Janice M. Morse
MAKING USEFUL THEORY: MAKING THEORY USEFUL
There was a dean at my university who was a scientist with high intelligence but low boiling point. One day at the faculty meeting, after I said something that displeased him, he replied. “Peter, you have never made a contribution of interest to scientists.” Naturally, my first thought was to take offence. But trying to maintain a generous spirit, and believing that a highly intelligent dean offers personal insults only in private, I decided what he really meant was not the singular “you”, but the plural one. “You philosophers of science,” he meant, “have nothing to offer us scientists.”
This interpretation at least took some of the sting out of his remarks and enabled me to think about them more clearly. Perhaps the dean is right, I now speculated. Although philosophers of science have carefully worked out views about a range of general concepts scientists employ—such as, evidence, explanation and law, to name just three of many—scientists seem to take little heed of them.
—Peter Achinstein (2001, p. 3)
In this chapter, I address the why and how of theory development in qualitative inquiry. Let us explore the processes of constructing mid-range theory, modeling, and moving toward the certainty of our models; most important, let us explore modes of clinical application for theory. First, we discuss “ Making Useful Theory”—rather than esoteric theory, or theory for its own sake—and in the second part of the chapter, “Making Theory Useful,” we deal with some of the clinical applications of theory itself.
MAKING USEFUL THEORY
Qualitative research methods are usually implemented by targeting a particular concept or group of concepts. An exception to this less focused approach may be a researcher who wants to find out “what is going on” in a particular setting, or a researcher who is interested in a particular phenomenon (but not in a concept per se). However, the majority of researchers are interested in some thing—that is, a concept at some level of development. Of course, concepts in the research setting do not occur in isolation—but rather occur with other concepts that influence the target concept’s “performance.”
WHAT IS MID-RANGE THEORY?
A mid-range theory is a theoretical scheme that reveals, enlightens, and provides understanding clinically. It is the analysis of data and organization of the subsequent derived concepts into a network that explains reality, predicts future actions and outcomes, and guides actions. Mid-range theory ranges in the level of abstraction from a theory that is closely associated with the data, to one that is reasonably abstract. It ranges from theory that is quite narrow in scope to those that are quite comprehensive, broad, and abstract. If the theory is developed from linked and interrelated concepts1 (as in Chapter 22), the theory may itself range from connected small concepts, to laterally connected studies, to horizontally connected larger concepts. There is no formula dictating how many concepts are required to make a theory, what level of abstraction is “ideal,” and how it guides practice. Mid-range theories contain whatever is necessary to understand the phenomenon and fulfill its role, in nursing or elsewhere, for guiding practice. Recall that a concept is anatomically static, an elaborate label of sorts, so it is the physiology of the concepts and the interaction of their components, or the interaction between concepts themselves, which make mid-range theory dynamic—as theory needs to be if it is to fit reality.
The most important feature of mid-range theory is that the theory makes the research potential obvious and exciting, useful, and powerful. Mid-range theories are generally about phenomena of clinical significance, and are sufficiently “applied” enough as to guide care.
Authors have argued about the interrelationship of concepts and theory, and I talked about this in Chapter 7. Some authors considered concepts of the “building blocks of theory” (Walker & Avant, 1995), and Paley (1996a) describes concepts as “notches in theory.” These discussions are important for describing quantitative theories, which are deliberately constructed by linking concepts according to the researchers’ needs, and are modified in response to subsequent research results. But when developing qualitatively derived theory, these discussions are rather extraneous, for the concepts within a theory, and the theoretical form, are developed as they need to be to represent the phenomena, and the necessary conceptual components and the theoretical form may change from project to project. Often the concepts remain identified, but their form is still at the level of description. You cannot always quickly find the anatomical structure of some concepts. In quantitative inquiry it is different. Their concepts have been developed, defined, and operationalized before having been placed in the theory. Therefore, the method used to develop mid-range theory differs between qualitative and quantitative inquiry; here, the discussion focuses on the modes of development of qualitatively derived, mid-range theories.
What, then, is mid-range theory? It is something that is full of enough misgivings and misunderstandings to make a student shudder? Let us start with what it is not.
Mid-range theory is not:
• A collection of concepts linked with lines
• Something that must be learned by rote and rigidly applied
• Something that is fence-like, delineates, grabs, dictates
• Something that is fixed, immobile, and perceived to be right
Sometimes, in the process of developing concepts, researchers order the attributes and diagram the results to show a rudimentary process. We will refer to these as a within-concept theory. Recall in Chapter 16, by exploring the different patterns of the attributes, or different strengths of the attributes, we found different types of hope (hoping for a chance for a chance, and so forth). This conceptual scheme was a nice development of a concept working toward a theory, but does not fit our definition of a mature theory.
Definition of Mid-Range Theory
A mid-range theory is a theoretical schema that reveals, enlightens, and shows. It organizes complexity, labels, and thus enables communication and understanding. It shows how “things work,” and enables agreement and communication and discussion about complex phenomena. In some cases, as well as enlightening, the development of the mid-range theory will also allow understanding of what will happen (prediction) and sometimes for quantitative testing (Wuest & Hodgins, 2011).
THE ROLE OF THEORY
Burton (1974), succinctly describes the roles and uses of theory from the informal sorting of the “overwhelming number of observations” and “masses of data” that we accumulate in everyday life (“theory as cognition”) to formal theory that allows for hypothesis testing and formal knowledge development. Burton also states that theories cause tension and compete with each other in science; or as “counter-nihilism,” fills voids of “intellectual nothingness” of that which is not known. Theory as “temporality” is theory that enables us to place things in time (past, present, and future) and account for trends and critical periods in history.
Theories clarify pulling together, and fitting, discrepant data. But in the process, investigators privilege some data and ignore others, invalidating discrepant data so that such theory is “a provisional, imaginative structure” (Burton, 1974, p. 14). It is “on this schema that measurement hangs its hat, and leads the way for still more measurement … How one sees it depends on the context and the readiness with which one sees.”
Theory as prediction, as a predictive tool, is the most common perception of the use of theory: it is considered valid and a “good” theory if it holds when tested. But, Burton notes, theory does not “rise or fall on its predictive value alone,” and the “distinction between fact and theory has always been overdrawn” (p. 14) and the prediction must be valued by society. A theory in itself may “lead to a consequence or away from it; go in several directions simultaneously, or importantly, lead a researcher to new directions” (p. 15). Theory may be “silence”; where no theory necessarily exists or when “a phenomenon requires social incubation necessary for its application” (p. 15). Theories may conceal, rather than being open and direct and, as Burton notes, sometimes this concealment is the direct function of a theory. “The silence of the theory is a salient theory-property, in that society is not ready for the disclosure” (p. 15). Theories may reveal, “show,” disclose properties, and every theory has its “showing or disclosing properties, which, even without their propositions or postulates, change something” (p. 15).
The Roles of Mid-Range Theory
Where does mid-range theory fit into Burton’s (1974) scheme? Clearly mid-range theory is theory that shows, reveals, and organizes lay theories of cognition. Mid-range theory clarifies and, although it is internally verified in the process of construction, qualitatively derived mid-range theories are not usually tested in the quantitative sense (although we discuss this later in the chapter). Mid-range theories sometimes delve into areas of silence, much to the discomfort of others, so that researchers may report explicitly on sexual behaviors about which we may sometimes rather not know the details. Our norms and our values impinge on science when we develop such theory at the personalized, behavioral level.
How Abstract Should a Mid-Range Theory Be?
Mid-range theories range in the level of abstraction from a theory that is closely associated with data, to one that is reasonably abstract. The theory may be one that is quite narrow in scope, or one that is quite comprehensive and abstract. There is no formula—these studies are, in Glaser’s (1978) phrase, “whatever is demanded by the concept” to understand the phenomenon. Further, the concept, as we have discussed, is anatomically static; it is the physiology of the concepts and their interaction that make the theory dynamic—as theory needs to be.
However, note that there is a relationship between the level of abstraction and the scope of mid-range theory. If the theory is descriptive and close to data, then the scope is relatively narrow. If the mid-range theory is abstract, it will be broader in scope and have much wider application.
THE STRUCTURE OF MID-RANGE THEORIES
Mid-range theories that are developed usually contain reasonably developed concepts and explicitly show their relationship. Remember, concepts within a theory are developed as they are needed in the theory, and may change from project to project. Usually they are developed directly from data (and this we do that all the time in grounded theory or ethnography). Sometimes the concept is partway developed into a theory by ordering the attributes within a concept, and sometimes they start out by developing concept(s) and then develop those into a mid-range theory in a subsequent study. And sometimes the researcher commences the study with a targeted concept in mind, or one that is partially developed.
The difference between a qualitative project that develops the concept and one that directly develops a theory is somewhat blurred. The concept-developing qualitative study has as its product a concept, which is by definition abstract and somewhat static, but may be an end goal in itself. Qualitatively derived theory is not usually static, and may have movement (often describing transitions). Frequently qualitatively derived theory contains broader explanations and is more abstract than concept development research. Qualitatively derived theory may contain more than one concept, and in this case it will explore the relationship between those concepts.
TYPES OF QUALITATIVELY DERIVED THEORY
In Chapter 2, I discussed various types of theory as theories were originally used in nursing research and practice. We discussed philosophy as it gives research a gaze or an agenda, such as feminism, culture, or social justice. We discussed what has become known as “nursing theory,” developed by the nurse theorists to give focus or frameworks for practice. We discussed theoretical frameworks that are used to justify your qualitative project (without impeding induction), or for a quantitative study used deductively to organize concepts in preparation for measurement. Finally, we discussed conceptual frameworks used in qualitative inquiry that provide context for qualitative research: that is, research that is intended to develop concepts or mid-range theory. Recall that these frameworks may include a philosophical stance, or concepts or even theories that the researcher is building upon or even using as a scaffold and skeleton.
What Are Theoretical Models?
The term model is used in a number of ways: A model may be a schematic diagram that shows the relationships of concepts or attributes within the theory. It may or may not show a part of the theory, or the theory in its entirety. Theoretical models, however, are diagrammed to illustrate results, and depict the form of the theory obtained; often the form of the model represents the type and structure of the method used.
Are models the same as theories? Models come in various types. When they are schematic representations of the attributes within a concept, they are best called a model. If they are to be considered as some type of a theory, they are relatively low-level theory.
THE PROCESS OF QUALITATIVELY DEVELOPING MID-RANGE THEORIES
Mid-range theories are developed either by processes of data analysis and synthesis in the process of qualitative inquiry or, as in quantitative inquiry, by identifying, analyzing, and linking concepts.
PROCESSES OF THEORY CONSTRUCTION
How Is Mid-Range Theory Constructed?
Most commonly, mid-range theory is developed from qualitative data extending qualitative strategies beyond the stage of simple description or developing concepts. Several qualitative methods are designed to bring analysis to the level of mid-range theory—most commonly grounded theory (Glaser, 1978; Corbin & Strauss, 2008; Charmaz, 2006), or some types of ethnography. Other methods of developing mid-range theory are by the linking of concepts, either deliberately looking for shared attributes, or by placing concepts associated with the topic, at different levels of abstraction, and looking for connections.
Continuing Qualitative Inquiry
While it is not my intention to write a textbook on qualitative methods at this point, I will quickly provide an overview listing the major analytic strategies for moving one’s study to the level of mid-range theory.
• Induction, deduction, and abduction: Cognitive process inherent in qualitative inquiry, induction (i.e., thinking up from data), deduction (for testing conjectures, however small), and abduction (a process of induction–deduction, building and testing and then building again) all play a part in project theory construction. By constantly asking analytic questions of these data, by developing conjectures and “testing” these conjectures both within categories and newly gathered data, the research systematically builds categories and concepts. This constant internal incremental testing ensures validity. Other processes in theory construction are as follows.
• Processes of synthesis: As similar data are gathered, they are synthesized within the categories, within attributes, and within concepts. This process of synthesis decontextualizes, meaning that it removes the variation that pertains to individual nuances in the context.
• Emergence: Ideas are derived both from the data and from the investigator. Important features within data demand the investigator’s attention by occurring over and over until the researcher takes notice and incorporates that finding into the emerging conceptual scheme. Ideas also arise from the researcher’s observations and when seeking answers to the analytic questions. These are noted, tested and incorporated into the emerging theory.
• Category formation: As data collection continues, data are sorted into broad categories. As these data become dense, copious, the researcher is able to sort each category and ask: Is this one category or more than one thing? Is this one concept, or different types of the same concept? Or, is this one concept with different attributes in this category?
• Identification of themes: Not all data are analyzed with categories. Themes run right through the data, and are sometimes foregrounded; at other times they are backgrounded and must be reached by inference. In grounded theory the core variables for the basic social processes are usually derived from a theme.
• Fit and fittingness: As the analysis proceeds, categories and concepts fit together in a cohesive whole. Unlike quantitative data in which linkages are drawn by a line with a + sign for a positive relationship or a – sign for a negative relationship, the theory is linked by logic, by common sense, and by similarity.
• Micro–macro linkages: Not all linkages are horizontal; some are vertical, with some processes being a microanalytic analysis of a larger process. Alternatively, several smaller microdescriptive processes may be a part of a larger (macro) process.
• Identifying relationships: Recognizing that this goes with that, this is a part of that, and when, why, and how is the essence of fitting the pieces together into a cohesive whole.
• Indices of causality: We recognize that this causes that; and that that co-occurs with that; when this is present, that is absent, and so forth.
• Forming the structure: Once the researcher can tell the story in a general form, as, for instance, “these people do this and that, and on the other hand, these people do that because ________” analytic procedure may commence. Being able to address the topic in this narrative-schematic style enables the outline to the theory to be developed. Hold seminars to let the researcher articulate the emerging concepts and relationships, and to recognize where he or she should be theoretically sampling, where the parts that require more thinking or development are, and which parts are rich and well understood.
• Functions of exemplars: Exemplars are illustrations only. Excellent theory has descriptive passages that can stand alone and convince the reader; the exemplars are representative sections of text that illustrate the text. The reason that the text must explain in a self-contained way is that the participants do not speak for themselves—the analyst/researcher/investigator does that, and if the analysis is good, then readers will see more in the text than they would if they had simply read the quotation without the explanation.
• Revisiting saturation and negative cases: Saturation is achieved when the investigator is bored; this means when all dimensions of the concept have been explicated, when the boundaries are clear, when the concepts are well developed, and when the definitions hold across numerous instances and cases.
• Issues in validity and verification: Some researchers think that data should be coded by more than one person. Although interrater reliability is appropriate for descriptive research, elsewhere I argue that such a strategy makes interpretative analysis superficial, obvious, and trite—“perfectly healthy but dead” (Morse, 1997). Earlier, we spoke of internal verification—how ideas are tested in the process of doing research, so that as the theory is built, the theory is internally solid. But, the constructed theory must always stand up to external scrutiny. How?
Some researchers suggest that the theory be returned to the participants for verification. Again, I argue, “Who’s the analyst here?” You have gone to school for umpteen years and learned social science theory and researcher methods. If your interpretation is a great theory, synthesized and abstracted from the context, it is highly likely that they will not recognize their own interview or story in the analysis.