polit chapter 6

6

Good research usually integrates researchfindings into an orderly, coherent system.Such integration typically involves linking newresearch and existing knowledge through a thor-ough review of prior research on a topic (seeChapter 5) and by identifying or developing anappropriate conceptual framework. Both activitiesprovide an important context for a research project.This chapter discusses theoretical and conceptualcontexts for nursing research problems.

THEORIES, MODELS,AND FRAMEWORKS

Many terms have been used in connection withconceptual contexts for research, including theo-ries, models, frameworks, schemes, and maps.There is some overlap in how these terms are used,partly because they are used differently by differentwriters, and partly because they are interrelated.We offer guidance in distinguishing these terms,but note that our definitions are not universal.

Theories

The term theory is used in many ways. For example,nursing instructors and students frequently use theterm to refer to the content covered in classrooms,as opposed to the actual practice of performing

Developing a ConceptualContext

nursing activities. In both lay and scientific usage,the term theory connotes an abstraction.

In research circles, the term theory is used differ-ently by different authors. Classically, scientists haveused theory to refer to an abstract generalization thatoffers a systematic explanation about how phenomenaare interrelated. The traditional definition requires atheory to embody at least two concepts that are relatedin a manner that the theory purports to explain.

Others, however, use the term theory less restric-tively to refer to a broad characterization of a phe-nomenon. According to this less restrictive definition,a theory can account for (i.e., thoroughly describe) asingle phenomenon. Some authors specifically referto this type of theory as descriptive theory. Forexample, Fawcett (1999) defines descriptive theoriesas empirically driven theories that describe or clas-sify specific dimensions or characteristics of individ-uals, groups, situations, or events by summarizingcommonalities found in discrete observations (p. 15).Descriptive theory plays an especially important rolein qualitative studies. Qualitative researchers oftenstrive to develop a conceptualization of the phenom-ena under study that is grounded in the actual obser-vations made by researchers.

Components of a Traditional TheoryAs traditionally defined, scientific theories involvea series of propositions regarding interrelationships

CHAPTER 6 Developing a Conceptual Context 115

among concepts. The writings on scientific theoryinclude a variety of terms such as proposition, pos-tulate, premise, axiom, law, principle, and so forth,some of which are used interchangeably, and othersof which introduce subtleties that are too complexfor our discussion. Here, we present a simplifiedanalysis of the components of a theory.

Concepts are the basic building blocks of atheory. Examples of nursing concepts are adapta-tion, health, anxiety, and nurseclient interaction.Classical theories comprise a set of propositionsthat indicate a relationship among the concepts.Relationships are denoted by such terms as isassociated with, varies directly with, or is con-tingent on. The propositions form a logicallyinterrelated deductive system. This means that thetheory provides a mechanism for logically arrivingat new statements from the original propositions.

Let us consider the following example, whichillustrates these points. The Theory of PlannedBehavior (TPB; Ajzen, 1988), which is an exten-sion of an earlier theory called the Theory ofReasoned Action (Ajzen & Fishbein, 1980), pro-vides a framework for understanding peoplesbehavior and its psychological determinants. Agreatly simplified construction of the TPB consistsof the following propositions:

1. Behavior that is volitional is determined bypeoples intention to perform that behavior.

2. Intention to perform or not perform a behavioris determined by three factors:

Attitudes toward the behavior (i.e., the overallevaluation of performing the behavior)

Subjective norms (i.e., perceived social pres-sure to perform or not perform the behavior)

Perceived behavioral control (i.e., anticipatedease or difficulty of engaging in the behavior)

3. The relative importance of the three factors ininfluencing intention varies across behaviorsand situations.

The concepts that form the basis of the TPBinclude behaviors, intentions, attitudes, subjectivenorms, and perceived self-control. The theory,which specifies the nature of the relationship

among these concepts, provides a framework forgenerating many hypotheses relating to healthbehaviors. We might hypothesize on the basis ofthe TPB, for example, that compliance with a med-ical regimen could be enhanced by influencingpeoples attitudes toward compliance, or byincreasing their sense of control. The TPB has beenused as the underlying theory in studying a widerange of health decision-making behaviors, includ-ing contraceptive choice, AIDS preventionbehavior, condom use, vaccination behavior, andpreventive health screening.

Example of the TPB:Aminzadeh and Edwards (2000) conducted

a study, guided by the TPB, in which they exam-ined factors associated with the use of a caneamong community-dwelling older adults. Theirresults provided further evidence of the utility ofthe TPB in understanding health behaviors andhave implications for the design of theory-basedfall prevention interventions.

Types of Traditional TheoriesTheories differ in their level of generality. So-called grand theories or macrotheories purport todescribe and explain large segments of the humanexperience. Some learning theorists, such as ClarkHull, or sociologists, such as Talcott Parsons,developed general theoretical systems to accountfor broad classes of behavior.

Within nursing, theories are more restricted inscope, focusing on a narrow range of experience.Such middle-range theories attempt to explainsuch phenomena as decision-making, stress, self-care, health promotion, and infant attachment.

Conceptual Models

Conceptual models, conceptual frameworks, orconceptual schemes (we use the terms inter-changeably) represent a less formal attempt atorganizing phenomena than theories. Conceptualmodels, like theories, deal with abstractions (con-cepts) that are assembled by virtue of their rele-vance to a common theme. What is absent from

116 PART 2 Conceptualizing a Research Study

Statistical models are playing a growing rolein quantitative studies. These models use symbolsto express quantitatively the nature of relationshipsamong variables. Few relationships in the behav-ioral sciences can be summarized as elegantly as inthe mathematic model F ma (force mass acceleration). Because human behavior is complexand subject to many influences, researchers typi-cally are able to model it only in a probabilisticmanner. This means that we are not able to developequations, such as the example of force frommechanics, in which a human behavior can be sim-ply described as the product of two other phenom-ena. What we can do, however, is describe theprobability that a certain behavior or characteristicwill exist, given the occurrence of other phenomena.This is the function of statistical models. Anexample of a statistical model is shown in thefollowing:

where

Y nursing effectiveness, as measured by asupervisors evaluation

X1 nursing knowledge, as measured by a stan-dardized test of knowledge

X2 past achievement, as measured by grades innursing school

X3 decision-making skills, as measured by theParticipation in Decision ActivitiesQuestionnaire

X4 empathy, as measured by the MehrabianEmotional Empathy Scale

e a residual, unexplained factor1, 2, 3, and 4 weights indicating the

importance of X1, X2, X3, and X4, respectively,in determining nursing effectiveness

Each term in this model is quantifiable; that is,every symbol can be replaced by a numeric value,such as an individuals score on a standardized testof knowledge (X1).

What does this equation mean and how does itwork? This model offers a mechanism for under-standing and predicting nursing effectiveness. Themodel proposes that nurses on-the-job effectiveness

Y 1X1 2X2 3X3 4X4 e

conceptual models is the deductive system ofpropositions that assert and explain a relationshipamong concepts. Conceptual models provide a per-spective regarding interrelated phenomena, but aremore loosely structured than theories. A conceptualmodel broadly presents an understanding of thephenomenon of interest and reflects the assump-tions and philosophic views of the models designer.Conceptual models can serve as springboards forgenerating research hypotheses.

Much of the conceptual work that has beendone in connection with nursing practice falls intothe category we call conceptual models. These mod-els represent world views about the nursing processand the nature of nurseclient relationships. A sub-sequent section of this chapter describes some con-ceptual models in nursing and illustrates how theyhave been used in nursing research.

Schematic and Statistical Models

The term model is often used in connection withsymbolic representations of a conceptualization.There are many references in the research literatureto schematic models and mathematic (or statistical)models. These models, like conceptual models, areconstructed representations of some aspect of real-ity; they use concepts as building blocks, with aminimal use of words. A visual or symbolic repre-sentation of a theory or conceptual frameworkoften helps to express abstract ideas in a conciseand readily understandable form.

Schematic models, which are common inboth qualitative and quantitative research, representphenomena graphically. Concepts and the linkagesbetween them are represented through the use ofboxes, arrows, or other symbols. An example of aschematic model (also referred to as a conceptualmap) is presented in Figure 6-1. This model,known as Penders Health Promotion Model, isa multivariate paradigm for explaining and pre-dicting the health-promotion component oflifestyle (Pender, Walker, Sechrist, & Frank-Stromborg, 1990, p. 326). Schematic models of thistype can be useful in clarifying associations amongconcepts.


might be 2 to 1, respectively (i.e., two parts empa-thy to one part past achievement). The e (or errorterm) at the end of the model represents all thoseunknown or unmeasurable other attributes thataffect nurses performance. This e term would beset to a constant value; it would not vary from onenurse to another because it is an unknown elementin the equation. Once values for the weights and e

is affected primarily by four factors: the nursingknowledge, past achievement, decision-makingskill, and empathy of the nurse. These influencesare not presumed to be equally important. Theweights (s) associated with each factor representa recipe for describing the relative importance ofeach. If empathy were much more important thanpast achievement, for example, then the weights

F I G U R E 6 . 1 The Health Promotion Model (From Pender, N. J., Walker, S. N., Sechrist, K. R., & Frank-Stromborg, M. [1990]. Predicting health-promoting lifestyles in the workplace. Nursing Research, 39, 331.)


have been established (through statistical proce-dures), the model can be used to predict the nurs-ing effectiveness of any nurse for whom we havegathered information on the four Xs (standardizedtest scores and so forth). Our prediction of who willmake an especially effective nurse is probabilisticand thus will not always be perfectly accurate, inpart because of the influence of the unknown fac-tors summarized by e. Perfect forecasting is rarelyattainable with probabilistic statistical models.However, such a model makes prediction ofnursing effectiveness less haphazard than mereguesswork or intuition.

Frameworks

A framework is the overall conceptual underpin-nings of a study. Not every study is based on a the-ory or conceptual model, but every study has aframework. In a study based on a theory, the frame-work is referred to as the theoretical framework;in a study that has its roots in a specified conceptualmodel, the framework is often called the concep-tual framework (although the terms conceptualframework and theoretical framework are frequentlyused interchangeably).

In many cases, the framework for a study is notan explicit theory or conceptual model, but rather isimplicitthat is, not formally acknowledged ordescribed. The concepts in which researchers areinterested are by definition abstractions of observ-able phenomena, and our world view (and views onnursing) shape how those concepts are defined andoperationalized. What often happens, however, isthat researchers fail to clarify the conceptual under-pinnings of their research variables, thereby makingit more difficult to integrate research findings. Asnoted in Chapter 2, researchers undertaking a studyshould make clear the conceptual definition of theirkey variables, thereby providing information aboutthe studys framework.

Waltz, Strickland, and Lenz (1991) describe afive-step process for developing conceptual defini-tions. These steps include (1) developing a prelim-inary definition, (2) reviewing relevant literature,(3) developing or identifying exemplary cases, (4)

mapping the meaning of the concept, and (5) stat-ing the developed conceptual definition.

Example of developing a conceptual definition:

Beck (1996) provides an example of developing aconceptual definition of the concept panic:

1. Preliminary definition: Panic is a sudden, unpre-dictable rush of overpowering terror that is associ-ated with a marked physiological uproar along witha loss of reasoning capacity and fears of dying andgoing crazy.

2. Literature review: Inter- and intra-disciplinary reviewof panic is undertaken.

3. Developing exemplary cases: Linda is married and themother of two children, a 3-year-old daughter and 3-month-old son. She is in her late 20s and is a collegegraduate.... After her first delivery, Linda experiencedpostpartum panic disorder. With her first child, the panicbegan 6 weeks postpartum during her daughters chris-tening. The panic suddenly came out of nowhere. Herheart started racing, her hands got sweaty, and she couldnot stop crying. For Linda, it was unbearable trying tostay seated in church. She wanted to run outside verybadly. Linda shared that the feelings inside her headwere so painful she thought she was going crazy andcould not focus on anything. As a result of the panic,Linda altered her lifestyle. She spent a tremendousamount of energy trying to appear normal. She went togreat lengths to avoid a panic attack. For example, when-ever she went to church, she would make sure she sat inthe back and at the end of the aisle so she could quicklyexit if she felt panic beginning. While experiencingpanic, even if only for 5 or 10 minutes, Linda felt min-utes were like hours. When panicking, Linda loses her-self. She feels like she is not sitting there (pp. 271272).

4. Mapping the concepts meaning: By integrating liter-ature with empirical observations, a conceptual mapwas developed that organized the various meanings ofpanic in the literature.

5. Stating the revised theoretical definition: Panic is a sud-den, unpredictable rush of overpowering terror that hasan all-or-nothing quality and is associated with (1) amarked physiological uproar such as palpitations, faint-ness and sweating; (2) a distortion of time and loss ofreasoning capacity, which engenders fearful cognitionsof dying, impending doom, losing control, and/or goingcrazy; and (3) an intense desire to flee the situation andavoid it in the future (Beck, 1996, pp. 271272).


might integrate new observations with an existingtheory to yield a more parsimonious explanation ofa phenomenon.

Theories and models that are not congruentwith a cultures values and philosophic orientationalso may fall into disfavor over time. It is not unusualfor a theory to lose supporters because some aspectsof it are no longer in vogue. For example, certainpsychoanalytic and structural social theories, whichhad broad support for decades, have come to bechallenged as a result of changes in views aboutwomens roles. This link between theory and valuesmay surprise you if you think of science as beingcompletely objective. Remember, though, that theo-ries are deliberately invented by humans; they arenot totally free from human values and ideals,which can change over time.

Thus, theories and models are never consideredfinal and verified. There always remains the possi-bility that a theory will be modified or discarded.Many theories in the physical sciences have receivedconsiderable empirical support, and their well-accepted propositions are often referred to as laws,such as Boyles law of gases. Nevertheless, we haveno way of knowing the ultimate accuracy and utilityof any theory and should, therefore, treat all theoriesas tentative. This caveat is nowhere more relevantthan in emerging sciences such as nursing.

Purposes of Theories and Conceptual Models

Theoretical and conceptual frameworks play sev-eral interrelated roles in the progress of a science.Their overall purpose is to make research findingsmeaningful and generalizable. Theories allowresearchers to knit together observations and factsinto an orderly scheme. They are efficient mecha-nisms for drawing together accumulated facts,sometimes from separate and isolated investiga-tions. The linkage of findings into a coherentstructure can make the body of accumulatedevidence more accessible and, thus, more useful.

In addition to summarizing, theories and mod-els can guide a researchers understanding of notonly the what of natural phenomena but also the

Quantitative researchers in general are moreguilty of failing to identify their frameworks thanqualitative researchers. In most qualitative studies,the frameworks are part of the research traditionwithin which the study is embedded. For example,ethnographers usually begin their work within atheory of culture. Grounded theory researchersincorporate sociologic principles into their frame-work and their approach to looking at phenomena.The questions that most qualitative researchers askand the methods they use to address those questionsinherently reflect certain theoretical formulations.

THE NATURE OFTHEORIES ANDCONCEPTUAL MODELS

Theories and conceptual models have much incommon, including their origin, general nature,purposes, and role in research. In this section, weexamine some general characteristics of theoriesand conceptual models. We use the term theory inits broadest sense, inclusive of conceptual models.

Origin of Theories and Models

Theories and conceptual models are not discov-ered; they are created and invented. Theory build-ing depends not only on the observable facts in ourenvironment but also on the originators ingenuityin pulling those facts together and making sense ofthem. Thus, theory construction is a creative andintellectual enterprise that can be engaged in byanyone who is insightful, has a solid knowledgebase, and has the ability to knit together observa-tions and evidence into an intelligible pattern.

Tentative Nature of Theories and Models

Theories and conceptual models cannot be proved.A theory is a scientists best effort to describe andexplain phenomena; todays successful theory maybe discredited tomorrow. This may happen if newevidence or observations undermine a previouslyaccepted theory. Or, a new theoretical system


text for their studies. This section briefly discussesseveral frameworks that have been found useful bynurse researchers.

Conceptual Models of Nursing

In the past few decades, several nurses have formu-lated a number of conceptual models of nursingpractice. These models constitute formal explana-tions of what the nursing discipline is and what thenursing process entails, according to the modeldevelopers point of view. As Fawcett (1995) hasnoted, four concepts are central to models of nurs-ing: person, environment, health, and nursing. Thevarious conceptual models, however, define theseconcepts differently, link them in diverse ways, andgive different emphases to relationships amongthem. Moreover, different models emphasize differentprocesses as being central to nursing. For example,Sister Calista Roys Adaptation Model identifiesadaptation of patients as a critical phenomenon(Roy & Andrews, 1991). Martha Rogers (1986), bycontrast, emphasizes the centrality of the individualas a unified whole, and her model views nursing asa process in which clients are aided in achievingmaximum well-being within their potential.

The conceptual models were not developedprimarily as a base for nursing research. Indeed,these models have had more impact on nursingeducation and clinical practice than on nursingresearch. Nevertheless, nurse researchers haveturned to these conceptual frameworks for inspi-ration in formulating research questions andhypotheses. Table 6-1 (p. 122) lists 10 prominentconceptual models in nursing that have beenused by researchers. The table briefly describesthe models key feature and identifies a studythat has claimed the model as its framework.Two nursing models that have generated particu-lar enthusiasm among researchers are describedin greater detail.

Roys Adaptation ModelIn Roys Adaptation Model, humans are viewedas biopsychosocial adaptive systems who copewith environmental change through the process of

why of their occurrence. Theories often provide abasis for predicting the occurrence of phenomena.Prediction, in turn, has implications for the controlof those phenomena. A utilitarian theory has poten-tial to bring about desirable changes in peoplesbehavior or health.

Theories and conceptual models help to stimulateresearch and the extension of knowledge by providingboth direction and impetus. Many nursing studies havebeen generated explicitly to examine aspects of aconceptual model of nursing. Thus, theories may serveas a springboard for advances in knowledge and theaccumulation of evidence for practice.

Relationship Between Theory andResearch

The relationship between theory and research is recip-rocal and mutually beneficial. Theories and models arebuilt inductively from observations, and an excellentsource for those observations is prior research, includ-ing in-depth qualitative studies. Concepts and relation-ships that are validated empirically through researchbecome the foundation for theory development. Thetheory, in turn, must be tested by subjecting deductionsfrom it (hypotheses) to further systematic inquiry.Thus, research plays a dual and continuing role in the-ory building and testing. Theory guides and generatesideas for research; research assesses the worth of thetheory and provides a foundation for new theories.

It would be unreasonable to assert, however, thatresearch without a formal substantive theory cannotcontribute to nursing practice. In nursing research,many facts still need to be accumulated, and purelydescriptive inquiries may well form the basis for sub-sequent theoretical developments. Research that doesnot test a theory can potentially be linked to one at alater time. Suggestions for linking a study to a con-ceptual framework are presented later in this chapter.

CONCEPTUAL MODELSUSED IN NURSINGRESEARCH

Nurse researchers have used both nursing and non-nursing frameworks to provide a conceptual con-

Other Models Developed by Nurses

In addition to conceptual models that are designedto describe and characterize the entire nursingprocess, nurses have developed other models andtheories that focus on more specific phenomena ofinterest to nurses. Two important examples arePenders Health Promotion Model and MishelsUncertainty in Illness Theory.

The Health Promotion ModelNola Penders (1996) Health Promotion Model(HPM) focuses on explaining health-promotingbehaviors, using a wellness orientation. Accor-ding to the model (see Figure 6-1), health promo-tion entails activities directed toward developingresources that maintain or enhance a personswell-being. The HPM encompasses two phases: adecision-making phase and an action phase. Inthe decision-making phase, the model empha-sizes seven cognitive/perceptual factors that com-pose motivational mechanisms for acquiring andmaintaining health-promoting behaviors and fivemodifying factors that indirectly influence pat-terns of health behavior. In the action phase,barriers and cues to action trigger activity inhealth-promoting behavior. Nurse researchershave used the HPM in numerous studies of healthpromoting behaviors.

Example using the HPM:McCullagh, Lusk, and Ronis (2002) used

the Pender HPM to identify factors affectingfarmers use of hearing protection devices. Thefindings offered further support of the HPM.

Uncertainty in Illness TheoryMishels Uncertainty in Illness Theory (Mishel,1988) focuses on the concept of uncertaintytheinability of a person to determine the meaning ofillness-related events. According to this theory, peo-ple develop subjective appraisals to assist them ininterpreting the experience of illness and treatment.Uncertainty occurs when people are unable to rec-ognize and categorize stimuli. Uncertainty results inthe inability to obtain a clear conception of the situ-ation, but a situation appraised as uncertain will

adaptation. Within the human system, there arefour subsystems: physiologic needs, self-concept,role function, and interdependence. These subsys-tems constitute adaptive modes that provide mech-anisms for coping with environmental stimuli andchange. The goal of nursing, according to thismodel, is to promote client adaptation; nursing alsoregulates stimuli affecting adaptation. Nursinginterventions usually take the form of increasing,decreasing, modifying, removing, or maintaininginternal and external stimuli that affect adaptation.

Example using Roys model:Cook, Green, and Topp (2001) explored the

incidence and impact of physician verbal abuse onperioperative nurses, using the Roy Adaptation Modelas their conceptual framework. The researchers exam-ined how nurses used adaptive coping behaviors andproblem-focused skills to deal with the abuse.

Orems Self-Care ModelOrems Self-Care Model focuses on each indi-viduals ability to perform self-care, defined asthe practice of activities that individuals initiateand perform on their own behalf in maintaininglife, health, and well-being (1985, p. 35). Abilityto care for oneself is referred to as self-careagency, and the ability to care for others isreferred to as dependent-care agency. In Oremsmodel, the goal of nursing is to help people meettheir own therapeutic self-care demands. Oremidentified three types of nursing systems: (1)wholly compensatory, wherein the nurse compen-sates for the patients total inability to performself-care activities; (2) partially compensatory,wherein the nurse compensates for the patientspartial inability to perform these activities; and (3)supportiveeducative, wherein the nurse assiststhe patient in making decisions and acquiringskills and knowledge.

Example using Orems model:McCaleb and Cull (2000) studied the influ-

ence of sociocultural characteristics and economiccircumstances on the self-care practices of middleadolescents, using Orems model as the framework.



THEORIST AND NAME OF KEY THESIS OF THEREFERENCE MODEL/THEORY MODEL RESEARCH EXAMPLE

Imogene King, Open Systems Personal systems, interper- Doornbos (2000) based her1981 Model sonal systems, and social framework on Kings model; she

systems are dynamic and tested the prediction that family interacting, within which stressors, coping, and other transactions occurs. factors affected family health

with young adults with seriousmental illness.

Madeline Leininger, Theory of Culture Caring is a universal Raines and Morgan (2000) 1991 Care Diversity and phenomenon but varies studied the culturally grounded

Universality transculturally. meanings of the concept ofcomfort, presence, and involvement in the context of thechildbirth experience of blackwomen and white women.

Myra Levine, Conservation Conservation of integrity Deiriggi and Miles (1995) 1973 Model contributes to maintenance based their study of the effects

of a persons wholeness. of waterbeds on heart rate inpreterm infants on Levines concept of conservation.

Betty Neuman, Health Care Each person is a complete Brauer (2001) described1989 Systems Model system; the goal of nursing common patterns of person

is to assist in maintaining environment interaction inclient system stability. adults with rheumatoid arthritis,

based on Neumans model.

Margaret Newman, Health as Expanding Health is viewed as an Endo and colleagues (2000)1994 Consciousness expansion of consciousness used Newmans theory to study

with health and disease pattern recognition as a caringparts of the same whole; partnership between nurseshealth is seen in an evolving and families of ovarian cancerpattern of the whole in time, in Japan.space, and movement.

Dorothea Orem, Self-Care Model Self-care activities are what Anderson (2001) explored, with1985 people do on their own a sample of homeless adults,

behalf to maintain health the relationship between self-and well-being; the goal of care, self-care agency, and nursing is to help people well-being.meet their own therapeutic self-care demands.

TABLE 6.1 Conceptual Models of Nursing Used by Nurse Researchers


Other Models Used by Nurse Researchers

Many phenomena in which nurse researchers areinterested involve concepts that are not unique to nursing, and therefore their studies are some-times linked to conceptual models that are notmodels from the nursing profession. In addition to the previously described Theory of Planned

mobilize individuals to use their resources to adaptto the situation. Mishels conceptualization ofuncertainty has been used as a framework for bothqualitative and quantitative nursing studies.

Example using Uncertainty in Illness Theory:Santacroce (2001) studied uncertainty in 25

mothers during their infants diagnosis; the infantswere HIV seropositive.

THEORIST AND NAME OF KEY THESIS OF THEREFERENCE MODEL/THEORY MODEL RESEARCH EXAMPLE

Rosemarie Rizzo Theory of Human Health and meaning are Mitchell and Lawton (2000) Parse, 1992, Becoming co-created by indivisible studied how diabetic patients1995 humans and their experienced the consequences

environment; nursing of personal choices about livinginvolves having clients with restrictions, and discussedshare views about the emerging concepts withinmeanings. Parses theory.

Martha Rogers, Science of Unitary The individual is a unified Using Rogers framework,1970, 1986 Human Beings whole in constant interaction Bays (2001) explored the

with the environment; nurs- phenomenon of hope anding helps individuals achieve associated factors in oldermaximum well-being patients who hadwithin their potential. experienced a stroke.

Sr. Callista Roy, Adaptation Model Humans are adaptive Roys Adaptation Model1984, 1991 systems that cope with provided the framework

change through adaptation; for Johns (2001) studynursing helps to promote of whether perceptionsclient adaptation during of quality of life changehealth and illness. over time in adults who

receive curative radiation therapy.

Jean Watson, Theory of Caring Caring is the moral ideal, Using Watsons 10 carative1999 and entails mindbodysoul factors, Baldursdottir and

engagement with Jonsdottir (2002) studied theone another. importance of nurse caring

behaviors as perceived bypatients receiving care at an emergency department.

TABLE 6.1 Conceptual Models of Nursing Used by Nurse Researchers (continued)


or her well-being. The model posits that copingstrategies are learned, deliberate responses used toadapt to or change stressors. According to thismodel, a persons perception of mental and physi-cal health is related to the ways he or she evaluatesand copes with the stresses of living.

Example using the Theory of Stress andCoping:

Maurier and Northcott (2000) used the Lazarus andFolkman theory as the conceptual framework in astudy that examined whether job uncertainty, work-ing conditions, cognitive appraisal, and copingstrategies affected the health of nurses during therestructuring of health care in Alberta.

Banduras Social Cognitive TheorySocial Cognitive Theory (Bandura, 1986, 1997)offers an explanation of human behavior using theconcepts of self-efficacy, outcome expectations, andincentives. Self-efficacy expectations are focusedon peoples belief in their own capacity to carry outparticular behaviors (e.g., smoking cessation). Self-efficacy expectations, which are context-specific,determine the behaviors a person chooses to per-form, their degree of perseverance, and the qualityof the performance. Bandura identified four factorsthat influence a persons cognitive appraisal of self-efficacy: (1) their own mastery experience; (2) ver-bal persuasion; (3) vicarious experience; and (4)physiologic and affective cues, such as pain andanxiety. The role of self-efficacy has been studied inrelation to numerous health behaviors such asweight control, self-management of chronic illness,phobic reactions, and smoking.

Example using Social Cognitive Theory:Using social cognitive constructs, Resnick

(2001) tested a model of factors that influence theexercise behavior of older adults.

Theoretical Contexts and Nursing Research

As previously noted, theory and research have rec-iprocal, beneficial ties. Fawcett (1978) describedthe relationship between theory and research as a

Behavior, three non-nursing models or theorieshave frequently been used in nursing researchinvestigations: the Health Belief Model, Lazarusand Folkmans Theory of Stress and Coping, andBanduras Social Cognitive Theory.

The Health Belief ModelThe Health Belief Model (HBM; Becker, 1978) hasbecome a popular conceptual framework in nursingstudies focused on patient compliance and preven-tive health care practices. The model postulates thathealth-seeking behavior is influenced by a personsperception of a threat posed by a health problemand the value associated with actions aimed atreducing the threat. The major components of theHBM include perceived susceptibility, perceivedseverity, perceived benefits and costs, motivation,and enabling or modifying factors. Perceived sus-ceptibility is a persons perception that a healthproblem is personally relevant or that a diagnosis isaccurate. Even when one recognizes personal sus-ceptibility, action will not occur unless the individ-ual perceives the severity to be high enough to haveserious organic or social implications. Perceivedbenefits are the patients beliefs that a given treat-ment will cure the illness or help prevent it, andperceived costs are the complexity, duration, andaccessibility of the treatment. Motivation is thedesire to comply with a treatment. Among the mod-ifying factors that have been identified are person-ality variables, patient satisfaction, and sociodemo-graphic factors.

Example using the HBM:Petro-Nustas (2001) used the HBM as the

theoretical framework for a study of youngJordanian womens health beliefs toward mammog-raphy as a screening procedure for breast cancer.

Lazarus and Folkmans Theory of Stress and CopingThe Theory of Stress and Coping (Folkman &Lazarus, 1988; Lazarus, 1966) is an effort toexplain peoples methods of dealing with stress,that is, environmental and internal demands thattax or exceed a persons resources and endanger his


TEST ING, USING, ANDDEVELOPING A THEORYOR FRAMEWORK

The manner in which theory and conceptual frame-works are used by qualitative and quantitativeresearchers is elaborated on in the following sec-tion. In the discussion, the term theory is used in itsbroadest sense to include both conceptual modelsand formal theories.

Theories and Qualitative Research

Theory is almost always present in studies that areembedded in a qualitative research tradition such asethnography or phenomenology. As previouslynoted, these research traditions inherently providean overarching framework that give qualitativestudies a theoretical grounding. However, differenttraditions involve theory in different ways.

Sandelowski (1993) makes a useful distinctionbetween substantive theory (conceptualizations ofthe target phenomena that are being studied) and the-ory that reflects a conceptualization of human inquiry.Some qualitative researchers insist on an atheoreticalstance vis--vis the phenomenon of interest, with thegoal of suspending a priori conceptualizations (sub-stantive theories) that might bias their collection andanalysis of data. For example, phenomenologists arein general committed to theoretical naivet, andexplicitly try to hold preconceived views of the phe-nomenon in check. Nevertheless, phenomenologistsare guided in their inquiries by a framework or phi-losophy that focuses their analysis on certain aspectsof a persons lifeworld. That framework is based onthe premise that human experience is an inherentproperty of the experience itself, not constructed byan outside observer.

Ethnographers typically bring a strong culturalperspective to their studies, and this perspectiveshapes their initial fieldwork. Fetterman (1989) hasobserved that most ethnographers adopt one of twocultural theories: ideational theories, which sug-gest that cultural conditions and adaptation stemfrom mental activity and ideas, or materialistictheories, which view material conditions (e.g.,

double helix, with theory as the impetus of scien-tific investigations, and with findings from researchshaping the development of theory. However, thisrelationship has not always characterized theprogress of nursing science. Many have criticizednurse researchers for producing numerous isolatedstudies that are not placed in a theoretical context.

This criticism was more justified a decade agothan it is today. Many researchers are developingstudies on the basis of conceptual models of nurs-ing. Nursing science is still struggling, however, tointegrate accumulated knowledge within theoreti-cal systems. This struggle is reflected, in part, inthe number of controversies surrounding the issueof theoretical frameworks in nursing.

One of these controversies concerns whetherthere should be one single, unified model of nursingor multiple, competing models. Fawcett (1989) hasargued against combining different models, notingthat before all nurses follow the same path, thecompetition of multiple models is needed to deter-mine the superiority of one or more of them (p. 9).Research can play a critical role in testing the utilityand validity of alternative nursing models.

Another controversy involves the desirabilityand utility of developing theories unique to nursing.Some commentators argue that theories relating tohumans developed in other disciplines, such as phys-iology, psychology, and sociology (so-called bor-rowed theories), can and should be applied to nurs-ing problems. Others advocate the development ofunique nursing theories, claiming that only throughsuch development can knowledge to guide nursingpractice be produced. Fawcett (1995) argues thatborrowed theories are sometimes used without con-sidering their adequacy for nursing inquiry. When aborrowed theory is tested and found to be empiricallyadequate in health-relevant situations of interest tonurses, it becomes shared theory.

Until these controversies are resolved, nurs-ing research is likely to continue on its currentpath of conducting studies within a multidiscipli-nary and multitheoretical perspective. We areinclined to see the use of multiple frameworks asa healthy and unavoidable part of the developmentof nursing science.


In grounded theory studies, theory is producedfrom the inside, but theory can also enter aqualitative study from the outside. That is, somequalitative researchers use existing theory as aninterpretive framework. For example, a number ofqualitative nurse researchers acknowledge that thephilosophic roots of their studies lie in conceptualmodels of nursing such as those developed byNeuman, Parse, and Rogers. Other qualitativeresearchers use substantive theories about the tar-get phenomenon as a comparative context for inter-preting data after the researcher has undertaken apreliminary analysis. Sandelowski (1993) notesthat, in this manner, previous substantive theoriesor conceptualizations are essentially data them-selves, and can be taken into consideration, alongwith study data, as part of an inductively drivennew conceptualization.

Example of using existing theory as an interpretive framework:

In Yehs (2001) qualitative study of the adaptationprocess of 34 Taiwanese children with cancer, sheused Roys Adaptation Model as a guide for in-depth interviews and also for her data analysis.

An integrative review of qualitative researchstudies in a specific topic is another strategy that canlead to theory development. In such integrativereviews, qualitative studies are combined to identifytheir essential elements. These findings from differ-ent sources are then used for theory building.Paterson (2001), for example, used the results of 292qualitative studies that described the experiences ofadults with chronic illness to develop the shiftingperspectives model of chronic illness. This modeldepicts living with chronic illness as an ongoing,constantly shifting process in which individualsperspectives change in the amount to which illness isin the foreground or background in their lives.

Theories in Quantitative Research

Quantitative researchers, like qualitative researchers,link research to theory or models in several ways.The classic approach is to test hypotheses deducedfrom a previously proposed theory.

resources, money, production) as the source of cul-tural developments.

The theoretical underpinning of grounded theorystudies is symbolic interactionism, which stressesthat behavior is developed through human interac-tions, through ongoing processes of negotiation andrenegotiation. Similar to phenomenologists, however,grounded theory researchers attempt to hold priorsubstantive theory (existing knowledge and concep-tualizations about the phenomenon) in abeyance untiltheir substantive theory begins to emerge. Once thetheory starts to take shape, grounded theoristsdo not ignore the literature; rather, previous literatureis used for comparison with the emerging and devel-oping categories of the theory. The goal of groundedtheory researchers is to develop a conceptualiza-tion of a phenomenon that is grounded in actualobservationsthat is, to explicate an empiricallybased conceptualization for integrating and makingsense of a process or phenomenon. Theory develop-ment in a grounded theory study is an inductiveprocess. Grounded theory researchers seek to identifypatterns, commonalities, and relationships throughthe scrutiny of specific instances and events. Duringthe ongoing analysis of data, the researchers movefrom specific pieces of data to abstract generalizationsthat synthesize and give structure to the observed phe-nomenon. The goal is to use the data, grounded inreality, to provide a description or an explanation ofevents as they occur in realitynot as they have beenconceptualized in preexisting theories. Grounded the-ory methods are designed to facilitate the generationof theory that is conceptually dense, that is, withmany conceptual patterns and relationships.

Example of a grounded theory study:Schreiber, Stern, and Wilson (2000) devel-

oped a grounded theory of how black West Indian-Canadian women manage depression and its stigma.These women from a nondominant cultural back-ground used the process they called being strongto manage depression. Being strong includedthree subprocesses of dwelling on it, divertingmyself, and regaining composure. As illustratedin their graphic model (Figure 6-2), these sub-processes overlap.


TIP: If you are testing a specific theory orconceptual model, be sure to read about it

from a primary source. It is important to understandfully the conceptual perspective of the theorist, andthe detailed explication of key constructs.

The focus of the testing process involves acomparison between observed outcomes with thosepredicted in the hypotheses. Through this process,a theory is continually subjected to potential dis-confirmation. If studies repeatedly fail to discon-firm a theory, it gains support and acceptance (e.g.,the Theory of Planned Behavior). The testingprocess continues until pieces of evidence cannotbe interpreted within the context of the theory butcan be explained by a new theory that alsoaccounts for previous findings. Theory-testingstudies are most useful when researchers deviselogically sound deductions from the theory, designa study that reduces the plausibility of alternative

Testing a TheoryTheories often stimulate new studies. For exam-ple, a nurse might read about Orems Self-CareModel. As reading progresses, conjectures such asthe following might arise: If Orems Self-CareModel is valid, then I would expect that nursingeffectiveness can be enhanced in environmentsmore conducive to self-care (e.g., a birthing roomversus a delivery room). Such a conjecture,derived from a theory or conceptual framework,can serve as a starting point for testing the theorysadequacy.

In testing a theory, researchers deduce impli-cations (as in the preceding example) and developresearch hypotheses, which are predictions aboutthe manner in which variables would be related ifthe theory were correct. The hypotheses are thensubjected to empirical testing through systematicdata collection and analysis.

F I G U R E 6 . 2 Beingstrong: how black West-IndianCanadian women managedepression and its stigma.(Adapted with permissionfrom Schreiber, R., Stern, P. N.,& Wilson, C. [2000]. Beingstrong: How black West-IndianCanadian women managedepression and its stigma.Journal of NursingScholarship, 32, p. 41.)

Male-femaleroles and

relationships

Culturalstigma of

depression

Belief inChristiandoctrine

Divertingmyself

Regainingmy

composure

Dwellingon it

Visible minoritystatus within a

Eurocentric society

Trying new approaches

Tryi

ngne

wap

proa

ches

Tryingnew

approaches


is mediated by primary appraisalthe perception ofan experience as stressful or nonstressful. Chang rea-soned that an intervention that affects primaryappraisal could positively affect caregiver anxiety anddepression. She developed a cognitive-behavioralintervention designed to provide caregivers withknowledge and skills to improve the PWDs eatingand dressing abilities, and also to increase caregiverknowledge of coping strategies. In a careful studythat compared caregivers who received the interven-tion with those who did not, Chang found thatdepression decreased more in the intervention group.

Researchers sometimes combine elementsfrom more than one theory as a basis for generatinghypotheses. In doing this, researchers need to bethoroughly knowledgeable about both theories tosee if there is an adequate conceptual and empiricalbasis for conjoining them. If underlying assump-tions or conceptual definitions of key concepts arenot compatible, the theories should not be com-bined (although perhaps elements of the two can beused to create a new conceptual framework with itsown assumptions and definitions). Two theoriesthat have given rise to combinatory efforts are theHBM and the Theory of Reasoned Action.

Example of testing two combined models:Poss (2001) combined the HBM and the

Theory of Reasoned Action to examine the factorsassociated with the participation by Mexicanmigrant farm workers in a tuberculosis screeningprogram. Figure 6-3 illustrates how Poss integratedthese two models/theories in her study.

Testing Two Competing TheoriesResearchers who directly test two competing theo-ries to explain a phenomenon are in a particularlygood position to advance knowledge. Almost allphenomena can be explained in alternative ways, assuggested by the alternative conceptual models ofnursing. There are also competing theories for suchphenomena as stress, child development, andgrieving, all of which are important to nursing.Each competing theory suggests alternativeapproaches to facilitating a positive outcome orminimizing a negative one. In designing effective

explanations for observed relationships, and usemethods that assess the theorys validity undermaximally heterogeneous situations so that poten-tially competing theories can be ruled out.

Researchers sometimes base a new study on atheory in an effort to explain earlier descriptive find-ings. For example, suppose several researchers hadfound that nursing home patients demonstrategreater levels of anxiety and noncompliance withnursing staff around bedtime than at other times.These findings shed no light on underlying causes ofthe problem, and consequently suggest no way toimprove it. Several explanations, rooted in modelssuch as Lazarus and Folkmans Stress and CopingModel or Neumans Health Care Systems Model,may be relevant in explaining nursing homepatients behavior. By directly testing the theory in anew study (i.e., deducing hypotheses derived fromthe theory), a researcher might learn why bedtime isa vulnerable period for the elderly in nursing homes.

TIP: It may be useful to read researchreports of other studies that were based on a

theory in which you are interestedeven if theresearch problem is not similar to your own. Byreading other studies, you will be better able to judgehow much empirical support the theory has receivedand perhaps how the theory should be adapted.

Tests of a theory sometimes take the form oftesting a theory-based intervention. If a theory iscorrect, it has implications for strategies to influ-ence peoples attitudes or behavior, includinghealth-related ones. The impetus for an interventionmay be a theory developed within the context ofqualitative studies, as in the example of Swansonstheory of caring described later in this chapter. Theactual tests of the effectiveness of the interven-tionwhich are also indirect tests of the theoryare usually done in structured quantitative research.

Example of theory testing in an interven-tion study:

Chang (1999) used Lazarus and Folkmans Theoryof Stress and Coping to develop and test an inter-vention for homebound caregivers of persons withdementia (PWD). According to the theory, the rela-tionship between stress and a persons coping ability


a studyan enterprise that may be difficult forbeginning researchers. For advanced students,Chinn and Kramer (1999) and Fawcett (1999) pre-sent criteria for assessing conceptual frameworks.Box 6-1 presents a few basic questions that can beasked in a preliminary assessment of a theory ormodel.

The researcher who directly tests two (ormore) competing theories, using a single sample ofsubjects and comparable measures of the keyresearch variables, is in a position to make power-ful and meaningful comparisons. Such a studyrequires considerable advance planning and theinclusion of a wider array of measures than wouldotherwise be the case, but such efforts are impor-tant. In recent years, several nurse researchers haveused this approach to generate and refine ourknowledge base and to provide promising newleads for further research.

Example of a test of competing theories:Yarcheski, Mahon, and Yarcheski (1999)

tested three alternative theories of anger in earlyadolescents: one relating anger to stress; anotherattributing anger to differential emotions; and athird relating anger to personality traits. The find-ings suggested that all three theories are sound andrelevant explanations, but the trait theory providedthe most powerful explanation.

Using a Model or Theory as an Organizing StructureMany researchers who cite a theory or model astheir framework are not directly testing the theory.Silva (1986), in her analysis of 62 studies that used5 nursing models, found that only 9 were direct andexplicit tests of the models cited by the researchers.She found that the most common use of nursingmodels in empirical studies was to provide an orga-nizing structure. In such an approach, a researcherbegins with a broad conceptualization of nursing(or stress, health beliefs, and so on) that is consis-tent with that of the model developers. Theseresearchers assume that the models they espouseare valid, and then use the models constructs orproposed schemas to provide a broad organizational

nursing interventions, it is important to know whichexplanation has more validity.

Typically, researchers test a single theory (orone combined model) in a study. Then, to evaluatethe worth of competing theories, they must com-pare the results of different studies. Such compar-isons are problematic because each study design isunique. For example, one study of stress might usea sample of college students taking an examination,another might use military personnel in a combatsituation, and yet another might use terminally illpatients with cancer. Each study might, in additionto having divergent samples, measure stress differ-ently. If the results of these studies support alterna-tive theories of stress to different degrees, it wouldbe difficult to know the extent to which the resultsreflected differences in study design rather than dif-ferences in the validity of the theories.

TIP: It is often suggested that theories firstbe evaluated before being used as a basis for

F I G U R E 6 . 3 Combined Health Belief Model andTheory of Reasoned Action. (Adapted with permissionfrom Poss, J. E. [2001]. Developing a new model forcross-cultural research: Synthesizing the Health BeliefModel and the Theory of Reasoned Action. Advances inNursing Science, 23, p. 12.)

HBM + TRA

Susceptibility

Cues toaction

Behavioralbeliefs

Severity

Normativebeliefs

Attitude

Intention

Subjectivenorm

HBM

TRA

Behavior


To our knowledge, Silvas study has not beenreplicated with a more recent sample of studies.However, we suspect that, even today, most quanti-tative studies that offer models and theories as theirconceptual frameworks are using them primarily asorganizational or interpretive tools. Silva (1986)offered seven evaluation criteria for determiningwhether a study was actually testing a theory,rather than simply identifying an organizationalcontext. Box 6-2, broadly adapted from Silvas cri-teria, presents a set of evaluative questions to deter-mine if a study was actually testing a theory.

Example of a study meeting seven theory-testing criteria:

Woods and Isenberg (2001) tested one aspect of amiddle-range theory based on the Roy Adaptation

or interpretive context. Some use (or develop) datacollection instruments that are allied with themodel. Silva noted that using models in this fash-ion can serve a valuable organizing purpose, butsuch studies offer little evidence about the validityof the theory itself.

Example of using a model as organizingstructure:

Resnick and Jenkins (2000) used Banduras socialcognitive theory as an organizing structure to revisean unpublished instrument to measure self-efficacybarriers to exercise. Focusing on Banduras con-struct of self-efficacy, Resnick and Jenkins devel-oped the Self-Efficacy for Exercise Scale. Theyassessed their new measure with 187 older adultsliving in a continuing care retirement community.

BOX 6.1 Some Questions for a Preliminary Assessment of a Model or Theory

ISSUE QUESTIONS

Theoretical clarity Are key concepts defined and are the definitions sufficiently clear? Do all concepts fit within the theory? Are concepts used in the theory in a

manner compatible with conceptual definitions? Are basic assumptions consistent with one another? Are schematic models helpful, and are they compatible with the text? Are

schematic models needed but not presented? Can the theory be followedis it adequately explained? Are there ambiguities?

Theoretical complexity Is the theory sufficiently rich and detailed to explain phenomena of interest? Is the theory overly complex? Can the theory be used to explain or predict, or only to describe phenomena?

Theoretical grounding Are the concepts identifiable in reality? Is there an empirical basis for the theory? Is the empirical basis solid? Can the theoretical concepts be adequately operationalized?

Appropriateness of the Does the theory suggest possibilities for influencing nursing practice?theory Are the tenets of the theory compatible with nursings philosophy?

Are key concepts within the domain of nursing?

Importance of the theory Will research based on this theory answer critical questions? How will testing the theory contribute to nursings evidence base?

General issues Are there theories or models that do a better job of explaining phenomenonof interest?

Is the theory compatible with your world view?


theoretical context. Although in some situationssuch an approach may be appropriate, we never-theless caution that an after-the-fact linkage of the-ory to a problem may add little to the studys worthand, of course, no evidence of the theorys validity.(An exception is when the researcher is strugglingto make sense of findings, and calls on an existingtheory to help explain or interpret them.)

If it is necessary to find a relevant theoryafter selecting a problem, the search for such atheory must begin by first conceptualizing theproblem on an abstract level. For example, takethe research question: Do daily telephone con-versations between a psychiatric nurse and apatient for 2 weeks after discharge from the hos-pital result in lower rates of readmission byshort-term psychiatric patients? This is a rela-tively concrete research problem, but it mightprofitably be viewed within the context ofOrems Self-Care Model, a theory of reinforce-ment, a theory of social influence, or a theory ofcrisis resolution. Part of the difficulty in findinga theory is that a single phenomenon of interestcan be conceptualized in a number of ways and,depending on the manner chosen, may refer theresearcher to conceptual schemes from a widerange of disciplines.

Model. Their purpose was to test the efficacy ofadaptation as a mediator of intimate abuse andtraumatic stress in battered women. They testedthe following two relational statements: (1) phys-ical abuse, emotional abuse, and risk of homicideare focal stimuli that elicit the response of posttraumatic stress disorder (PTSD) in batteredwomen and (2) adaptation in the physiologic, self-concept, role, and interdependence modes acts as amediator between the focal stimuli of intimatephysical abuse, emotional abuse, and risk of homi-cide and the response of PTSD in women(p. 215). As a result of empirical testing, Woodsand Isenberg reported that adaptation in three ofthe four modes partially mediated the relationshipbetween intimate abuse, the focal stimuli, and theresponse of PTSD.

Fitting a Problem to a TheoryThe preceding sections addressed the situation inwhich a researcher begins with a specific theory ormodel and uses it either as the basis for developinghypotheses or for an organizational or interpretivepurpose. Circumstances sometimes arise in whichthe problem is formulated before any considerationis given to a conceptual framework. Even in suchsituations, researchers sometimes try to devise a

BOX 6.2 Criteria to Determine if a Theory/Model is Being Tested

1. Is the purpose of the study to determine the validity of a theorys assumptions or propositions?2. Does the report explicitly note that the theory is the framework for the research?3. Is the theory discussed in sufficient detail that the relationship between the theory on the one hand and the

study hypotheses or research questions on the other is clear?4. Are study hypotheses directly deduced from the theory?5. Are study hypotheses empirically tested in an appropriate manner, so as to shed light on the validity of the

theory?6. Is the validity of the theorys assumptions or propositions supported (or challenged) based on evidence from

the empirical tests?7. Does the report discuss how evidence from empirical tests supports or refutes the theory, or how the theory

explains relevant aspects of the findings?

Adapted from Silva M. C. (1986). Research testing nursing theory: State of the art. Advances in Nursing Science, 9, 111.


model, be willing to adapt or augment your originalresearch problem as you gain greater understandingof the theory. The linking of theory and researchquestion may involve an iterative approach.

Developing a Framework in a Quantitative StudyNovice researchers may think of themselves asunqualified to develop a conceptual scheme of theirown. But theory development depends much less onresearch experience than on powers of observation,grasp of a problem, and knowledge of prior research.There is nothing to prevent an imaginative and sensi-tive person from formulating an original conceptualframework for a study. The conceptual scheme maynot be a full-fledged formal theory, but it should placethe issues of the study into some broader perspective.

The basic intellectual process underlying theorydevelopment is inductionthat is, the process ofreasoning from particular observations and facts togeneralizations. The inductive process involves inte-grating what one has experienced or learned intosome conclusion. The observations used in theinductive process need not be personal observations;they may be (and often are) the findings and conclu-sions from other studies. When relationships amongvariables are arrived at this way, one has the makingsof a theory that can be put to a more rigorous test.The first step in theory development, then, is to for-mulate a generalized scheme of relevant concepts,that is, to perform a conceptual analysis.

Let us consider the following simple example.Suppose that we were interested in understandingthe factors influencing enrollment in a prenataleducation program. We might begin by consideringtwo basic sets of forces: those that promote enroll-ment and those that hinder it. After reviewing theliterature, discussing the problem with colleagues,and developing ideas from our own experiences,we might arrive at a conceptual scheme such as theone presented in Figure 6-4. This framework iscrude, but it does allow us to study a number ofresearch questions and to place those problems inperspective. For example, the conceptual schemesuggests that as the availability of social supportsdeclines, obstacles to participation in a prenatal

TIP: If you begin with a research problemand are trying to identify a suitable framework,

it is probably wise to confer with othersespeciallywith people who may be familiar with a broad rangeof theoretical perspectives. By having an open discus-sion, you are more likely to become aware of yourown conceptual perspectives and are thus in a betterposition to identify an appropriate framework.

Textbooks, handbooks, and encyclopedias inthe chosen discipline usually are a good startingpoint for the identification of a framework. Thesesources usually summarize the status of a theoreti-cal position and document the efforts to confirmand disconfirm it. Journal articles contain more cur-rent information but are usually restricted todescriptions of specific studies rather than to broadexpositions or evaluations of theories. Perhaps ourbrief overview of frameworks that have been usefulto nurses can serve as a starting point for identify-ing a suitable model or theory.

Fitting a problem to a theory after-the-factshould be done with circumspection. It is true thathaving a theoretical context can enhance the mean-ingfulness of a study, but artificially cramming aproblem into a theory is not the route to scientificutility, nor to enhancing nursings evidence base.There are many published studies that purport tohave a conceptual framework when, in fact, the ten-uous post hoc linkage is all too evident. In Silvas(1986) previously mentioned analysis of 62 studiesthat claimed a nursing model as their underpin-nings, approximately one third essentially paidonly lip service to a model. If a conceptual frame-work is really linked to a research problem, thenthe design of the study, the selection of data collec-tion methods, the data analysis, and (especially) theinterpretation of the findings flow from that con-ceptualization. We advocate a balanced and rea-soned perspective on this issue: Researchers shouldnot shirk their intellectual duties by ignoringattempts to link their problem to broader theoreti-cal concerns, but there is no point in fabricatingsuch a link when it does not exist.

TIP: If you begin with a research questionand then subsequently identify a theory or


education program increase. We might then makethe following hypothesis: Single pregnant womenare less likely to participate in a prenatal educationprogram than married pregnant women, on theassumption that husbands are an important sourceof social support to women in their pregnancy. (Ofcourse, this example is contrived; in reality, severalexisting theories such as the HBM or TPB could beused to study enrollment in prenatal care educa-tion.) Many nursing studies involve conceptualframeworks developed by the researchers.

Example of model development:Stuifbergen, Seraphine, and Roberts (2000)

conducted a study based on their own conceptualmodel of quality of life in persons with chronic dis-abling conditions. The model represented a syn-thesis of findings from extant literature and a seriesof preliminary qualitative and quantitative investi-gations (p. 123).

RESEARCH EXAMPLES

Throughout this chapter, we have described studiesthat involved various widely used conceptual and the-oretical models. This section presents two examplesof the linkages between theory and research from

the nursing research literatureone from a quantita-tive study and the other from a qualitative study.

Research Example From aQuantitative Study: Testing Orems Self-Care Model

Renker (1999) used Orems Self-Care Model of nurs-ing to study the relationships between self-care, socialsupport, physical abuse, and pregnancy outcomes ofolder adolescent mothers and their infants. Thestudys research variables included measures of themajor constructs in Orems model, including basicconditioning factors, self-care agency, and self-care.Orems basic conditioning factors (factors that affectpeoples ability to engage in self-care) include (1)social-environmental factors and (2) resource avail-ability and adequacy factors. In Renkers study, phys-ical abuse represented the key social environmentalfactor, and social support represented the resourceavailability and adequacy factor. Denyes Self-CareAgency Instrument was used to measure pregnancyself-care agency, and Denyes Self-Care PracticeInstrument measured pregnancy self-care.

Based on Orems model, Renker hypothesizedthat the absence of physical abuse and the presence ofsocial support increased self-care agency. Increasedlevels of self-care agency were expected to enhance

F I G U R E 6 . 4 Conceptual modelfactors that influence enrollment in a prenatal education program.

Age/maturity

Wantedness/intendednessof pregnancy

Enrollment in prenataleducational program

Maternal/paternaleducational level

Lack of social supports

Practical impediments(cost, accessibility)

Competing time demands

Positive Factors Negative Factors


ing way of relating to a valued other toward whomone feels a personal sense of commitment and respon-sibility (p. 165).

According to Swansons theory, the five caringprocesses are as follows:

Knowingstriving to understand an event as it hasmeaning in the life of the other

Being Withbeing emotionally present to the other Doing Fordoing for the other as he or she would

do for the self if it were at all possible Enablingfacilitating the others passage through

life transitions and unfamiliar events Maintaining Beliefsustaining faith in the others

capacity to get through an event or transition andface a future with meaning

In presenting her theory, Swanson described thefive processes, supporting each with rich excerpts fromher in-depth interviews. Here is an example of theexcerpt illustrating the process of knowing:

When things werent right, I could say that things were fineand it was only a matter of time. I mean the nurse wouldask certain questions and there would be no way that Icould be consistent without telling the truth. And then wewould talk, and pretty soon instead of saying it was fine, Iwould start out with what was really wrong. (p. 163)

Swansons theory of caring, in addition to beingused in the development and testing of a caring-basednurse counseling program for women who miscarry(Swanson, 1999), has been used by other researchers,including a qualitative study of the interactions ofAIDS family caregivers and professional health careproviders (Powell-Cope, 1994) and a study of theinvolvement of relatives in the care of the dying(Andershed & Ternestedt, 1999).

SUMMARY POINTS

A theory is a broad abstract characterization ofphenomena. As classically defined, a theory isan abstract generalization that systematicallyexplains relationships among phenomena.Descriptive theory thoroughly describes aphenomenon.

In a research context, the overall objective of the-ory is to make findings meaningful, to summa-rize existing knowledge into coherent systems,

self-care practices, which in turn were expected toresult in increased infant birth weight and decreasedpregnancy complications. Renker tested her hypothe-ses in a sample of 139 pregnant teenagers.

The results lent support to Orems model. Abusedpregnant teenagers gave birth to infants with signifi-cantly lower birth weights than the teenagers whowere not abused. Social support, self-care agency, andself-care practice were all significantly related toinfant birth weight.

A particular strength of this study is that OremsSelf-Care Deficit Model was interwoven throughoutits design. Renker developed hypotheses based on themodel and included all major constructs of Oremsmodel as research variables. Moreover, several datacollection instruments were specifically developed toassess components of Orems theory.

Research Example From a Qualitative Study: Development of a Theory of Caring

As noted earlier in this chapter, many qualitative stud-ies have theory development as an explicit goal. Herewe describe the efforts of a qualitative researcher whodeveloped an empirically derived theory of caring,and has used the theory in the development of acaring-based counseling intervention (Swanson,1999). Although the qualitative studies were doneover a decade ago, they are an excellent illustration oftheory development.

Using data from three separate qualitative inves-tigations, Swanson (1991) inductively derived andthen refined a theory of the caring process. Swansonstudied caring in three separate perinatal contexts: asexperienced by women who miscarried, as providedby parents and professionals in the newborn intensivecare unit, and as recalled by at-risk mothers who hadreceived a long-term public health nursing interven-tion. Data were gathered through in-depth interviewswith study participants and also through observationsof care provision. Data from the first study led to theidentification and preliminary definition of five caringprocesses. The outcome of the second study was con-firmation of the five processes and refinement of theirdefinitions. In the third study, Swanson confirmed thefive processes, redefined one of them, developed sub-dimensions of each process, and derived a definitionof the overall concept of caring: Caring is a nurtur-


tions of the phenomena under study, but never-theless there is a rich theoretical underpinningassociated with the tradition itself.

Some qualitative researchers specifically seek todevelop grounded theories, data-driven expla-nations to account for phenomena under studythrough inductive processes.

In the classical use of theory, researchers testhypotheses deduced from an existing theory. Aparticularly fruitful approach is the testing oftwo competing theories in one study.

In both qualitative and quantitative studies,researchers sometimes use a theory or model asan organizing framework, or as an interpretivetool.

Researchers sometimes develop a problem,design a study, and then look for a conceptualframework; such an after-the-fact selection of aframework is less compelling than the systematictesting of a particular theory.

STUDY ACTIVIT IES

Chapter 6 of the Study Guide to AccompanyNursing Research: Principles and Methods, 7thedition, offers various exercises and study sugges-tions for reinforcing concepts presented in thischapter. In addition, the following study questionscan be addressed:

1. Read the following article: Liken, M. A. (2001).Caregivers in crisis. Clinical Nursing Research,10, 5268. What theoretical basis does theauthor use in this study? Would you classify thetheoretical basis as a theory or as a conceptualframework? Did Liken use the framework totest hypotheses formally, or was the frameworkused as an organizational structure?

2. Select one of the nursing conceptual frame-works or models described in this chapter.Formulate a research question and twohypotheses that could be used empirically totest the utility of the conceptual framework ormodel in nursing practice.

3. Four researchable problems follow. Abstract ageneralized issue or issues for each of these

to stimulate new research, and to explain phe-nomena and relationships among them.

The basic components of a theory are concepts;classically defined theories consist of a set ofpropositions about the interrelationships amongconcepts, arranged in a logically interrelatedsystem that permits new statements to bederived from them.

Grand theories (or macrotheories) attempt todescribe large segments of the human experi-ence. Middle-range theories are more specificto certain phenomena.

Concepts are also the basic elements of concep-tual models, but concepts are not linked in alogically ordered, deductive system. Conceptualmodels, like theories, provide context for nurs-ing studies.

Schematic models (sometimes referred to asconceptual maps) are representations of phe-nomena using symbols or diagrams. Statisticalmodels use mathematic symbols to expressquantitatively the nature and strength of rela-tionships among variables.

A framework is the conceptual underpinningof a study. In many studies, the framework isimplicit, but ideally researchers clarify the con-ceptual definitions of key concepts. In qualita-tive studies, the framework usually springs fromdistinct research traditions.

Several conceptual models of nursing have beendeveloped and have been used in nursingresearch. The concepts central to models ofnursing are person, environment, health, andnursing.

Two major conceptual models of nursing usedby nurse researchers are Orems Self-CareModel and Roys Adaptation Model.

Non-nursing models used by nurse researchers(e.g., Lazarus and Folkmans Theory of Stressand Coping) are referred to as borrowed theo-ries; when the appropriateness of borrowed the-ories for nursing inquiry is confirmed, the theo-ries become shared theories.

In some qualitative research traditions (e.g.,phenomenology), the researcher strives to sus-pend previously held substantive conceptualiza-


Fawcett, J. (1989). Analysis and evaluation of conceptualmodels of nursing (2nd ed.). Philadelphia: F. A. Davis.

Fawcett, J. (1995). Analysis and evaluation of conceptualmodels of nursing (3rd ed.). Philadelphia: F. A. Davis.

Fawcett, J. (1999). The relationship between theory andresearch (3rd ed.). Philadelphia: F. A. Davis.

Fetterman, D. M. (1989). Ethnography: Step by step.Newbury Park, CA: Sage.

Flaskerud, J. H. (1984). Nursing models as conceptualframeworks for research. Western Journal of NursingResearch, 6, 153155.

Flaskerud, J. H., & Halloran, E. J. (1980). Areas ofagreement in nursing theory development. Advancesin Nursing Science, 3, 17.

Folkman, S., & Lazarus, R. S. (1988). Coping as a medi-ator of emotion. Journal of Personality and SocialPsychology, 54, 466475.

Hardy, M. E. (1974). Theories: Components, develop-ment, evaluation. Nursing Research, 23, 100107.

King, I. M. (1981). A theory for nursing: Systems, con-cepts, and process. New York: John Wiley and Sons.

Lazarus, R. (1966). Psychological stress and the copingresponse. New York: McGraw-Hill.

Leininger, M. (1991). Culture care diversity and univer-sality: A theory of nursing. New York: NationalLeague for Nursing.

Levine, M. E. (1973). Introduction to clinical nursing(2nd ed.). Philadelphia: F. A. Davis.

Marriner-Tomey, A. (Ed.). (1998). Nursing theorists andtheir work (4th ed.). St. Louis: C. V. Mosby.

Mehrabian, A., & Epstein, N. (1972). A measure of emo-tional empathy. Journal of Personality, 40, 525543.

Meleis, A. I. (1997). Theoretical nursing: Developmentand progress (3rd ed.). Philadelphia: Lippincott-Raven.IQ2

Mishel, M. H. (1988), Uncertainty in illness. Image:Journal of Nursing Scholarship, 20, 225232.

Neuman, B. (1989). The Neuman Systems Model (2nd ed.).Norwalk, CT: Appleton & Lange.

Newman, M. (1994). Health as expanding conscious-ness. New York: NLN.

Nicoll, L. H. (1996). Perspectives on nursing theory.Philadelphia: J. B. Lippincott.

Orem, D. E. (1985). Concepts of practice (3rd ed.). NewYork: McGraw-Hill.

Parse, R. R. (1992). Human becoming: Parses theory.Nursing Science Quarterly, 5, 3542.

Parse, R. R. (1995). Illuminations: The HumanBecoming Theory in practice and research. NewYork: NLN Pub. #15-2670.

problems. Search for an existing theory thatmight be applicable and appropriate.a. What is the relationship between angina

pain and alcohol intake?b. What effect does rapid weight gain during

the second trimester have on the outcome ofpregnancy?

c. Do multiple hospital readmissions affect theachievement level of children?

d. To what extent do coping mechanisms ofindividuals differ in health and illness?

4. Read the following article: Kelly-Powell, M. L.(1997). Personalizing choices: Patients expe-riences with making treatment decisions.Research in Nursing & Health, 20, 219227.What evidence does the researcher offer tosubstantiate that her grounded theory is a goodfit with her data?

SUGGESTED READINGS

Theoretical References

Andrews, H. A., & Roy, C. (1986). Essentials of the RoyAdaptation Model. Norwalk, CT: Appleton-Century-Crofts.

Ajzen, I. (1988). Attitudes, personality, and behavior.Milton Keynes, United Kingdom: Open UniversityPress.

Ajzen, I., & Fishbein, M. (1980). Understanding atti-tudes and predicting social behavior. EnglewoodCliffs, NJ: Prentice-Hall.

Bandura, A. (1986). Social foundations of thought andaction: A social cognitive theory. Englewood Cliffs,NJ: Prentice-Hall.

Bandura, A. (1997). Self-efficacy: The exercise of con-trol. New York: W. H. Freeman.

Becker, M. (1978). The Health Belief Model and sickrole behavior. Nursing Digest, 6, 3540.

Chinn, P. L., & Kramer, M. K. (1999). Theory and nurs-ing: Integrated knowledge development (5th ed.). St. Louis: C. V. Mosby.

Craig, S. L. (1980). Theory development and its rele-vance for nursing. Journal of Advanced Nursing, 5,349355.

Fawcett, J. (1978). The relationship between theory andresearch: A double helix. Advances in NursingScience, 1, 4962.


Brauer, D. J. (2001). Common patterns of person-envi-ronment interaction in persons with rheumatoidarthritis. Western Journal of Nursing Research, 23,414430.

Chang, B. L. (1999). Cognitive-behavioral interventionfor homebound caregivers of persons with dementia.Nursing Research, 48, 173182.

Cook, J. K., Green, M., & Topp, R. V. (2001). Exploringthe impact of physician abuse on perioperative nurses.AORN Journal, 74, 317327.

Deiriggi, P. M., & Miles, K. E. (1995). The effects ofwaterbeds on heart rate in preterm infants. ScholarlyInquiry for Nursing Practice, 9, 245262.

Doornbos, M. M. (2000). Kings systems framework andfamily health: The derivation and testing of a theory.Journal of Theory Construction and Testing, 4,2026.

Endo, E., Nitta., N., Inayoshi, M., Saito, R., Takemura,K., Minegishi, H., Kubo, S., & Kondo, M. (2000).Pattern recognition as a caring partnership in familieswith cancer. Journal of Advanced Nursing, 32,603610.

John, L. D. (2001). Quality of life in patients receivingradiation therapy for non-small cell lung cancer.Oncology Nursing Forum, 28, 807813.

Maurier, W. L., & Northcott, H. C. (2000). Job uncertaintyand health status for nurses during restructuring ofhealth care in Alberta. Western Journal of NursingResearch, 22, 623641.

McCaleb, A., & Cull, V. V. (2000). Sociocultural influ-ences and self-care practices of middle adolescents.Journal of Pediatric Nursing, 15, 3035.

McCullagh, M., Lusk, S. L., & Ronis, D. L. (2002).Factors influencing use of hearing protection amongfarmers. Nursing Research, 51, 3339.

Mitchell, G. J., & Lawton, C. (2000). Living with the con-sequences of personal choices for persons with dia-betes. Canadian Journal of Diabetes Care, 24, 2330.

Paterson, B. L. (2001). The shifting perspectives modelof chronic illness. Journal of Nursing Scholarship,33, 2126.

Pender, N. J., Walker, S. N., Sechrist, K. R., & Frank-Stromborg, M. (1990). Predicting health-promotinglifestyles in the workplace. Nursing Research, 39,326332.

Petro-Nustas, W. (2001). Young Jordanian womenshealth beliefs about mammography. Journal ofCommunity Health Nursing, 18, 177194.

Poss, J. E. (2001). Developing a new model for cross-cultural research: Synthesizing the Health Belief

Pender, N. (1996). Health promotion in nursing practice(3rd ed.). Englewood Cliffs, NJ: Prentice-Hall.

Rogers, M. E. (1970). An introduction to the theoreticalbasis of nursing. Philadelphia: F. A. Davis.

Rogers, M. E. (1986). Science of unitary human beings.In V. Malinski (Ed.), Explorations on Martha Rogersscience of unitary human beings. Norwalk, CT:Appleton-Century-Crofts.

Roy, C. (1984). Introduction to nursing: An adapta-tion model (2nd ed.). Englewood Cliffs, NJ:Prentice-Hall.

Roy, C. Sr., & Andrews, H. (1991). The Roy AdaptationModel: The definitive statement. Norwalk, CT:Appleton & Lange.

Sandelowski, M. (1993). Theory unmasked: The usesand guises of theory in qualitative research. Researchin Nursing & Health, 16, 213218.

Silva, M. C. (1986). Research testing nursing theory:State of the art. Advances in Nursing Science,9, 111.

Silva, M. C., & Sorrell, J. M. (1992). Testing of nursingtheory: Critique and philosophical expansion.Advances in Nursing Science, 14, 1223.

Stevens, B. J. (1994). Nursing theory: Analysis, applica-tion, evaluation (4th ed.). Philadelphia: J. B.Lippincott.

Waltz, C., Strickland, O., & Lenz, E. (1991). Measurementin nursing research. Philadelphia: F. A. Davis.

Watson, J. (1999). Postmodern nursing and beyond. NewYork: Churchill Livingston.

Studies Cited in Chapter 6

Aminzadeh, F., & Edwards, N. (2000). Factors associat-ed with cane use among community-dwelling olderadults. Public Health Nursing, 17, 474483.

Andershed, B., & Ternestedt, B. (1999). Involvement ofrelatives in care of the dying in different care cultures.Nursing Science Quarterly, 12, 4551.

Anderson, J. A. (2001). Understanding homeless adultsby testing the theory of self-care. Nursing ScienceQuarterly, 14, 5967.

Baldursdottir, G., & Jonsdottir, H. (2002). The impor-tance of nurse caring behaviors as perceived bypatients receiving care at an emergency department.Heart & Lung, 31, 6775.

Bays, C. L. (2001). Older adults descriptions of hopeafter stroke. Rehabilitation Nursing, 26, 1820.

Beck, C. T. (1996). A concept analysis of panic, Archivesof Psychiatric Nursing, 10, 265275.


manage depression and its stigma. Journal of NursingScholarship, 32, 3945.

Stuifbergen, A. K., Seraphine, A., & Roberts, G. (2000).An explanatory model of health promotion and qual-ity of life in chronic disabling conditions. NursingResearch, 49, 122129.

Swanson, K. M. (1991). Empirical development of amiddle range theory of caring. Nursing Research, 40,161166.

Swanson, K. M. (1999). Effects of caring, measurement,and time on miscarriage impact and womens well-being. Nursing Research, 48, 288298.

Woods, S. J., & Isenberg, M. A. (2001). Adaptation as amediator of intimate abuse and traumatic stress inbattered women. Nursing Science Quarterly, 14,215221.

Yarcheski, A., Mahon, N. E., & Yarcheski, T. J. (1999).An empirical test of alternate theories of anger inearly adolescents. Nursing Research, 48, 317323.

Yeh, C. H. (2001). Adaptation in children with cancer:Research with Roys model. Nursing ScienceQuarterly, 14, 141148.

Model and the Theory of Reasoned Action. Advancesin Nursing Science, 23, 115.

Powell-Cope, G. M. (1994). Family caregivers of peoplewith AIDS: Negotiating partnerships with professionalhealth care providers. Nursing Research, 43,324330.

Raines, D. A., & Morgan, Z. (2000). Culturally sensitivecare during childbirth. Applied Nursing Research, 13,167172.

Renker, P. R. (1999). Physical abuse, social support, self-care, and pregnancy outcomes of older adolescents.Journal of Obstetric, Gynecologic, and NeonatalNursing, 28, 377388.

Resnick, B. (2001). Testing a model of exercise behavior in older adults. Research in Nursing & Health, 24,8392.

Resnick, B., & Jenkins, L. S. (2000). Testing the reliabil-ity and validity of the Self-Efficacy for ExerciseScale. Nursing Research, 49, 154159.

Santacroce, S. J. (2001). Measuring parental uncertaintyduring the diagnosis phase of serious illness in achild. Journal of Pediatric Nursing, 16, 312.

Schreiber, R., Stern, P. N., & Wilson, C. (2000). Beingstrong: How black West-Indian Canadian women

polit chapter 6

Documents