ARTICLE

10.1177/0894318405274823
Scholarly
Nursing Science
Dialogue
Quarterly , 18:2, April 2005

Scholarly Dialogue Jacqueline Fawcett, Contributing Editor

Criteria for Evaluation of Theory

Jacqueline Fawcett, RN; PhD; FAAN
Professor, College of Nursing and Health Sciences, University of Massachusetts–Boston

This column presents criteria for evaluation of nursing theories specified by Jacqueline Fawcett and Rosemarie
Rizzo Parse. Fawcett’s criteria are significance, internal consistency, parsimony, testability, empirical adequacy,
and pragmatic adequacy. Some of those criteria are differentiated for grand theories and middle-range theories but
are not differentiated by type of data—qualitative or quantitative— used to develop the theory. Parse’s criteria are
structure and process. Structure encompasses historical evolution, foundational elements, and relational
statements. Process encompasses correspondence, coherence, and pragmatics. Parse’s criteria are appropriate for
the critical appraisal of all frameworks and theories, regardless of level of abstraction. Parse also presents a
comparison of her own and Fawcett’s criteria.

Several different sets of criteria for evaluation of theories
have been published (Barnum, 1988; Duffy & Muhlenkamp,
1974; George, 2002; Marriner-Tomey & Alligood, 2002;
Meleis, 1997; Parse, 1987). Just one set of criteria, however,
differentiates between grand theories and middle-range theo-
ries (Fawcett, 2000, 2005). Although none of the authors of
those publications have indicated that the source of the data
for a theory influences the selection of evaluation criteria, a
recent conversation with a colleague raised the question of
whether theories grounded in qualitative data should be eval-
uated using criteria that differ from those used to evaluate the-
ories grounded in quantitative data. That conversation
resulted in an invitation to Rosemarie Rizzo Parse to engage
in a dialogue about what criteria are appropriate for evaluat-
ing grand theories and middle-range theories and whether
those criteria can be applied to theories regardless of the type
of data (qualitative or quantitative) in which a theory is
grounded.
Consensus exists that theories are made up of ideas called
concepts and statements about the concepts, called proposi-
tions (King & Fawcett, 1997). Consensus also exists that
components of nursing knowledge, including theories, vary
in levels of abstraction (King & Fawcett). I regard grand theo-
ries as more abstract than middle-range theories but less ab-
stract than conceptual models (Fawcett, 2005). Accordingly,
my framework includes some differences in the evaluative
Editor’s Note: Any comments about this dialogue should be ad-
dressed to the Editor for possible inclusion in Letters to the Editor.
For other information, contact Jacqueline Fawcett, RN, PhD,
FAAN, 3506 Atlantic Highway, P. O. Box 1156, Waldoboro, ME
04572; phone: (207) 832-7398; E-mail: jacqueline.fawcett@
umb.edu
criteria for grand theories and middle-range theories. I do not,
however, differentiate criteria for evaluation of either grand
or middle-range theories based on the type of data (qualitative
or quantitative) used to develop the theory.
I developed a framework for both the analysis and evalua-
tion of nursing theories several years ago and refined the
framework twice (Fawcett, 1993, 2000, 2005). Analysis in-
volves objective and nonjudgmental descriptions of theories,
whereas evaluation involves judgments about the extent to
which nursing theories meet certain criteria. For the purposes
of this dialogue, my comments are limited to the criteria for
evaluation of grand theories and middle-range theories.
Those criteria are significance, internal consistency, parsi-
mony, testability, empirical adequacy, and pragmatic
adequacy.
The criterion of significance focuses on the context of the
theory. That criterion requires justification of the importance
of the theory to the discipline of nursing and is met when the
metaparadigmatic, philosophical, and conceptual origins of
the theory are explicit, when antecedent nursing and adjunc-
tive knowledge is cited (Levine, 1988), and when the special
contributions made by the theory are identified. The four
questions to be asked when evaluating the significance of a
theory, which are applicable to both grand and middle-range
theories, are listed in Table 1.
The criterion of internal consistency focuses on both the
context and the content of the theory. That criterion requires
all elements of the theorist’s work, including the philosophi-
cal claims, conceptual model, and theory concepts and propo-
sitions, to be congruent. The internal consistency criterion
also requires the concepts of the theory to reflect semantic
clarity and semantic consistency. The semantic clarity re-

Nursing Science Quarterly, Vol. 18 No. 2, April 2005, 131-135

DOI: 10.1177/0894318405274823 Keywords: evaluation criteria, grand theory, middle-
© 2005 Sage Publications range theory, nursing theories

Downloaded from nsq.sagepub.com at Bobst Library, New York University on May 24, 2015
132 Nursing Science Quarterly, 18:2, April 2005

Table 1
Fawcett’s Criteria for Evaluation of Nursing Theories and Pertinent Questions
Criteria Pertinent Questions

Significance Are the metaparadigm concepts and propositions addressed by the theory explicit?
Are the philosophical claims on which the theory is based explicit?
Is the conceptual model from which the theory was derived explicit?
Are the authors of antecedent knowledge from nursing and adjunctive disciplines acknowledged and
are bibliographical citations given?

Internal Consistency Are the context (philosophical claims and conceptual model) and the content (concepts and proposi-
tions) of the theory congruent?
Do the concepts reflect semantic clarity and semantic consistency?
Do the propositions reflect structural consistency?

Parsimony Is the theory content stated clearly and concisely?

Testability: Grand Theories Is the research methodology qualitative and inductive?

Is the research methodology congruent with the philosophical claims and content of the grand theory?
Will the data obtained from use of the research methodology represent sufficiently
in-depth descriptions of one or more personal experience(s) to capture the essence of the grand theory?

Testability: Middle-Range Theories
Does the research methodology reflect the middle-range theory?
Are the middle-range theory concepts observable through instruments that are appropriate empirical in-
dicators of those concepts?
Do the data analysis techniques permit measurement of the middle-range theory
propositions?

Empirical Adequacy: Grand Theories Are the findings from studies of descriptions of personal experiences congruent with the concepts and
propositions of the grand theory?

Empirical Adequacy: Middle-Range Theories Are theoretical assertions congruent with empirical evidence?

Pragmatic Adequacy Are education and special skill training required before application of the theory in nursing practice?
Has the theory been applied in the real world of nursing practice?
Is it generally feasible to implement practice derived from the theory?
Does the practitioner have the legal ability to implement and measure the effectiveness of theory-based
nursing actions?
Are the theory-based nursing actions compatible with expectations for nursing practice?
Do the theory-based nursing actions lead to favorable outcomes?
Is the application of theory-based nursing actions designed so that comparisons
can be made between outcomes of use of the theory and outcomes in the same situation when the the-
ory was not used?
Are outcomes measured in terms of the problem-solving effectiveness of the theory?

NOTE: From Fawcett (2005 p. 447-448. Copyright 2005) by F. A. Davis. Adapted with permission.

quirement is more likely to be met when a theoretical defini-
tion is given for each concept than when no explicit defini-
tions are given. The semantic consistency requirement is met
when the same term and the same definition are used for each
concept in all of the author’s discussions about the theory. Se-
mantic inconsistency occurs when different terms are used
for a concept or different meanings are attached to the same
concept. In addition, the internal consistency criterion re-
quires that propositions reflect structural consistency, which
means that the linkages between concepts are specified and
that no contradictions in relational propositions are evident.
The three questions to be asked when evaluating the internal
consistency of a theory, which is applicable to both grand and
middle-range theories, are listed in Table 1.
The criterion of parsimony focuses on the content of the
theory. Parsimony requires a theory to be stated in the most
economical way possible without oversimplifying the phe-
nomena of interest. This means that the fewer the concepts
and propositions needed to fully explicate the phenomena of
interest, the better. The parsimony criterion is met when the
most parsimonious statements clarify rather than obscure the
phenomena of interest. The question to be asked when evalu-
ating the parsimony of a theory, which are applicable to both
grand and middle-range theories, is given in Table 1.
The criterion of testability also focuses on the content of
the theory. That criterion frequently is regarded as the major
characteristic of a scientifically useful theory. Marx (1976)
declared, “If there is no way of testing a theory it is scientifi-

Downloaded from nsq.sagepub.com at Bobst Library, New York University on May 24, 2015

cally worthless, no matter how plausible, imaginative, or in-
novative it may be” (p. 249). Testability typically is regarded
as an empirically based criterion. Yet the relatively abstract
and general nature of grand theories means that their concepts
lack operational definitions stating how the concepts are mea-
sured, and their propositions are not amenable to direct em-
pirical testing. Therefore, I identified different criteria for the
evaluation of the testability grand theories and middle-range
theories.
Description of personal experiences may be used to evalu-
ate the testability of grand theories (Silva & Sorrell, 1992).
That approach requires specification of an inductive, qualita-
tive research methodology that is in keeping with the philo-
sophical claims and content of the grand theory and that has
the capacity to generate middle-range theories. The product
of the descriptions of personal experiences approach is “gen-
eralities that constitute the substance of [middle-range] nurs-
ing theories” (Silva & Sorrell, 1992, p. 19). In essence, then,
evaluation of the testability of a grand theory involves deter-
mining the middle-range theory-generating capacity of a
grand theory. The criterion of testability is met when the
grand theory has led to the generation of one or more middle-
range theories. Three questions, which were adapted from re-
quirements proposed by Silva and Sorrell (1992), are asked
when evaluating the testability of a grand theory (Table 1).
The relatively concrete and specific nature of middle-
range theories means that their concepts can have operational
definitions and their propositions are amenable to direct em-
pirical testing. Consequently, an approach called traditional
empiricism is used to evaluate the testability of middlerange
theories. That approach requires the concepts of a middle-
range theory to be observable and the propositions to be mea-
surable. Concepts are empirically observable when opera-
tional definitions identify the empirical indicators that are
used to measure the concepts. Propositions are measurable
when empirical indicators can be substituted for concept
names in each proposition and when statistical procedures
can provide evidence regarding the assertions made. The cri-
terion of testability for middle-range theories, then, is met
when specific instruments or experimental protocols have
been developed to observe the theory concepts and statistical
techniques are available to measure the assertions made by
the propositions. Three questions, which were adapted from
requirements identified by Silva (1986) and Fawcett (1999),
are asked when evaluating the testability of a middle-range
theory (Table 1).
The criterion of empirical adequacy requires the asser-
tions made by the theory to be congruent with empirical evi-
dence. The extent to which a theory meets that criterion is de-
termined by means of a systematic review of the findings of
all studies that have been guided by the theory. The logic of
scientific inference dictates that if the empirical data conform
to the theoretical assertions, it may be appropriate to tenta-
tively accept the assertions as reasonable or adequate. Con-
versely, if the empirical data do not conform to the assertions,
Downloaded from nsq.sagepub.com at Bobst Library, New York University on May 24, 2015
Scholarly Dialogue 133
it is appropriate to conclude that the assertions are false. Eval-
uation of the empirical adequacy of a theory should take into
consideration the potential for circular reasoning. More spe-
cifically, if data always are interpreted in light of a particular
theory, it may be difficult to see results that are not in keeping
with that theory. Indeed, if researchers constantly uncover,
describe, and interpret data through the lens of a particular
theory, the outcome may be limited to expansion of that the-
ory and that theory alone (Ray, 1990). Therefore, unless alter-
native theories are considered when interpreting data or the
data are critically examined for both their fit and nonfit with
the theory, circular reasoning will occur and the theory will be
uncritically perpetuated. Circular reasoning can be avoided if
the data are carefully examined to determine the extent of
their congruence with the concepts and propositions of the
theory, as well as from the perspective of alternative theories
(Platt, 1964). In other words, evaluation of a theory always
should take alternative theories into account when
interpreting data collected within the context of the theory in
question.
It is unlikely that any one test of a theory will provide the
definitive evidence needed to establish its empirical ade-
quacy. Thus decisions about empirical adequacy should take
the findings of all related studies into account. Meta-analysis
and other formal procedures can be used to integrate the re-
sults of related studies. It is important to point out that a the-
ory should not be regarded as the truth or an ideology that
cannot be modified. Indeed, no theory should be considered
final or absolute, because it is always possible that subsequent
studies will yield different findings or that other theories will
provide a better fit with the data. Thus the aim of evaluation of
empirical adequacy is to determine the degree of confidence
warranted by the best empirical evidence, rather than to deter-
mine the absolute truth of the theory. The outcome of evalua-
tion of empirical adequacy is a judgment regarding the need
to modify, refine, or discard one or more concepts or
propositions of the theory.
The extent to which a grand theory meets the criterion of
empirical adequacy is determined by a continuation of the de-
scription of personal experiences approach discussed earlier
in the section on testability of grand theories. The data used to
determine the empirical adequacy of a grand theory may
come from multiple personal experiences of an individual or
similar personal experiences of several individuals. The ex-
tent to which a middle-range theory meets the criterion of em-
pirical adequacy is determined by a continuation of the tradi-
tional empirical approach discussed earlier in the section on
testability of middle-range theories. The questions to be
asked when evaluating the empirical adequacy of grand and
middle-range theories are listed in Table 1.
The criterion of pragmatic adequacy focuses on the utility
of the theory for nursing practice. The extent to which a grand
theory or a middle-range theory meets this criterion is deter-
mined by reviewing all descriptions of the use of the theory in
practice. The pragmatic adequacy criterion requires that
134 Nursing Science Quarterly, 18:2, April 2005
nurses have a full understanding of the content of the theory,
as well as the interpersonal and psychomotor skills necessary
to apply it (Magee, 1994). Although that may seem obvious,
it is important to acknowledge the need for education and spe-
cial skill training before theory application.
The pragmatic adequacy criterion also requires that the
theory actually is used in the real world of nursing practice
(Chinn & Kramer, 1995). In addition, the pragmatic adequacy
criterion requires that the application of the theory-based
nursing actions is generally feasible (Magee, 1994). Feasibil-
ity is determined by an evaluation of the availability of the hu-
man and material resources needed to establish the theory-
based nursing actions as customary practice, including the
time needed to learn and implement the protocols for nursing
actions; the number, type, and expertise of personnel required
for their implementation; and the cost of in-service education,
salaries, equipment, and protocol-testing procedures. More-
over, the willingness of those who control financial resources
to pay for the theory-based nursing actions, such as
healthcare system administrators and third-party payers,
must be determined. In sum, the nurse must be in a setting that
is conducive to application of the theory and have the time and
training necessary to apply it.
Furthermore, the pragmatic adequacy criterion requires
the nurse to have the legal ability to control the application
and to measure the effectiveness of the theory-based nursing
actions. Such control may be problematic in that nurses are
not always able to carry out legally sanctioned responsibili-
ties because of resistance from others. Sources of resistance
against implementation of theory-based nursing actions in-
clude attempts by physicians and healthcare system adminis-
trators to control nursing practice, financial barriers imposed
by healthcare institutions and third-party payers, and skepti-
cism by other health professionals about the ability of nurses
to carry out the proposed actions (Funk, Tornquist, & Cham-
pagne, 1995). The cooperation and collaboration of others
may, therefore, have to be secured.
Moreover, the pragmatic adequacy criterion requires that
theory-based nursing actions be compatible with expecta-
tions for practice (Magee, 1994). Compatibility should be
evaluated in relation to expectations held by the public and
the healthcare system. If the actions do not meet existing ex-
pectations, they should be abandoned or people should be
helped to develop new expectations. Johnson (1974) com-
mented, “Current [nursing] practice is not entirely what it
might become and [thus people] might come to expect a dif-
ferent form of practice, given the opportunity to experience
it” (p. 376).
The pragmatic adequacy criterion also requires the theory-
based nursing actions to be socially meaningful by leading to
favorable outcomes for those who participate in the actions.
Examples of favorable outcomes include a reduction in com-
plications, improvement in health conditions, and increased
satisfaction with the theory-based actions on the part of all
who participate.
Downloaded from nsq.sagepub.com at Bobst Library, New York University on May 24, 2015
The outcomes of theory-based nursing actions are further
judged by use of what Silva and Sorrell (1992) called the
problem-solving approach. That approach emphasizes the
problem-solving effectiveness of a theory and seeks to deter-
mine “whether what is purported or experienced accom-
plishes its purpose” (Silva & Sorrell, 1992, p. 19). The prob-
lem-solving approach is based on the position that theories
are developed “to solve human and technical problems and to
improve practice” (Kerlinger, 1979, p. 280). It requires delib-
erative application of a theory. Chinn and Kramer (1995) ex-
plained that the application “involves using research methods
to demonstrate how a theory affects nursing practice and
places the theory within the context of practice to ensure that
it serves the goals of the profession . . . [and] provides evi-
dence of the theory’s usefulness in ensuring quality of care”
(p. 164). The problem-solving approach can be used with all
types of theories but is most effective when applied to middle-
range predictive theories. In that case, the application seeks to
determine the effects of interventions specified in middle-
range predictive theories on the health conditions of the hu-
man beings who participate in the interventions (Hegyvary,
1992). The eight questions to be asked when evaluating prag-
matic adequacy are given in Table 1. Two last two questions,
which were adapted from requirements identified by Silva
and Sorrell (1992), are asked when evaluating the problem-
solving effectiveness of nursing theories.
References
Barnum, B. J. S. (1998). Nursing theory: Analysis, application, eval-
uation (5th ed.). Philadelphia: Lippincott.
Chinn, P. L., & Kramer, M. K. (1995). Theory and nursing. A system-
atic approach (4th ed.). St. Louis, MO: Mosby.
Duffy, M., & Muhlenkamp, A. F. (1974). A framework for theory
analysis. Nursing Outlook, 22, 570-574.
Fawcett, J. (1993). Analysis and evaluation of nursing theories. Phil-
adelphia: F. A. Davis.
Fawcett, J. (1999). The relationship of theory and research (3rd ed).
Philadelphia: F. A. Davis.
Fawcett, J. (2000). Analysis and evaluation of contemporary nursing
knowledge: Nursing models and theories. Philadelphia: F. A.
Davis.
Fawcett, J. (2005). Contemporary nursing knowledge: Analysis and
evaluation of nursing models and theories (2nd ed.). Philadel-
phia: F. A. Davis.
Funk, S. G., Tornquist, E. M., & Champagne, M. T. (1995). Barriers
and facilitators of research utilization. Nursing Clinics of North
America, 30, 395-407.
George, J. B. (Ed.). (2002). Nursing theories: The base for profes-
sional nursing practice (5th ed.). Upper Saddle River, NJ:
Prentice Hall.
Hegyvary, S. T. (1992). From truth to relativism: Paradigms for doc-
toral education. In Proceedings of the 1992 forum on doctoral ed-
ucation in nursing (pp. 1-15). Baltimore: University of Maryland
School of Nursing.
Johnson, D. E. (1974). Development of theory: A requisite for nurs-
ing as a primary health profession. Nursing Research, 23, 372-
377.
Kerlinger, F. N. (1979). Behavioral research: A conceptual ap-
proach. New York: Holt, Rinehart, & Winston.

King, I. M., & Fawcett, J. (Eds.). (1997). The language of nursing
theory and metatheory. Indianapolis, IN: Sigma Theta Tau Inter-
national Center Nursing Press.
Levine, M. E. (1988). Antecedents from adjunctive disciplines: Cre-
ation of nursing theory. Nursing Science Quarterly, 1, 16-21.
Magee, M. (1994). Eclecticism in nursing philosophy: Problem or
solution? In J. F. Kikuchi & H. Simmons (Eds.), Developing a
philosophy of nursing (pp. 61-66). Thousand Oaks, CA: Sage.
Marriner-Tomey, A., & Alligood, M. R. (2002). Nursing theorists
and their work (5th ed.). St. Louis, MO: Mosby.
Marx, M. H. (1976). Formal theory. In M. H. Marx & F. E. Goodson
(Eds.), Theories in contemporary psychology(2nd ed., pp. 234-
260). New York: Macmillan.
Parse’s Criteria for Evaluation of Theory With a
Comparison of Fawcett’s and Parse’s Approaches
Rosemarie Rizzo Parse, RN; PhD; FAAN
Professor and Niehoff Chair, Loyola University Chicago, Chicago, Illinois
I am pleased to share with Dr. Fawcett my criteria for evalua-
tion of theory and a comparison of our work. First, I should
clarify my position on middle-range theory, since it is in di-
rect contrast to Dr. Fawcett’s. The term middle-range theory
is ubiquitous in the nursing literature without any substantive
definition. Cody (1999) said that there is “a lack of clarity as
to what constitutes middle-range” (p. 10). He raised this ques-
tion: Are those “working in the middle range on myriad top-
ics . . . really developing nursing science or merely
elaborating the vast patchwork quilt of applied-science nurs-
ing” (Cody, 1999, p. 11)? Dr. Fawcett and I generally agree
that nursing is a basic science with its own unique body of
knowledge and that theory is defined as a set of concepts com-
bined uniquely and written at an abstract level to describe, ex-
plain, or predict phenomena (Parse, 1997). My view is that
propositional statements written at lower levels of abstrac-
tion, often called middle-range theories, are really hypothe-
ses that can be tested only through quantitative research
methods using appropriate instrumentation (Parse, 2000).
Thus, my comments here do not address middle-range
theories and are relevant only to theory as described above.
I set forth criteria for the evaluation of nursing theory in
my 1987 book, Nursing Science: Major Paradigms, Theo-
ries, and Critiques. At that time, I developed a design for criti-
cal appraisal appropriate for all frameworks and theories, no
matter how they are constructed. I have modified these ideas
considerably to be consistent with my current thinking, but I
still believe in the basic premise that criteria for evaluation of
theory should be broad enough to accommodate all perspec-
Nursing Science Quarterly, Vol. 18 No. 2, April 2005, 135-137
DOI: 10.1177/0894318405275860
© 2005 Sage Publications
Downloaded from nsq.sagepub.com at Bobst Library, New York University on May 24, 2015
Scholarly Dialogue 135
Meleis, A. I. (1997). Theoretical nursing: Development and prog-
ress (3rd ed.). Philadelphia: Lippincott.
Parse, R. R. (1987). Nursing science: Major paradigms, theories,
and critiques. Philadelphia: Saunders.
Platt, J. R. (1964). Strong inference. Science, 146, 347-353.
Ray, M. A. (1990). Critical reflective analysis of Parse’s and
Newman’s research methodologies. Nursing Science Quarterly,
3, 44-46.
Silva, M. C. (1986). Research testing nursing theory: State of the art.
Advances in Nursing Science, 9(1), 1-11.
Silva, M. C., & Sorrell, J. M. (1992). Testing of nursing theory: Cri-
tique and philosophical expansion. Advances in Nursing Science,
14(4), 12-23.
tives in the discipline. The two major areas of critical ap-
praisal in my design are structure and process (Parse, 1987).
Questions applicable to both areas are listed in Table 1.
Structure criteria refer to the physiognomy of the theory,
that is, the historical evolution, foundational elements, and re-
lational statements. Historical evolution refers to the details
of the development of the theory including the philosophical
and theoretical antecedents and the changes in the theory over
time. Foundational elements refer to the philosophical as-
sumptions underpinning the theory and the major concepts of
the theory. These are written at an abstract level and are the
theorist’s beliefs about the phenomenon of concern to the dis-
cipline, the human-universe-health process. Relational state-
ments refer to the principles that are created from the unique
weaving of the concepts into descriptions of the human-
universe-health process, and they also are written at an
abstract level.
The process criteria encompass correspondence, coher-
ence, and pragmatics (Parse, 1987). Correspondence refers to
semantic integrity and simplicity. Semantic integrity is recog-
nized by the consistency of meanings among the terms used
to explain the human-universe-health process in the philo-
sophical assumptions and in the definitions of the concepts
and principles. Two aspects of semantic integrity are sub-
stance and clarity (Parse, 1987). Substance refers to the dura-
bility of the meaning assigned to terms, the breadth of the de-
scriptions, and the consistency in levels of discourse within
and among the assumptions, concepts, and principles. Clarity
refers to the distinctness and mutually exclusive nature of the
definitions. Simplicity is recognized by the uncluttered
abstract descriptions and economy of words used to explain
the theory.