J O U R N A L F O R N U R S E S I N S T A F F D E V E L O P M E N T • Volume 19, Number 3, 130–142 • © 2003 • Lippincott Williams & Wilkins, Inc.

U se of new nursing knowledge

has improved nursing practice and
patient care outcomes. However,
the busy bedside duties nurses
face constitute barriers to reading
and evaluating research. The
nurse may want to use new
knowledge about pain assessment,
wound care, preoperative fasting,
and depression screening to
improve practice but may lack the
skills and know how to evaluate
CRITICAL ANALYSIS OF RESEARCH
the research critically. Barriers to
analysis of the research include
PAPERS
confusing technical terminology,
statistics, and unclear research
reports. The nurse needs to
understand and examine the steps,
cohesiveness, merits, and
Sharon Valente, PhD, RN
weaknesses of the research to
assess his or her level of
confidence in the findings. This
article describes the criteria for
analysis of the sections of a
research report and illustrates
ways that sound research has
improved patient care.
R esearch in a clinical practice
profession enhances schol-
arly knowledge, improves
practice, and helps clinicians de-
liver care with quality, efficiency,
retention of nurses; research offers
educational perks that enhance
professional growth and develop-
ment. Nurses want to show admin-
istrators they are cutting-edge pro-
and effective outcomes (Dunning, fessionals with a practice based on
1999) Participation in research is the research evidence, but they
often a requirement for perfor- need to upgrade their competence
mance evaluations and promo- in evaluating research. However,
tions. Nurses who want to improve many nurses lack experience with
their care, show their effective research terminology, critique,
practices, and advance in nursing methods, and findings. Reading re-
must read and use research. Re- search requires learning a new lan-
search also provides evidence for guage because scientific terms and
effective practice and demon- concepts appear confusing, and
strates a need for a change in prac- nurses confess that research seems
tice (Collins & Robinson, 1996; dull, boring or “it’s all Greek to
Dunn, Crichton, Roe, Seers, & me”—particularly the statistics.
Williams, 1997). Nursing research This article offers strategies for
has led to significant improve- developing competence in read-
ments in nursing care (see Table ing, understanding, and critically
1). Research opportunities also im- analyzing the research. The focus
K E Y W O R D S
prove marketing, recruitment, and is on identifying nursing problems
ANALYSIS
RESEARCH CRITQUE
RESEARCH UTILIZATION
Sharon Valente, PhD, RN, is Assistant Professor at the University of Southern California, Los Angeles,
California.

130 May/June 2003


Research Findings That Changed Practice
Topic
Parents Visiting Hospitalized
Children
Pain Evaluation and Management
Sexuality in Patients With
Alzheimers Disease
Violence in Rape Trauma
and Serial Murders
Nurse Practitioners’ Care Delivery
Nurse Midwives’ and Residents’
Obstetric Care
and analyzing the problems, meth-
ods, data collection, and results.
One of the first challenges in
understanding research is learning
the language of science (see Table
2). Researchers use common
words, such as “significance,” but
they apply a very narrow defini-
tion (e.g., significance ⫽ a statistic
showing the probability of getting
JOURNAL FOR NURSES IN STAFF DEVELOPMENT
T A B L E 1
Summary
Pillitteri studied children’s response
to having parents visit compared
to a traditional limited visiting
schedule; more liberal visiting
policies improved childrens’
recovery.
Ferrell studied pain in oncology
patients and set standards for
improved treatment of pain.
Shapira studied women who
cared for their husbands who
had Alzheimer’s Disease and
identified the hypersexual behavior
of Alzheimer’s patients and
management strategies.
Burgess studied women who had
been raped and defined the rape
trauma syndrome and nursing
care needed. Her research
changed legal standards for a
woman’s defense. Her forensic
research on murders established
the profiling system for identifying
serial killers.
In 21 studies comparing primary
care by nurse practitioners with
MDs, there was no difference in
healthcare, outcomes, cost, and
quality. An NP clinic is more cost
effective than other clinics or an
MD on call. Patients were more
satisfied with NP’s education and
counseling.
In studies comparing the quality of
routine obstetrical care from nurse
midwives and residents, results show
major reductions in prematurity,
neonatal mortality, percentage of
low birthweight, and increases in
symptom-free babies. Nurse mid-
wives provide a cost savings.
these results by chance). Other
terms may be unfamiliar (e.g.,
quasi-experimental design). Learn-
ing the terms and specific word
definitions is an important step to-
ward comprehending research.
When reading research, nurses
need to be critical of the research
project and expect the author to
“show “each step clearly and log-
Citation
Seidl, F. W., & Pillitteri, A. (1967).
Development of an attitude scale
on parent participation. Nursing
Research, 16(1), 71–73.
Ferrell, B. R., & Rivera, L. M.
(1997). Cancer pain education
for patients. Seminars in On-
cology Nursing, 13(1), 42–48.
Roper, J. M., Shapira, J., &
Chang, B. L. (1991). Agitation in
demented patients. Journal of
Gerontological Nursing, 17(3),
17–21.
Burgess, A. W. (1985). Rape
trauma syndrome: A nursing
diagnosis. Occupational Health
Nursing, 33(8), 405–406.
Horrocks, S., Anderson, E., &
Salisbury C. (2002). Systematic
review of whether nurse
practitioners working in primary
care can provide equivalent
care to doctors. British Medical
Journal, 324(7341), 819–823.
Diers, D. (1981). Nurse midwifery
as a system of care. In L. Aiken
(Ed.), Health policy and nursing
practice (pp. 73–89). New York:
McGraw-Hill.
ically. Some novices hesitate to
critique a senior investigator be-
cause they respect the senior’s
knowledge or wisdom. However,
it is the researcher’s cohesive and
logical argument that requires cri-
tique. The nurse also questions
whether this study, its proce-
dures, and its findings would
apply to clinical practice. For ex-
131
 T A B L E 2
Frequently Used Research Terms and Definitions
Research Terms Definitions
Hypothesis A statement of anticipated relationship
between two or more variables that
may be tested empirically. Directional
hypotheses specify the direction of the
relationship (e.g., as exercise
increases, weight will decrease).
Research Question A statement of the topic under investigation.
Method The research steps, procedures, strategies,
and activities used to gather and analyze
the data in a research study.
Research Design The overall blueprint or plan for collecting
and analyzing data; it includes the para-
meters or specifications for increasing the
internal and external validity of the study.
Significance A statement of the importance and value
of the study to the profession, knowledge
development, research, and theory.
Significance (statistical) A statement of the probability that the results
could have occurred by chance.
Variable A characteristic or attribute of a person or
object that takes on different values within
the population under study (e.g., pain,
fatigue, depression, knowledge).
Independent Variable The variable (e.g., treatment or intervention)
that is manipulated by the researcher. In a
descriptive study, this may be the variable
that comes first.
Dependent Variable The outcome or result of the treatment or
intervention, also the criterion or outcome
variable.
Reliability of The extent to which an instrument is
Measurement Validity a consistent measure over time.
of Measurement The extent to which an instrument
accurately measures what it is
supposed to measure (e.g., a scale
accurately measures weight).
Conceptual Framework The researcher provides a rational
organization of how the variables in
the study relate to each other.
132
Example
A brief educational presentation about
advance directives will increase the
number of patients who complete
advance directives.
What are the factors that influence
completion of advance directive
documents by patients?
In a quasiexperimental study, a pre-
and postsurvey and retrospective chart
review will be used to document presence
of advance directives in a control
and experimental patient teaching group.
A quasiexperimental design is effective
for studying the impact of patient
education on patient’s knowledge and
completion of advance directives.
Advance directive education will
decrease unwanted, distressing, and
intrusive treatments, reduce family bur-
den in making choices for a loved one,
and improve patient’s autonomy at
end of life. Avoiding unwanted and
highly technical treatments is a cost
savings for families.
In a comparison of people receiving
a teaching intervention about advance
directives with a control group, the
patient teaching group had an average
of 1 advance directive on file as
compared with controls (x ⫽ 0.24).
Findings were significant at p ⫽ .01.
Knowledge and completion of advance
directives are the variables investigated
in this study.
Patient teaching about advance directives
is the intervention or the independent
variable in this study (or the cause).
Knowledge and completion of advance
directives are the dependent variables or
outcomes.
Blood pressure or temperature are
reliable if they are consistent. A pulse
oximeter is valid if it accurately
measures oxygenation.
Education and learning are
included in a model of behavior
change.
May/June 2003
ample, the landmark findings that
a daily aspirin reduced heart at-
tacks were conducted on men
and not women. A critical evalua-
tor would note that findings might
only be applied to the men. In a
landmark study, Tissot concluded
that masturbation leads to insan-
ity, although he limited his study
to psychiatric patients in an asy-
lum (Bullough, 1994). He made
an invalid assumption (e.g., mas-
turbation causes insanity) and
used a biased sample. Research
studies are rarely infallible; they
have limits and advantages. Find-
ings may not apply to all settings.
Common problems include the
lack of an ethnically diverse pop-
ulation, use of English speakers
only, and use of only college stu-
dents or volunteers. The re-
searcher’s rationale for the sample
selection should match the study
purpose and plan. For example, a
study using a veteran population
that was 98% male would be ap-
propriate if they are the clinical
population.
The purpose of the research cri-
tique is to evaluate the steps, co-
hesiveness, merits, and weak-
nesses of the study. The critique
should contain a statement of the
purpose, design, methods, and re-
sults of the research before a dis-
cussion of strengths and weak-
nesses. The goal is to develop a
balanced evaluation of the study
and its contribution to scholarly
knowledge and practice.
Evaluation Criteria
The researcher’s qualifications to
conduct this type of research may
include expertise in the topic, ed-
ucational degrees, credentials,
and previous publications. Con-
sultants often have expertise in
such specialized areas as statisti-
cal analysis or cultural issues. Re-
search teams studying pain and
JOURNAL FOR NURSES IN STAFF DEVELOPMENT
Participation in research is
often a requirement for
performance evaluations and
promotions.
pharmacological treatments should
have expertise in pain manage-
ment and pharmacology. A study
on nursing care should involve in-
vestigators with nursing expertise.
An article may include a brief bi-
ography of authors or may pro-
vide scant detail about qualifica-
tions. The reviewer looks for
evidence of the researcher’s com-
petence throughout the study and
for accuracy or inconsistencies in
the report.
The title, abstract, and report
should be clear and accurately de-
pict the problem under study. A
poorly worded title makes it diffi-
cult for scholars to locate the
topic in a computerized index,
confuses readers, and the result is
the article is indexed in an obtuse
way. For example, a computer-
ized literature search for articles
on suicide among children pro-
duced an amazing array of articles
on “children” who were adults
age 20 to age 55 and nothing on
suicide among those under age
12. In the articles from the litera-
ture search, suicide occurred dur-
ing adulthood, but the titles may
have reflected the bereaved par-
ents’ perspective that “my child
committed suicide.” The reader
who looked for an article on sui-
cide among children would be
surprised to discover the articles
were about “adult children” be-
cause the title suggested other-
wise. A brief summary of the sec-
tions of a research report is found
in Table 3.
Categories of a Research
Critique
Although these sections are stan-
dard, authors can organize and
subtitle them in diverse ways. The
reader looks for the information in
these categories when the re-
searcher leaves the section unti-
tled. The background may also be
called introduction or may lack a
subtitle. Results may be labeled
findings. The author may expand
the detail of some topics more
than others. A research article will
contain more detail on methods
and procedures and analysis,
while a clinical journal will proba-
bly explain the research questions
and illustrate how the results can
be applied in practice. Reading the
original article or primary source
rather than relying on an abstract
or another author’s review of liter-
ature or summary of existing liter-
ature is essential for an accurate
review.
Background of the
Study/Introduction
This section usually follows a brief
abstract and contains the review of
the literature. The authors typically
state the problem clearly, define
their terms, and describe the ques-
tion under investigation. They il-
lustrate the importance of the re-
search problem or topic to
knowledge development, practice,
or research.
The research problem identifies
the topic under investigation that
may arise from clinical practice,
the research literature, or theory.
The reader expects a clear depic-
tion of what the study is about,
why it was done, and how it was
carried out (Fleming & Hayter,
1974, 1975). A nurse may test a
theory of pain or complicated grief
or apply a nursing theory such as
Orem’s Self-Care theory or Roy’s
133
 T A B L E 3
Overview of Sections of a Research Report
A determination of the merits and worth of the study reflects several dimensions,
and a research report typically contains these sections.
Background of the Study—explains the problem under investigation within the
context of the current research. The gaps in the published literature provide a
foundation for the study. The discussion introduces and defines the major
variables to be studied, the theory or framework for understanding relationships
among variables, and the question to be studied. If the researcher investigated
pain, the background would introduce pain and its aspects under study and
explain how the proposed study will answer important questions and contribute
to knowledge and practice.
Method—describes the design, sample, data collection procedures, and analysis. If
the study involves tools or instruments, an explanation of their consistency or
reliability and accuracy is included.
Results—capture the findings of the study for each research question or hypothesis.
Discussion—offers an interpretation of the results and compares them with other
research findings and their relationship to practice, theory, policy, and
research.
Implications and Application—examines ways that the findings of the study can be
applied to practice and the implications of the findings for further research.

Adaptation Theory in the research
study. Theory-based research
makes a substantive contribution
to knowledge development. A
nurse midwife might study pain re-
lief and want to compare the gold
standard of injectable lidocaine
with a topical anesthetic. The
problem should be clear, research-
able, and significant.
The review of the literature may
be untitled but it presents the most
current research and theory. Typi-
cally, most literature is recent, usu-
ally from the last 5 years, and cov-
ers the most important studies. In
a historical study or a study with
little recent research, older studies
are necessary. If most of the refer-
ences are over 10 years old and
the problem (e.g., teen pregnancy)
is a frequently researched topic,
the reader wonders why newer in-
formation is absent. Sometimes the
literature is briefly summarized.
The goal is to explain the gaps in
the literature and show how this
proposed study will expand
knowledge. The definition of the
research topic or problem sets the
stage for the design and methods.
The literature review explains a ra-
tionale for the choice of the most
effective design, methods, and in-
struments. An outdated or skimpy
literature review raises questions
about whether the researcher
knew the current scope of the
problem. However, common sense
prevails. For example, if the author
reviews a topic that was last stud-
ied 30 years ago (e.g., anthrax),
we would expect articles from 30
years ago. The reader who is well
versed in the topic may recognize
that some classic or landmark
studies or pertinent results were
not cited. Otherwise, the reader
asks if the problem, variables, and
background appeared organized,
logical, and cohesive.
Significance
Although it is often untitled, this
section should provide a clear, or-
ganized, and cohesive argument
for how the problem is significant
to nursing and patient care. A
study may be important or signifi-
cant because the problem (e.g.,
pain, depression, infection, or sui-
cide) leads to substantial morbidity
or mortality. Solid arguments for
significance include cost savings,
quality care delivery, prevention of
disease or death, knowledge de-
velopment, and improved safety
(e.g., prevent medication errors),
decreased length of stay, and in-
creased customer satisfaction. Sig-
nificance is not established by say-
ing the problem is interesting. For
instance, a student wanted to
study which types of nursing posi-
tions had the highest salary. She
stated her rationale for the study
was to expand her personal
knowledge. She missed the
point—the significance should be
to the profession or practice. For
example, the nurse who reads a
report on bicycle helmet safety,
smoking cessation, or foot fetishes
should not have to guess how this
relates to nursing or how the study
can enhance patient care.

134 May/June 2003

Research Questions and
Hypotheses Guide the Study
A research question occurs fre-
quently in descriptive studies or
when scant research exists and
does not clarify the important vari-
ables. Sometimes the research
question is hidden within the pur-
pose of the study. If the literature
identifies the variables, then the
researcher may pose a hypothesis
or a best estimate about how two
or more variables correlate. For
example, the concept of compli-
cated grief is poorly understood so
a research question might be,
“What are the psychological and
physical characteristics of people
with complicated grief?” If, how-
ever, two types of face lift (or
childbirth method) are both well
respected but we do not know
which is most effective, the hy-
pothesis might be “Patients who
receive face lift A versus B will
have greater satisfaction after
surgery, less discomfort and
swelling, and faster healing.” Of
course the type of face lift would
be precisely described.
Theory or Theoretical
Framework
Some researchers use a theory or
theoretical framework for their re-
search and relate all the variables,
measures, and hypotheses to this
framework—often nursing theory
such as Roy’s theory of adaptation
in patients recovering from stroke
or Orem’s Self-Care theory in can-
cer patients is used. A research
study does not necessarily need a
theory, but many studies benefit
from a framework illustrating how
the variables relate to each other.
Alternatively, some research stud-
ies are atheoretical and gather de-
scriptive data in order to establish
a theory to explain the problem.
For example, Saunders studied
widow bereavement after modes
JOURNAL FOR NURSES IN STAFF DEVELOPMENT
When reading research,
nurses need to be critical of
the research project and
expect the author to “show “
each step clearly and
logically.
of death. Her study did not apply
or relate to a nursing theory (Saun-
ders, 1982). If the researcher is
studying the relationship among
variables, a theoretical framework
showing how the variables relate
is useful (e.g., the relationship of
exercise and metabolism to obe-
sity). Alternatively, the researcher
may describe the relationship
among variables using a concep-
tual framework, a set of abstract
concepts that broadly explains the
phenomena of interest and that il-
lustrates or maps how variables re-
late. A researcher might apply
models including Peplau’s Inter-
personal Relations in Nursing and
Orem’s Self-Care Concepts of Prac-
tice. In a study of how exercise re-
lates to wound healing or immune
function, a conceptual framework
shows the relationship among
variables (e.g., exercise, healing,
immune function).
The background sets the stage
for whether this will be a histori-
cal, naturalistic, or descriptive
study, a replication of a prior
study, or a quantitative study.
Method: How the Study
Was Designed and
Carried Out
This section describes the blue-
print or design for the study and
the procedures for sample selec-
tion, data collection, and analysis
(see Table 4). The authors explain
the population and the selection
criteria for the people or informa-
tion that will be studied. The re-
search design should follow logi-
cally from the background and be
clear and appropriate for the prob-
lem. Each design has strengths and
weaknesses that should be de-
scribed. The researchers should
provide a rationale for the selec-
tion of the design. Descriptive
studies are useful if little is known
about a subject or if qualitative ap-
proaches are the effective way to
examine experiences. For an ex-
perimental design, the reader asks
if the potential influences on the
results are well controlled. If a
control group is used, is it the ap-
propriate control?
Some designs use meta-analy-
sis, a technique to re-examine and
analyze data from prior research
studies to determine the results
over many studies. Other studies
use a consensus approach to re-
viewing the literature and summa-
rizing the conclusions. In each
study, the problem, procedures,
and design are evaluated for
strengths and weaknesses. The
reader asks, “Is this the best de-
sign for the study; does it allow
enough control and flexibility; and
what are the limits of this design
and the effectiveness of execution
of the design?” Other questions in-
clude, “How will you get your
sample; what will you use to se-
lect the sample; what are the cri-
teria for including someone in the
sample?” The issues of control of
outside influences (extraneous
variables), sampling, measure-
ment, analysis, and ethical issues
require discussion. Some authors
with multiple publications will cite
an earlier publication that details
the method or instruments. This
can be frustrating and irritating to
the reader who is trying to critique
a study.
In general, designs tend to be
naturalistic or qualitative or con-
trolled (quantitative) and some re-
135
 T A B L E 4
Types of Design
Design Characteristics of Study Illustration

Descriptive Research/ To depict the characteristics of individuals, A descriptive study of health

Nonexperimental groups, or phenomena of interest
Qualitative Design A naturalistic approach to gathering,
organizing, and interpreting non-
numerical information to understand
a person’s or group’s experience.
Quasiexperimental To examine the results of an experimental
intervention without meeting the criteria
of a true experiment (e.g., subjects cannot
be randomly assigned or controlled)
Experimental The investigator aims to test an intervention
or treatment with rigorous control by
randomly selecting subjects and controlling the
independent variable.
Clinical Trials Describing phenomena by manipulating
numerical data to determine the extent
and degree of relationships among the
variables
Case Study A research method that uses a thorough,
in-depth analysis of an individual, or group.
Retrospective To link a current problem with earlier
influences on the problem (e.g., a study
of medical records of suicides after their
deaths)
Meta-analysis A statistical approach to reviewing
several research studies, combining
their results, and analyzing the results
of multiple studies on a topic
seeking behavior among
college students who have
had unprotected sex
An observational study of the use
of restraints on inpatients; a
field study; narratives of women
who report their reactions to rape
A study of the impact of music
and foot massage on reducing
nausea and pain (no control
group was used)
Randomized clinical trials of
two different drug treatments
for obesity
Statistical analysis of randomized
drug trials
A case study of a nurse’s
response to the suicide of
a patient
A survey of people with HIV about
their earlier behaviors (e.g., safe
sex, number of partners, blood
transfusion, IV drug use); a
chart review of antecedents of
completed suicide
A meta-analysis of critical
incident debriefing studies

searchers use a combination. The
naturalistic study investigates the
topic in its natural setting without
controls. Anthropologists, social
scientists, and nurses often use this
method of investigating a person’s
experiences or lifestyle. The quan-
titative study creates a simple and
very controlled situation and mea-
sures a precise aspect of the situa-
tion. A good example is a clinical
drug trial. More recently, studies
have used both quantitative and
qualitative approaches to obtain a
broader perspective of the prob-
lem. Types of research design are
listed in Table 4.
Sampling is often an untitled
section that refers to the relation-
ship between the individuals,
group, or items to be studied and
the larger population they repre-
sent. The authors explain the crite-
ria for the sample and who will be
included and how they will be lo-
cated. The reader questions
whether the sample will allow the
researcher to answer the research
question or test the hypotheses.
Another question is whether the
sample is representative of the
larger population of all people
with the problem and whether any
bias exists that may influence the
sample. Selecting only students
who received an A grade to evalu-
ate the effectiveness of an instruc-
tor’s teaching abilities would likely
be a biased sample. If the goal of
the study is to generalize the find-
ings from a small group of people
to the larger population, then
quantitative researchers prefer ran-
dom sampling. Random sampling
gives everyone in the population
of interest, or target population, an
equal chance of participating. If
we studied breast cancer in
women, and wanted a random
sample, every woman of diverse
age or ethnicity would need a

136 May/June 2003

chance to participate. Often, the
researcher uses a convenience
sample of those who volunteer
from a healthcare setting. If the re-
searcher only studies women over
age 18 who speak English, the re-
sults cannot be generalized to non-
English speaking or younger
women. Although quantitative re-
searchers prefer random sampling,
it is not always the most appropri-
ate or feasible approach. Alterna-
tive sampling strategies exist. For
instance, when one studies les-
bians or a hidden group where the
total number of the population is
not known, one might use net-
work sampling. Network sampling
may involve distributing flyers
inviting people to participate in a
study and asking those who see
the flyer to give it to their associ-
ates.
Next, the reader considers the
sample size of the study. In a
quantitative study, generally a
large sample is preferred. Al-
though a survey of 500 or 1,000
nurses provides a broader view of
different nurses’ responses than a
study of 12, the qualitative study of
12 might provide more in-depth
information. However, some types
of naturalistic or qualitative re-
search are designed to study small
groups. The question is whether
the sample will provide the suffi-
cient data. Researchers can use a
large database, such as the na-
tional crime statistics or national
suicide statistics to study a large
population, a statistically less com-
mon problem or national trends in
healthcare data. These large
datasets help answer questions
about healthcare trends such as,
“Are more C-sections performed in
certain parts of the country?” or
“How do patients rate their health
care from and satisfaction with
nurse practitioners versus physi-
cians?” Analyzing an existing set of
data is an intriguing opportunity.
On one hand, analysis is limited to
the original data collected for the
JOURNAL FOR NURSES IN STAFF DEVELOPMENT
Data collection measures are
a “yardstick” that needs to
measure consistently.
survey’s specific goals, but it pro-
vides a national and large dataset
without the challenges of data col-
lection. For example, the
health/weight standards were de-
veloped from insurance company
statistics collected from mainly
white males who had life insur-
ance policies. These gold stan-
dards were generalized to many
diverse populations. Although so-
ciologists and public health and
health services researchers use
large and previously collected
datasets, few nurses have done so.
Replicating a previous study and
analysis of existing datasets also
offers useful and efficient ap-
proaches for nursing research.
Consistently throughout an ap-
plied research article the nurse
asks, “Will this study gather useful
information about a problem that is
important to nursing?” and “Will it
improve nursing practice, research,
or theory?” Applied research is
practical research and is a scientific
study that aims to generate knowl-
edge that can apply to practice. If
the researcher neglects to provide
enough information about the sam-
ple, this is a weakness.
Data Collection Procedures
The procedures for gathering the
data may include surveys, observa-
tion, questionnaires, biophysio-
logic measures (e.g., blood pres-
sure, lab assays), interviews, and
standardized tests. The reader
wants to know when, where, how,
and how often these data will be
collected and the accuracy and re-
liability of data collection. For ex-
ample, if researchers survey pa-
tient satisfaction with nursing care,
they should explain the accuracy
and reliability of the measurement.
Data collection measures are a
“yardstick” that needs to measure
consistently. If the yardstick were
not tested or calibrated, it might
have the properties of a rubber
band and measure haphazardly.
Another issue is the compe-
tence, training, and appropriate-
ness of the data collectors. The re-
searcher needs to detail their
training and qualifications to show
that the data collectors were care-
ful, accurate, knowledgeable, and
trustworthy. In one study, I care-
fully trained the students in gather-
ing data from the chart records
and had the students rate their per-
cent agreement on 33% of charts.
All went well until I verified one
student’s data and discovered that
she was “guessing” at the patient’s
income—she recorded that a pa-
tient on state disability earned
$40,000 yearly. Her unreliable data
were discarded.
A description of the instruments
in the section on Method(s),
should include their reliability or
consistency, validity, and trustwor-
thiness. Often the measures are
pain or symptom rating scales, at-
titude, teaching, or patient satisfac-
tion measures or knowledge rating
scales. The researcher needs to de-
scribe each instrument, its use in
this population, and its statistics
for reliability and validity (accu-
racy to measure what it says it
measures). Failure to do so is a
weakness. The reader is not ex-
pected to “guess” how well the
measure worked or its properties.
Sometimes the wording of a ques-
tion or the order of research in-
struments makes a big difference.
This is true in research in different
language groups. For example,
137
one researcher reports translating
an English language survey of
young women’s knowledge and
use of contraceptives into another
language. The word “condom”
was translated as “a pocket of
love,” which was an idiosyncratic
term that no one understood and
would have invalidated the ques-
tion if not corrected. Procedures
for translation and reverse transla-
tion typically solve this problem. A
pilot study testing the instruments
and measures helps solve some of
these dilemmas. The researcher
needs to acknowledge the limits of
the design.
Analysis
Although some readers are
tempted to skip statistical analysis
and go straight to the conclusions,
this section is important. The re-
searcher explains the procedures
used to analyze the data for each
research question or hypothesis.
These analytical procedures may
be descriptive or inferential. De-
scriptive data (e.g., frequencies,
percentiles, totals, averages, range
of scores and their distribution)
help depict the characteristics of
the sample and count narrative re-
sponses. These numbers may be
displayed in a table (e.g., educa-
tion, age, ethnicity, income, reli-
gion, and occupation of the sam-
ple). Descriptive statistics do not
inform the reader about any cause
or effect relationships, but they il-
lustrate qualities or characteristics
of a person or phenomenon. Table
5 provides a brief summary of
common statistical tests.
Inferential statistics allow the re-
searcher to draw cause and effect
conclusions and to relate the sam-
ple findings to the target popula-
tion. These statistics reflect the
probability that the results in the
study occurred by chance. As the
risk of chance diminishes, confi-
dence in the results increases. The
138
Critical evaluation of research
is a skill that requires time to
develop and refine.
researcher presents the statistics
that support but do not prove the
hypothesis. When the researcher
sets the limit for the risk of results
occurring by chance, this is called
the level of significance (p ⫽ .05 or
p ⫽ .01 are common). This num-
ber and the concept of statistical
significance can be confusing. The
simple way to remember its mean-
ing is to subtract the number (ig-
nore the decimal) from 100—the
result is 95—or the researcher can
be 95% confident that the results
are a true finding and not an error.
This number identifies the risk the
researcher is willing to take that
the null hypothesis (e.g., no differ-
ence between groups) will be re-
jected when it actually was true. If
the researcher wants to know if
knowledge improves after a lec-
ture, 95% confidence is acceptable.
If, however, the topic under inves-
tigation is life threatening such as
the evaluation of suicide risk, the
researcher will want to be 99%
confident or have a higher signifi-
cance (p ⫽ .01 level).
In experiments, the researcher
provides a formula for statistical
analysis of the data for each hy-
pothesis. Often a statistical consul-
tant helps define the analytic tests.
To compare the average scores of
two groups of subjects to deter-
mine if a difference existed on test
scores ranging from 0–100, the re-
searcher uses a t-test. In a qualita-
tive research study, the researcher
illustrates how the interviews or
narratives or observations will be
gathered, recorded, and then inter-
preted.
Protection of Human Subjects
Recently, researchers have placed
more emphasis on documenting
the risks and benefits, harms, and
safeguards for human subjects. Al-
most all academic and healthcare
institutions require that all research
on living beings be reviewed for
protection for living subjects, both
human and animal. The researcher
should show the precautions for
protecting human subjects and
meeting ethical research standards.
Often the researcher notes that the
study received Institutional Review
Board (IRB) approval and that the
subjects provided informed con-
sent. The reader needs to ask if ad-
equate procedures were followed
to ensure that the subjects under-
stood and consented to the study.
Researchers should provide a ra-
tionale if women and minorities
were excluded and should explain
how they protected the rights and
safety of any vulnerable subjects
(e.g., children, cognitively im-
paired persons, prisoners, elderly
persons, or stigmatized group).
The researcher needs to make sure
that a subject can stop participat-
ing in the study at any time, and
that explanations of the study are
provided in lay language (e.g., we
will take 3 teaspoons of blood—
not 18 ml of blood).
Results
The researcher repeats the hy-
potheses or research questions be-
fore stating the data analysis and
findings. Commonly, the results
will include a description of the de-
mographics of the sample. Next,
the statistical analysis will be re-
ported. However, in a qualitative
study, the results may not be statis-
May/June 2003

Categories
Univariate Descriptive Statistics
Inferential Statistics
Chi Square Test of
Independence
T-Tests
Pearson’s Correlation
Analysis of Variance (ANOVA)
Probabilities
Level of Significance
Degrees of Freedom
Power Analysis and Effect Size
All statistical tests specify the rules that must be met before the test is valid. For example, the type of distribution and the level
of measurement.
tical. For example, in a study of in-
terventions to reduce fatigue, the
findings might include themes or
narratives describing the signs,
symptoms, and patients’ reports of
their fatigue and what reduced fa-
tigue. In a study of oncology
nurses, researchers expected that
knowledge and attitudes toward
suicide would predict the effective-
JOURNAL FOR NURSES IN STAFF DEVELOPMENT
T A B L E 5
Common Statistical Tests
Illustrations
Measures of central tendency
(averages)
Frequencies
Percentiles
Range
Test hypotheses
Tests whether two variables being
examined are independent or related
Measure the difference between two
group means; describe how different
the two groups are on a specific
characteristic; parametric test used to
analyze interval data for independent
groups
Common test used to measure the
extent of linear relationship
between two variables
Tests for differences among three or
more means
Probability measure; identifies the
probability of predicting an event
without error and the risk of error;
risk of making a Type I error (rejecting
the null hypothesis when it is true)
Freedom of the score’s value to vary
given the other existing scores
Statistical analysis can indicate how
many subjects are needed to test a
hypothesis at a given level of power
(e.g., how certain you want to be).
ness of patient care (Valente &
Saunders, 2000). Knowledge and
attitudes were weak predictors and
more research to explain effective
care was needed. Hence, negative
or weak findings can be important
(e.g., If the cause is not knowledge
and attitudes, then education is not
the solution.) Tables, graphs, and
pictures help display the results.
Description and Limits
Describe the sample or characteristics
of the sample; frequencies are often
used in qualitative and quantitative
studies to count responses
May allow generalization of findings
from a random sample to a population
Can use if there is one entry of data
per subject in the sample—not if the
subjects have repeated measures
Must meet criteria for t-test that
includes the type of data you have
and a normal distribution of
data
0 shows no relationship,
.1 – .3 shows a weak relationship,
and .3 – .5 a moderate and above
.5 a strong relationship; a positive
number shows the increase of one
variable when the other goes up
(e.g., determines if outcomes of three
interventions were different)
p ⫽ .05
df ⫽ 10
A sample of 200 will be needed to
test the hypotheses at 80% level.
This section should omit any dis-
cussion or interpretation of the
findings.
Discussion
The authors compare and contrast
their results with other studies and
interpret the findings. They may
139
 T A B L E 6
Guidelines for Critique
Topic Sample Criteria for Critique
Problem and Purpose • Is the problem clear, precise, and
well defined?
• Does it match the purpose?
• Is it current and important to nursing?
• Is the problem appropriate and feasible
for research?
• What is the topic under investigation?
Significance • Is a good argument made for the
importance of the problem to clinical
practice, research, theory, or knowledge
and policy development?
Literature Review • Citations include relevant and current
research.
• Current knowledge and gaps in
knowledge of the problem are explained.
• Key literature is cited accurately.
• Citations are complete.
Human Subjects • Are the rights of human subjects
protected? Has the study been approved
by an appropriate IRB? Can the subjects
give consent and understand the study?
• Are minorities or women excluded
without a solid rationale?
Method • Does the research design match the
problem and appear appropriate?
Would another design have been better?
What are the limits of the design?
Interventions • Interventions and controls are described
in detail.
Sample • Target population is clear, described,
and appropriate.
• Eligibility or exclusion criteria, recruitment,
and sampling procedures are clear.
• Sample size is appropriate.
• Did any of the sample drop out and
why?
• Were there any biases?
• Was a convenience or random sample
used?
Instruments • The reliability and validity of instruments
are cited; the rationale for these
instruments is clear and convincing.
• Limits to the instruments are clear
(e.g., will the instruments gather
140
Written Critique (Sample)
The authors clearly defined an
important and underexamined topic
for research.
The problem, low use of advance
directives, is a costly patient care
problem that places undue burden on
patients, families, and nursing staff.
Understanding use of advance
directives is necessary to improve
clinical practice and research, or
authors fail to explain the
significance.
The literature is over 10 years old
and outdated; few references are
from healthcare journals. Many
citations are country or state reports
or Internet sites that may not be
available over time.
The authors clearly described
sound procedures for protection of
human subjects and informed
consent.
Women and minorities are included.
Research design, problem, and
methods are sound and well
matched. Rationale explains
selection of these approaches.
This is a descriptive study; no
intervention.
The researchers used a convenience
sample of psychiatric patients at an
urban clinic. Rationale for lack of a
random sample from diverse settings
is absent. Rationale for using English
speakers is absent.
The Beck Depression Inventory is a
well-known instrument. Rationale and
psychometrics for use in primary care
is absent. Argument is lacking for why
this is the best instrument.
(Continues)
May/June 2003
 T A B L E 6
Guidelines for Critique (Continued)
Topic Sample Criteria for Critique Written Critique (Sample)

the data needed; will there be

problems with poor recall, lying
about sensitive information?)
Data Collection • Data collection procedures are clear. Data collection procedures are
• How many times were data collected? unclear. Description is lacking for
• Did the researcher control external frequency of data collection, controls,
influences? and training of data collectors.
• Will the data answer the question?
• Data collectors are trained, reliable,
and accurate
• Circumstances of data collection
were conducive to data collection
(e.g., confidentiality).
• Did the researchers minimize any risks,
stress, distress, or burden to subjects?
Data Analysis • Procedures match the research question, Descriptive statistics describe the
design, and data collection. sample. t tests match hypotheses.
• Statistical tests match the data and However, how error will be managed
hypothesis. with multiple t tests is unclear.
Results • Results are clear, complete, and Results are complete and match
identified; they answer the research research questions. Significance is
questions. stated.
Discussion • Results are compared and contrasted Discussion is absent.
with other research findings.
Conclusions/Implications • Conclusions accurately reflect the data. Conclusions accurately reflect the
• Implications for practice, research, and data. Implications for practice,
theory are discussed. research, and knowledge are absent.
• No unwarranted conclusions or
interpretations are made

confirm other researchers’ findings
or contradict them and search for
an explanation for the contradic-
tions. If these study findings con-
tradict earlier results, the re-
searchers discuss this. The
researcher can also examine trends
in the data that were promising but
did not reach significance and
speculate why the trends were not
significant. The limits of the study
are outlined. The researcher may
also discuss problems encountered
during the research such as the dif-
ficulty obtaining an ethnically di-
verse sample despite intensive ef-
forts to do so. If a natural disaster
occurs during a study on depres-
sion, the authors need to discuss
the impact of the disaster on their
findings.
Conclusions/Implications
The researcher summarizes the
major findings and conclusions
and their application to practice,
research, theory, and knowledge
development. Implications for fur-
ther research are mentioned. Cau-
tion in generalizing the findings
beyond the data is emphasized.
The reader asks whether the con-
clusions represent the data accu-
rately. Often even the most senior
researchers can draw a conclusion
that is not based on the data even
though the conclusion may appear
plausible. For example, if the re-
searcher studies men and con-
cludes that aspirin reduces risk of
stroke, the researcher cannot con-
clude that this also is true for
women (who were not studied).
The implications need to describe
how the research findings can be
applied to nursing practice, can
suggest future research studies,
and can improve theory and
knowledge.
Summary
Critical evaluation of research is a
skill that requires time to develop
and refine. In addition, some re-
search reports simply provide less
than ideal information for the cri-
tique. If the researcher does not
explain the sample criteria, ratio-
nale for sample selection or some
other aspect, the reader cites this

JOURNAL FOR NURSES IN STAFF DEVELOPMENT 141

as a flaw. It may not be possible to
determine if this is just a defect in
the written report or a flaw in the
research design or implementa-
tion. Sometimes, the publisher’s
guidelines and space limit what is
put in print. Some aspects of the
critique rely on the reader’s judg-
ment, experience, and common
sense. A critique should contain a
systematic assessment, be unbi-
ased, and identify the strengths
and weaknesses of the study (see
Table 6). The critique can identify
whether the research report con-
tained sufficient data for critique,
was well written, well organized,
clear, concise, and lacked jargon.
Current standards also require that
authors avoid sexist language, pro-
vide citations for strong state-
ments, offer complete and accu-
rate statistics, and write a concise,
logical, and orderly report.
ACKNOWLEDGMENTS
This material is the result of work
supported with the resources and
use of the facilities at the Depart-
ment of Veterans Affairs, Greater
Los Angeles Healthcare System.
142
Appreciation is extended to Sandi
Strome and Ahnnya Slaughter for
review and editing suggestions.
REFERENCES
Bullough, V. L. (1994). Science in the
bedroom. New York: Basic Books.
Burgess, A.W. (1985). Rape trauma
syndrome: A nursing diagnosis. Occu-
pational Health Nursing, 33(8),
405–406.
Collins, E. G., & Robbinson, D. (1996).
Bridging the research-practice gap.
Nursing Standard, 10, 25.
Diers, D. (1981). Nurse midwifery as a
system of care. In L. Aiken (Ed.),
Health policy and nursing practice
(pp. 73–89). New York: McGraw-Hill.
Dunning, M. (1999). Experience, evi-
dence, and everyday practice. London,
Kings Fund.
Dunn, V., Crichton, M., Roe, B., Seers,
K., & Williams, K. (1997). Using re-
search for practice. Journal of Ad-
vanced Nursing, 26, 1203–1210.
Ferrell, B. R., & Rivera, L. M. (1997).
Cancer pain education for patients.
Seminars in Oncology Nursing, 13(1),
42–48.
Fleming, J. W., & Hayter, J. (1975).
Reading research critically. Kango Ky-
oiku— Japanese Journal of Nurses Ed-
ucation, 16(5), 260–265.
Fleming, J. W., & Hayter, J. (1974).
Reading research reports critically.
Nursing Outlook, 22, 172–175.
Horrocks, S., Anderson, E., & Salisbury
C. (2002). Systematic review of whether
nurse practitioners working in primary
care can provide equivalent care to
doctors. British Medical Journal,
324(7341), 819–823.
Roper, J. M., Shapira, J., & Chang, B. L.
(1991). Agitation in demented patients.
Journal of Gerontological Nursing,
17(3), 17–21.
Saunders, J. M. (1982). Uncoupled
identity: A process of bereavement res-
olution. Western Journal of Nursing
Research, 3(4), 319–332.
Seidl, F. W., & Pillitteri, A. (1967). De-
velopment of an attitude scale on par-
ent participation. Nursing Research,
16(1), 71–73.
Valente, S. M., & Saunders, J. M.
(2000). Oncology nurses’ difficulties
with suicidal patients. Medicine and
Law, 19(4) 793–814.
Address for reprints: Sharon Valente, 346 N.
Bowling Green Way, Los Angeles, CA 90049
(e-mail: Sharon.valente@med.va.gov).
May/June 2003