (a process called triangulation). Definitions of the term theory support the notion that a theory provides for an integration of ideas within a context of a phenomenon. "A theory is a set of interrelated principles and definitions that present a systematic view of phenomena by specifying relationships among variables with the purpose of explaining natural phenomena."4,5 Another definition suggests that a theory is "any set of hypotheses or principles linked by logical or mathematical arguments which is advanced to explain an area of empirical reality or type of phenomenon."6,7 As discussed previously, the terms model and hypothesis nudge their way into the discussion; this is especially true in the study of leadership. Theory as a Conceptual Model In this section you will learn to visualize theory as a conceptual model. A conceptual model is a conceptual description of something abstract. It is not uncommon in business practices for executives to develop models to represent certain ideas and concepts. The model itself, like a photograph, breaks down barriers of communication, thereby making it easier for individuals to view the model and understand complex relationships. Models serve as representations of the world or phenomena around us; they are particularly useful for understanding, analyzing, and evaluating how the world works around us. Figure B-1 shows an example of a conceptual model for a hospital. In this basic model, hospital performance is achieved as a result of an organization's inputs and outputs. In this case, the inputs are constructs (to be discussed later) of hospital types, structural units, and the environment. The outputs in this model are the result of the previous three constructs combining to form some sort of recognized output. Finally, the performance of the hospital is based on the efficiency of those outputs being earmarked as high or low. If the output is high, there is no need for leadership to take action. If the output is low, the conceptual model guides leaders back to the relationship between constructs so that additional action can be taken. FIGURE B-1 A conceptual model. The basic conceptual model shown in Figure B-1 helps us to evaluate hospital efficiency in a manner that is complex, yet easily understood by outside agents and stakeholders. OVERVIEW OF THEORY This section presents a brief overview of theory and explores how a theory can be deconstructed into constructs, variables, and measures. Model building is a necessary precursor to performance-based management development, health system examination, policy formulation, and the conducting of quantitative analysis, to name only a few of its potential uses. Nevertheless, understanding the process of building empirically measurable healthcare and leadership models may be one of the more underrated aspects of leader development in today's healthcare system. This chapter places specific emphasis on creating conceptual models of leadership that can be used to measure the outcomes of various theories. Differentiating Between Theories and Models The terms theory and model are used interchangeably in much of the literature. A theory is a hypothesis that has undergone scientific and practical scrutiny, albeit at various levels of intensity, to determine its value, truth, and validity. A theory does not "just happen" quickly, but rather develops over time after significant scrutiny by multiple scholars, practitioners, and researchers has proven the model is worthy of theory status. A model is a simplified abstraction of reality; it has not yet met a level of academic and practical scrutiny for scientific validation, so its value has yet to be determined. Both theories and models require ecological validity to represent and present reality well, so that scholars and practitioners can study and use them. It is more difficult to prove value, truth, and validity in the leadership discipline than it is in a basic science such as chemistry or physics. Consequently, one could argue that leadership theories and models are really hypothesesthat is, speculations about how a phenomenon actually behaves or works: A [social] scientific theory is a synthesis of well-tested and verified hypotheses about some aspect of the world around us. When a [social] scientific hypothesis has been confirmed repeatedly by experiment, it may become known as a [social] scientific law or scientific principle.8 In the leadership literature, the terms theory and model are much more prevalent than the terms law and principle, mostly due to the social requirement and nature of leadership: People's beliefs, values, attitudes, and behaviors often change, and situations are dynamic. Going back to the structure of theory, constructs, and variables, an example can explain this social sciences-based methodology. Many leadership theories include interpersonal relationships as a category within their theories. Within the category of interpersonal relationships, one of the constructs is communication. As a construct, communication is broad and has many possible variables that can be measured. For example, conflict management style may be a variable within the construct of communication. Conflict management style, as a variable, can be measured using a survey to determine the leader's dominant style from six previously identified conflict management styles. Because a theory, to be useful, needs to adjust and work well in a dynamic world, recommending a conflict management style based on a particular situation would be more helpful than just relying on chance to pick the correct conflict management style out of six possibilities. Thus, in this example, a situational variable (observations from the situation) can interact with a variable from the interpersonal relationships category and communication construct, and specifically the conflict management style variable, to provide a recommended conflict management style. From a social sciences and academic perspective, this approach serves as a structure to facilitate studying, teaching, and learning about leadership. From an application viewpoint, however, the leadership theory in practice would be transformed into knowledge, skills, and abilities. Continuing with the same example, the practicing leader understands that interpersonal relationships are important and can describe and explain why this subject is important in general, but also based on the constructs included under this category. This is an example of knowledge. Focusing on the communication construct and the conflict management style variable, the practicing leader can identify the type of situation at hand and apply observations from the environment to the leadership theory to reveal a recommendation for the appropriate conflict management style to use in this situation. This is an example of ability. Even though the recommended conflict management style may not be the leader's dominant or preferred style, the leader uses the recommended style in this particular situation. This is an example of a skill. If the leader can use any of the six conflict management styles dictated by the situation as appropriate, he or she would be more skillful than a leader who can use just two or three conflict management styles. To increase the value of studying leadership, building your knowledge of theories and models is critical. Developing an ability to take apart leadership theories and models and separate them into their component domains, constructs, and variables is vital. Also, developing the ability to assess the situation or environment of leadership is important. Lastly, developing, refining, and maintaining skills associated with analyzing and using constructs and variables of leadership theories and models are paramount. As you study and master the knowledge and practice the abilities and skills of the various theories and models, your leadership acumen will improve. THEORIES, MODELS, CONSTRUCTS, VARIABLES, AND MEASUREMENTS History has recorded leaders' exploits for thousands of years. Anthropology, archeology, social anthropology, political science, business, communication, and other disciplines have all made valuable contributions to the basic foundations of leadership theory and practice. The health professions place leadership foundational theories and models and the practice of leadership into the complex environment of promoting health, preventing disease and injury, diagnosing and treating disease and injury, rehabilitating human bodies and functions, and facilitating dignity at the end of life. In this arena, leadership, as a complex topic, is coupled with a complex environment, the health industry, which creates a multilayered and integrative system in which the health leader must perform. To use leadership theories and models, it is important to understand the scholarly building blocks of these foundations and to extend or bridge scholarly and theoretical perspectives to the applied or practical use of leadership knowledge, skills, and abilities. Leadership is an interdisciplinary field of study of the social sciences. Leadership is a social phenomenon: It involves individuals, groups, and populations and focuses on how those people interact given the multitude of beliefs, values, attitudes, and behaviors in society. Leadership can be simple or very complex depending on how complicated the social environment appears to both the leaders and the led. As a science (in this case, a social science), leadership is explained, taught, learned, and documented in the literature. Within this knowledge base, theories and models combine domains, constructs, and variables that can be measured to describe, prescribe, or both describe and prescribe how to think about, practice, and evaluate leadership. Given that leadership knowledge comes from several disciplines, it is important to use social sciences-based methods to provide clarity to the study of leadership. That clarity is provided through a structure that uses theories, models, constructs, variables, and measurement as a common language and guide to leadership inquiry, practice, and evaluation. Anatomy of Theory The anatomy of theory can be broken down into specific units of analysisnamely, the theory itself, followed by subordinate constructs, variables, and operationalized measures. Surrounding these elements is the environment of discussion, an enclosure called bounded rationality. When discussing theoretical constructs, variables, and measures, it is first necessary to frame these elements within a plausible discussion group. By framing constructs, variables, and measures in a bounded rationality, an "out of bounds" area is revealed that helps researchers stay within certain parameters of discussion.9 Contained within this bounded rationality is the physiology of theory (discussed in the next section). The physiology of theory describes the interaction among constructs, variables, measures, and other elements. In this regard, the interaction of constructs within theory is helpful for developing propositional statements. This consideration is important in the early stages of qualifying theoretical relationships before quantitative data become available for testing or disconfirmation of the theory. Forming more concrete and testable relationships within the theory are the relationships between variables known as hypotheses. Propositional and hypothetical relationships are discussed in greater detail later in this chapter. Also contained within the bounded rationality of theory are contextual factors and confounders. Contextual factors are generally known elements that exist in the same environment as constructs, variables, and measures. The interaction of contextual factors on certain constructs and variables may be known in advance and can be controlled for through awareness and intervention. Confounders are properties in the environment that are generally not known in advance and may interact with theory to produce unanticipated effects. Figure B-2 depicts theory as a conceptual model for visual representation and understanding. A conceptual model is a "conceptual description" of the key elements of a phenomenon under study. The conceptual model should be parsimonious (simple) and offer graphic representations of theoretical elements that help outsiders understand the issue(s) being investigated at a glance. A conceptual model may include the actual theory, as well as constructs, variables, measures, confounders, and contextual factors. Some conceptual models, however, consider only specific constructs as well as certain elements specific to the unit of analysis under study. FIGURE B-2 Anatomy and physiology of a theory. Physiology of Theory The physiology of theory can be described in terms of the relationships within the theoretical model. Two of these relationships are expressed in terms of hypotheses and propositions (Figure B-3). A proposition is a statement of opinion, based on some degree of preliminary study or heuristics, which is offered as a true or valid statement. Although the statement may not always be true or valid, it is offered as such until evidence of disconfirmation is provided. The aphorism "time is endless" is an example of a propositional statement between constructs. Another classic example of a propositional statement combining constructs is "the right to bear arms." This propositional statement is offered as a statement of fact as if it were true; however, the interpretive nature of this proposition continues to be repeatedly addressed in the United States. FIGURE B-3 Conceptual model of theory physiology. A hypothesis is a testable relationship between two variables. The purpose of hypothesis testing is to discover causal relationships or associations between variables. Hypothesis statements are the foundation for all social sciences research and form the basis for the advancement of knowledge. The words model, hypothesis, and theory are each used quite differently in science. Their use in science is also quite different [than] in everyday language. To scientists the phrase "the theory of . . ." signals a particularly well-tested idea. A hypothesis is an idea or suggestion that has been put forward to explain a set of observations. It may be expressed in terms of a mathematical model. The model makes a number of predictions that can be tested in experiments. After many tests have been made, if the model can be refined to correctly describe the outcome of all experiments, it begins to have a greater status than a mere suggestion. Scientist do not use the term "the theory of . . ." except for those ideas that have been so thoroughly tested and developed that we know there is indeed some range of phenomena for which they give correct predictions every time. (But, language being flexible, scientists may use "a theory" as a synonym for "a hypothesis," so listen carefully.) Today, any set of scientific ideas referred to as "the theory of . . ." is a well-tested and well-established understanding of an underlying mechanism or process. Such a theory can never be proved to be that is why we no longer call it a "law." However, it is the same kind of well-tested set of rules, with an established area of applicability, as the older ideas called "laws."10 In essence, a theory is a representation of the world that has been confirmed to be reasonably true, valuable, and valid. Some theories (and some models) are better than others; some are very specific to particular situations, whereas others are broad or even universal. Theories and "those hoping to be theories" (i.e., models) represent an aspect of our world or a methodology to improve our world through a structured process grounded in language and approaches of the scientific method used by scholars and theorists. At the next lower level, theories (as well as models) integrate and combine constructs. Constructs The building blocks of theory are constructs. In our proposed conceptual model building, constructs are visualized as circles. It is critical to the study of organizational behavior to have a clear understanding of what a construct is. Failing to have a clear understanding of constructs will result in an inability to understand many important leadership concepts. By definition, a construct is a latent variable that lacks empiricism (taste, touch, see, smell, hear). Elements that are empirical have tangible, physical properties; for example, an apple possesses empiricism insofar as it can be tasted, touched, seen, and smelled. It does not matter that the apple does not make noise. Because an apple possesses attributes related to four of the five empirical senses, we can say that an apple is not a construct because it can be rationalized through empirical properties. Any physical element or property that can be described through at least one of the five senses does not qualify technically as a construct. Said another way, a construct is an organizing device that captures a topic, subject area, or smaller and specific theory or model within a larger theory or model. For example, communication is a construct within a larger leadership theory. Constructs organize and combine multiple variables that are closely linked into one grouping or subheading within the structure of the theory or model. The critical feature of constructs derives from the term closely linked. Constructs must make sense in relation to the real world; thus they must have ecological validity. Constructs, then, are the basic building blocks for grouping variables within a model or theory. It is commonplace to see the term concept associated with constructs. A concept is a method of organizing or categorizing an abstract topic or reality under a construct. A concept should be linked logically to the construct. A group of concepts under a construct must have cohesion that is logical or holds true to reality (ecological validity). Concepts, in essence, are either variables or constants. Variables and constants are located at the lowest level within the organization of the construct. Another classic example of a construct is the term quality. Quality is a construct that cannot be discussed without identifying it through other measurable properties or variables. The well-known statement, "Quality is in the eye of the beholder," generalizes the difficult problem we have describing quality. Other latent variables, such as health, love, happiness, pain, efficiency, effectiveness, performance, satisfaction, organizational survival, leadership, success, and motivation, are all examples of constructs. What makes constructs difficult to study is that different people may have different opinions of how to define them. Providers encounter this issue frequently when trying to understand a patient's level and tolerance for "pain." Pain is clearly a construct (or said another way, a personal concept and/or belief) that is difficult to communicate, may not be visible to another person, and clearly is perceived differently by different age groups. Although adults may have greater success in describing pain to a provider in regard to how it affects activities of daily living (ADLs), as well as frequency and duration of pain, children have less of a vocabulary and experience with describing pain. As a result, many pediatricians use a model called the pain rating scale (Figure B-4).11 FIGURE B-4 Pain rating scale. As you can see, the model describes the concept of pain using a 10-point scale. Associated with the scale are emoticons. Studies have shown that children as young as 4 years old can associate levels of happiness and sadness using emoticons in a manner more consistent with their peers than they can using ordinary language. This example shows how the use of modeling in health care is essential for standardizing relatively simple concepts (like pain) between a patient and a provider. Unfortunately, many other communication efforts dealing with constructs in leadership are not as easily modeled and accepted. As a result, it is often necessary for health leaders to develop standardized models and measures in order to communicate exact intent. Next we present a universal technique for communicating leadership constructs. by defining constructs with empirical properties we call variables and measures. Variables Flowing from constructs are variables. In model building we use rectangles to visually depict a variable. A variable is "a property associated with a concept that varies when measured."12 Variables are also empirical units that can be identified through one of the five senses. Accordingly, a variable is an element that has precise meaning in the physical world. Generally, variables are universally understood and easily described. Weight is a good example of a variable. When discussing weight with peers or colleagues, universally understood concepts of pounds, ounces, or tons are immediately recognized as valid descriptors of weight. If a variable is incapable of being defined with a generalized descriptor, it may cause problems in communication with leadership peers. Take, for example, a colleague who suggests that her new china cabinet is very "heavy." Based on this description alone, can we accurately describe the weight of the cabinet, or do we need more information? How about a provider who suggests that he just encountered a "large customer" at the pharmacy. "Large" might mean physically obese, tall, or perhaps even muscular. For this reason, variables by themselves are not adequate to communicate concepts in leadership and model building. The notion of "something that can be measured, and whose measurement can vary or change" is the essence of a variable. An example of a common variable in the health professions is patient body type, which is a variable that can be described by both height and weight. Height can be measured, yet varies by person, as does weight. In contrast, a constant is a property that "does not vary";13 it is unchanging. Variables are essential to leadership theory confirmation through model and hypothesis testing because they are the properties or items that are tested to see whether the theory is confirmed. The importance of variables is that they vary; they change so that empirical testing can be accomplished. If the model is full of constants that do not vary, how could you test whether the model is true to reality? Moreover, how could such a model help someone to be a better leader if everything stayed the same? Constants Constants are those concepts that do not vary; however, depending on what you are studying, a concept may be either a constant or a variable. An example would be the study of prostate cancer: Would gender be a variable or a constant in this study? Although gender is important, only males get prostate cancer. Given that gender would not vary, it is a constant in this example. High-risk pregnancy would be another example where gender would not vary, because only females would be included in the study. In contrast, in a population-based heart disease study, gender could vary, because each observation could be either male or female; thus gender would be a variable in this situation. The variablegender in this casewould be a nominal, mutually exclusive, and categorically exhaustive (you can be either male or female, but not both, and you have to be one or the other) variable that could vary. Many times the situation or context dictates whether a concept is a variable or a constant. Operationalization Each variable has an operational definition and an operational process for measurement. An operational definition is the exact description of the conceptual propertythe variableunder empirical study: An operational definition identifies one or more specific observable conditions [or characteristics] or events and then tells the researcher how to measure that event. Typically, there are several operational definition possibilities for variables and values. The operation chosen will often have an immediate impact on the course of the research, especially the findings.14 The operational process for measurement is important because it lists the steps in the measurement of the variable as it is operationally defined. Some variables are observed and measured with a tape measure or ruler (such as height); some are categorized (such as shirt size being small, medium, or large); still others use survey measurement (such as conflict management styles). This brings us to the important construct of measurement. Measures Derived from variables are operationalized measuresthat is, operationalized descriptions of variables that must be capable of numerical identification. Operationalizing is the process of quantifying a variable using appropriate, numerical, descriptive terms. Additionally, measures must be universally understood and are classically categorized as continuous (1 to n, with n equal to infinity), dichotomous or binary (e.g., yes or no; on or off), or categorical (e.g., Caucasian, African American, Latino, multiracial). If measures are incapable of being operationalized through continuous, dichotomous, or categorical identification, the measure may not have enough precise significance to be valid in describing or testing the theoretical model under study. Another aspect of measurement is numbering taxonomy. Measurements can be categorized into four distinct types (a taxonomy): nominal, ordinal/categorical, interval, and ratio (discussed later in this section). Each type of number has distinct properties associated with it. For example, if you asked a health leader how many subordinates he or she leads, you could get four different answers based on the numbering taxonomy employed; all four answers would be true and correct. If you asked, "Do you have more than 10 subordinates?" you could get a yes or no (nominal) answer. If you asked, "Which group best describes your number of subordinates: 5 or fewer, 6 to 10, 11 to 15, 16 to 20, and so on," you would get a categorical answer. If you asked, "How many subordinates do you lead exactly?" you could get an answer of 17. If you had no subordinates then you would have ratio data (from the taxonomy) because ratio data require an absolute zero. Depending on the number taxonomy used, data can be changed. You can go down the numbering taxonomy, such as from ratio to ordinal/categorical, but you cannot group or change data to go up the taxonomy, as nominal data cannot ever be changed to interval or ratio data once nominal data are the only data captured. With an answer of 17 subordinates, you could group the data (16 to 20) or categorize the answer into the "yes, more than 10" grouping; however, you cannot manipulate the data from a yes or no answer into a hierarchical group or a pure number. Why not always use exact numbers? There are two reasons: Some variables are not exact, but rather are inherently hierarchically grouped (e.g., place in a race) or categorical (e.g., gender). It tends to cost more in resources, time, money, attention, and materials to obtain precision or higher-order data such as interval or ratio taxonomical data. As already mentioned, the four types of number taxonomy data are nominal, ordinal, interval, and ratio.15,16 Interval and ratio data are frequently identified as "continuous" data in the literature. Of note, ratio data can be transformed into interval, ordinal, or nominal data; interval data can be transformed into ordinal or nominal data; and ordinal data can be transformed into nominal data. Nominal data, however, cannot be transformed. A researcher can transform data down the taxonomy (ratio down to nominal), but never up the taxonomy. Table B-1 presents the distinctions of each type of numbering taxonomy. Table B.1 Number Taxonomy Measures Type Definition Scale Transformation Nominal Codes are assigned as labels to observations and are mutually exclusive and categorically exhaustive, such as gender (male = 1, female = 2). Also known as categorical or binary variables. One-to-one, where each observation assigned a label code must be recoded as a group Ordinal Numbers coded to observations are in rank or sequential order, such as the result of a race (e.g., first, second, third place). Monotonic increasing or decreasing as long as order is maintained Interval A quantitative score is given to an observation where the interval between scores holds meaning (e.g., observation B is 2 times as much as observation A, observation C is 3 less than observation D). Examples include height in meters or temperature on a Celsius scale. Positive linear Ratio Observations are given a quantitative score as in the interval scale number, but an absolute zero is possible (e.g., on the Kelvin temperature scale). Positive similarities such as multiplication Variables should be capable of several different methods of operationalization. For example, the variable age can be operationalized as a continuous, categorical, or binary variable, depending on how the researcher chooses to define and measure this variable. The following section discusses this process in more detail. Operationalizing Measures To operationalize the variable age, we first must associate the variable with a specific unit of analysis (such as an organization, a team, or an individual). Next, a brief definition of the variable age that supports the unit of analysis under study; for example, we may say that age is defined as the number of years associated with a human individual's life. This statement includes two important features. First, it qualifies age in terms of years. Second, it provides a reference group for age, where a potential range for a life span is universally understood. Given this information, it becomes possible to operationalize age in several different ways: (1) as a continuous variable (interval or ratio data); (2) as a binary (nominal) variable; and (3) as a hierarchical group variable (ordinal), as illustrated in Table B-2. Table B.2 Operationalized Variable: Age Age is defined as the number of years associated with a human individual's life. Continuous (interval or ratio data): 1 to n (e.g., 105) Hierarchical group or category (ordinal data): 0-10, 11-20, 21-30, 31-40, 41-50, 51-60, 61-70, 71-80, 81-90, 91-100, n > 101 Binary (nominal data): Medicare age eligible? Yes or No (Medicare eligibility = 65 years of age or older) In Table B-2, we arbitrarily established cutoff points for categorical and dichotomous variables that help support the issue under study. The categorical description of age also could have been classified using quarter-century marks or 5-year increments. The selection of a category is up to the executive analyzing the data. Finally, if the executive is interested in partitioning age-eligible Medicare recipients from those not eligible for Medicare, setting a break point at age 65 provides an opportunity to analyze the two different groups. Measurement is vital to empirical assessment and evaluation, and measurements taken must be both reliable and valid. Reliability is the "consistency of an operationalized measure."17 This concept applies to both the measurement apparatus or method and the person or machine doing the measuring. Validity focuses on the notion of measuring what is intended to be measured. An unreliable measure is also invalid, whereas an invalid measure can still be reliablethat is, you can measure the wrong thing correctly with repetitive and consistent results. Within the concept of validity, specific kinds of validity are distinguished: Face validity refers to whether a measure, on its face, seems to be related to the concept that is presumably being measured. Predictive validity refers to whether a new measure of something has the same predictive relationship with something else that the old measure had. Convergent validity refers to whether two different measures of presumably the same thing are consistent with each otherwhether they converge to give the same measurement. Criterion validity is a test of a measure when the measure has several different parts or indicators in itcompound measures. Each part, or criterion, of the measure should have a relationship with all the parts in the measure for the variable to which the first measure is related in a hypothesis. Content validity tests whether the measure has sufficient content [does it cover or contain enough measurement of the property or characteristic] in it to be acceptable.18 Not all variables lend themselves to operationalized measures through all three metrics. For example, the variable racial category does not lend itself to a continuous measure. The most appropriate nomenclature for race is a category, where the researcher selects racial categories of interest in the study for analysis. The process of operationalizing measures is critically important in the business world and in the health industry. A requirement to gather, manage, and measure outcomes from data implies a continuous process. Before data can be collected and analyzed, however, they must be operationalized in a consistent and logical manner. The Balanced Scorecard developed by Kaplan and Norton is an example of a quality-oriented tool that requires executives to select and define not only constructs of interest to measure, but also valid and reliable variables capable of identification through operationalized measures that provide meaningful data for trend analysis.19 The last step in operationalizing variables is the construction of a code sheet. Whenever variables are tested in a hypothetical relationship, it is not the variables themselves that are actually tested, but rather the operationalized units of the variables. This step is a necessary precursor to loading data into a statistical software program such as SPSS, SAS, Minitab, or Excel. When we operationalize data for statistical software manipulation, we create code sheets and coding methodology. A code sheet is a very simple explanation of how operationalized units of a variable will be used in the study. The researcher must keep in mind the assumptions of the test when building a code sheet. For example, parametric and nonparametric tests require different assumptions, which should be incorporated into the code sheet. Although numerous examples of building code sheets are available that demonstrate how data may be operationalized in a study, Table B-3 provides one proven example of successin this case, a code sheet is created for the variable education. Table B.3 Example Code Sheet for the Variable Education Label Description Operationalized Education Highest education degree obtained by member Taxonomy = Ordinal 1 = High school degree 2 = Associate's degree 3 = Bachelor's degree 4 = Master's degree 5 = JD (law degree) 6 = MD (medical degree) 7 = PhD (doctorate) A MODEL OF HEALTH Leaders in the health professions will most certainly discuss the construct of health on a daily basis. Figure B-5 presents a conceptual model of health that should look relatively familiar. As we have discussed, health is a concept (i.e., construct) that has numerous meanings to several different brackets of patients. For example, an 82-year-old male of average weight and height may describe being healthy in a very different manner than an 18-year-old male of average weight and height. How might a provider standardize these two definitions of health status in a manner that provides meaning across a population of patients? this easily with the modeling of constructs, variables, and measures. Although it may be seamless to a patient, a provider collects variable and measurement information on a patient with each and every encounter. FIGURE B-5 Conceptual model of health using constructs as circles, variables as rectangles, and measures as operationalized units. Let us consider your last visit to your healthcare provider. During this encounter your provider collected variable information such as height, weight, heart rate, blood pressure, and body mass index. Your provider also may have collected blood to evaluate your cholesterol level and blood sugar. All of this variable information is then operationalized and measured in a manner that is consistent within the health profession. Based on the operationalized information gathered from the measurements, your provider can present a universally accepted and well-understood concept of the construct of health. For example, the following are two outcomes for measuring health from two patient encounters: Example 1: A 91-year-old white female presents to the provider. She is 5'2", 250 pounds, has a blood pressure of 220/140, and presents with an above average heart rate (tachycardia). Blood work confirms high blood sugar and high cholesterol. Example 2: A 24-year-old Latino male presents to the provider. He is 5'9", 170 pounds, has a blood pressure of 120/70, and presents with a normal heartbeat. Lab results suggest normal levels of cholesterol and blood sugar. Based on these two examples, which patient is in better "health"? Furthermore, which patient might require follow-up tests and/or education to achieve a "greater level of health" over a period of time? Construct, variable, and measurement information is collected by a variety of stakeholders in the health professions all the time.20,21 Leaders collect similar types of information when evaluating issues such as employee performance and productivity. Such information may be used later in rating an employee for annual merit raises and/or promotion. Similar organization metrics can be collected by leaders to measure organizational productivity, efficiency, effectiveness, and success. KEY RELATIONSHIPS: LEADERSHIP MODELS TO THEORIES TO ACCEPTED THEORIES As noted earlier in this chapter, a theory has undergone scientific and practical scrutiny, albeit at various levels of intensity, to determine its value, truth, and validity. In contrast, a model has not yet met a level of academic and practical scrutiny for scientific validation, so its value is yet to be determined. Both theories and models require ecological validitythat is, the ability to represent and present reality wellfor scholars and practitioners to study and utilize them. Once a leadership model achieves the designation of a theory, however, how is the theory justified and placed within the greater context of leadership studies and understanding? There is no perfect answer to this question, although the system of evaluation presented here is ecologically valid: The word "criteria" is the plural of the term "criterion." A criterion is a standard of judgment such as an examination score of 85% is a grade of "B" or 95% is a grade of "A." Here are five criteria that are generally used when comparing theories, and a new theory satisfying these will then replace a previously accepted theory. The previously accepted theory gives an acceptable explanation of something so the new theory must give the same results; New theory explains something that the previously accepted theory either got wrong or, more commonly, did not apply; The new theory makes a prediction that is later verified; The new theory is elegant, has aesthetic quality, is simple, [is] powerful, and includes universal symmetries that are simple, easy to remember or apply, and/or are expressed as some symmetry of nature, and/or are powerful enough to be used in many applications; and Provide a deeper insight or link to another branch of knowledge.22 DESCRIPTIVE AND PRESCRIPTIVE LEADERSHIP MODELS REVISITED: A CONCEPTUAL MODEL OF A LEADERSHIP THEORY ON MOTIVATION Leadership theories and models can be descriptive, prescriptive, or both descriptive and prescriptive. Descriptive theories and models illustrate, define, and capture the description of leadership phenomena but do not recommend or prescribe actions, behaviors, or processes to employ. Prescriptive theories and models provide recommendations to the leader practitioner with regard to actions, behaviors, or processes to employ in order to be a successful leader. Some leadership theories and models both describe and prescribe. As an example, let us explore an application of a leadership theoryin this case, a motivational theory introduced by Edwin Locketo measure organizational outcomes such as satisfaction and performance and investigate this theory's connection to other motivational theories. As you read about Locke's theory, try to list the constructs, concepts, and variables from this leadership motivational theory. Goal-setting theory23 was first introduced by Edwin A. Locke in 1968, when he published the classic article "Toward a Theory of Task Motivation and Incentives" in the journal Organizational Behavior and Human Performance. Since the late 1960s, considerable attention has been given to applying goal-setting theory in industry and management situations. Locke, along with other contributors such as Gary Latham, performed laboratory and field experiments; the most widely publicized were field studies conducted in the 1970s in conjunction with the logging industry.24 During the past few decades, Locke's emphasis has focused on the area of applied goal-setting theory with regard to improved performance in complex business tasks. In 1992, Locke and others studied the relationship between goal-setting and expectancy theory.25 Studies over the past 35 or more years were performed in an effort to learn more about the potential for improving performance by using goal setting as a motivational technique. This theory can be disassembled into constructs, concepts, and variables. (Try to list or draw the elements of this theory on your own.) First, Locke's basic assumptions and ideas are important to understand. Goals are the aim of an action or behavior. Goals can be set for any verifiable or measurable outcome. Locke's basic assumption is that goals are immediate regulators of human action.26 An individual synthesizes direction, effort, and persistence to accomplish goals. To maximize goal setting, specific and challenging goals are set to focus action and effort over time to accomplish tasks. From 1968 to 1980, 90% of studies showed that specific, well-defined, and challenging goals led to greater improved performance than did vague and easy goals.27-32 Individuals must commit to set goals to produce results; the more difficult (challenging yet reasonable) the goal, the better the individual will perform. Individuals need management support (feedback, reward mechanisms, and required resources [time, training, and material goods]) to maximize performance when applying goal setting.33 To apply this theory, Locke suggests seven steps34 to follow to optimize goal setting: Specify objectives or tasks to be done. Specify how performance will be measured. Specify the standard to be reached. Specify the time frame involved. Prioritize goals. Rate goals as to difficulty and importance. Determine the coordination requirements. Managers must ensure they set goals that do not conflict with each other or conflict with organizational goals. For groups, every group member should have verifiable specific goals, as well as a group goal to counter the tendency toward "social loafing." Smaller groups (three to eight people) are more effective than larger ones. Potential negative issues related to this theory include excessive risk taking, excessive competition, and goal failure, all of which can diminish members' confidence and create unwanted stress. A graphic illustration, like that shown in Figure B-6, should assist in understanding this theory. FIGURE B-6 Goal-setting theory. Can you list the constructs, variables, and process of goal-setting theory? How would you operationalize the variables? How would you measure them if you conducted a study to see how goal setting works in a healthcare organization? Goal-setting theory is readily integrated with other motivational theories as well. Although goal setting is a principal attribute of many motivational and performance theories, recent research has largely focused on locus of control theory influences and expectancy theory relationships. Locus of control theory suggests that people acquire motivation either through an internal catalyst (internalizers) or via an external catalyst (externalizers). Integration of goal-setting theory with locus of control theory has revealed that internalizers tend to have better performance than externalizers.35 Both of these theories also relate very well to expectancy theory, in that the goals we set are based on the outcomes we hope to achieve. Another motivational theory, expectancy theory, may be integrated with goal-setting theory in that goal setting is negatively related to valence (setting low goals does not satisfy individuals as well as setting high goals), and instrumentality is positively related to goal setting (achieving difficult goals gives the individual a greater sense of achievement, self-efficacy, and skill improvement than achieving easy goals).36 To illustrate the connection of goal-setting theory to expectancy theory, the following summary is provided. Can you see how the constructs relate? Expectancy theory was developed by Victor Vroom in 1964. This theory combines motivation and the process of leadership.37 With expectancy theory, subordinate behavior and action are seen as the result of the subordinate making conscious choices among alternatives such as a reward or no reward. Constructs of this theory are valence, expectancy, and instrumentality. Valence concerns the affective/emotional orientations that subordinates have regarding rewards based on outcomes; this construct focuses on the extent to which the subordinate has extrinsic motivation (see Herzberg's two-factor theory or hygiene theory; extrinsic motivation refers to money, promotion, time off, or some other external reward) or intrinsic motivation (e.g., satisfaction, self-esteem, self-efficacy). Leaders must understand which type of reward will best motivate the subordinate, given his or her wants and needs. Expectancy is the subordinate's expectations, confidence, knowledge, skills, and abilities to perform a task or action. Leaders should remove barriers and enhance resources appropriately to allow the subordinate to perform to his or her maximum ability. Instrumentality is the perception of the subordinate concerning receiving the reward for the task or action performed based on the resulting outcome. Leaders should fulfill promises to subordinates; that is, leaders should reward subordinates for their performance outcomes. A formula for this theory would be Motivation = Valence Expectancy (Instrumentality) Expectancy theory is closely tied to social exchange theory and the transactional leadership model. Figure B-7 illustrates its major precepts. FIGURE B-7 Expectancy theory. Which constructs make up goal-setting theory? Clearly, persistence, direction or focus, effort, and performance are constructs. Also, managerial feedback, resources and rewards, and organizational behaviors and tasks are constructs. Concepts from each construct may be identified, such as time on task and attention, and then operationally defined as variables. Because each variable observation changes or varies, the total model is evaluated against performance. Can you operationally define the variables that make up the concepts of the constructs of goal-setting theory? How would you operationally define the variables of this theory? Could you operationally define the variables of goal-setting theory and expectancy theory to merge these theories into a larger model? Can you synthesize goal-setting theory, expectancy theory, and the theory of planned behavior (see Figure B-8)? Review the graphics from the three theories and think about where constructs of the theories connect. any of the constructs in the three frameworks contradict other constructs? If there is contradiction, can you resolve the tension between constructs? What variables could you operationalize and measure? Would the variables move in one direction or both directions (increase and decrease)? What hypotheses could you develop based on the movement of the variables? FIGURE B-8 Theory of planned behavior. Reproduced from Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179-211, with permission from Elsevier.

Why is it helpful to understand leadership as a theory while managing a healthcare-orientated organization? Justify your stance using two examples.

What factors do you think should appear in a model of leadership?Why?Summarize and justify the relationships between the model's theoretical elements and the application of leadership principles.