Title: Testing a decision making model for nursing
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00100745/00001
 Material Information
Title: Testing a decision making model for nursing
Physical Description: Book
Language: English
Creator: Coble, Daniel Bruce, 1949-
Publisher: State University System of Florida
Place of Publication: Florida
Florida
Publication Date: 2000
Copyright Date: 2000
 Subjects
Subject: Nursing thesis, Ph. D   ( lcsh )
Dissertations, Academic -- Nursing -- UF   ( lcsh )
Genre: government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )
 Notes
Abstract: ABSTRACT: The purpose of this study was to test a decision making model for nursing developed by the author in a previous qualitative study. Concepts in the decision making model were creativity, experience, leadership, education, risk taking, and informatics. The Chinese philosophy of Cosmogony and chaos theory provided the theoretical framework for the model. Analysis of Decision Making in Nursing (ADMN), a tool developed for this study, incorporated a bipolar scale of 81 items. Pilot testing showed adequate reliability and validity of the ADMN. The 5,000 subjects were randomly selected from a list of registered nurses licensed and residing in Florida. A total of 510 questionnaires was returned, with 491 providing usable information. Correlation analysis, multiple regression, and path analysis were performed on an over-identified model. Four path coefficients had non-significant t values and were removed.
Abstract: The goodness of fit indices for the reduced model provided a mixed review of the effectiveness of the model. The significant chi-square test result (chi square = 35.47, df = 7, p = 0.0001) suggested the model did not adequately represent the decision making framework. However, the Normed Fit Index (NFI) and Comparative Fit Index (CFI) values above 0.95 indicated a relatively high fit of the model to the data. In addition a Lagrange multiplier test did not suggest further improvements to the model that had not previously been considered or rejected. The reduced model explained 86% of the variance in decision making. Leadership furnished the largest direct effect (0.33), indirect effect (0.19), and total effect (0.52) on decision making. Experience (0.32), creativity, (0.24) and education (0.24) followed in total effect. Risk taking (0.14) and informatics (0.16) had the smallest direct effects on decision making.
Abstract: The study contributed to the body of knowledge in nursing about decision making in the four domains of nursing practice: clinical, administration, research, and education. Along with opportunities for additional research in model development, the results of this project may advance an informatics-based understanding of nursing.
Summary: KEYWORDS: decision, nursing, theory, creativity, informatics, leadership
Thesis: Thesis (Ph. D.)--University of Florida, 2000.
Bibliography: Includes bibliographical references (p. 100-131).
System Details: System requirements: World Wide Web browser and PDF reader.
System Details: Mode of access: World Wide Web.
Statement of Responsibility: by Daniel Bruce Coble.
General Note: Title from first page of PDF file.
General Note: Document formatted into pages; contains xi, 133 p.; also includes graphics.
General Note: Vita.
 Record Information
Bibliographic ID: UF00100745
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 81623605
alephbibnum - 002678719
notis - ANE5946

Downloads

This item has the following downloads:

Coble ( PDF )


Full Text











TESTING A DECISION MAKING MODEL FOR NURSING


By

DANIEL BRUCE COBLE













A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2000

































Copyright 2000

by

Daniel Bruce Coble



























In memory of
Amanda Christine Coble
October 17, 1980- July 15, 1999















ACKNOWLEDGMENTS

The author wishes to express his sincere gratitude and appreciation to Dr. Lois J.

Malasanos, chair of his supervisory committee, for her guidance, encouragement, and

insightful comments throughout his research and in the preparation of this dissertation.

Special thanks are extended to Dr. Hossein Yarandi for his expert assistance in the design

and analysis of this research study. The author would also like to thank the other

members of the supervisory committee, Drs. Claydell Horne and Patrick Thompson, for

their valuable advice and support throughout this study and in the preparation of this

manuscript.

Gratitude is extended to former supervisory committee chairs, Dr. Marjorie

White, who sparked the idea for this research and fostered its development, and Dr. Sally

Knox, who provided encouragement during difficult phases of the study. The

contributions of time, information, and scholarship in this project are gratefully

acknowledged from former committee members, Drs. Kathleen Smyth, Ira Horowitz, and

Selwyn Piramuthu.

Particular appreciation is given to Dr. Imogene M. King who offered nonstop

encouragement to complete this study and to Dr. Lou Carey of the University of South

Florida College of Education who provided assistance and expert advice during

instrument development.









A special note of thanks is merited by the pastor, music ministry, and

congregation of St. James United Methodist Church for the timely release from

responsibilities to finish the project.

The cheerful encouragement and generous understanding from colleagues and

students at The University of Tampa Department of Nursing provided significant

motivation to complete this project. The author would like to thank Dr. Cathy Kessenich

for her time and for her expertise in reviewing this document.

Special recognition and grateful acknowledgment are given to the author's wife,

Patricia, and to his sons, Matt and Tim, for their unceasing love and support throughout

the many challenges encountered while completing this dissertation.
















TABLE OF CONTENTS

page

ACKNOW LEDGM ENTS .................................. ......... . iv

LIST OF TABLES ................ .............................. viii

LIST OF FIGURES ................ ............................... ix

ABSTRACT ................ ................. ............. .... .. x

CHAPTERS

1 INTRODUCTION .......... ............................... 1

Rationale, Significance, or Need for the Study ........................ . 1
Theoretical Framework for the Study ................................ 2
Statement of the Problem ............... ........................ 10
Research Questions .................. ................. . ....... 10
Delimitations of the Study ......... ............... ........ . .11
Definition of Terms .... ........... ..... .......... .......... 11
Summary of Chapter 1 ......... ............................... 12

2 REVIEW OF THE LITERATURE .............................. .. 13

Theory and Model Concepts .............. ................... 13
Chaos Theory ............. .................. ........ 13
Decision Making ......... ...................... ........ 15
Decision M making in Nursing ........................... .. 20
Decision Making Tools ......... ...................... 29
Creativity ......... ..... ................ ... ..........32
Education .......... .......................................36
Leadership .......... ......................................41
Experience .......... ......................................48
R isk T making ................ .................... ......... 52
Informatics .......... ............................... .......56
Summary of Chapter II .............. .......................... 66
Critique of the Review of Literature ........................... .. 68











3 M ETHOD S ............................................... ......... 69

Research Methodology ............... .......................... 69
Specific Procedures ............... ............................ 69
Research Population or Sample ................. ................. 69
Instrumentation ................. ............................ 70
Pilot Study ................ ................................ 71
Data Collection ................ ........................... 71
Treatment of the Data ............................................ 73
Summary of Chapter 3 ............... .......................... 73

4 R E SU L T S ....................................................... 74

Sample Characteristics ............... .......................... 74
Instrument Testing ................. ......................... 79
Path Analysis ................ .............................. 80

5 DISCUSSION ................. ................ ............... 86

Relating the Findings to the Original Conceptualization of the Problem ...... 86
Relationships among Model Concepts ................. ............. 88
Practical Implications of Research Results .......................... . 90
Limitations of the Study ........................................... 92
Implications for Further Research .................................. 93
Conclusions ........ ........................................... 93

APPEND IX ................................................. ....... . 95

ANALYSIS OF DECISION MAKING IN NURSING .................... 96

R E FE R EN C E S ....................................................... 100

BIOGRAPHICAL SKETCH ............... .......................... 132















LIST OF TABLES


Table p

1. Number and Frequency of the Variables Gender, Ethnicity, Marital Status, Area of
Practice, Specialty, Basic Nursing Education, Highest Level of Education,
Level of Practice, Employment Status and Current Position for the Total
Sample ..................... ...........................75

2. Summary Measures of the Variables Age (in years), Length in Current Position (in
years), and Total Years of Nursing Experience for the Total Sample ..... 79

3. Correlations among M odel Variables ................................. 81

4. Summary Measures for Model Variables ............................ 82

5. Covariance Structure Analysis: Maximum Likelihood Estimation
G oodness of Fit .................................... ......... 85

6. Direct, Indirect, and Total Effect on Decision Making ................. .. 85















LIST OF FIGURES


Figure a

1. Chinese Cosmogony ............... ............................. 6

2. Decision Model for Nursing, with the Imbedded Nursing Process ............ 7

3. Hypothesized (Over-Identified) Decision Making Model for Nursing ........ 72

4. Hypothesized Decision Making Model for Nursing ................... .. 83

5. Reduced Decision Making Model for Nursing .......................... 84















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

TESTING A DECISION MAKING MODEL FOR NURSING

By

Daniel Bruce Coble

December 2000


Chairperson: Lois J. Malasanos, RN PhD
Major Department: Nursing

The purpose of this study was to test a decision making model for nursing

developed by the author in a previous qualitative study. Concepts in the decision making

model were creativity, experience, leadership, education, risk taking, and informatics.

The Chinese philosophy of Cosmogony and chaos theory provided the theoretical

framework for the model.

Analysis of Decision Making in Nursing (ADMN), a tool developed for this study,

incorporated a bipolar scale of 81 items. Pilot testing showed adequate reliability and

validity of the ADMN.

The 5,000 subjects were randomly selected from a list of registered nurses

licensed and residing in Florida. A total of 510 questionnaires was returned, with 491

providing usable information.









Correlation analysis, multiple regression, and path analysis were performed on an

over-identified model. Four path coefficients had non-significant t values and were

removed. The goodness of fit indices for the reduced model provided a mixed review of

the effectiveness of the model. The significant chi-square test result (X2 = 35.47, df = 7, p

= 0.0001) suggested the model did not adequately represent the decision making

framework. However, the Normed Fit Index (NFI) and Comparative Fit Index (CFI)

values above 0.95 indicated a relatively high fit of the model to the data. In addition a

Lagrange multiplier test did not suggest further improvements to the model that had not

previously been considered or rejected.

The reduced model explained 86% of the variance in decision making.

Leadership furnished the largest direct effect (0.33), indirect effect (0.19), and total effect

(0.52) on decision making. Experience (0.32), creativity, (0.24) and education (0.24)

followed in total effect. Risk taking (0.14) and informatics (0.16) had the smallest direct

effects on decision making.

The study contributed to the body of knowledge in nursing about decision making

in the four domains of nursing practice: clinical, administration, research, and education.

Along with opportunities for additional research in model development, the results of this

project may advance an informatics-based understanding of nursing.















CHAPTER 1
INTRODUCTION

Rationale, Significance, or Need for the Study

Nurses face many challenges in the early years of the 21st Century. Health care

reform has produced dramatic changes in health care financing and delivery mechanisms

(Spitzer, 1998) in an environment that is chaotic, with many threats to quality (Forrest,

1999). As nurses are poised to provide a unique role in the reformed system, a challenge

is how to make effective decisions in the unglued environment (Flarey, 1993). In this

new millennium, nurses serve as information brokers, using the Internet for obtaining and

communicating information, and as information generators, accessing knowledge for

decision support (Clark, 2000). Decisions made in the enactment of the professional

nursing role combine complex human interactions, leadership skills, allocation of

resources, and prioritization of needs within an overriding concern for quality outcomes

(Boblin-Cummings, Baumann, & Deber, 1999). Decisions made during this time of

chaos and turbulence will transform nursing, its leadership, and its practices to meet

future challenges.

Decision making is the choosing of options directed toward the resolution of

problems and the achievement of goals (Kerrigan, 1991). Decisions that nurses make

influence patient care in all four domains of nursing: (a) health through mobilization of

human and environmental resources to preserve health and to promote healing;









2

(b) human beings through a humanistic ethic to preserve dignity, uniqueness and freedom

of choice; (c) environment through creation of a culture that promotes health, self respect

and professional identity; and (d) nursing through professional practice models, research,

networks, and partnerships (R. Anderson, 1993; Flarey, 1993; Kerrigan, 1991; M. Smith,

1993).

In a previous qualitative research study (Coble, 1995), a conceptual model for

decision making in nursing was constructed. A sample of 30 registered nurses wrote

stories on making a decision which affected patient care or patient outcomes. The stories

were analyzed for themes and concepts; in depth interviews were conducted with 10 of

these nurses to clarify the concepts and themes used in decision making. Six concepts

related to decision making emerged from these analyses: creativity, experience,

leadership, education, risk taking and communication. After a secondary review of

literature, communication was changed to a more global term to reflect both

communication and information management: informatics. The model required

additional testing to determine relationships among the concepts. The purpose of this

study was to test this decision making model using correlation and path analyses to

determine the relationships among the concepts and with decision making in nursing.

Theoretical Framework for the Study

The theoretical framework for this study was chaos theory. Chaos theory was first

developed in mathematics but has become increasingly popular in science to understand

complex nonlinear systems (Abraham & Gilgen, 1995; Bassingthwaighte, Beard,

Percival, & Raymond, 1995; Center for Nonlinear Science, 1999; Grebogi & Yorke,









3

1997; Murray, 1998; Phillips, 1991; Tvede, 1997; Zbilut & Staffileno, 1994). Poincare in

1890 disavowed the Newtonian explanations of the earth and the moon around the sun

(N. K. Hayles, 1990). Hurst and Mandelbrot (Tvede, 1997) separately discovered self-

similarity: nonlinear systems often repeat the same behavior in different scales.

Mandelbrot further distinguished two dynamic properties of complex systems: the Noah

effect, in which small movements are interrupted by violent, disruptive jumps; and the

Joseph effect, in which movement is unidirectional and trendy.

The world view of chaos is that variation, change, and unpredictability are the

centers of the knowledge process. Gleick (1987) proposed that chaos is a paradox: an

apparent irregular, unpredictable behavior which is deterministic, nonlinear, and dynamic.

For nursing, chaos theory allows the understanding of complex, variable events such as

decision making. While the action of some variables can be predicted under controlled

circumstances, the real world is complex, disorganized, interconnected, individualistic,

and unique.

A nonlinear or chaotic system is any system that evolves over time (N. Hayles,

1999; Herbert, 1995). Nonlinear systems are rich and complex spectrums of recurrent

behaviors (from equilibria, periodic to periodic) and modes of changes in behavior

(bifurcations) which may be continuous, discontinuous or explosive. Chaos is

characterized by short-term predictability and long-term unpredictability because of a

sensitivity for initial conditions. Sensitive dependence on initial conditions is a

phenomenon common to chaos theory in which a small change in the initial conditions

can drastically change the long-term behavior in a system.











Chaotic systems are in continual fluctuation; patterns never repeat themselves yet

stay within bounded parameters. Nurses expect change and variation in the environment

and consider these positive processes or desirable outcomes of nursing intervention.

While deviations may be judged exceptions or even errors, chaos theory allows the nurse

an infinite variety of decision outcomes within a framework of holistic, humanistic care.

An attractor is the behavior of a system during equilibrium or disequilibrium

(Herbert, 1995); this set of behaviors defines a system's steady state motion (Mark,

1994). An attractor can be single, closed curve or strange. A closed curve attractor

fluctuates in a repetitive manner, while a strange attractor is characterized by non-

repetitive fluctuations.

For this study, chaos theory described the behavior of the six concepts within the

framework of the decision model. For a nurse making a decision, the process followed, or

the steps initiated, or the sequencing of events is inconsequential. The nurses in the

earlier study (Coble, 1995) reported that the best decisions were made when a number of

the concepts converged from the chaotic environment. The nurse's decisions and

behavior were a consequence of the alignment of the six concepts in some array. In the

model, the six concepts are floating within the decision framework, with a process of

decision making in a circular, open systems configuration. A more apt version may be

that of a sphere containing the six concepts in a primordial soup. Some concepts are

linked at times, while others are independent at times. Nurses in the study reported both

conscious and unconscious awareness of the mix and alignment of the concepts at the

time of decision making. In this study, the conceptual framework of chaos theory











allowed the model to be explored for patterns and relationships, while maintaining

integrity with the nurses' description of behaviors leading to decision making.

The decision making model for nursing being tested in this study was composed

of open, interactive, dynamic systems, based upon a world view of nursing, health, human

beings, and environment. The Chinese philosophy of Cosmogony (Gleick, 1987;

Mahdihassan, 1990a, 1990b; Rossbach, 1983; Spear, 1995; Wensel, 1980) was utilized to

depict the decision making process in an open system and in an interactive environment.

The Chinese believed that five essences of life each had its own basic energy, color,

emotion, posture, and voice. Fire, Water, Earth, Wood, and Air (also known as Metal)

are depicted by the author in Figure 1, along with their colors and voices. In addition,

relationships between model components and across the model were utilized by Chinese

acupuncturists as the philosophical base for their profession. Today, the Chinese art of

placement (Feng Shui) is founded on this philosophy. Opposites in the model may be

antagonistic or may be used for control of phases out of bounds.

In the model, the decision making process is the nursing or scientific process.

These five steps are (a) problem identification; (b) assessment; (c) planning; (d)

implementation; and (e) evaluation and are depicted as constructed by the author in

Figure 2.



















~i It


b.c


r
/ -- %
[J,


Figure 1 Chinese Cosmogony


i. ~
~
i
if
~....~~...I i
~
';""


t I '




























rG.


So


5:5 .5


\ ~A0. i


F


5?**


/T-:.i~m x~r.'






fl
41
t"-

\_ .-';

. .. .I
S i i.
^- .-... / ..9.


t









X 4
4^ 1 /


Figure 2 Decision Model for Nursing, with the Imbedded

Nursing Process









8

Problem identification. In the model this step defines the purpose, motivation and

boundaries of the problem or opportunity. Resources (people, methods, technology,

materials) are identified for future consultation or utilization. When problem

identification is out of bounds or control, the nurse becomes swamped with data, which in

turn diminishes the success of the decision or hinders the ability to complete the process.

Assessment. This step alters information from the problem identification step to

synthesize a structured approach to the problem. Specific data may be clarified or

additional information requested. Potential conflicts in methods and outcomes are

addressed. Controls are devised to guarantee that the focus of the decision making

process remains correctly placed on the root problem or opportunity. When not managed

in a planned methodology, conflict becomes subterranean and may surface as an

unwelcomed outcome or as a challenge to the decision itself.

Planning. This step provides an opportunity to stimulate creativity through the

generation of ideas. The role of each individual is valued and appreciated as an important

contribution of the team. Building teamwork fosters an atmosphere of group effort and

consensus. After ideas are generated, priorities are established based on factors such as

goals and mission of the organization or individual; probability of success; resource

intensity required; and probability of addressing the root cause of the problem. Too many

ideas without reaching a consensus may be exhaustive and may cause communication

problems in that the mission and goals of the organization or individual are not brought to

bear on possible solutions, thus clouding the relationship between the activity and

attention to the problem with the perceived outcome and benefit.











Intervention. The purposes of this step are to produce stability and to increase

productivity of the organization if the problem has been corrected. Support for the

solution may require contributions by a few or by many individuals, groups, or

departments. The service delivery system may be affected by the proposed changes and

alternative solutions being tested. When a number of solutions are implemented in a

short time frame, the work may be exhausting, depleting the energy of individuals and the

organization, resulting in sickness rather than health.

Evaluation. In this step a new paradigm or standards of practice may be

established. Individual and organizational goals are evaluated based upon the

achievement of problem resolution. Values of the organization or individual may change

as the new methods become the accepted standards. Quality control measures are

instituted to assure continued compliance with the new standards. Grieving may occur as

the old ways are discarded. Individuals and organizations may be reluctant to terminate

successful past practices, resulting in a rejection of the decision rather than acceptance.

Assumptions for the model include the following:

1. The design of the model allows for interactions and opposing forces among the

five steps.

2. Behavior is a reflection of decisions that people make.

3. The decision making process normally starts with problem identification and

proceeds clockwise through the model.

4. Nurses possess the power and authority to make decisions about patient care and

within organizations.











5. Nurses adapt behavior to the demands of the situation.

Propositions for the model include the following:

1. Opposing forces can be used to balance. For example, planning may be related

back to problem identification to assure a focus on the correct variables is maintained.

2. A return to a previous step is used to recapture the path. If evaluation steps

languish for lack of commitment, a return to planning may provide a more acceptable

solution.

3. A pattern of Perpetual Start Up is identified when problem identification

alternates with assessment but no progress to other steps is detected.

4. A pattern of Co-dependency is established when the triangle of assessment-

planning-intervention is detected. Without initial problem identification and proceeding

through the model to evaluation, the pattern becomes a self-fulfilling prophecy.


Statement of the Problem

The purpose of this study was to determine the relationships among the concepts

of a decision making model for nursing: creativity, experience, leadership, education,

risk taking, and informatics. The population was registered nurses licensed in the State of

Florida.

Research Questions

Is the hypothesized model adequate in explaining decision making in nursing?

What are the relationships among the concepts in the decision making model?











Delimitations of the Study

Generalization of this study is limited by the non-experimental design, the size of

the sample, and the return rate of the questionnaire. The registered nurses may not have

valued research into a theoretical model; therefore, their responses may not have reflected

true decision making in their practice. The use of a bipolar scale to measure the six

concepts in the model may not have been the best approach to measure the effects in the

model or the relationships among the model concepts.

Definition of Terms

Decision making is the choosing of options directed toward the resolution of

problems and the achievement of goals (Kerrigan, 1991).

Creativity is the application of novel solutions which change the context in which

the problem is embedded (Kirton, 1989).

Experience is the refinement of preconceived notions and acquired theory through

encountering actual situations that add nuances or shades of difference to theory (Benner,

1991; Benner, Tanner, & Chesla, 1992, 1996).

Leadership is the process of influencing people to accomplish goals (McCloskey

& Molen, 1987).

Education is investigational inquiry or learning by critiquing, proposing, doing, or

doing in collaboration (Wheeler, Fasano, & Burr, 1995).

Risk taking is a behavior characterized by increased tolerance of possible error or

mistake (R. Jones & Beck, 1996).











Informatics (American Nurses Association, 1994) includes the management and

processing of data, information and knowledge (Graves & Corcoran, 1989); information

seeking behaviors (Henry, 1995c); communication methods (verbal, nonverbal,

messaging); analytical methods (identifying, standardizing, modeling)(Henry & Mead,

1997); engineering methods (design, evaluation, adapting, customizing, applications,

tools); and managerial methods (strategic planning, change, collaboration, accessibility,

project management) (Goossen, 1996; Goossen, Epping, & Abraham, 1996).

Summary of Chapter 1

Nurses face a challenging time of transition in the 21st Century. Decisions made

in practice, education, and management will profoundly influence the profession for

years. Decision making is a concept that has not been examined by nurse theorists or

researchers in detail. The purpose of this study was to test the relationships among

decision making concepts: creativity, experience, leadership, education, risk taking, and

informatics. Using the framework of chaos theory, decision making was viewed as a

random, unique pattern which is unlimited, yet deterministic. Operational definitions

were presented for model concepts. The generalizability of this study may be limited by

small sample size, rate of return of questionnaires, and ability of the bipolar scale to

measure the model concepts.















CHAPTER 2
REVIEW OF THE LITERATURE

Theory and Model Concepts

Chaos Theory

The utilization of Chaos Theory in nursing has been both philosophical and

empirical. Ray (1994) defined chaos as belonging to the universe and as

knowledge-in-process. Since all phenomena in the universe are interconnected and

dynamic, all knowledge is limited and approximate. Creativity takes place at the

boundaries between chaos and order. Tension exists between chaos and order; this

tension stimulates change and a creative reordering into self-organizing systems. The

edge of chaos is a communication and information process which feeds back on itself and

promotes systems behavior though continual, mutual interaction.

The laws of information patterns were developed using chaos theory (Sabelli,

Carlson-Sabelli, & Messer, 1994). Symmetric flux is the state of information which is

always mixed with uncertainty, ignorance, and error. Processes contain random,

symmetric flux, which is unobservable except as an absence. Observable patterns of

matter, information and energy are asymmetric within the symmetric flux. Action

asymmetry exists as a flow of energy in time and is unidirectional. Union of opposites is

the duality in nature; everything contains a difference and that difference implies

information. Opposites alternate in predominance so they never coexist in the same











place, same time, and same respect. One process may have higher priority and

supremacy, yet can coexist with the opposite state. Organizational diversification is the

result of interaction of two opposites which produces processes of three or more

dimensions. Although many aspects of the world appear random, the randomness is

really a surface manifestation of deep, complex structures (Bolland & Atherton, 1999). A

key feature of chaos is flexibility (Pediano, 1996). For nursing, chaos theory may bridge

the gap between nursing theories (the scientific representation of reality) and research

utilization in individuals and communities (reality with variety, diversity, irregularity,

uncertainty-in other words, chaos) (Rambihar et al., 1999; Rambur, 1999).

Chaos theory has been utilized in nursing research to explain complex behaviors

of individuals, groups, and organizations. Researchers in Texas (Hamilton, West, Cherri,

Mackey, & Fisher, 1994) identified patterns of births in adolescent girls. Time series

analysis revealed a data pattern that was noisy (chaotic) and superficially random. The

complexity was marked by ragged fluctuations in the process over time and by the

constantly changing background features of the population.

Mark (1994) examined chaotic events related to nursing vacancy rates. The

impact of a small change in nursing vacancy rates at an institution created a far-from

equilibrium state in the system. The organization may institute a change in salaries which

would describe a bifurcation point; this change in salaries may return the organization to

the previous steady state. With nursing vacancies, a cyclical pattern of highs and lows

may be observable (a closed curve attractor). Symmetry-breaking occurs with the failure

of tactics to regain equilibrium. Redesign, removal of service barriers, and new











technology may be creative energies which force the system to evolve into a new

organization or a new steady state (a strange attractor).

Puskar and Lucke (1999) employed chaos theory to explain nursing care of stroke

patients. Early stages of care focused on stable patterns of information processes and

cognitive functioning. Progression to more complex activities and abilities followed

gradually. Following stroke, neural impulses transmit at random intervals, overloading

the patient's ability to focus on a single task. Small changes in environment or routines

may have a negative effect by contributing confusion, agitation, or anxiety.

Other studies ascertained nonlinear relationships between spiritual disequilibrium

and traumatic life events (Dombeck, 1996); between exercise capacity and heart rate

variability (Cowan, Hathaway, & Kreider, 1999); with suicide in the elderly (Winreich,

1999); in public health and epidemiological patterns (Coppa, 1993); and among

temperature, pulse and respiration (Vicenzi & Hamilton, 1999).

Decision Making

Decision making has been found to be the discovery and selection of alternatives

(Loke, 1996); the intuitive judgments made by humans; and the relationship of priorities

to goals in the decision process (Slade, 1994). Decisions may be well structured, routine,

and require few resources (tactical decisions) or may be ill structured, unique, and require

substantial resources (strategic decisions) (Kline, 1994). People are generally unaware of,

or concerned with, the nature of decision making, or why one alternative is preferred to

another, or the quality of the decision making process (Hogarth, 1987; Hogarth &

Kunreuther, 1992). Judgment about choices is a persuasive activity, in which the











decision maker must be persuaded to choose a course of action from alternatives. Most

decisions are based on the anticipations people make about the immediate and/or distant

future.

Decision approaches can be classified as descriptive/behavioral or

prescriptive/normative (Fox, 1994, 1997; Fox & Tversky, 1998; Koutsoukis, Mitra, &

Lucas, 1999; Loke, 1996; Luce & von Winterfeldt, 1994; Raiffa, 1994; Vazsonyi, 1990).

The descriptive mode delineates the actual decision process, the players and the rules of

the game. Descriptive techniques include decision trees (stages of response across time)

and regression analysis (identifying individual factors and their weights). The normative

mode compares actual behavior with that predicted by laws of probabilities. The

utilitarian normative mode takes the action with the largest sum of predicted subjective

value and objective probability of the payoff. The dynamic normative mode accounts for

risky behavior by choosing actions with the largest product of subjective value and

subjective probability of the payoff. The prescriptive mode requires thinking reflectively

about serious choices. Values and beliefs about uncertainties are elicited as well as

confronting incoherence. The foundation of the prescriptive concept is the subjective

expected utility theory, a doctrine of choice based on psychology (subjective), Bayesian

probability theory (expected), and economics (utility). Through this art and science

interaction, a comprehensive overview of the situation guides actions.

The decision process in the real world is complex, ill-defined, and disorderly;

human behavior is at best quasi-rational, with deep iterative thinking a myth, or at least a

rare occurrence (Quinn, Anderson, & Finkelstein, 1996). Values and interests of the











decision players (if players are even known) are not usually common knowledge. With

limited information, individuals often choose the first alternative that gives a satisfactory

solution.

The individual's perception of information may be selective rather than

comprehensive (Hogarth, 1987). Memory works by association with past events; thus,

through an active process of reconstruction, memory can change. Processing is mainly

done sequentially and across time. However, an unstable environment leads to deficient

judgmental strategies. Processing capacity and mental effort are reduced by using

heuristics (subjective operational knowledge of the world). Several heuristics have been

found to produce systematic errors in the decision process (Levin & Reicks, 1996; Loke,

1996): representativeness; availability; or adjustment and anchoring. Errors in

representativeness arise when similarities between events and properties rather than the

frequencies or base rates influence the subjective evaluation of important descriptive

features. Other errors in representativeness occur when the decision process is insensitive

to sample size; when the effect of chance is disregarded or underestimated; when the

reliability of the description is not questioned; or when there is an illusion of validity

based on the goodness of fit between the predicted outcome and the input information.

Availability heuristic is the situation in which individuals judge the frequency of an event

based on prior experiences or other related, known or imaginary experiences. Adjustment

and anchoring bias occur when an individual fails to consider different individuals and

different initial points; adjustment along a continuum of choices would improve

sensitivity for underestimations and overestimations of the correct answer.











Utilizing a pyramid to represent various approaches to decision making,

Schoemaker (1994) found that people tended to use the quickest and the easiest process.

The base of the pyramid was formed by the concept of intuitive judgment. Intuition is a

quick and easy approach, sometime brilliant. However, intuition is open to inconsistency

and distortion, is difficult to articulate, and is difficult to apply consciously or

unconsciously. The second level of heuristic procedures consists of rules, tailored and

generic shortcuts. These rules sort out information and may be generic or specific to a

situation. While quick and clever, these rules can be articulated and applied consciously.

Importance weighting allows that some factors have more importance or weight than

other factors in a decision. These weights can be articulated, tested and stored for future

use. Bootstrapping is a variant of importance weighting in which the model is derived

from an expert but then the model out performs the expert. Value analysis is the peak of

the pyramid. Value analysis is a comprehensive assessment refining weighting by

considering how factors affect broader objectives and values added to the decision. Key

objectives and nonlinear values (an increase in some given factor does not always

increase the value of the factor) are uncovered. Value analysis requires the use of experts

and consideration of multiattribute utility. Value analysis, as the epitome of decision

making, commands resources and time, resulting in increased costs.

Styles of decision making have been explored using the Myers Briggs Type

Indicator (MBTI) (Nutt, 1993). An individual's preferences for types of data and ways to

process that data have been incorporated into the MBTI. A sensing (S) individual prefers

hard data, dealing with what is and with specifics. The intuitive (N) person considers











possibilities, qualitative values, and what might be. Thinking (T) stresses logic and

formal modes of reasoning, while feeling (F) considers the decision in personal terms, or

how people are affected. Combining the preference for one of the data acquisition types

with one of the data processing approaches creates four styles: ST (sensation-thinking),

NT (intuition-thinking), SF (sensation-feeling), and NF (intuition-feeling). The preferred

modes of understanding (S, N, T, and F) suggest how people like to make judgments and

choices. Therefore, the ST style applies to an analytic approach to decisions, while the

NT implies a speculative approach. The SF style creates a consultative approach, such as

a decision group; NF style suggests a charismatic approach, catering to the whims of

powerful individuals. The ideal decision-maker would use all four modes of

understanding (S, N, T, and F), resulting in an auxiliary style of SNTF or synthesizer.

Synthesizers have high tolerance for ambiguity and uncertainty, were the most prone to

adopt a solution, and saw the least risk in the decision. Linkers have access to three

modes, producing SNT, SNF, STF, and NTF styles. Tolerance for ambiguity and

uncertainty was less pronounced in linkers, who also identified more risk in decision.

Pure types evoke one strong preference for data and one for processing (ST, NT, SF, and

NF) and are called analyzers. ST and SF types had a low tolerance of ambiguity, while

NT and NF tolerated a moderate amount of ambiguity. Observers have dual function

(NS) with strong preference for perceptive action; data processors have dual function

(TF) with strong preference for judgmental action. Observers tended to be action-oriented

with a pessimistic view of the decision's risk. The observer had much more tolerance to

ambiguity and uncertainty than data processors. Overall, conservative outlooks toward









20

adoption of a decision were profiled by ST, TF, and STF styles. Action-oriented postures

were found in SN, SNT, and SNF styles.

Decision Making in Nursing

The review of literature for decision making in nursing revealed several areas of

effort: decision process identification and refinement; theory utilization; decision maker

characteristics; and decision application in practice. These areas will be discussed

separately.

Decision process. The routine process for decision making in nursing has been

identified by several authors (Corcoran, 1986; Grobe, Drew, & Fonteyn, 1991; Hannah,

Reimer, Mills, & Letourneau, 1987; Harbison, 1991; R. Jones & Beck, 1996; Joseph,

Matrone, & Osborne, 1988; Kerrigan, 1991; Laborde, Dando, & Hemmasi, 1989;

Marriner-Tomey, 1992, 1993; Narayan & Corcoran-Perry, 1997; Nixon, 1995; Rowland

& Rowland, 1997; Ruland, 1996; M. Smith, 1993; Steele & Maraviglia, 1981; Tappen,

1989; Tschikota, 1993; Turley, 1996a; Wangsness, 1991; Watkins, 1998) and follows the

well-established nursing process: assessment (problem identification); planning

(identifying alternatives); intervention (implementing decision); and evaluation

(evaluating response or outcome). Variations to the decision making process have been

devised to enhance creativity (Steele & Maraviglia, 1981) by inserting an incubation and

illumination phase and to deal with emergency procedures (Marriner-Tomey, 1992) by

increasing the pace, limiting alternatives, and immediate evaluation.

Theory utilization. Ruland (1996) delineated three main decision-making theories

in the nursing literature: (a) information processing, (b) decision theory or decision









21

analysis; and (c) intuitive clinical judgment. First, information processing uses concepts

of memory, processing, and problem solving and is the basis for most of nursing decision

making studies. Decision making within the information processing framework involves

interaction between the problem solver and the task in a complex environment. Humans

have limited short-term memory but have access to infinite long-term memory, a key

resource of knowledge gained from study and experience. The aim of information

processing is to achieve diagnosis, assessment or classification. Hypotheses in diagnostic

reasoning are made early in the search for further information. Information is confirmed;

potential solutions are eliminated; possibilities are finely discriminated from one another;

and specific hypotheses are explored for related or unrelated disorders or complications.

Task complexity increases the demands on the information processing skills of the

decision maker (M. Lewis, 1997). Individuals tend to simplify the decision-making

requirements and this tendency increases as the complexity increases. Components of the

decision-making task are defined in terms of content (signs, symptoms and behaviors)

and context (physical and social environment). Diagnostic decision making incorporates

critical thinking and data collection by clinicians to identify and classify phenomena in

clinical situations (G. Davis, 1994; Hamers, Abu-Saad, & Halfens, 1994; Hannah et al.,

1987; Koch & McGovern, 1993). This form of decision making incorporates the nursing

process model with the medical diagnostic model.

Narayan and Corcoran-Perry (1997) developed a line of reasoning model for

clinical decision making based upon information processing theory. A clinical situation

was analyzed for its task demands. Three types of requisite knowledge were identified:











general factual knowledge about human conditions or situations; specific factual

knowledge about differences in those with and without the characteristic under

consideration; and procedural knowledge surrounding the specific requirements of the

situation. Using a think-aloud method, the investigators found triggering cues

(symptoms), domain concepts (specific disease profiles) which activated hypotheses,

intermediate conclusions and actions, and a final line of reasoning conclusion. The

transition from cues to hypotheses reflects adaptations to short-term memory limitations

and the process of long-term memory retrieval.

The second theoretical perspective of decision theory focuses on the clinician

making decisions of a probabilistic nature, which includes the majority of clinical

decisions (Panniers & Walker, 1994; Anderson, Rungtusanatham, Schroeder, & Devaraj,

1995; Boblin-Cummings, 1996; Ash & Smith-Daniels, 1999; Ruland, 1999). The

purpose of decision theory is to arrive at a correct decision in less than optimal

environments. Decision analysis considers the preferences and values of the decision

maker, conflicting objectives, and the ambiguous nature of information. The goals are to

overcome uncertainty and to reduce risk by producing recommendations for decision

making rather than modeling human behavior.

The development of an algorithm, protocol, or decision tree for clinical decisions

is an example of decision theory. Decision analysis provides a structure for depicting

relationships between actions and outcomes, as well as a procedure for combining

information with values or preferences for outcomes (Corcoran, 1986). Corcoran

identified four steps in the decision analysis process: (a) structuring a decision flow











diagram; (b) assigning values (worth or utility) to possible outcomes; (c) assigning

probabilities to chance events; and (d) working backwards through the decision tree to

reveal the best alternative action. This last step calculates the expected value of each

alternative; the alternative with the highest expected value becomes the prescribed choice

of the analysis.

In a study of oncology nurses, Akers (1991) found algorithms to be flexible, safe

and effective, with feedback to reinforce behavior and references to standardize care. The

process included expert development, critiques by users, and delineation of progressive,

escalating decision making throughout the branches of the decision tree. Use of the

algorithm led to a more thorough patient assessment and a more informed nursing

response, ultimately resulting in better patient management.

Other researchers (Brennan, 1995; J. Clarke, 1999; Edwards & Barron, 1994;

Garre, 1992; Kokol, Zorman, & Medoo, 1999) found a system of attributes and their

corresponding weights or utility (worth, payoff, psychological value, or level of

satisfaction) had usefulness. Users of the complex process identified a list of attributes

potentially relevant to the decision and, by consensus, generated a relative weight for each

attribute. Scoring rules for physical value-related measures, if available, were determined

in advance. Rules of thumb were instituted when physical value-related measures were

not available. In this additive model, the physical value-related measures were adjusted

by the weights and then totaled. Analysis of the contribution of each individual value to

the decision as well as the cumulative effect was assessed.









24

The third theoretical perspective on decision making in nursing is that of clinical

judgment using an intuitive process (Ruland, 1996). Intuitive clinical judgment uses

pattern recognition, commonsense, skilled knowhow, and deliberate rationality to present

and to implement solutions (Benner, 1984, 1991; Benner et al., 1992, 1996; English,

1993; Nixon, 1995). Key elements of reflection, dialogue, and dialectical thinking enable

the reconstruction of relationships and roles into partnerships which optimize health care

delivery (Duchscher, 1999).

Cruz (1994) found evidence of three qualitatively different styles of decision

making in a group of home health care nurses: skimming, surveying, and sleuthing.

Experienced nurses switched from one style to another depending on the situation. A

nurse using the skimming style delivers the required, but minimal service; skimming

allows the nurse to expedite predetermined and well-defined tasks based upon priorities

and available time frame. Surveying concentrates on gathering of concrete, observable

data to infer specific problems, to formulate a reactive and short-term plan of care, and to

address routine and recurrent situations. Sleuthing is used to manage ambiguous,

complex, and unstructured problems by taking a flexible approach to gathering and

evaluating information. On receiving incoming information, the decision maker redirects

subsequent questions, following where the clues lead. Sleuthing often evokes heuristics

based upon experience and clinical knowledge. Sleuthing also has the additional

characteristic of future orientation, resulting in a proactive, adaptive, and realistic

long-term plan.











Additional patterns of knowing beyond cognition, intuition, and experience

inherent in clinical decision making were identified by Jenks (1993). The establishment

of interpersonal relationships (knowing patients, peers, and physicians) with individuals

during the decision making experience was a major influencing factor in the nurses'

clinical decision making ability. In addition, factors in the patient care environment

influenced the decision making abilities.

In a contrasting study, Wells (1995) concluded that decision making was mediated

largely by systemic forces and that the patient's clinical trajectories were not a key

element in shaping the process. In the majority of cases, decisions were made without the

knowledge or understanding of the patients' disease experience; patient and family were

approached only when decision had already been made. Decision making and

communication were one-sided and systematically distorted. Social factors and

organizational imperatives were a major concern to the professionals. Professional

perceptions were functional or instrumental in orientation rather than holistic.

In a study of nursing executives (Johnson, 1990), beliefs, ideals, knowledge, and

values of the nurses combined to form three sets of assumptions which guided decisions:

mechanistic, organismic, and systemic. Mechanistic assumptions are derived from the

belief that nursing is a closed entity, having no interaction with its environment.

Mechanistic modes seek to keep communication closed within nursing, to foster loyalty

and obedience, and to build a strong infrastructure. Organismic systems operate from a

predetermined blueprint; nursing is inherently unstable in a passive environment. Goals

and structures are faithfully maintained, ever increasing resources and contributing to









26

growth and stability of the department. Systemic nursing is viewed as a set of interrelated

roles, functions, and processes. Nurses have unique competencies and knowledge for

participating in decisions that affect work, communication, environment, and change

processes.

Offredy (1998) identified strategies of decision making in an advanced practice

setting. Hypothesis generation started from a known starting point and worked toward an

unknown end point. Hypothesis generation allowed the nurse to transform a seemingly

unmanageable problem into a manageable one by generating a set of possible diagnoses

and then testing their appropriateness in further data collection. Decision trees provided

methods to isolate a finding and to explore all possibilities. By following the tree

branches, clinicians analyzed individual actions then reassembled them in a systematic

way to provide an option. Pattern recognition focused on making a judgment based on a

few critical pieces of information. Pattern recognition occurred on two levels:

analytically, where information was chunked or intuitively, where the whole situation was

grasped. Heuristics facilitated reasoning and often resulted in more effective decision

making. Intuition, a judgment based upon individual experience, gave importance to

recognizing the processes at an unconscious level for rapid, subliminal judgment of visual

and verbal cues. Important factors related to decision making were the ability to

recognize patterns in clinical situations; an appreciation of the consequences of

inappropriate action; and the ability to concentrate simultaneously on complex and

sometimes masked patient cues.











Decision maker characteristics. Watkins (1998) determined that expert nurses

engaged in both rational and intuitive decisions. Characteristics of the expert decision

maker embodied courage in spite of fear or uncertainty; caring and commitment to caring

for the patient first; willingness to take ownership. Expert nurses cited their clinical

experience and their nursing education as factors which developed skill in decision

making.

Decision application in practice. Two other models of decision making have

emerged in the literature. These models expand decision making to a group process and

to a professional level (R. Jones & Beck, 1996).

The collaborative decision model incorporates a group or an interdisciplinary

level to decision making (Baggs, 1994; Collins, 1996; Cruz, 1994; Glendon & Ulrich,

1992; Higgins, 1999; Luker, Hogg, Austin, Ferguson, & Smith, 1998; Marriner-Tomey,

1993; McMurray, 1994; Parsons, 1999; Porter-O'Grady, 1997a; 1997b; Taccetta-

Chapnick, 1996). The collaborative model emphasizes participation, respect, and

consensus. Land (1994) outlined an approach to collaborative decision making using a

soft systems theory. Soft systems theory acknowledges that attitudes and views of

participants within a system may present differing perspectives of problems and differing

satisfaction with solutions. Using the mnemonic CATWOE (customers, actors,

transformation process, Weltanschauung or world view, owners, and environmental

constraints) pinpoints the key stake holders in a system and, subsequently, all the related

activities which enable the system to function. In this holistic and realistic technique,

optimal models may be compared with the real-world in one of four ways: finding









28

differences between the model and the real-world; applying criteria to specific activities

in the model and in the real-world to identify possible change; operating the model

system on paper and comparing with what might occur to what happens in the real-world;

and overlaying the theoretical model with the real-world situation. The collaborative

model works well in complicated, complex situations where other models have failed. A

hallmark of the model is the ability to achieve consensus and commitment to change

through participation by the players: customers, actors, and owners.

The scope of practice model is based upon the assessment of competencies to

determine possible paths on the decision tree (DeSimone, 1999; Heller, 1992; Krejci &

Malin, 1997; Maynard, 1996; P. Nolan, 1998). Competency is the individual's actual

performance, integrating knowledge, skill, attitudes, and behavior; competence is the

individual's capacity to perform job functions regardless of knowledge, skills, behaviors

and personal characteristics (P. Nolan, 1998). Depth and breadth of knowledge to

perform acts are evaluated in the light of safety and accountability. Training and

experience lead to higher competency.

For nurses to work efficiently and effectively, the areas of competency must be

identified for specific roles and settings (P. Nolan, 1998). Using the Nursing

Interventions Classification system to select relevant areas of practice, nurses identified

core competencies, with class labels and definitions. Performance criteria were

determined, and actions or behaviors were defined to assist patients to reach a desired

outcome. Competencies described the specific job requirements and job environment,

useful for managers in selecting staff. Orientation of staff provided assessment and











validation of high-risk, high-frequency interventions, while ongoing education and

training addressed high-risk, low-frequency interventions. Teaching strategies were

planned to accommodate large number of learners, to reduce duplication of the same

information, and to maximize resources. Planning on a regional level rather than a local

level integrated core competencies into nursing curricula and, therefore, equipped the new

graduate with adequate patient-care skills directly from the academic setting.

Decision Making Tools

Tools to measure aspects of decision making in nursing have been developed with

a focus on decision types, attitudes, collaboration, critical thinking, participation, quality,

and task analysis. None of the tools found in the literature contained all of the variables

in the decision model being tested.

An early instrument was the Joseph Decision Making Tool (Joseph, 1982; Joseph

et al., 1988) which measured attitudes and beliefs rather than decision making practices of

registered nurses. The relationships examined were sex-role stereotypes, years of

experience and education upon attitudes toward decision making. The instrument which

has known reliabilities of Cronbach's alpha 0.79 consisted of 20 short scenarios that

required nursing decisions. Similar findings have occurred in subsequent studies

(Catolico, Navas, Sommer, & Collins, 1996).

The Clinical Decision Making in Nursing Scale (CDMN) (Jenkins, 1985) was

developed to measure perceptions of clinical decision making by students. The scale is a

40-item Likert-type, with four subscales: search for alternatives; canvassing of objectives

and values; evaluation and reevaluation of consequences; and search for information.











Cronbach's alpha was 0.78. Subsequent research has combined use of the CDMN with

the Watson-Glaser Critical Thinking Appraisal (WGCTA) (Girot, 2000). The WGCTA is

an 80-item questionnaire divided into five sections: inference, recognition, deduction,

interpretation, and evaluation. No relationship could be found between the development

of critical thinking and decision making in practice, using the two scales.

The Actual Decision Making instrument (Joseph et al., 1988) consists of 27 items

which relate to nursing actions. Subjects choose an answer using a 5-point Likert-type

scale. An alpha coefficient of 0.88 was obtained during preliminary testing.

The Perceptions of Collaboration instrument (Joseph et al., 1988) was developed

as a 10-item questionnaire on a Likert-type format. Subjects rate their relationship with

physicians. A preliminary reliability of 0.63 was obtained.

The Collaboration and Satisfaction About Care Decisions (CSACD) tool was

developed by Baggs (1994) to measure collaboration and satisfaction of providers at the

general practice level and at the level of a specific decision with another instrument,

Decision About Transfer (DAT). The CSACD contained seven questions concerning

collaboration, rated on a Likert-type scale. Cronbach's alpha score was 0.93. The DAT

asked two questions about satisfaction during the decision about transfer; Cronbach's

alpha was 0.63. The DAT was used subsequently by Higgins (1999) to measure nurses'

perceptions of collaborative nurse-physician transfer decision making.

Hollen (1994) tested two scales: Decision Making Quality Scale (DMQS) and

Decision Making Quality Inventory (DMQI). The DMQS was a 7-item Likert-type scale









31

used to screen decisions according to seven quality indicators. The DMQI was a 24-item

Likert-type scale, used to assess decisions by teens and parents.

To evaluate line of reasoning methods in clinical decision making, a task analysis

approach was developed (Narayan & Corcoran-Perry, 1997). Task analysis comprised a

review of the literature on the subject, interviewing an expert nurse, and using the

think-aloud process to identify decision tasks. The methodology was utilized to test a

model for decision making task complexity (M. Lewis, 1997).

Parsons (1999) created a measure of delegation decision by use of a visual analog

scale (VAS). The 10-point VAS ranged from no confidence to highly confident ratings.

Reliability was not reported.

Anthony (1999) constructed a Participation in Decision Activities Questionnaire

(PDAQ) to measure level of authority used in decision and level of involvement by nurse.

The PDAQ listed 42 decision activities, 21 related to patient care and 21 related to unit

operation. Cronbach's alphas ranged from 0.85 to 0.86.

Decision types were a focus of a study on decision-making activity and influence

of nurses in teams (Banaszak-Holl et al., 1999). The researchers developed four

questions on involvement in decisions and six questions on the quality of decisions.

Cronbach's alpha ranged from 0.74-0.89.

Lauri and Salantera (1995, 1998; Lauri, Salantera, Gilje, & Klose, 1999)

developed and tested a tool to distinguish between decision theory (information

processing theory) and intuition (Dreyfus model) in nursing practice. The questionnaire

consisted of 56 items on a five-point Likert-type scale. Half the items describe a











systematic-analytical and rule-based approach to decision making, and half a

holistic-interpretive approach. Alpha coefficients ranged form 0.85 to 0.90 in initial

testing, with similar values in subsequent studies.

Creativity

Creativity has been described since ancient civilizations; Plato understood the

creative person as inspired and possessed by divine influence (D. Miller, 1989). Torrance

described creativity as the process of becoming sensitive to problems, deficiencies, gaps

in knowledge, missing elements, disharmonies; identifying the difficulty; searching for

solutions, making guesses, or formulating hypotheses about the deficiencies (Ferguson,

1992). The creative process is teachable and allows for the development of more creative

behaviors in all individuals. The creative process involves both conscious and

unconscious thought processes, with much of the restructuring of thoughts or perceptions

taking place on an unconscious level (Ackerman, 2000; James, Clark, & Cropanzano,

1999; Therivel, 1999).

Miller (1992) found that scientific creativity has two parts: (a) network thinking

that leads to (b) the nascent moment of creativity. In network thinking, concepts from

apparently disparate disciplines are combined by proper choice of mental image or

metaphor to catalyze the nascent moment of creativity. This nonlinear thought process

can occur unconsciously and is not necessarily in real time.

Simon's approach to creativity commences with a search along the branches and

nodes of a decision tree (D. Miller, 1989). Simon's law of discovery means only finding

patterns in the data that have been observed; whether the pattern will continue to hold for









33

new data that are observed subsequently will be decided in the course of testing the new

law, not in discovering it.

Runco and Albert (1990) found several characteristics that led to creative

behaviors: (a) begin with decisions; (b) knowledge of self; (c) highly intentional; (d)

creativeness and personal identity are emergent in that both grow and change; (e)

creativeness and personal identity both drive each other; and (f) engage the individual on

the personal level of identities, abilities, and differences. Creativity may be viewed as an

opposite of stasis; a manifestation of power; a special form of leadership; the ability to

generate ideas; or the process of becoming.

Kirton (1989) developed the Adaption-Innovation Theory in which an individual

difference construct represents different approaches to problem solving, decision making,

and creativity. Adapters pose solutions which apply accepted, normal procedures, while

innovators offer novel solutions which change the context in which the problem is

embedded.

In nursing, Steele and Maraviglia (1981) added an idea-finding step to the nursing

process to promote creativity in decision making. Deferred judgment, brainstorming,

incubation, and divergent-convergent thinking are processes which facilitate creativity.

These processes are currently reflected in the domain of critical thinking (Abegglen &

Conger, 1997; M. Adams, Whitlow, Stover, & Johnson, 1996; Birx, 1993; Case, 1994;

Daly, 1998; Dexter et al., 1997; Duchscher, 1999; N. Facione & Facione, 1996; Gendrop,

1996; Jacobs, Ott, Sullivan, Ulrich, & Short, 1997; Leppa, 1997; Maynard, 1996;

Reynolds, 1994; Videbeck, 1997). The goals of the creative process outlined by Steele











and Maraviglia are in concert with those of the critical thinking genre: to find answers

that have utility and perhaps novelty for meeting challenges and brining them to a

successful conclusion.

In numerous studies, Amabile (1988, 1996, 1997, 1998) has ascertained

characteristics of both individual and organizational creativity. Individual creativity is

based upon expertise, creative thinking, and intrinsic task motivation. The work

environment utilizes resources, management practices and organizational motivation to

foster innovation. This work environment has a direct impact on individual creativity,

while individual creativity feeds innovation in the work environment.

Li (1997) challenged the creator-centered model by viewing creativity as both a

horizontal and a vertical process. Horizontal domains of creativity allow novelty to occur

in all dimensions, resulting in divergent developments of the domain. Vertical domains,

in contrast, possess certain stable elements that are existentially fundamental, thus

permitting alteration only around certain dimensions. Some dimensions can be both

horizontal and vertical.

Rogers (1996) recognized that an individual's decision is not an instantaneous act,

but rather a process over time of actions and decisions. Rogers' model of diffusion of

innovation consists of five stages: (a) a knowledge stage, in which insight or

understanding develops; (b) a persuasion stage, during which an attitude (favorable or

unfavorable) is formed about the innovation; (c) a decision stage, in which adoption or

rejection of the innovation occurs; (d) an implementation stage, when an innovation is put











into use; and (e) a confirmation stage, when there is reinforcement of a decision already

made.

Innovation, imagination, insight, initiative, critical thinking, fluency, flexibility,

and value judgments are characteristics for the 21st Century nurse sought by nursing

administrators (Barton, 1994; Cassey & Savalle-Dunn, 1994; Daly, 1998; Gendrop, 1996;

Gillmore, 1993; Gilmartin, 1999; Glendon & Ulrich, 1992; Kerfoot, 1998; Koerner,

1996; Vaughan, 1997). Globalization, change, and constraints of time and money

influence the creative ability of nurses. Nurses have moved into a broader arena of

managing the care of aggregate groups and systems, establishing partnerships and

collaboration. The issue is not control but rather connectedness. This link to chaos theory

through quantum mechanics (Porter-O'Grady, 1997b; 1999) enhances the creativity of

nurses by boosting courage to take risk (Greiner & Valiga, 1998), to act outside the

norms, and to be willing to receive criticism (May, 1975). Organizations tend to protect

the status quo and fail to invest the time, effort, energy and money into creative processes

(Losee, 2000); yet thinking outside of the box leaves one open to many ideas, not bound

by tradition, and not prejudging ideas prematurely. The nursing profession has promoted

creativity and innovation for more than 30 years (Gilmartin, 1999). In today's health

environment, the nursing professional ethic fosters the expertise of nurses, the openness

to creativity and change, the development of creative skills, and the management of

professional values within complex social systems.











Education

Carper (1978) proposed four fundamental ways of knowing in nursing: the

empirical, ethical, personal, and aesthetic ways of knowing. Empirical knowing reflects

knowledge that is systematically organized and consists of laws and theories to describe,

explain and predict phenomena unique to the discipline of nursing. Aesthetics

encompasses the art of nursing: the ability of the practitioner to perceive and grasp unique

perceptions of patient's behavior within a specific context. Aesthetics incorporates

envisioning valid modes of helping and responding with controlled balance, rhythm,

proportion, integration and articulation of the whole which is observed as mastery of

knowledge. Personal knowledge involves the perception and management of the self s

feelings and prejudices within a situation in order to respond appropriately and to manage

anxiety. This personal knowledge enables the practitioner to be connected within the

situation and to demonstrate caring. The ethical way of knowing determines what is right

or wrong and commits the practitioner to take action. When tension and conflicts exist

between the ethical dimensions and the clinical components of a situation, the potential

outcomes of these dissonances are significant determinants of the aesthetic response.

Fox (1997) concluded that Carper's ways of knowing could be separated into two

groups: academic and practical knowledge. Academic knowledge includes formal

education in clinical symptoms, medications, equipment, education of patients. Tacit

knowledge is knowledge learned either through experience or through observations.

Tacit knowledge is not formally taught during nursing education because each situation is

complex and context dependent; thus, subtle changes are significant only in light of the











patient's history and current symptoms. The Practical Knowledge Inventory for Nurses

(Fox, 1994) contained scenarios of typical practical problems for nurses. Each scenario

fits into one of three categories, with either a local or global context: managing self,

managing others, or managing tasks. Testing of the instrument indicated that both

academic knowledge and tacit knowledge are required for effective decision making.

While nursing education mainly focuses on knowledge of patients and working with

patients, Fox recommended that nurses should be exposed to learning situations (such as

mentoring relationships) in which tacit knowledge is acquired implicitly.

Kim (1993) found a widening gap between theory and practice in nursing. In a

proposed model, practitioners bring both knowledge in the public domain and knowledge

in the private domain to bear on the specific clinical situation. The complex clinical

situation presents multiple phenomena which the practitioner perceives and then frames

against knowledge from the two domains. Choices in intervention theories and strategies

are made and implemented based upon four different modes of theory application: (a)

coherence; (b) integrative; (c) pragmatic/eclectic; and (d) reflective. In the coherence

mode nurses align with theories and philosophies which have ideological congruity with

their own personal orientations. This mode may result in a conflict between varying

philosophies of nurse and client. The integrative mode is an evolutionary mode of

practice in which new theories and approaches from diverse sources are interwoven with

previously established knowledge and experience in an ever-expanding context. In the

pragmatic/eclectic mode, the focus is not on the practitioner but on the client and client's

problems. This meta-model requires the practitioner to reject the idea of a single model











approach for nursing practice and instead to rely on a process of choices for application,

rather than relying on habit. The reflective mode allows the practitioner to adopt new

theories for their congruence with the situation and meanings gained from the

practitioner's thoughts, actions, and interactions with the client's thoughts and actions.

This action science method leads to continuous learning of new theories-in-use. The

focus of the reflective mode is on the practitioner, the situation, and the client.

Approaches to nursing education are rooted in four historical views of the teacher

and of the learner (Babcock & Miller, 1994): behaviorist, Gestaltist, humanist, and

processor. The behaviorist views thinking as covert trial and error; learning changes

behavior. The teacher directs the learning and the learner complies with the teacher's

direction. The Gestaltist believes thinking to be knowing and perception; the aim of

learning is to understand. The Gestaltist teacher designs programs in which the learner

participates. The humanist discovers meaning through the thinking process; learning

focuses on actualization. The teacher responds, while the learner initiates and evaluates

the experience. The processor perceives thinking as processing information; learning

occurs when information is registered, retained, or recalled. The teacher programs the

educational content and the learner absorbs inputs of information and displays outputs of

the same or transformed information.

Benner (1984; 1991; Benner et al., 1992; 1996) found that knowledge and

education are hallmarks of expertise in nursing. Nurses may be categorized along a

continuum from novice, or inexperienced, to expert.









39

Relating theories of knowledge and education into practice has been the focus of

research in nursing education (Baker, 2000; Carty & Rosenfeld, 1998; G. Davis, 1994;

Dierckx de Casterle, Janssen, & Grypdonck, 1996; Hart, 1999; Hawks, 1999; Ingersoll,

1998; Koch & McGovern, 1993; Koerner, 1996; Lindholm & Uden, 1999; MacCallum,

Roznowski, & Necowitz, 1992; Minnick, 1998; Princeton, 1993; Rambur, 1999; Reinert

& Fryback, 1997; Stevenson, Doorley, Moddeman, & Benson-Landau, 1995; Tschikota,

1993; Wheeler et al., 1995). Major themes in the findings of these studies were that (a)

students processed information sequentially and in small amounts; (b) demonstration and

validation were primary methods of nursing education; (c) mentorship played an

increasing role to broaden the clinical experiences of students; (d) integration of theory

and academic knowledge with practical, tacit knowledge stimulated learning and growth;

(e) the culture, curriculum and educational strategies in the specific nursing education

organization strongly influenced the theoretical and ethical choices of the students; (f)

nursing education was a social event, integrating structure in the teacher-student

relationship, enhancing communication and contact between the faculty and students, and

creating connectedness or affiliation of students in a class structure; (g) successful

strategies were based upon research utilization, centralized resources, supportive

infrastructure for technology and information literacy for both students and faculty, and

interdisciplinary collaboration to provide economical benefits and exciting educational

opportunities; and (h) complex, conflicting worldviews, along with the information

explosion, produced tensions and resultant ambiguity unless an atmosphere of

questioning was created to critique existing practices.











A major focus of nursing education in the past three decades has been critical

thinking (Birx, 1993; Dexter et al., 1997; N. Facione & Facione, 1996; Jacobs et al.,

1997; Leppa, 1997; Maynard, 1996; Reynolds, 1994; Videbeck, 1997; Wells, 1995).

Critical thinking incorporates both academic knowledge and tacit knowledge. Critical

thinking is the purposeful, self-regulatory, judgmental process which incorporates

interpretation, analysis, evaluation and inference (N. Facione & Facione, 1996). Critical

thinking may be viewed as an educational outcome. The tools developed for measuring

this outcome are the Watson-Glaser Critical Thinking Appraisal (WGCTA) (Watson &

Glaser, 1980), California Critical Thinking Skills Test (CCTST) (P. Facione, 1990),

Ennis-Weir Critical Thinking Essay Test (EWCTET) (Ennis & Weir, 1985), and Cornell

Critical Thinking Test (CCTT) (Ennis, Millman, & Tomko, 1985). The WGCTA has

been found to be the mostly widely used tool in nursing education with the CCTST

serving as an alternative (M. Adams et al., 1996). All four tests have limitations because

of small samples and lack of psychometric data; in addition, none of the tests are specific

for the domain of nursing. Thus, critical thinking remains an elusive outcome evaluation

standard for nursing programs, while at the same time critical thinking is an integral

component of nursing education strategies focusing on clinical decision making. The role

of critical thinking in decision making is to provide a careful analysis and judgment, to

seek reasons and alternatives, to perceive the total situation, and to change one's view

based upon evidence (Reynolds, 1994). Making a decision is an outcome of critical

thinking and a component of professional nursing practice.











Educators, who simply transmit knowledge and ways of thinking only,

exemplified an inertia of habit (Greiner & Valiga, 1998). To foster creativity in the

decision making process, educators broke with tradition; helped learners to be aware of

the world; heightened the consciousness of students; engaged learners as distinctive,

questioning persons; and cultivated the perception of constant change in the world

(Greene, 1995).

Leadership

Research on the concept of leadership has been focused on predicting leadership,

determining leadership skills, and evaluating leadership effectiveness (Altieri & Elgin,

1994; McCloskey & Molen, 1987). Leadership in nursing has been traditionally viewed

as the process of influencing people to accomplish goals. Nurse leaders exhibit roles as

change agents, builders, and innovators. Trait theory has been used to describe the inborn

characteristics of a leader, while behavioral theory has classified the functional aspects of

leadership into four stages: autocratic, bureaucratic, participatory, and free-rein (Hanna,

1999). Another classification of leadership exposed four types (Rothschild, 1993): (a)

risk takers are leaders who have a passion and genius to make dreams happen beyond the

reach of others; (b) caretakers are leaders undergoing evolutionary growth rather than

revolutionary; (c) surgeons are leaders who remove areas no longer needed or productive;

and (d) undertakers are leaders who harvest, close, or merge operations, and show

concern for survivors.

An alternative perspective is the behavioral approach to leadership and decision

making (Rowland & Rowland, 1997): (a) assertive types decide what is right based on











one's own priorities but accounting for others' perspectives; (b) aggressive types decide

what is right regardless of others' perspective; (c) manipulative types persuade others on

common, shared priorities; and (d) submissive types decide according to other people's

expectations. Leadership requires a learning environment to engage people in

confronting challenges, to adjust values and perspectives, and to acquire new habits

(Heifetz & Laurie, 1996).

Irurita (1994) found leaders optimize the situation, making the best of the options

possible. Optimizing occurs in an environment prone to retardation (delay, impedance)

and to turbulence. Leaders provide solutions that optimize survival, invest in the future,

and transform the present environment. Failure to optimize led to a floundering state or

inability to cope or to meet expectations. Optimizing leaders displayed caring values,

were client-centered, and committed. Commitment was demonstrated by a pronounced

identification with the decision, involvement in the processes and displayed loyalty to the

work unit.

Transformational leadership (Barker, 1992; Marriner-Tomey, 1993; Manfredi,

1995; Taccetta-Chapnick, 1996; Pesut & Herman, 1998; Sosik, Kahai, & Avolio, 1998;

Dixon, 1999; Perra, 1999) demonstrates a vision, increases creativity, and focuses on

change and conflict resolution. Transformational leaders create not only the vision, but

the climate for change; the vision creates conflict because change is inherent.

Transformational leaders develop new leaders. Implementing the vision leads to

professional satisfaction and growth. Five charismatic behaviors of transformational

leaders were identified (Taccetta-Chapnick, 1996): (a) focusing attention on planned











actions; (b) encouraging risk taking and creativity; (c) listening and providing feedback;

(d) demonstrating trustworthy behavior and commitment to the vision; and (e) expressing

concern for others.

Quantum thinking proposed a dramatic departure from industrial age thinking in

late 20th Century research on leadership and decision making (Bassingthwaighte et al.,

1995; Bolland & Atherton, 1999; Goswami, 1996; Herbert, 1995, 1996; Ingersoll, 1998;

Kupperschmidt, 1998; Lindholm & Uden, 1999; McMurray, 1994; A. Miller, 1992;

Mishel, 1990; M. Morris, Speier, & Hoffer, 1999; Neuman, Newman, & Holder, 2000; P.

Nolan, 1998; Northhouse, 1997; Offredy, 1998; Parsons, 1999; Perra, 1999; Pesut &

Herman, 1998; Porter-O'Grady, 1997a, 1997b, 1999; Rambihar et al., 1999; Speier,

Valacich, & Vessey, 1999; Starck, Warner, & Kotarba, 1999; Thompson, 1999a, 1999b;

Vicenzi & Hamilton, 1999; Vogelsang, 1999; Williams, 1998; Winreich, 1999; Woods,

1999; Zausner, 1998). In a quantum orientation, linear thinking becomes meta-thinking;

analyses shift from compartmental to whole systems; functions are subservient to

outcomes; and predictable events are replaced by variable effects. Quantum thinking

leaders work in teams, accept accountability for decisions, and provide an information

infrastructure with immediate feedback. Leadership is focused at the point-of-service

since the effectiveness of the decision process is in the hands of the provider. The

process leader integrates resources, information, and generates knowledge and support for

the provider to make effective decisions. Leadership is horizontal rather than vertical in

quantum systems; leadership is relationship-oriented rather than control focused.

Quantum thinking recognizes the integration of all things within a whole form of











reflection (Porter-O'Grady, 1997b; 1999). Issues of fit are paramount to issues of

function. The tools of technology, comprehensiveness of information and demand for

service integrity are the driving forces in the quantum age.

Leadership competencies for the 21st Century leader (Fralic & Denby, 2000;

Gauthier, 1996; Hansten & Washburn, 2000; Hillesheim, 1998; Mahaffey, Kaplan, &

Triolo, 1998; Neuman et al., 2000; Perra, 2000; Starck et al., 1999) were identified as

utilization of technical knowledge and systems thinking; resource allocation to achieve

goals, decision analysis; motivation; delegation; mentoring; conflict management;

communication skills; and consensus building. Personal characteristics of successful 21st

Century leaders were acknowledged as integrity, drive, dedication, flexibility,

influencing, interactive empowerer. Other necessary skills embraced acclimatization to

chaos, pattern recognition, coaching, and continuous learning.

Antrobus and Kitson (1999) ascertained that nursing leadership arose from

fundamental knowledge about nursing practice. This philosophical understanding of

nursing incorporated an ethic of care and legitimatized the leadership influence internally

and externally to nursing. Leaders in nursing provided the bridge between policy and

practice through interpretation and translation. Language was used within a context of

practice, administration, or academic settings; leaders who performed this translation

were viewed as credible and visible within multiple settings. A bi-cultural nature of

nursing leadership stemmed from the leader's ability to embrace the values of nursing

while recognizing and influencing the values of the organization. The skills repertoire for

nursing leadership was summarized into five profiles: a powerful influential operator; a











strategic thinker; a developer of nursing knowledge; a reflexive thinker; and a process

consultant. The powerful influential operator worked with others to empower them,

creating and sustaining a work environment with common values and value-driven

relationships. The strategic thinker created meaning and facilitated learning. Leaders

enabled an emergent process by identifying patterns within the process and by shaping

and articulating those patterns into a collective vision in a changing environment. A

developer of nursing knowledge integrated research evidence with practice and explicated

tacit knowledge from practice, using strategic processes to channel and translate nursing

knowledge from the grassroots to collective efforts. The reflexive thinker understands

self and values, along with purpose and personal meaning, within a complex and

ever-changing environment. The reflexive thinker establishes support mechanisms and

process to enable structured reflection within the environment. The process consultant

works through and with others to intervene in human processes as appropriate to achieve

transformational change. In-depth knowledge of human processes, communication

patterns, problem solving and decision making is required. Thus, nursing leadership is a

resource and a vehicle to influence and shape both policy and practice in political,

managerial, academic, and clinical domains.

Recent researchers (Goleman, 1998, 2000; Perra, 1999, 2000; Strickland, 2000)

have expanded the leadership styles to six categories: (a) coercive leaders who demand

immediate compliance; (b) authoritative leaders who mobilize people toward a vision; (c)

affiliative leaders who create emotional bonds; (d) democratic leaders who build a

consensus through participation; (e) pacesetting leaders who expect excellence and











self-direction; and (f) coaching leaders who develop people for the future. Emotional

intelligence is the ability to self-manage and to manage relationships effectively consists

of four fundamental capabilities: self-awareness, self-management, social awareness, and

social skill. Each capability is composed of sets of competencies. Self-awareness fosters

the leader's ability to understand emotions and the impact of emotions on work

performance and relationships; to perform a realistic self-assessment of strengths and

limitations; and to portray self-confidence or a strong sense of positive self-worth.

Self-management skills include self-control; trustworthiness, honesty and integrity;

adaptability; achievement orientation; initiative; and conscientiousness or the ability to

manage oneself and one's responsibilities. Social awareness concentrates on empathy,

organizational awareness, and a service orientation to meet customers' needs. Social

skills in leaders embody the vision to others through inspiration, influence, persuasion,

communication, as well as through roles in developing others, in being a change catalyst,

in managing conflict, and in building bonds, teamwork, and collaboration. Leaders need

many styles; the most effective leaders switch flexibly among the leadership styles as

needed. Leaders who have mastered four or more (especially the authoritative,

democratic, affiliative, and coaching styles) have the best performance. Leaders do not

mechanically match style to fit a checklist but rather are sensitive to the impact on others

and seamlessly adjust style to get the best results. This fluid leadership in action is

spontaneous, nimble and rapid in assessing the situation and in determining the best

approach to employ to maximize results.











Studies using a qualitative approach have produced a rich understanding of

leadership in nursing. Exploratory studies described leadership in autonomous

professional practice (Ferguson-Pare, 1997; Spooner, Keenan, & Card, 1997); quality

(Williams, 1998); transformational processes (Pesut, 1998; Pesut & Herman, 1998);

beliefs and perceptions (D. Allen, 1998); image of leaders (Lehna et al., 1999); skills

(Starck et al., 1999); and changes in leadership (Bleich, 1998). Ethnographic and

grounded theory methods resulted in identification of leadership styles and characteristics

(Antrobus & Kitson, 1999; Ferguson-Pare, 1997; Irurita, 1994). Case analysis, with

reflexive comparison and reframing techniques, was used to describe the aspects of

leadership within a decision framework (Pesut & Herman, 1998).

Tools to measure leadership in the literature focus on leadership behaviors,

leadership factors, leadership competency, leadership styles, and leadership environment.

The Ohio State Leader Behavior Description Questionnaire (W. Allen, 1995; Taunton,

Boyle, Woods, Hansen, & Bott, 1997) was developed to determine behaviors exhibited by

leaders. The Multifactor Leadership Questionnaire (Morrison, Jones, & Fuller, 1997;

Stordeur, Vandenberghe, & D'hoore, 2000; Tepper & Percy, 1994) was used for

evaluating transformational behaviors. Cronbach's alpha across these studies was 0.92,

with intercorrelations among the transformational scales ranging from 0.68 to 0.85.

Competencies were assessed in two studies with researcher developed scales: Leadership

Competency Instrument (Krejci & Malin, 1997) and Leadership Competency Profile

(DeSimone, 1999). Reliability for competency instruments was not reported. The

Achieving Styles Inventory (Klakovich, 1996) examined styles of leadership used, while











the Job Activity Scale (Laschinger, Sabiston, & Kutszcher, 1997) gauged leadership

activities. The effect of environment has been estimated using Chandler's Conditions for

Work Effectiveness Questionnaire (A. Adams, Bond, & Hale, 1998; Laschinger et al.,

1997; McDermott, Laschinger, & Shamian, 1996). Several studies employed tools

created by the researcher without reported reliability (Bond & Fiedler, 1999; Ellefsen,

1998).

Experience

The value of experience in nursing has been explored by researchers (Benner,

1991; Benner et al., 1992, 1996; Bleich, 1998; C. Davis, 1995; Grobe et al., 1991; Leppa,

1997; M. Morris et al., 1999; Perra, 1999; Radwin, 1998; Stevenson et al., 1995; Tabak,

Bar-Tal, & Cohen-Mansfield, 1996; Tishelman et al., 1999). Benner (1984; Benner et al.,

1996) established through pioneering endeavors that the experienced nurse utilized

intuitive, engaged, proactive reasoning more than scientific, disengaged, theoretical

reasoning. Benner established five levels of practitioner based on reasoning and

experience: (a) the novice nurse uses theoretical knowledge and rules; (b) the advanced

beginner employs more objective facts and more sophisticated rules, learning to

anticipate; (c) the competent nurse devises new rules, is responsible for outcomes,

remains less detached, and formulates hierarchical choices; (d) the proficient nurse

assimilates theoretical, intuitive experiences and replaces reasoned responses with easier

actions; (e) the expert nurse knows only what needs to be achieved, based on the practical

and situational determinants of problem, skillful coping, and purposeful goal attainment.











Habitual practices and skills grow through reflection on experiences, using analytic

clinical thinking.

The community of nursing is created through storytelling by nurses (Benner,

1991); these narratives share the embodiment of common nursing experiences to new

nurses and to peers. Narratives of learning create an openness to experience. The

narrative of learning may expose evidence of failure, in knowledge, in relational skills, or

in judgments. A corrective narrative highlights poor clinical judgment in one example so

other nurses may avoid similar errors in the future. The story becomes integrated with the

feelings, thoughts, perceptual recognition, and memory of the individuals who recognize

or project themselves as the characters in the story. Narratives of learning the skill of

involvement foster relational skills appropriate to the practice of nursing. These

narratives convey knowledge about getting the right kind of involvement and

interpersonal distance to fit specific situations. The relational skills are context

dependent, so biases and exclusions are encountered; therefore, new possibilities for

connection and distance may be discovered. Nurses have an elaborate discourse on the

right kind, level, and amount of involvement in the clinical situation; nurses identify

over-involvement in which the ability to offer alternatives may be lost. A rich tradition of

practical knowledge develops over time in situations and in dialogue with patients,

families, and colleagues. Narratives of disillusionment convey the limits of others'

knowledge. These stories related times when rules, policies, and procedures do not match

the clinical situation or environment. These narratives are often characterized with

humor, self-discovery, new insights, and sometimes wisdom. Disillusionment is











frequently a snapshot of the confrontation between theoretical, idealized visions of

practice and the reality of true life situations. Narratives about death and suffering enable

nurses to learn moral and practical lessons about presence, comfort, courage, anger, failed

expectations, hope, and power. Liberative narratives depict nurses discovering the worth

of their work and the importance of their voice for the patient and family. Narratives of

liberation often contain narratives of disillusionment within them. Liberation themes

include stories about breaking free from bias, intimidation, inhibition, fear of risk,

vulnerability, visibility, responsibility, rules and procedures. The narratives promote a

sense of nursing community in which stories, habits, practices, concerns, and experiential

wisdom are exchanged through multiple perspectives.

Novices often rely on information and earlier examples from within the same or

related domains (M. Morris et al., 1999); knowledge may be transferred from a known

domain to a new problem space. However, the individuals may devise analogies that are

not valid or relevant in the new domain.

Experience need not be entirely viewed as intuitively derived and validated

(English, 1993; Radwin, 1998; Rolfe, 1997; Sutherland, 1997; Tabak et al., 1996).

Researchers have found other characteristics of experience that foster development of

clinical knowledge in nurses. Radwin (1998) found that confidence was a key component

of experience. Confidence was the ability to ask difficult questions, to listen actively, and

to learn of patient's perceptions. Confidence enabled the nurse to consider a broader

range of interventions and enhanced the individual selection of interventions for the

situation. Habitualization was more likely to develop as the nurse moved toward expert











levels of performance (Heath, 1998). Negative habitualization resulted in routine and

coping dominated practice in response to external pressures. Habitual practices, and its

awareness, originated spontaneously and contradicted efforts to improving practice over

time.

Tabak et al. (1996) discovered that the complexity of the task and the consistency

(or uncertainty) of information differentiated novice and expert decision making.

Novices failed to see contradictions in the information and therefore maintained a posture

of certainty and confidence, while the expert identified uncertainty in the scenario and

offered more alternatives, with less confidence in any one pathway. Novice nurses

differed from experts not only in relying on more effortful processing but also in basing

their decisions on biased information and strategies that could lead them to wrong

conclusions. Experienced nurses used intuition because their experience enabled them to

develop the appropriate schemata. Analytical process was used when appropriate

schemata was not available or when the problem was not too difficult to solve by

piecemeal processing.

A reexamination of Benner's model found that expertise may be explained by

abductive reasoning, or fuzzy logic, rather than on the basis of intuition alone (Rolfe,

1997). As reported by Benner, experts are often unable to verbalize their expertise;

Benner relates this to intuitive deductive methods. Adaptive fuzzy systems concentrated

not on what the expert verbalizes but rather what the expert does. A fuzzy system

generates its own fuzzy rules based upon accumulated experience. Instead of applying

the rules in a linear, step by step fashion, the fuzzy system applies the rules all at once but









52

to different degrees, depending on the situation. Thus, expert behavior can be reproduced

with relatively few rules. The key is not to dictate how experts should practice (decision

trees, protocols, algorithms can do that) but rather to illustrate how experts do practice.

The strength of the fuzzy model is in descriptive ability, not in prescriptive power.

Experience, viewed as historical narratives or archival data, is amenable to

cognitive process of induction, deduction, and retroduction (moving back and forth

between induction and deduction to generate interpretations) (Sutherland, 1997).

Retroduction is most often employed where new concepts or new associations are being

evaluated. Information exists as a result of actual accounts perceived, interpreted, and

reported by nurses. These accounts are temporal and contiguous. Creative and critical

thinking are crucial intellectual modes for visualizing new patterns in behavior and in

interpreting relationship and linkages among experiences.

Risk Taking

Risks are decisions made with an element of potential loss or danger (Umiker,

1997); action or inaction is taken under conditions of uncertainty. Risk is everywhere and

unavoidable. Risk is a property of options that affects choices among the options

(Pollatsek & Tversky, 1970; Raiffa, 1994). Options can be ordered with respect to their

riskiness and their utility. The risk of an option is related to the variance of its outcomes.

Perceived risk may be viewed as a vague and subjective notion; however, studies have

found that people are consistently able to order choices with respect to riskiness (Jia,

Dyer, & Butler, 1999; Zickar & Highhouse, 1998). Perceived risk increases when there is

an increase in range, variance or expected loss. For the risk averter, the ordering of









53

choices by risk is the inverse of preference ordering based upon expected value. For the

risk seeker, the ordering of choices is based upon expected value, coupled with the

perceived risk, to maximize the outcome benefits. Outcome history influences risk taking

(Sitkin & Weingart, 1995; Thaler & Johnson, 1990): a history of success leads to higher

risk taking; a history of failure leads to lower risk taking. Decisions which have the

potential to affect an individual's or organization's performance and position become

strategic decisions because of the significance of actions taken, the levels of resources

utilized, or the precedents set (Mittal & Ross, 1998). Data mining may find unexpected

or hidden patterns in data (Padmanabhan & Tuzhilin, 1999); these unexpected findings

challenge conventional wisdom. Patterns contradictory to prior knowledge are by

definition unexpected. The interestingness of a pattern depends on the decision maker

and not solely on the strength of the pattern. An interesting pattern may nonetheless have

little added value to decision making.

Information reduces the perception of risk or, at least, increases the tolerance of

risk in decision making (Jia et al., 1999; Tversky & Fox, 1995). Decision theory

distinguishes between risky choices in which the probabilities of possible outcomes are

known and uncertain possibilities in which the probabilities are not assumed to be known.

When there is no risk, there is no uncertainty involved.

Using chaos theory, Mishel (1990) conceptualized uncertainty related to nursing

practice. Two processes determine the value of uncertainty in a situation: inference and

illusion. Inferences are evaluated under conditions of uncertainty based upon examples

of related situations. If the inferences are as positive ones, the uncertainty may be









54

interpreted as an opportunity; if the inferences are threatening types, then the uncertainty

may be interpreted as a danger. Opportunity and danger are parallel concepts, indicating

the choice of one and only one path. Illusion may be explained as the construction of

beliefs that have a positive outlook. Because of the vague and formless nature of

uncertainty, an event may be transformed into a positive illusion. Coping mechanisms

are activated to reduce uncertainty under conditions of danger, while coping mechanisms

are activated to maintain uncertainty under conditions of opportunity.

Wurzbach (1991) discovered several judgmental heuristics and biases related to

the inherent uncertainty of a decision outcome. Inferences may not be readily available to

the decision maker or the decision maker may have overconfidence in their own

erroneous reasoning. Accuracy in decisions is enhanced when all possible choices are

outlined. However, people favor positive rather than negative evidence for choices and

tend to disregard evidence inconsistent with the chosen path. The frame a decision maker

adopts is controlled partly by the problem itself and partly by the norms, habits, and

characteristics of the decision maker. Spurious judgments during decision making may

be made by nurses who are unaware of their decision processes, who do not differentiate

between inference and illusion, and who do not consider alternative explanations.

An exploration of personal risk taking by Dobos (1997) ascertained that nurses

were guided by strongly held values and confident knowledge, as well as the utilitarian

ethic. Problems were perceived as personal threats. Consequently, the nurses in the study

tended to act in isolation rather than to collaborate with others to minimize risk for all.

The research concluded that nurses who placed themselves at risk did so for the patient









55

and utilized a process of risk taking that was thoughtful, constructive, goal oriented, and

value driven.

Umiker (1997) found that risk avoiders utilize self-protection and apathy in the

face of change to reduce the fear of failure and loss of colleague support. Risk aversion is

often a lifelong habit, characterized by procrastination, silent or noncommital stances,

cynicism, and over-caution. Risk takers tend to seek increased responsibility and

accountability, characterized by ownership, commitment, innovation, and sense of reality.

Some people are risk obsessive, getting a thrill from the danger of risk taking, marked by

explosive energy and change, exceeding authority, argumentative and insensitive

treatment of others. Risks can be minimized with confidence from past successes,

achievable goals, sufficient information, and preparation for some failures or setbacks.

A constraint to risk taking is the professional role conception (Kubsch, 1996) in

which individual and organizational values about nursing influence the level of autonomy

and power in decision making. Successful risk takers are role-breakers who increase

tolerance for risk within the work environment. Using role switching and self-reflection,

along with continual juxtaposition of roles within a nursing environment, a risk taker may

question the adequacy of an individually based conceptualization of autonomy

(Tishelman et al., 1999). Difficulties exist in opposing decisions assumed to be valid for

a group, especially when supported by persons with a higher rank on a professional or

organizational hierarchy. Conscious and systematic exploration of issues can offer new

perspectives with the basic and often existential issues embedded in the situational

decision process.











Informatics

Nursing informatics was first described as a combination of computer science,

information science, and nursing science. The management of nursing information and

knowledge supports the practice of nursing and the delivery of nursing care (Graves &

Corcoran, 1989). Hannah (1988) expanded nursing informatics to include information

technologies related to any function of nursing or to activities carried out by nurses. The

boundaries of nursing informatics are dynamic, changing, and contiguous with those of

nursing (Strachan, 1996). Nursing informatics has been visualized as the superhighway

(Nagle & Ryan, 1996) which provides comprehensive access to and integration of

information and knowledge in a holistic approach that transcends time, settings, and

providers. A broad definition by the Nursing Informatics Working Group of the

International Medical Informatics Association asserted that nursing informatics is the

integration of nursing, its information and information management with information

processing and communication technology to support health (Yensen, 1997).

The scope of nursing informatics was delineated by a task force on nursing

informatics (American Nurses Association, 1994) as a speciality group within nursing

that integrates nursing science, computer science, and information since in identifying,

collecting, processing, and managing data to support nursing practice, administration,

education, research, and the expansion of nursing knowledge. The practice of nursing

informatics includes the development of applications, tools, processes and structures

which assist nurses with the management of data. Activities involved in nursing

informatics are identifying, naming, organizing, grouping, collecting, processing,











analyzing, storing, retrieving, or managing data and information. Nursing informatics

intersects with supporting disciplines (such as computer science, psychology,

management information science, organizational science, engineering, library science,

information management, communication theory, and decision sciences) and with

collaborating health providers which share programs and common data.

The functional areas of nursing informatics (American Nurses Association, 1994)

encompass system analysis and design; system implementation and support; system

testing and evaluation; human factors; computer technology; information management;

professional practice, trends, and issues; and theories about informatics and related areas.

In system analysis and design, a comprehensive needs assessment is conducted and turned

into a set of functional specifications. The process of information management is

depicted in algorithms, decision trees and flow charts. In the design phase, hardware and

software requirements are determined along with data and file structures, information and

process sequencing, and report definition and generation. Safety and security measures

are integrated with a risk analysis and disaster recovery plan. Audits serve as checks for

errors, provide exception reports, and a trail for problem solving. Systems

implementation and support personnel develop strategies for education and training,

policy and procedure developments, documentation materials, implementation strategies,

and ongoing support mechanisms. The systems testing and evaluation foci are upon all

areas of the application; the effectiveness and efficiency of the system are evaluated in

terms of cost benefits, cost effectiveness, benefits, realization, and social impact. Human

factors to consider in nursing include ergonomics and environment (physical aspects);











visual display, user interfaces, and data representation (software aspects); and user

satisfaction. Computer technology is used to examine the hardware and software

components as well as the processing and storage functions. Distribution and

communication of information resources involve user access, telecommunications,

networks, and reports. Information management is concerned with data analysis; data

may be transformed, aggregated, or presented to users in a variety of ways. Updates,

additions and deletions of the database ensure currency and purging of unnecessary files.

Professional practice issues discuss the role, education, and ethics of nursing informatics.

Trends and issues in health care, technology, and regulatory-accreditation requirements

alter the design and implementation of nursing information systems. Nursing informatics

employs borrowed theories from nursing and other sciences, such as communication,

information, computer, behavior, management and systems. Group dynamics, change

theory, organizational behavior and learning theory are major influences in the practice of

nursing informatics. Within nursing informatics, theory development focuses on

nomenclatures/vocabulary, taxonomies, and coding schemes.

The Staggers and Parks Nurse-Computer interaction framework represents nurses

and computers interacting in a system of mutual influences with information as the

medium of exchange between them (Staggers, 1996; Staggers & Parks, 1993). The dyad

interacts within a context or environment and across time. Nurse characteristics and

behaviors affect the initiation and response of the information system in the task of

information exchange.











Goossen (1996; Goossen et al., 1996) expanded the definition by Graves and

Corcoran to include the multidisciplinary endeavors of nursing informatics. The model

created included the analyzing, formalizing, and modeling of how nurses collect and

manage data; processing data into information and knowledge; making knowledge-based

decisions and inference for patient care; and using empirical and experiential knowledge

to enhance the quality of professional practice. Nursing informatics investigates

determinants, conditions, elements, models, and processes in order to design and

implement effective, efficient computerized systems. The framework for nursing

informatics integrates the process of transforming data into information and knowledge to

make decisions for actions and evaluation within the umbrella nursing information

management system.

Turley (1996b) designed a three-dimensional model for nursing informatics in

which cognitive science, information science and computer science are integrated to form

informatics within the greater nursing science environment. Cognitive science is

combined with the fields of psychology, linguistics, computer science, philosophy, and

neuroscience. Cognitive science encompasses a wide range of perception, thinking,

understanding, and remembering. Information processing theory is embedded within the

cognitive science domain. Information science provides an understanding about

organizations and information flow in the environment, while computer science includes

hardware and software development. The convergence of these three intersected spheres

(cognitive science, information science, and computer science) describes informatics and

rests on a base of nursing science.











Humans are complex, adaptive systems in which information is a unifying

concept (P. Jones, 1996). Information connects the mechanistic and the humanistic

worlds of nursing. The sheer volume of information creates a bottleneck and requires

compression of the data itself. However, individuals, from a chaotic environment, extract

information that is significant and relevant for the individual. Human knowledge is

doubling at an estimated rate of 33 years, while medical knowledge is doubling about

every 19 years (Healthfield & Louw, 1999).

Perception of information in the environment may be viewed as two stages

(Delvin, 1991): analog and digital. The first stage of perception is that in which

information is directly accessible by way some type of sensor; this information flow is

analog. The second stage of cognition involves the extraction of a specific item of

information, in a conversion along a continuum from analog to digital information.

Redundancy is a fundamental characteristic of information systems (P. Jones, 1996); this

redundancy allows humans freedom of choice in constructing and interpreting message.

Errors and uncertainty are created within the message but the overall message can be

deciphered based upon past experiences, coding and classification schemes. Information

technology is forging machines that are very small, even invisible to the average observer.

This nanotechnology produces a medium for the development of quantum theories of

information that apply across nanometers and macro scales. As data are gathered and

manipulated into information, facts are deduced and concepts identified which lead to

models, theories and laws of information. Knowledge is thus contained within this

hierarchal paradigm.









61

Communication is a process within informatics: the exchange of information for

some purposes (Caris-Verhallen, Kerkstra, & Bensing, 1997). Enormous diversity exists

with respect to participants, settings, and types of exchanges. Most communicative

behavior is classified as one of two types: verbal (conveying messages with language) and

nonverbal (conveying messages without the use of language). The goal of effective

communication is interpreting the messages and responding in an appropriate manner.

Communication in health care has instrumental and affective aspects. Instrumental refers

to task-related behaviors necessary in assessing and solving problems. Instrumental

behavior is mainly verbal in nature. Affective refers to socio-emotional behaviors

required to establish relationships, such as respect, comfort, and trust. This kind of

behavior is transferred both verbally and nonverbally. Levels of communication may be

categorized into four contexts of care: (a) rare or very brief task oriented communication

(instrumental); (b) two-way discussion about care needs and instruction (instrumental);

(c) communication about care or social talk in which the patient directs his own care

(affective); and (d) intense dialog and supportive statements by the nurse for the purpose

of having the patient understand the illness and treatment (mixed instrumental and

affective).

Information systems rely on a taxonomy of data, information, and knowledge

(Bliss-Holtz, 1990). Data are entities void of interpretation while information has

structure, organization, and interpretation. Knowledge is synthesized information in

clearly defined and formalized relationships. Mylopoulos (Wand, Monarchi, Parsons, &

Woo, 1995) suggested that there are four types of knowledge related to information











systems: (a) the subject world or represented domain; (b) the usage world or the

environment within which the system is used; (c) the development world or the process

and environment with which the system is developed; and (d) the system work, or the

information system itself. Analysis transforms the perceived real-world system into a

model of the subject and usage worlds. Design transforms the model of the subject world

into a model of the information system. Implementation transforms the model of the

information systems into the implemented information system.

Information systems facilitate rapid, less expensive, more precise, and more

selective communication across time and geographic location (Huber, 1990; Henry,

Warren, Lange, & Button, 1998; Fralic & Denby, 2000). Information systems, in the

context of decision making, allow the storage and retrieval of large amounts of

information and the access of information outside of the organization. New information

can be reconfigured with speed and accuracy to compactly store for use in decision

models and to inexpensively report information about organizational transactions.

Quality of decisions is direction dependent on the quality of analysis of information.

Information systems affect structure and processes by reducing personnel and other

resources required to maintain information. The organizational memory is standardized,

indexed for retrieval and exchanged across boundaries. Internal and external

environmental scanning examines problems and opportunities. Messages are screened,

packaged, and interpreted by information systems; however, users may be vulnerable to

overload of irrelevant and unintelligible messages.











The use of technology by nurses is contextual, affected by social, political, and

economic forces (Bernardo, 1998). Technology enables nurses to perform functions

accurately, quickly, and efficiently. The strength of technology lies in its potential to

explain and predict effective health care for persons and families. Bernardo points outs

that nursing is vulnerable in several areas in which technology may dehumanize the

nurse-client relationship: (a) technology alienates the dyad through interference in the

nurse's goals to assist the patient; (b) human parameters are interpreted by machines (use

of monitoring probes); (c) the art of nursing is based on technology rather than human

ecology; (d) technology focuses on the visible and tangible; (e) language connected with

the technology is contrived and commonplace, de-emphasizing the individual's

perspective.

Technological optimism views technology as beneficial for nursing, while

technological romanticism views technology as disruptive and even dangerous to nursing

(Sandelowski, 1993, 1996, 1997). Technological optimists reject the notion that

technology is harmful to nursing, and in fact, believe the rejection of technology to be

harmful. Nursing could be depicted as technology. For technological romantics,

technology resists incorporation into the body and spirit of nursing because the seducing,

deceiving, and diverting power of technology detracts from the art of nursing. Thus,

romantics view technology at odds with nursing. In the real world of practices, nurses are

pulled and pushed between these two opposing forces. Technology pushes or pulls in one

direction, while the nurses pull back or push in other directions.









64

In a technology-driven world, division of labor occurs when problems are broken

into successfully smaller parts (Ketner, 1996). This specialization may result in the

regarding of hypotheses as facts. Technology is often tailored to work specifically within

narrowly confined environments; applications do not work in all problem settings. While

the main purpose of technology is control of a process and such control implies a causal

environment, the nursing environment is chaotic, nonlinear, and dynamic. Science itself

is not analogous to technology. Having a technological frame of reference may deny or

neglect significant qualities of humanness in the nursing domain.

A review of the 181 research presentations at an international nursing informatics

conference (Nursing Informatics New Zealand, 2000) pinpointed current research in 11

major areas: (a) education of nurses (53); (b) use of the internet (19); (c) management

issues (18); (d) patient care applications (18); (e) medical records (17); (f) computer

hardware and software development (12); (g) nursing languages and taxonomies (12); (h)

nursing process implementation (12); (i) nurse attitudes and characteristics (10); (j) theory

development and application in informatics (5); and (k) decision making and decision

support services (5).

An examination of the 127 nursing research studies on informatics published

during the last decade discovered the following general topics:

1. Language and taxonomy development which was the most frequent topic (22)

(Bliss-Holtz, Taylor, & McLaughlin, 1992; Bowles, 1999; Button et al., 1998; Clark,

1998; M. Clarke, 1998; Elfrink, 1999; Goossen et al., 1998; Hardiker & Rector, 1998;

Henry, Holzemer, Reilly, & Campbell, 1994; 1997; 1998; M. Johnson, 1998; Jones-









65

Baucke, 1997; Lowen, 1999; Martin, 1999; McCloskey & Bulechek, 1998; McCormick et

al., 1994; Ozbolt, Fruchtnicht, & Hayden, 1994; Wand et al., 1995; Warren & Coenen,

1998; Westra & Solomon, 1999; Zielstorff, 1998).

2. Management functions (21) (Chase, 1994; Clifford, 1997; Coleman, 1997; Eck,

1999; Edgar, 1997; Fabray, 1992; Fortenberry, 1995; Hemman, 1998; Henry, 1995b;

Hovenga, 1995; Ireson, 1995; Merryman, Sharbaugh, & Roberts, 1999; M. Nolan et al.,

1997; Parson, 1995; Pollock, 1996; Preuss, 1998; Schonenman, 1999; Sun, 1994; Werth,

1998; Westbrook, 1994; Zia, 1997).

3. Education (18) (Bachman & Panzarine, 1998; Brown, 1999; Carty & Rosenfeld,

1998; Cobb, 1999; Connolly, Huynh, & Gorney-Moreno, 1999; Corliss, 1994; Drury,

1997; Edel, 1998; Graveley, Lust, & Fullerton, 1999; Graves, Amos, Huether, Lange, &

Thompson, 1995; Hague & Gibson, 1998; E. Jones, 1992; D. Lewis, 1999; Mastrian &

McGonigle, 1997; T. Morris & McCain, 1998; Ribbons, 1998; C. Smith, Young-Cureton,

Hooper, & Deamer, 1998; Travis & Brennan, 1998);

4. Decision making and decision support (17) (Barton, 1993; Berner, Maisiak,

Cobbs, & Taunton, 1999; Botter, 1998; Breitfeld, Weisburd, Overhage, Sledge, &

Tierney, 1999; Bremner & Brannan, 2000; Gassert, 1988; Grand, 1997; B. Johnson,

1998; Kohli, 1994; Kolodner, 1991; Lamond, Crow, Chase, Doggen, & Swinkels, 1996;

V. Lewis, 1996; Nelson, 1995; Prin, 1996; Rossi, 1991; Ruland, 1999; von Tettenborn,

1990);

5. Patient care issues (15) (Bosque, 1995; Bowman, 1991; Caris-Verhallen et al.,

1997; Carlton, 1997; Clevenger, 1994; Downing, 1994; Ganong & Coleman, 1997;











Gassert, 1998; Gorman, 1995; Knestrick, 1999; Larrabee, 1999; Merrigan, 1998;

Nightingale Tracker Field Test Nurse Team, 1999; Schrot, Foulis, Morrison, & Farese,

1999; Woolery, 1992);

6. Nurse attitudes and characteristics (14) (Boldreghini & Larrabee, 1998; Burke,

1993; Cuyar, 1998; Gregor, Alm, Arnott, & Newell, 1999; Leavenworth, 1994; Ngin,

1993; Roberts, While, & Fitzpatrick, 1995; Rolfe, 1997; Simpson, 1997; L. Smith, 1998;

Staggers, 1991; Staggers & Kobus, 2000; Van Wynen, 1997; Weiner et al., 1999).

7. Computer hardware and software (8) (Duford, 1991; Sorensen, 1991; Carty, 1993;

Newton, 1993; Romano, 1993; Rose, 1993; Charters, 1998; Chocholik, Bouchard, Tan, &

Ostrow, 1999);

8. Nursing process (6) (Bakken, Cashen, Eneida, O'Brien, & Zieniewicz, 2000;

Boswell, 1995; Finfgeld, 1999; Henry, 1995a; Maher, 1992; Renfro, 1995);

9. Use of the internet (4) (Fitch, 1999; Gomez, DuBois, & King, 1998; Hovenga,

Hovel, Klotz, & Robins, 1997; Westberg & Miller, 1999);

10. Patient records (2) (Aronsky & Haug, 2000; Lauer, Joshi, & Browdy, 2000).

Summary of Chapter II

The review of literature encompassed the theoretical framework of chaos theory

and the concepts in the model being tested: decision making, creativity, experience,

leadership, education, risk, and informatics. Decision making approaches were found in

the two broad areas of descriptive/behavioral or prescriptive/normative. Factors affecting

decision making were the environment; information; heuristics; values; personal

characteristics (styles); and processes utilized.









67

Nursing studies of decision making processes arose in the early second half of the

20th Century and incorporated both quantitative and qualitative methods. Decision

making has not been a concept embraced by nursing theorists to describe the practice

environment. The process of decision making in nursing primarily followed the nursing

process, with allowances made for creativity and critical thinking. Information

processing, decision theory, and intuition were the major theories utilized in exploring

nursing decision making. Some effort has been made to distinguish individual, group and

professional levels of decision making in nursing.

Decision tools in the literature focused on attitudes and beliefs; perception;

collaboration or participation; quality indicators; or task analyses. Where reported,

reliability coefficients ranged from 0.74 to 0.90. Most scales were Likert-type.

The review of literature by the author showed that model concepts have limited

integration in nursing research. Creativity was usually visualized as a part of the nursing

or critical thinking process, with individual and organizational components. Creativity in

decision making was an attribute of nurse leaders sought by administrators but rarely

addressed in nursing research. Nursing education was grounded in fundamental ways of

thinking and in the critical thinking process. Leadership research concentrated on the

qualities and behaviors of the leader; styles of leadership, and leadership competencies.

The experience of the nurse in decision making has been found to be a useful

classification tool for education and professional practice. The perception of risk or

uncertainty in nursing situations was modified by information-seeking behaviors of the

nurse. The framework for informatics evolved around the interaction of systems; the











management of information; the classification (taxonomy) of nursing practice; and the

influence of technology.

Critique of the Review of Literature

No studies were found to have incorporated all of the concepts in the model being

tested. Most studies were limited to one or two of the concepts and, rarely, to three of the

concepts. No study utilized four or more of the seven concepts in the model. Study

designs incorporated both qualitative and quantitative methodologies, with quantitative

methods predominating. Samples were mostly non-random, convenience samples in

limited settings. Tools which have been developed were limited in scope to measuring a

single concept in the model. No tools were found which measured all seven concepts in

the model.

Most studies reviewed focused on individual achievement in decision making.

The main tendency was to evaluate individual responses or personal characteristics, rather

than to discover the conceptual framework within the decision environment.















CHAPTER 3
METHODS

Research Methodology

The research design for this study was quantitative, descriptive, and

non-experimental. A questionnaire consisting of demographic data and a bipolar scale on

decision making was distributed to a random sample of registered nurses licensed in

Florida. The data were analyzed using descriptive and inferential statistics. Correlations

and path analysis were performed to answer the research question about the relationships

among the study variables.

Specific Procedures

The study was approved by the Institutional Review Boards (IRB) at the

University of Florida and at the University of Tampa. Names and addresses of registered

nurses licensed in Florida were obtained from the Florida Department of Professional

Regulation. Participants were informed of consent procedures and had the right to refuse

participation in this study. No participant identifiers were included on the data collection

tool; thus, confidentiality of the data was maintained.

Research Population or Sample

A random sample was selected from a list of all 162,705 registered nurses

licensed in the State of Florida. Nurses who had a mailing address outside of Florida

were excluded from the study, reducing the population to 140,202. Using a computer









70

software program, a random sample of 5,000 registered nurses was selected to participate

in the study.

To estimate the sample size required for a covariance structure model, a procedure

based on goodness-of-fit (Hoelter, 1983) and modified by Matsueda and Bielby (1986)

was utilized as implemented in EX-SAMPLE (Brent, Mirielli, & Thompson, 2000).

Using this procedure, for an alpha of 0.05, six degrees of freedom, and an effect size of

0.20, the sample size required to detect this effect size was estimated at 151. A rule of

thumb (Bentler, 1989) recommendation that sample size be no lower than five times the

number of parameters (in this study, 81) indicated that the final sample must be more than

405 responses to achieve a power of 0.80.

Instrumentation

A tool developed by the researcher was used to collect data from the participants.

The first section was designed to collect demographic data about participant

characteristics, such as age, experience, and education. In the second section, the

investigator used a bipolar scale to evaluate the participant's degree of agreement with

words which reflect the model concepts. The word list of 272 terms was developed

through analysis of nurses' stories on decision making. The word list was reviewed by a

panel of experts for content validity and reduced to 91 word pairs. These word pairs were

examined for ability to distinguish concepts in the model, and a coding scheme was

developed.

The instrument was tested in a sample of 67 baccalaureate and master's students

at a local university. Inter-item reliability analysis yielded a Cronbach's alpha of 0.97 for









71

the full scale and a split-half reliability of 0.98. Word pairs were deleted if the total item

correlation was below 0.30 or above 0.80 and if a subsequent increase in alpha would be

detected if the item were deleted. Nine word pairs were deleted to form the final tool.

The revised tool of 81 word pairs had a Cronbach's alpha of 0.88 for the full scale and a

split-half reliability of 0.91.

Pilot Study

A pilot study of 67 undergraduate and graduate nursing students tested study

procedures, data gathering techniques, and analysis of data. Correlation analysis

demonstrated that all concepts in the model were significantly associated with the other

concepts (r = 0.44-0.97). Using analysis of variance, a significant relationship was found

on the effect of the six model concepts with decision making (F = 287.04, p = 0.000001).

The six model concepts were regressed on the dependent concept of decision making,

with a resulting adjusted R2 of 0.96. Changes were made to the tool and to the research

design and procedures. An over-identified path model was constructed for testing (see

Figure 3).

Data Collection

The researcher mailed the questionnaire to the selected participants. Nurses

completed and returned the questionnaire to the researcher with prepaid postage. Data

collection was estimated at 20-30 minutes for each participant; the data collection phase

was completed within one month of initiation of the study.




















V2
Creativity


S V3
P? Risk Taking

II:


I V4
P?, Informatics


P?

P?


/ ,


VI
Decision Making


P = Path Coefficient
VAR? = Variances
C?= Covriances
E, Error Term


Figure 3 Hypothesized (Over-Identified) Decision Making Model for Nursing


Leadership


V6
Education


Experience


P?
/ R











Treatment of the Data

The data were coded for computer entry. Subscores for each concept and a total

decision making score were calculated based on the scoring grid for the tool. Tests for

internal consistency and inter-rater reliability were repeated to confirm previous findings

about the tool.

Descriptive data were displayed in tables and analyzed using descriptive statistics.

Inferential statistics performed were correlation, regression, and path analyses to test the

decision making model as posed in the research question. Significance was set at the

0.05 level.

Summary of Chapter 3

Using a quantitative, non-experimental design and a random sample of registered

nurses, this research study tested a decision making model for nursing. The data

collection tool was constructed to collect demographic data on participants and used a

bipolar scale to measure relationships among the six concepts in the model with decision

making. Measures to protect anonymity and confidentiality were constructed, along with

review by several IRB's. Data were coded for computer entry. Descriptive and inferential

statistical tests included correlation and path analyses.















CHAPTER 4
RESULTS

Sample Characteristics

Of the 5,000 questionnaires distributed, 510 or 10.2% were returned. Of these, 19

were incomplete, leaving 491 usable questionnaires for a 9.8% return rate. The

respondents were 92.9% female (n = 456) and 7.1% male (n = 35) (see Table 1). Most

respondents (88.0%) were White, with the remainder African-American (5.5%), Hispanic

(3.5%), Asiatic-Pacific Islander (2.6%), and Native American (0.4%). The majority of

the respondents were married (66.6%); however, 31.1% of the respondents were single

status (single 13.0%; divorced 14.1%; widowed 2.4%; and separated 1.6%). Only 11 or

2.3% of the sample reported a partnership.

Hospital nursing represented the area of practice for the largest number of

respondents (57.0%). Home health had the second largest number of respondents with

8.0%, while clinic practice followed with 5.0% of respondents. Nursing home, public

health, and school settings each composed less than 5% of the reported areas of practice.

Critical care was the largest speciality area represented, with 21.0% of

respondents. Adult medical/surgical (15.1%), pediatrics (9.2%), women's health (7.1%),

geriatrics (5.9%) outpatient areas (4.7%), psychiatric/mental health (4.5%), and

rehabilitation (4.1%) were listed as speciality areas by the respondents; 28.5% of sample

listed other areas not defined on the questionnaire.











Table 1: Number and Frequency of the Variables Gender, Ethnicity, Marital
Status, Area of Practice, Specialty, Basic Nursing Education, Highest
Level of Education, Level of Practice, Employment Status and Current
Position for the Total Sample (N = 491)

Variable Number Percent

Gender

Female 456 92.9

Male 35 7.1

Ethnicity

White 432 88.0

African-American 27 5.5

Hispanic 17 3.5

Asiatic-Pacific Islander 13 2.6

Native American 2 0.4

Marital Status

Married 327 66.6

Divorced 69 14.1

Single 64 13.0

Widowed 12 2.4

Partnership 11 2.3

Separated 8 1.6

Area of Practice

Hospital 280 57.0

Home Health 39 8.0

Clinic 25 5.0

Nursing Home 18 3.7

Public Health 14 2.9

College/University 13 2.7

Elementary/High School 10 2.0

Other 92 18.7











Table 1 continued

Variable Number Percent

Specialty
Critical Care 103 21.0

Adult Medical/Surgical 74 15.0
Pediatrics 45 9.2

Women's Health 35 7.1

Geriatrics 29 5.9

Outpatient 23 4.7

Psychiatric/Mental Health 22 4.5
Rehabilitation 20 4.1

Other 140 28.5

Basic Nursing Education
Diploma/Associate Degree 300 61.1

Baccalaureate 191 38.9

Highest Level of Education
Diploma/Associate 209 42.6

BSN 125 25.5

Baccalaureate (Other Field) 46 9.4
MSN 59 12.0

Master's (Other Field) 68 7.7
Doctorate 11 2.2

Other 3 0.6

Level of Practice
Novice 6 1.2

Advanced Beginner 18 3.7

Competent 72 14.7
Proficient 182 37.0

Expert 213 43.4












































Basic nursing education was predominately at the diploma/associate degree

(61.1%), while the baccalaureate degree was the entry level for 38.9% of the respondents.

A large portion (42.6%) of the sample did not hold another degree above the

diploma/associate entry level. While 34.9% of the respondents stopped with a

baccalaureate degree in nursing (25.5%) or in another field (9.4%), 21.9% had achieved a

master's degree in nursing (12.0%), a master's degree in another field (7.7%), or a

doctoral degree (2.2%).

Respondents rated their level of practice using Benner's (1984) classification

scheme, with 1.2% at the novice level; 3.7% at the advanced beginner level; 14.7% at the


Table 1 continued

Variable Number Percent

Employment Status
Full Time 362 73.7

Part Time 81 16.5

Retired 24 4.9

Not Employed 24 4.9

Current Position
Staff 192 39.1
Administrator 56 11.4

Manager 54 11.1
Educator 37 7.5

Case Manager 35 7.1
Nurse Practitioner 25 5.1

Office/School/Industry 12 2.4

Private Duty 2 0.4
Other 78 15.9










competent level; 37.0% at the proficient level; and 43.4% at the expert level. Nearly

three-quarters of the sample were employed full time, with 16.5% in part time positions.

Less than 5% of the respondents were retired from full time positions; however, many

retired nurses indicated continued employment in part time positions. A small

component (4.9%) was not employed at the time of the survey.

Staff nurses comprised the largest segment (39.1%) of the sample. Administrators

(11.4%) and managers (11.1%) responded as well as educators (7.5%), case managers

(7.1%), nurse practitioners (5.1%), and office/school/industry nurses (2.4%). Private duty

nurses were only a small component of the respondents (0.4%), while a large segment (n

= 78 or 15.9%) listed other types of positions on the survey.

The average age of the respondents was 45.4 years, with a range from 23 to 78

years (see Table 2). Age was not significantly different between females (45.6 years) and

males (42.2 years) (F = 3.75, p = 0.053). Females tended to be in their current position

for a longer period than males (6.3 years vs. 3.7 years respectively; F = 5.38, p = 0.02).

Males had significantly fewer total years of nursing experience (14.7 years)

compared to females (19.8 years) (F = 7.13, p = 0.008). Other significant differences

were found between males and females in characteristics of marital status (X2= 19.54, p =

0.002) and practice areas (x2 = 20.94, p = 0.004).




































Instrument Testing

The instrument developed for this study, Analysis of Decision Making in Nursing,

was examined for reliability. Four word pairs were found with inter-item correlations

below 0.30 with a corresponding increase in alpha and were deleted from further

analyses; no items had correlations above 0.80. Cronbach's alpha for the full scale was

reported at 0.93, with split half reliability of 0.94.

Correlation analysis (see Table 3) demonstrated that all concepts in the model

were significantly associated with the other concepts. High correlations were found

between decision making and creativity (r = 0.60); decision making and informatics (r =

0.70); decision making and leadership (r = 0.76); decision making and education (r =

0.64); and experience (r = 0.71). Risk taking was weakly correlated with other model

variables (r = 0.10-0.31).


Table 2: Summary Measures of the Variables Age (in years), Length in Current
Position (in years), and Total Years of Nursing Experience for the Total
Sample (N = 491)

Variable Mean Min Max SD

Age 45.4 23 78 9.99
Female 45.6 23 78 9.97

Male 42.2 24 60 9.91

Length in Current Position 6.1 1 55 6.19
Female 6.3 1 55 6.32

Male 3.7 1 13 3.25

Total Years of Nursing Experience 19.4 1 55 10.85
Female 19.8 1 55 10.85

Male 14.7 2 39 9.79











Table 4 shows the summary measures for the model variables. The standard

deviation for the informatics variable was considerably larger than all the other

independent variables.

Path Analysis

Path analysis was performed to test the hypothesized theoretical model presented

in Figure 4. All analyses were conducted using the SAS System's CALIS procedure

(SAS, 1996). These analyses used the maximum likelihood method of parameter

estimation, and all analyses were performed on the variance-covariance matrix. The

direct, indirect and total effects of model variables on decision making were determined.

Four regression equations were constructed to represent model paths. First, decision

making was regressed on creativity, risk taking, informatics, leadership, education, and

experience. Second, creativity was regressed on leadership, education, and experience.

Third, risk taking was regressed on leadership, education, and experience. Finally,

informatics was regressed on leadership, education, and experience.

Four path coefficients were found to have t values less than the absolute value of

1.96 (p>0.05) and were removed from the model. All path coefficients in the reduced

model were statistically significant (p<0.05). (See Figure 5 for the reduced model.)

Adjusted R2 values were calculated for decision making (0.86), for informatics

(0.45), for creativity (0.20), and for risk taking (0.03). Covariances of the exogenous

variables were determined between leadership and education (0.41); between leadership

and experience (0.49); and between education and experience (0.54).













81







0






0 0


S. a



S 00

o o*




0 o o
o o0









o o C O
S0 000 (
Q o O O e m
<+- o O O O




S- 00 0











L Ma 0 oo C' 0 0 c


0







0 a a a





ua 3 3 '- O 5 V
.5 QS u p^ O O p-l p-l *







































The goodness of fit indices for the reduced model are presented in Table 5. These

indices provided a mixed review of the effectiveness of the model in explaining decision

making. The chi-square statistic included in this table tested the null hypothesis that the

reproduced covariance matrix has the specified model structure, or that the model fits the

data. Based upon the chi-square results the null hypothesis was rejected (X2 = 35.47, df=

7, p = 0.0001). Therefore, the chi-square test suggested that the model does not

adequately represent the decision making framework. However, Goodness of Fit Index

(GFI), the Normed Fit Index (NFI), and the Non-Normed Fit Index (NNFI), and the

Comparative Fit Index (CFI) values above 0.95 indicated a relatively high fit of the model

to the data. In addition a LaGrange multiplier test did not suggest further improvements

to the model which had not previously been considered or rejected.


Table 4: Summary Measures for Model Variables

Variable Mean Min Max SD

Decision 239.65 167 287 19.09
Making

Creativity 23.87 9 32 3.55


Risk 14.26 7 24 2.72
Taking

Informatics 102.74 72 126 9.27


Leadership 26.43 12 32 3.93


Education 26.88 19 32 2.35


Experience 30.25 15 36 3.14





















/. i
N / /


\\ / //
\ /


E2
VA




V2
Creativity


V3
Risk Taking


V4
Informatics


P?


P?
P? /


/ /

/ /P?
/ //

//

/

/


/


V1


Decision
Decision 'owi


P Path Coefficient
VAR? : Variances
C? CovEriances
E,, = Error Term


Figure 4 Hypothesized Decision Making Model for Nursing


V5
Leadership


V6
Education


V7
Experience


/

C?
/ /


\ ( C?

\ \.4


I


P? A


\P?

p?_
























V5
Leadership




41


V6
Education


054



\ V7
Experience


0.451
V2
Creativity

Adjusted R' = 0.20


V3
Risk Taking

0.33 Adjusted R2 0.03


\ E^'"--

E14

0.20


.- 0.29


S V4
Informatics

Adjusted R2 = 0.45


0.14
oz" i/


0 /
/ i
/ /


/ /2
/ /

/

/ 0.27
/


El
0,38




VI
Decision Making

Adjusted R2 0.86


Figure 5 Reduced Decision Making Model for Nursing


/


/49





0.49


\









85
The reduced model explained 86% of the variance in decision making. As shown

in Table 6, leadership provided the largest direct effect (0.33), indirect effect (0.19), and

total effect (0.52) on decision making. Experience (0.32), creativity, (0.24) and education

(0.24) followed in total effect on decision making. Risk taking (0.14) and informatics

(0.16) had the smallest direct effects on decision making.


Table 5: Covariance Structure Analysis: Maximum Likelihood Estimation
Goodness of Fit

Index Hypothesized Model Reduced Model

Chi-Square X2 = 27.51 X2 = 35.47
df= 3, p= 0.0013 df= 7, p= 0.0001


Comparative Fit Index 0.9858 0.9835
(CFI)

Goodness of Fit Index 0.9848 0.9803
(GFI)

GFI Adjusted for 0.8585 0.9211
Degrees of Freedom

Non-Normed Fit Index 0.9008 0.9506
(NNFI)

Normed Fit Index 0.9843 0.9797
(NFI)


Table 6: Direct, Indirect, and Total Effect on Decision Making

Variable Direct Indirect Total
Creativity 0.24 0.24
Risk Taking 0.14 0.14
Informatics 0.16 0.16
Leadership 0.33 0.19 0.52
Education 0.21 0.03 0.24
Experience 0.27 0.05 0.32















CHAPTER 5
DISCUSSION

Relating the Findings to the Original Conceptualization of the Problem

The purpose of this study was to determine the relationships among the concepts

of a decision making model for nursing: creativity, experience, leadership, education,

risk taking, and informatics. The population was registered nurses licensed in the State of

Florida. Path analysis was performed to answer the first research question: whether the

hypothesized model was adequate in explaining decision making in nursing. A reduced

model was produced. Although goodness-of-fit indices indicated mixed results, the

reduced model provided a minimally acceptable fit to the data using several criteria.

First, although the reduced model demonstrated a significant model chi-square

test, this statistic does not provide a valid test of model fit in most applied situations and

should be viewed more as a general goodness of fit index rather than as a statistical test

(Frees, 1996; Joreskog & Sorborn, 1989). In support of the reduced model, the NFI

exceeded 0.9 and the NFI and CFI were close to 1.0, indicating a good-to-superior fit

between model and data (Hatcher, 1998).

Second, the reduced model incorporated only one modification from the original

model, elimination of non-significant paths. The reduced model was more desirable in

that an increase in NNFI occurred through a relatively small number of modifications

(MacCallum et al., 1992) while CFI, GFI, and NFT remained stable.











Third, the reduced model remained clearly interpretable within the theoretical

framework of chaos theory. LaGrange multipliers did not propose any modifications

which were significant paths and which had not been previously considered in model

development (Hatcher, 1998; MacCallum et al., 1992).

Fourth, all of the path coefficients on the reduced model were significant. The

standardized path coefficients were not trivial in size, exceeding 0.05 in absolute value.

The R2 values for the endogenous variables of decision making and informatics were

substantial in size.

Therefore, the majority of goodness-of-fit indicators supported tentative

acceptance of the reduced model. Still, the reduced model was itself of admittedly

questionable validity, since the model was created from data-driven modifications made

to a rejected initial model and was based on a single sample of only moderate size. The

model may not generalize to other samples or to the population.

These findings are consistent with previous nursing research utilizing path

analysis to test model concepts. The adjusted R2 for the endogenous variables in this

study ranged from 0.03 to 0.86, while other nursing studies (Balneaves & Long, 1999;

Fox, 1997; Musil, Jones, & Warner, 1998; Resnick, Palmer, Jenkins, & Spellbring, 2000;

Richmond, 1997; Smyth & Yarandi, 1992; Trinkoff, Zhou, Storr, & Soeken, 2000) found

adjusted R2 values for the endogenous variables in a comparable range of 0.08 to 0.67.

Several of these studies (Fox, 1997; Resnick et al., 2000) reported an initial or final

model which had a significant value for the chi-square statistic, such as was found in this

study. Other indices (CFI, GFI, NFI, and NNFI) used in prior research to promote

acceptance of the final path models were the same ones used in this study.











Relationships among Model Concepts

The second research question explored the relationships among the concepts in

the decision making model. The reduced model explained 86% of the variance in

decision making. Leadership provided the largest total effect on decision making,

followed by experience, creativity and education. Risk taking and informatics had the

smallest direct effects on decision making.

All model variables were significantly correlated with each other. Moderate to

strong correlations were found between decision making and creativity, informatics,

leadership, education, and experience. Risking taking had only weak correlations with all

other model variables. Creativity and risk taking would have been expected to have an

inverse relationship; however, this study found a positive, albeit small, relationship.

Informatics was the only endogenous variable besides decision making which had

moderate correlations with the three exogenous variables in the model. Creativity was

moderately correlated with informatics and leadership.

No studies were found in the literature which combined all of the variables in the

decision making model tested in this study. Therefore, this study provided new

information about the relationship of decision making with creativity, education,

leadership, experience, risk taking, and informatics. The reduced model envisioned

decision making as a complex process, which may vary according to task and context

(Lauri & Salantera, 1998) and as a necessary skill for nursing leadership (Antrobus &

Kitson, 1999; Hemman, 1998; Krejci, 1999; Krejci & Malin, 1997; Porter-O'Grady,

1999).











The substantial contribution of leadership (directly and indirectly through

creativity) in the reduced model echoed the findings of other research studies on

leadership in nursing (J. Anderson, Rungtusanatham, Schroeder, & Devaraj, 1995; King,

2000; Krejci & Malin, 1997): decision making is the most predominant leader behavior.

Several leadership classifications (Antrobus & Kitson, 1999), such as developer of

nursing knowledge, process consultant, and user of information, may be visualized in the

reduced model pathways.

The relationships shown in the model between experience (tacit knowledge) and

decision making were reinforced through prior investigations by Anderson et al. (1995),

Fox (1997), Girot (2000), and Greenwood (2000). In those studies, tacit knowledge was

acquired by nurse leaders and staff nurses through mentoring and role modeling processes

in the work environment. The majority of decisions which nurses made were resolved by

managing tasks or others through tacit knowledge of the nursing environment. This tacit

knowledge combined with years of experience provided a foundation for information

seeking behaviors and for evaluation of consequences (Girot, 2000); both of these

behaviors are components of the informatics concept in the model. The direct and indirect

effects of education and experience on decision making underpin previous research which

found education and experience lead to more effective decision making (Girot, 2000) and

to a higher level of reasoning (Benner et al., 1996).

In the reduced model, creativity and risk taking presented a smaller (and

disappointing) role in decision making than the other model variables. Past research

(Anthony, 1999; Gilmartin, 1999; Porter-O'Grady, 1997a) has established that, as a group,

nurses reward conformity and punish creativity. Organizational constraints, workload




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs