Download Research Question preparation and more Essays (university) Research Methodology in PDF only on Docsity!
Research Questions, Hypotheses,
and Clinical Questions
Judith Haber
◗ K E Y T E R M S
clinical question complex hypothesis dependent variable directional hypothesis hypothesis
independent variable nondirectional hypothesis population purpose research hypothesis
research question statistical hypothesis testability theory variable
◗ L E A R N I N G O U T C O M E S
After reading this chapter, you should be able to do the following:
- Describe how the research question and hypothesis relate to the other components of the research process.
- Describe the process of identifying and refining a research question or hypothesis.
- Identify the criteria for determining the significance of a research question or hypothesis.
- Discuss the purpose of developing a clinical question.
- Discuss the appropriate use of the purpose, aim, or objective of a research study.
- Discuss how the purpose, research question, and hypothesis suggest the level of evidence to be obtained from the findings of a research study.
- Describe the advantages and disadvantages of directional and nondirectional hypotheses.
- Compare and contrast the use of statistical versus research hypotheses.
- Discuss the appropriate use of research questions versus hypotheses in a research study.
- Discuss the differences between a research question and a clinical question in relation to evidence-based practice.
- Identify the criteria used for critiquing a research question and hypothesis.
- Apply the critiquing criteria to the evaluation of a research question and hypothesis in a research report.
◗ S T U D Y R E S O U R C E S
Go to Evolve at http://evolve.elsevier.com/LoBiondo/ for review questions, critiquing exercises, and additional research articles for practice in reviewing and critiquing.
PROPERTY
After reading this chapter, you should be able to do the following:
PROPERTY
After reading this chapter, you should be able to do the following: Describe how the
PROPERTY
Describe how the research question and hypothesis relate to the other components of
PROPERTY
research question and hypothesis relate to the other components of the research process.
PROPERTY
the research process. Describe the process
PROPERTY
Describe the process Identify the criteria
PROPERTY
Identify the criteria
PROPERTY
Discuss the purpose
PROPERTY
Discuss the purpose Discuss the appropriate
PROPERTY
Discuss the appropriate Discuss how the
PROPERTY
Discuss how the
evidence to be obtained from the findings of a research study. evidence to be obtained from the findings of a research study.^ PROPERTY
After reading this chapter, you should be able to do the following: After reading this chapter, you should be able to do the following:^ OFOF^ OF
ELSEVIERELSEVIER
independent variable
ELSEVIER
independent variable nondirectional hypothesis
ELSEVIER
nondirectional hypothesis population
ELSEVIER
population purpose
ELSEVIER
purpose research hypothesis
ELSEVIER
research hypothesis
SAMPLE
Describe the process
SAMPLE
Describe the process Identify the criteria
SAMPLE
Identify the criteria for determining the significance of a research question or hypothesis.
SAMPLE
for determining the significance of a research question or hypothesis.
SAMPLE
Discuss the purpose
SAMPLE
Discuss the purpose of developing a clinical question.
SAMPLE
of developing a clinical question. Discuss the appropriate
SAMPLE
Discuss the appropriate Discuss how the
SAMPLE
Discuss how the
SAMPLE
evidence to be obtained from the findings of a research study.
SAMPLE
evidence to be obtained from the findings of a research study.
Describe the advantages Describe the advantages SAMPLE
CONTENT
research hypothesis
CONTENT
research hypothesis
CONTENT
After reading this chapter, you should be able to do the following:
CONTENT
After reading this chapter, you should be able to do the following: research question and hypothesis relate to the other components of
CONTENT
research question and hypothesis relate to the other components of
of identifying and refining a research question or hypothesis. of identifying and refining a research question or hypothesis. CONTENT
research hypothesisresearch hypothesis - -
NOT
nondirectional hypothesis
NOT
nondirectional hypothesis
FINALFINAL
28 paRt i OvErviEw Of rEsEarch and EvidEncE-BasEd PracticE
A
s you read each chapter remember that each step of the research process will be defined and discussed as to how that particular step relates to evidence-based practice. All research studies begin with questions and hypotheses. The first step of the evidence-based practice process also asks a question, but it is a clinical question. The research questions and hypotheses in a research study discussed in the beginning of this chapter have different purposes than the clinical questions found in an evidence-based practice project. In a research study the research question and hypothesis lead to the development of a research study; the clinical question in an evidence-based practice project is the first step in the development of an evidence-based practice project. At the beginning of this chapter you are going to learn about research questions and hypotheses from the perspective of the researcher, which, in the second part of this chapter, will help you to generate your own clinical questions that you will use to guide the develop- ment of evidence-based practice projects. From a clinician’s perspective you have to understand the research question and hypothesis as it aligns with the rest of the study. As a practicing nurse, the clinical questions you will develop (see Chapters 17 and 18) represent the first step of the evidence-based practice process. When nurses ask questions such as, “Why are things done this way?”, “I wonder what would happen if … ?”, “What characteristics are associated with … ?”, or “What is the effect of … on patient outcomes?”, they are often well on their way to developing a research ques- tion or hypothesis. Research questions are usually generated by situations that emerge from practice, leading nurses to wonder about the effectiveness of one intervention versus another for a specific patient population. For an investigator conducting a study, the research question or hypothesis is a key pre- liminary step in the research process. The research question (sometimes called the problem statement) presents the idea that is to be examined in the study and is the foundation of the research study. The hypothesis attempts to answer the research question. Hypotheses can be considered intelligent hunches, guesses, or predictions that help researchers seek a solution or answer a research question. Hypotheses are a vehicle for testing the validity of the theoretical framework assumptions and provide a bridge between theory (a set of interrelated concepts, definitions, and propositions) and the real world. In the scien- tific world, researchers derive hypotheses and research questions from theories and subject them to empirical testing. A theory’s validity is not directly examined. Instead, it is through the hypotheses that the merit of a theory can be evaluated. For a clinician making an evidence-informed decision about a patient care issue, a clinical question such as whether chlorhexidine or povidone-iodine is more effective in preventing central line catheter infections, would guide the nurse in searching for and retrieving the best available evidence that, combined with clinical expertise, and patient preferences, would provide an answer on which to base the most effective decision about patient care for this population. You will often find research questions or hypotheses at the beginning of a research article. However, because of space constraints or stylistic considerations in such publications, they may be embedded in the purpose, aims, goals, or even in the results section of the research report. Nevertheless, it is equally important for both the consumer and the producer of research to understand the importance of research questions and hypotheses as the founda- tional elements of a research study. This chapter provides a working knowledge of quantitative
PROPERTY
statement) presents the idea that is to be examined in the study and is the foundation of the
PROPERTY
statement) presents the idea that is to be examined in the study and is the foundation of the hypothesis
PROPERTY
hypothesis Hypotheses can be considered intelligent hunches, guesses, or predictions that help
PROPERTY
Hypotheses can be considered intelligent hunches, guesses, or predictions that help researchers seek a solution or answer a research question. Hypotheses are a vehicle for testing
PROPERTY
researchers seek a solution or answer a research question. Hypotheses are a vehicle for testing the validity of the theoretical framework assumptions and provide a bridge between
PROPERTY
the validity of the theoretical framework assumptions and provide a bridge between
PROPERTY
(a set of interrelated concepts, definitions, and propositions) and the real world. In the scien-
PROPERTY
(a set of interrelated concepts, definitions, and propositions) and the real world. In the scien- tific world, researchers derive hypotheses and research questions from theories and subject
PROPERTY
tific world, researchers derive hypotheses and research questions from theories and subject them to empirical testing. A theory’s validity is not directly examined. Instead, it is through PROPERTY
them to empirical testing. A theory’s validity is not directly examined. Instead, it is through the hypotheses that the merit of a theory can be evaluated. the hypotheses that the merit of a theory can be evaluated. PROPERTY
OF
For an investigator conducting a study, the research question or hypothesis is a key pre-
OF
For an investigator conducting a study, the research question or hypothesis is a key pre- liminary step in the research process. The
OF
liminary step in the research process. The statement) presents the idea that is to be examined in the study and is the foundation of the OF
statement) presents the idea that is to be examined in the study and is the foundation of the hypothesis hypothesis^ OF
ELSEVIER
When nurses ask questions such as, “Why are things done this way?”, “I wonder what
ELSEVIER
When nurses ask questions such as, “Why are things done this way?”, “I wonder what would happen if … ?”, “What characteristics are associated with … ?”, or “What is the effect
ELSEVIER
would happen if … ?”, “What characteristics are associated with … ?”, or “What is the effect of … on patient outcomes?”, they are often well on their way to developing a research ques-
ELSEVIER
of … on patient outcomes?”, they are often well on their way to developing a research ques- tion or hypothesis. Research questions are usually generated by situations that emerge from
ELSEVIER
tion or hypothesis. Research questions are usually generated by situations that emerge from practice, leading nurses to wonder about the effectiveness of one intervention versus another
ELSEVIER
practice, leading nurses to wonder about the effectiveness of one intervention versus another
For an investigator conducting a study, the research question or hypothesis is a key pre- ELSEVIER
For an investigator conducting a study, the research question or hypothesis is a key pre- liminary step in the research process. The liminary step in the research process. The^ ELSEVIER
SAMPLE
the validity of the theoretical framework assumptions and provide a bridge between
SAMPLE
the validity of the theoretical framework assumptions and provide a bridge between
SAMPLE
(a set of interrelated concepts, definitions, and propositions) and the real world. In the scien-
SAMPLE
(a set of interrelated concepts, definitions, and propositions) and the real world. In the scien- tific world, researchers derive hypotheses and research questions from theories and subject
SAMPLE
tific world, researchers derive hypotheses and research questions from theories and subject them to empirical testing. A theory’s validity is not directly examined. Instead, it is through
SAMPLE
them to empirical testing. A theory’s validity is not directly examined. Instead, it is through the hypotheses that the merit of a theory can be evaluated.
SAMPLE
the hypotheses that the merit of a theory can be evaluated.
SAMPLE
For a clinician making an evidence-informed decision about a patient care issue, a clinical SAMPLE
For a clinician making an evidence-informed decision about a patient care issue, a clinical question such as whether chlorhexidine or povidone-iodine is more effective in preventing question such as whether chlorhexidine or povidone-iodine is more effective in preventing^ SAMPLE
CONTENT
For an investigator conducting a study, the research question or hypothesis is a key pre-
CONTENT
For an investigator conducting a study, the research question or hypothesis is a key pre- liminary step in the research process. The
CONTENT
liminary step in the research process. The research question
CONTENT
research question statement) presents the idea that is to be examined in the study and is the foundation of the
CONTENT
statement) presents the idea that is to be examined in the study and is the foundation of the attempts to answer the research question.
CONTENT
attempts to answer the research question. Hypotheses can be considered intelligent hunches, guesses, or predictions that help
CONTENT
Hypotheses can be considered intelligent hunches, guesses, or predictions that help researchers seek a solution or answer a research question. Hypotheses are a vehicle for testing CONTENT
researchers seek a solution or answer a research question. Hypotheses are a vehicle for testing the validity of the theoretical framework assumptions and provide a bridge between the validity of the theoretical framework assumptions and provide a bridge between CONTENT
For an investigator conducting a study, the research question or hypothesis is a key pre-For an investigator conducting a study, the research question or hypothesis is a key pre-^ -^ -
NOT
would happen if … ?”, “What characteristics are associated with … ?”, or “What is the effect
NOT
would happen if … ?”, “What characteristics are associated with … ?”, or “What is the effect of … on patient outcomes?”, they are often well on their way to developing a research ques-
NOT
of … on patient outcomes?”, they are often well on their way to developing a research ques- tion or hypothesis. Research questions are usually generated by situations that emerge from
NOT
tion or hypothesis. Research questions are usually generated by situations that emerge from practice, leading nurses to wonder about the effectiveness of one intervention versus another NOT
practice, leading nurses to wonder about the effectiveness of one intervention versus another
FINAL
the research question and hypothesis as it aligns with the rest of the study. As a practicing
FINAL
the research question and hypothesis as it aligns with the rest of the study. As a practicing nurse, the clinical questions you will develop (see Chapters 17 and 18) represent the first step
FINAL
nurse, the clinical questions you will develop (see Chapters 17 and 18) represent the first step
When nurses ask questions such as, “Why are things done this way?”, “I wonder what
FINAL
When nurses ask questions such as, “Why are things done this way?”, “I wonder what would happen if … ?”, “What characteristics are associated with … ?”, or “What is the effect FINAL
would happen if … ?”, “What characteristics are associated with … ?”, or “What is the effect of … on patient outcomes?”, they are often well on their way to developing a research ques- of … on patient outcomes?”, they are often well on their way to developing a research ques-^ FINAL
30 paRt i OvErviEw Of rEsEarch and EvidEncE-BasEd PracticE
Area Influence Example critical appraisal of the scientific literature
the critical appraisal of research studies that appear in journals may indirectly suggest a clinical problem area by stimulating the reader’s thinking. the nurse may observe the outcome data from a single study or a group of related studies that provide the basis for developing a pilot study, quality improvement project, or clinical practice guideline to determine the effectiveness of this intervention in their own practice setting.
at a staff meeting, nurses, physicians, and other members of the interdisciplinary oncology team at a hospital specializing in treatment of cancer were discussing developing an algorithm to serve as an interdisciplinary protocol for the most effective interventions for treating adult cancer pain in specific treatment settings. their search for and critical appraisal of existing clinical practice guidelines led to development of an interdisciplinary Cancer Pain Practice Guideline, based on national cancer institute (nci) and national cancer consensus network (nccn) practice guidelines, for treatment of adult cancer pain in a variety of settings that were relevant to their patient population and clinical setting (Md anderson cancer center, 2008). Gaps in the literature a research idea may also be suggested by a critical appraisal of the literature that identifies gaps in the literature and suggests areas for future study. research ideas also can be generated by research reports that suggest the value of replicating a particular study to extend or refine the existing scientific knowledge base.
rural adults have higher rates of chronic illness and physical limitations that might be prevented by increased physical activity, yet few studies have been focused on helping people increase their regular physical activity in rural environments. the study used a telephone-only motivational interviewing (Mi) intervention that is different from other Mi studies that included one or more in-person Mi counseling sessions (Bennet, Lyons, winter-stone et al., 2008).
interest in untested theory
verification of an untested theory provides a relatively uncharted territory from which research problems can be derived. inasmuch as theories themselves are not tested, a researcher may consider investigating a particular concept or set of concepts related to a particular nursing theory or a theory from another discipline. the researcher would pose questions such as the following: “if this theory is correct, what kind of behavior will i expect to observe in particular patients and under which conditions?” “if this theory is valid, what kind of supporting evidence will i find?”
self-regulation theory (Johnson et al., 1997) proposes that individuals cope with illness according to their understanding of the experience. the theory emphasizes that patients need to have adequate information to gain knowledge and understanding of a specific health-related problem or risk (e.g., for breast cancer survivors, a health-related issue is lymphedema risk) and to make decisions and develop preventive or coping strategies (e.g., lymphedema risk reduction behaviors). accordingly, patient education interventions that provide accurate information may be a critical component of lymphedema risk reduction. the use of self-regulation theory to test the effect of providing breast cancer survivors with lymphedema information on clinical outcomes has not been explored. Using Johnson’s self-regulation theory to guide its development, the purpose of this study was to explore the effect of provision of lymphedema information on survivors’ symptom experiences and practice of risk reduction behaviors (fu et al., 2008).
taBLe 2-1 How Practical Experience, Scientific Literature, and Untested Theory Influence the Development of a Research Idea—cont’d
PROPERTY
extend or refine the existing
PROPERTY
extend or refine the existing scientific knowledge base.
PROPERTY
scientific knowledge base.
PROPERTY
an untested theory
PROPERTY
an untested theory provides a relatively uncharted
PROPERTY
provides a relatively uncharted territory from which research
PROPERTY
territory from which research problems can be derived.
PROPERTY
problems can be derived.
PROPERTY
as theories
PROPERTY
as theories tested, a researcher may consider
PROPERTY
tested, a researcher may consider investigating a particular concept or PROPERTY
investigating a particular concept or
OF
reports that suggest the value of
OF
reports that suggest the value of replicating a particular study to OF
replicating a particular study to extend or refine the existing OF
extend or refine the existing scientific knowledge base. scientific knowledge base.^ OF
ELSEVIER
c
ELSEVIER
center, 2008).
ELSEVIER
enter, 2008).
can be generated by research ELSEVIER
can be generated by research reports that suggest the value of reports that suggest the value of ELSEVIER
ural adults
ELSEVIER
ural adults have higher rates of chronic illness and physical
ELSEVIER
have higher rates of chronic illness and physical limitations that might be prevented by increased physical
ELSEVIER
limitations that might be prevented by increased physical activity, yet few studies have been focused on helping
ELSEVIER
activity, yet few studies have been focused on helping people increase their regular physical activity in rural
ELSEVIER
people increase their regular physical activity in rural environments.
ELSEVIER
environments.
SAMPLE
territory from which research
SAMPLE
territory from which research problems can be derived.
SAMPLE
problems can be derived.
SAMPLE
themselves are not
SAMPLE
themselves are not tested, a researcher may consider
SAMPLE
tested, a researcher may consider investigating a particular concept or
SAMPLE
investigating a particular concept or
SAMPLE
set of concepts related to a SAMPLE
set of concepts related to a particular nursing theory or a theory particular nursing theory or a theory SAMPLE from another discipline. SAMPLE from another discipline.
CONTENT
motivational interviewing (M
CONTENT
motivational interviewing (M from other M
CONTENT
from other M M
CONTENT
M
an untested theory
CONTENT
an untested theory provides a relatively uncharted CONTENT
provides a relatively uncharted territory from which research territory from which research CONTENT
motivational interviewing (Mmotivational interviewing (M - -
NOT
limitations that might be prevented by increased physical
NOT
limitations that might be prevented by increased physical activity, yet few studies have been focused on helping
NOT
activity, yet few studies have been focused on helping people increase their regular physical activity in rural NOT
people increase their regular physical activity in rural environments. environments. NOT
FINAL
guidelines, for treatment of adult cancer
FINAL
guidelines, for treatment of adult cancer pain in a variety of settings that were relevant to their
FINAL
pain in a variety of settings that were relevant to their patient population and clinical setting (M
FINAL
patient population and clinical setting (M enter, 2008).
FINAL
enter, 2008). have higher rates of chronic illness and physical FINAL
have higher rates of chronic illness and physical limitations that might be prevented by increased physical limitations that might be prevented by increased physical^ FINAL
CHapteR 2 research Questions, hypotheses, and clinical Questions 31
Defining the Research Question
Brainstorming with teachers, advisors, or colleagues may provide valuable feedback that helps the researcher focus on a specific research question area. For example, suppose a researcher told a colleague that her area of interest was pain as a prevalent problem for older adults. The colleague may have said, “What is it about the topic that specifically interests you?” This conversation may have initiated a chain of thought that resulted in a decision to explore the relationship between pain and functional disability in older adults (Horgas et al., 2008) (see Appendix C). Figure 2-1 illustrates how a broad area of interest (pain as a prevalent problem for older adults) was narrowed to a specific research topic (persistent pain and its relationship to functional disability in older adults).
EVIDENCE-BASED PRACTICE TIP
a well-developed research question guides a focused search for scientific evidence about assessing, diagno- sing, treating, or assisting patients with understanding of their prognosis related to a specific health problem.
Beginning the Literature Review
The literature review should reveal a relevant collection of individual studies and systematic reviews that have been critically examined. Concluding sections in such articles, that is, the recommendations and implications for practice, often identify remaining gaps in the literature, the need for replication, or the need for extension of the knowledge base about a particular research focus (see Chapter 3). In the previous example about persistent pain and functional disability in older adults, the researcher may have conducted a preliminary review of books and journals for theories and research studies on factors apparently critical to pain experience such as racial and/or ethnic differences in pain experience, pain treatment, and access to pain medications. These factors, termed variables in the language of research, should be potentially relevant, of interest, and measurable. Possible relevant factors mentioned in the literature begin with an exploration of the rela- tionship between self-reported pain intensity, acute versus chronic pain, pain management effectiveness, and functional disability. Other variables, called demographic variables, such as race, ethnicity, gender, age, income, education, and marital status, are also suggested as essen- tial to consider. This information can then be used by the researcher to further define the research question and address a gap in the literature, as well as extend the knowledge base related to relationships among race (black or white), pain, and functional disability (physical and social functioning) in older adults. At this point the researcher could write the following tentative research question: “What are the relationships among race, pain, and disability in older adults?” Readers can envision the interrelatedness of the initial definition of the question area, the literature review, and the refined research question. Readers of research reports examine the end product of this process in the form of a research question and/or hypothesis, so it is important to have an appreciation of how the researcher gets to that point in construct- ing a study (Horgas et al., 2008) (see Appendix C).
HELPFUL HINT
reading the literature review or theoretical framework section of a research article helps you trace the devel- opment of the implied research question and/or hypothesis.
PROPERTY
and journals for theories and research studies on factors apparently critical to pain experience
PROPERTY
and journals for theories and research studies on factors apparently critical to pain experience such as racial and/or ethnic differences in pain experience, pain treatment, and access to pain
PROPERTY
such as racial and/or ethnic differences in pain experience, pain treatment, and access to pain medications. These factors, termed
PROPERTY
medications. These factors, termed
PROPERTY
relevant, of interest, and measurable.
PROPERTY
relevant, of interest, and measurable. Possible relevant factors mentioned in the literature begin with an exploration of the rela-
PROPERTY
Possible relevant factors mentioned in the literature begin with an exploration of the rela- tionship between self-reported pain intensity, acute versus chronic pain, pain management
PROPERTY
tionship between self-reported pain intensity, acute versus chronic pain, pain management effectiveness, and functional disability. Other variables, called
PROPERTY
effectiveness, and functional disability. Other variables, called race, ethnicity, gender, age, income, education, and marital status, are also suggested as essen-
PROPERTY
race, ethnicity, gender, age, income, education, and marital status, are also suggested as essen- tial to consider. This information can then be used by the researcher to further define the tial to consider. This information can then be used by the researcher to further define the PROPERTY
OF
research focus (see Chapter 3). In the previous example about persistent pain and functional
OF
research focus (see Chapter 3). In the previous example about persistent pain and functional disability in older adults, the researcher may have conducted a preliminary review of books
OF
disability in older adults, the researcher may have conducted a preliminary review of books and journals for theories and research studies on factors apparently critical to pain experience OF
and journals for theories and research studies on factors apparently critical to pain experience such as racial and/or ethnic differences in pain experience, pain treatment, and access to pain such as racial and/or ethnic differences in pain experience, pain treatment, and access to pain^ OF
ELSEVIERELSEVIER
Beginning the Literature Review
ELSEVIER
Beginning the Literature Review
The literature review should reveal a relevant collection of individual studies and systematic
ELSEVIER
The literature review should reveal a relevant collection of individual studies and systematic reviews that have been critically examined. Concluding sections in such articles, that is, the
ELSEVIER
reviews that have been critically examined. Concluding sections in such articles, that is, the recommendations and implications for practice, often identify remaining gaps in the literature,
ELSEVIER
recommendations and implications for practice, often identify remaining gaps in the literature, the need for replication, or the need for extension of the knowledge base about a particular
ELSEVIER
the need for replication, or the need for extension of the knowledge base about a particular research focus (see Chapter 3). In the previous example about persistent pain and functional ELSEVIER
research focus (see Chapter 3). In the previous example about persistent pain and functional disability in older adults, the researcher may have conducted a preliminary review of books disability in older adults, the researcher may have conducted a preliminary review of books^ ELSEVIER
SAMPLE
Possible relevant factors mentioned in the literature begin with an exploration of the rela-
SAMPLE
Possible relevant factors mentioned in the literature begin with an exploration of the rela- tionship between self-reported pain intensity, acute versus chronic pain, pain management
SAMPLE
tionship between self-reported pain intensity, acute versus chronic pain, pain management effectiveness, and functional disability. Other variables, called
SAMPLE
effectiveness, and functional disability. Other variables, called race, ethnicity, gender, age, income, education, and marital status, are also suggested as essen-
SAMPLE
race, ethnicity, gender, age, income, education, and marital status, are also suggested as essen- tial to consider. This information can then be used by the researcher to further define the
SAMPLE
tial to consider. This information can then be used by the researcher to further define the research question and address a gap in the literature, as well as extend the knowledge base SAMPLE
research question and address a gap in the literature, as well as extend the knowledge base related to relationships among race (black or white), pain, and functional disability (physical related to relationships among race (black or white), pain, and functional disability (physical^ SAMPLE
CONTENT
the need for replication, or the need for extension of the knowledge base about a particular
CONTENT
the need for replication, or the need for extension of the knowledge base about a particular research focus (see Chapter 3). In the previous example about persistent pain and functional
CONTENT
research focus (see Chapter 3). In the previous example about persistent pain and functional disability in older adults, the researcher may have conducted a preliminary review of books
CONTENT
disability in older adults, the researcher may have conducted a preliminary review of books and journals for theories and research studies on factors apparently critical to pain experience
CONTENT
and journals for theories and research studies on factors apparently critical to pain experience such as racial and/or ethnic differences in pain experience, pain treatment, and access to pain
CONTENT
such as racial and/or ethnic differences in pain experience, pain treatment, and access to pain medications. These factors, termed
CONTENT
medications. These factors, termed
CONTENT
variables
CONTENT
variables relevant, of interest, and measurable.
CONTENT
relevant, of interest, and measurable. Possible relevant factors mentioned in the literature begin with an exploration of the rela- Possible relevant factors mentioned in the literature begin with an exploration of the rela- CONTENT
the need for replication, or the need for extension of the knowledge base about a particular
the need for replication, or the need for extension of the knowledge base about a particular research focus (see Chapter 3). In the previous example about persistent pain and functionalresearch focus (see Chapter 3). In the previous example about persistent pain and functional^ -
NOT
The literature review should reveal a relevant collection of individual studies and systematic
NOT
The literature review should reveal a relevant collection of individual studies and systematic reviews that have been critically examined. Concluding sections in such articles, that is, the
NOT
reviews that have been critically examined. Concluding sections in such articles, that is, the recommendations and implications for practice, often identify remaining gaps in the literature, NOT
recommendations and implications for practice, often identify remaining gaps in the literature, the need for replication, or the need for extension of the knowledge base about a particular the need for replication, or the need for extension of the knowledge base about a particular^ NOT
FINALFINALFINALFINAL
assisting patients with understanding of their prognosis related to a specific health problem.
FINAL
assisting patients with understanding of their prognosis related to a specific health problem.
The literature review should reveal a relevant collection of individual studies and systematic The literature review should reveal a relevant collection of individual studies and systematic^ FINAL^ FINAL
CHapteR 2 research Questions, hypotheses, and clinical Questions 33
Examining Significance
When considering a research question, it is crucial that the researcher has examined the ques- tion’s potential significance to nursing. The research question should have the potential to contribute to and extend the scientific body of nursing knowledge. Guidelines for selecting research questions should meet the following criteria.
- Patients, nurses, the medical community in general, and society will potentially benefit from the knowledge derived from the study.
- The results will be applicable for nursing practice, education, or administration.
- The results will be theoretically relevant.
- The findings will lend support to untested theoretical assumptions, extend or challenge an existing theory, fill a gap in the literature, or clarify a conflict in the literature.
- The findings will potentially provide evidence that supports developing, retaining, or revising nursing practices or policies. If the research question has not met any of these criteria, it is wise to extensively revise the question or discard it. For example, in the previously cited research question, the significance of the question includes the following facts:
- Pain is a persistent problem in the daily lives of approximately 50% of community- dwelling older adults.
- Pain is due to the high prevalence of acute and chronic health problems in this population.
- Pain from acute and chronic conditions is a key indicator of physical and social func- tioning and quality of life.
- Data on racial and/or ethnic differences in pain experience indicate that African Ameri- cans report more pain, more untreated pain, and have less access to pain medications.
- Few of the studies conducted have been conducted in older adult populations.
- This study sought to fill a gap in the related literature by examining the relationships among race, pain, and functional disability in older adults.
- This study sought to extend the knowledge base about this phenomenon, thereby pro- viding a foundation for the development and testing of interventions.
Determining Feasibility
The feasibility of a research question must be pragmatically examined. Regardless of how significant or researchable a question may be, pragmatic considerations such as time; availability of subjects, facilities, equipment, and money; experience of the researcher; and any ethical considerations may cause the researcher to decide that the question is inappropriate because it lacks feasibility (see Chapters 4 and 7).
EVIDENCE-BASED PRACTICE TIP
without a well-developed research question, the researcher may search for wrong, irrelevant, or unnecessary information. this will be a barrier to identifying the potential significance of the study.
PROPERTY
studies
PROPERTY
studies conducted
PROPERTY
conducted sought
PROPERTY
sought to
PROPERTY
to among race, pain,
PROPERTY
among race, pain, study
PROPERTY
study sought
PROPERTY
sought viding
PROPERTY
viding a foundation for the development and testing of interventions.
PROPERTY
a foundation for the development and testing of interventions.
PROPERTYPROPERTYPROPERTYPROPERTY
EVIDENCE-BASED PRACTICE TIP
PROPERTY
EVIDENCE-BASED PRACTICE TIP
ithout a well-developed research question, the researcher may search for wrong, irrelevant, or unnecessary ithout a well-developed research question, the researcher may search for wrong, irrelevant, or unnecessary PROPERTY
OF
ethnic
OF
ethnic eport more pain, more untreated pain, and have less access to pain medications. OF
eport more pain, more untreated pain, and have less access to pain medications. conducted conducted OF
ELSEVIER
daily
ELSEVIER
daily lives
ELSEVIER
lives of
ELSEVIER
of
prevalence
ELSEVIER
prevalence of
ELSEVIER
of acute
ELSEVIER
acute
conditions
ELSEVIER
conditions
differences differences^ ELSEVIER
SAMPLE
sought
SAMPLE
sought
SAMPLE
a foundation for the development and testing of interventions.
SAMPLE
a foundation for the development and testing of interventions.
SAMPLESAMPLESAMPLESAMPLESAMPLESAMPLESAMPLE
EVIDENCE-BASED PRACTICE TIP
SAMPLE
EVIDENCE-BASED PRACTICE TIP
ithout a well-developed research question, the researcher may search for wrong, irrelevant, or unnecessary
SAMPLE
ithout a well-developed research question, the researcher may search for wrong, irrelevant, or unnecessary information. SAMPLE
information. t SAMPLE
this will SAMPLE
his will
CONTENT
differences
CONTENT
differences in
CONTENT
in eport more pain, more untreated pain, and have less access to pain medications.
CONTENT
eport more pain, more untreated pain, and have less access to pain medications. conducted
CONTENT
conducted have
CONTENT
have fill
CONTENT
fill a
CONTENT
a gap
CONTENT
gap and functional disability in older adults. CONTENT
and functional disability in older adults. to CONTENT to extend CONTENT extend a foundation for the development and testing of interventions. CONTENT a foundation for the development and testing of interventions.
-^
NOT
acute
NOT
acute and
NOT
and
a a (^) NOT NOTkey key
FINAL
If the research question has not met any of these criteria, it is wise to extensively revise the
FINAL
If the research question has not met any of these criteria, it is wise to extensively revise the question or discard it. For example, in the previously cited research question, the significance
FINAL
question or discard it. For example, in the previously cited research question, the significance
approximately FINAL
approximately
34 paRt i OvErviEw Of rEsEarch and EvidEncE-BasEd PracticE
THE FULLY DEVELOPED RESEARCH QUESTION
When a researcher finalizes a research question, the following three characteristics should be evident:
- It clearly identifies the variables under consideration.
- It specifies the population being studied.
- It implies the possibility of empirical testing. Because each of these elements is crucial to the formulation of a satisfactory research ques- tion, the criteria will be discussed in greater detail. These elements can often be found in the introduction of the published article; they are not always stated in an explicit manner.
HELPFUL HINT
remember that research questions are used to guide all types of research studies, but are most often used in exploratory, descriptive, qualitative, or hypothesis-generating studies.
EVIDENCE-BASED PRACTICE TIP
the answers to questions generated by qualitative data reflect evidence that may provide the first insights about a phenomenon that has not been previously studied.
Variables
Researchers call the properties that they study variables. Such properties take on different values. Thus a variable is, as the name suggests, something that varies. Properties that differ from each other, such as age, weight, height, religion, and ethnicity, are examples of variables. Researchers attempt to understand how and why differences in one variable relate to differ- ences in another variable. For example, a researcher may be concerned about the variable of pneumonia in postoperative patients on ventilators in critical care units. It is a variable because not all critically ill postoperative patients on ventilators have pneumonia. A researcher may also be interested in what other factors can be linked to ventilator-acquired pneumonia (VAP). There is clinical evidence to suggest that elevation of the head of the bed is also associated with VAP. You can see that these factors are also variables that need to be considered in rela- tion to the development of VAP in postoperative patients. When speaking of variables, the researcher is essentially asking, “Is X related to Y? What is the effect of X on Y? How are X 1 and X 2 related to Y? ” The researcher is asking a question about the relationship between one or more independent variables and a dependent variable. ( Note: In cases in which multiple independent or dependent variables are present, subscripts are used to indicate the number of variables under consideration.) An independent variable, usually symbolized by X, is the variable that has the presumed effect on the dependent variable. In experimental research studies, the researcher manipulates the independent variable. For example, a nurse may study how different methods of admin- istering pain medication affect the patient’s perception of pain intensity. The researcher may manipulate the independent variable (i.e., the method of administering pain medication) by using nurse-controlled versus patient-controlled administration of analgesia (see Chapter 8). In nonexperimental research, the independent variable is not manipulated and is assumed to have occurred naturally before or during the study. For example, the researcher may be study- ing the relationship between gender and perception of pain intensity. The independent vari-
PROPERTY
is, as the name suggests, something that varies. Properties that differ
PROPERTY
is, as the name suggests, something that varies. Properties that differ from each other, such as age, weight, height, religion, and ethnicity, are examples of variables.
PROPERTY
from each other, such as age, weight, height, religion, and ethnicity, are examples of variables. Researchers attempt to understand how and why differences in one variable relate to differ-
PROPERTY
Researchers attempt to understand how and why differences in one variable relate to differ- ences in another variable. For example, a researcher may be concerned about the variable of
PROPERTY
ences in another variable. For example, a researcher may be concerned about the variable of
PROPERTY
pneumonia in postoperative patients on ventilators in critical care units. It is a variable because
PROPERTY
pneumonia in postoperative patients on ventilators in critical care units. It is a variable because not all critically ill postoperative patients on ventilators have pneumonia. A researcher may
PROPERTY
not all critically ill postoperative patients on ventilators have pneumonia. A researcher may also be interested in what other factors can be linked to ventilator-acquired pneumonia (VAP).
PROPERTY
also be interested in what other factors can be linked to ventilator-acquired pneumonia (VAP). There is clinical evidence to suggest that elevation of the head of the bed is also associated PROPERTY
There is clinical evidence to suggest that elevation of the head of the bed is also associated with VAP. You can see that these factors are also variables that need to be considered in rela- with VAP. You can see that these factors are also variables that need to be considered in rela- PROPERTY PROPERTY
OF
Researchers call the properties that they study variables. Such properties take on different
OF
Researchers call the properties that they study variables. Such properties take on different is, as the name suggests, something that varies. Properties that differ OF
is, as the name suggests, something that varies. Properties that differ from each other, such as age, weight, height, religion, and ethnicity, are examples of variables. from each other, such as age, weight, height, religion, and ethnicity, are examples of variables.^ OF
ELSEVIERELSEVIERELSEVIERELSEVIERELSEVIERELSEVIER
he answers to questions generated by qualitative data reflect evidence that may provide the first insights
ELSEVIER
he answers to questions generated by qualitative data reflect evidence that may provide the first insights that has not been previously studied.
ELSEVIER
that has not been previously studied.
Researchers call the properties that they study variables. Such properties take on different Researchers call the properties that they study variables. Such properties take on different^ ELSEVIER^ ELSEVIER
SAMPLE
pneumonia in postoperative patients on ventilators in critical care units. It is a variable because
SAMPLE
pneumonia in postoperative patients on ventilators in critical care units. It is a variable because not all critically ill postoperative patients on ventilators have pneumonia. A researcher may
SAMPLE
not all critically ill postoperative patients on ventilators have pneumonia. A researcher may also be interested in what other factors can be linked to ventilator-acquired pneumonia (VAP).
SAMPLE
also be interested in what other factors can be linked to ventilator-acquired pneumonia (VAP). There is clinical evidence to suggest that elevation of the head of the bed is also associated
SAMPLE
There is clinical evidence to suggest that elevation of the head of the bed is also associated
SAMPLE
with VAP. You can see that these factors are also variables that need to be considered in rela-
SAMPLE
with VAP. You can see that these factors are also variables that need to be considered in rela- tion to the development of VAP in postoperative patients. SAMPLE
tion to the development of VAP in postoperative patients. When speaking of variables, the researcher is essentially asking, “Is When speaking of variables, the researcher is essentially asking, “Is^ SAMPLE
CONTENTCONTENT
Researchers call the properties that they study variables. Such properties take on different
CONTENT
Researchers call the properties that they study variables. Such properties take on different is, as the name suggests, something that varies. Properties that differ
CONTENT
is, as the name suggests, something that varies. Properties that differ from each other, such as age, weight, height, religion, and ethnicity, are examples of variables.
CONTENT
from each other, such as age, weight, height, religion, and ethnicity, are examples of variables. Researchers attempt to understand how and why differences in one variable relate to differ-
CONTENT
Researchers attempt to understand how and why differences in one variable relate to differ- ences in another variable. For example, a researcher may be concerned about the variable of CONTENT
ences in another variable. For example, a researcher may be concerned about the variable of pneumonia in postoperative patients on ventilators in critical care units. It is a variable because pneumonia in postoperative patients on ventilators in critical care units. It is a variable because CONTENT
-^
NOTNOTNOTNOT
he answers to questions generated by qualitative data reflect evidence that may provide the first insights
NOT
he answers to questions generated by qualitative data reflect evidence that may provide the first insights that has not been previously studied.
NOT
that has not been previously studied.
FINALFINALFINALFINAL
emember that research questions are used to guide all types of research studies, but are most often used
FINAL
emember that research questions are used to guide all types of research studies, but are most often used
he answers to questions generated by qualitative data reflect evidence that may provide the first insights he answers to questions generated by qualitative data reflect evidence that may provide the first insights^ FINALFINALFINAL^ FINAL
36 paRt i OvErviEw Of rEsEarch and EvidEncE-BasEd PracticE
Although one independent variable and one dependent variable are used in the examples just given, there is no restriction on the number of variables that can be included in a research ques- tion. Remember, however, that questions should not be unnecessarily complex or unwieldy, particularly in beginning research efforts. Research questions that include more than one inde- pendent or dependent variable may be broken down into subquestions that are more concise. Finally, it should be noted that variables are not inherently independent or dependent. A variable that is classified as independent in one study may be considered dependent in another study. For example, a nurse may review an article about sexual behaviors that are predictive of risk for human immunodeficiency virus (HIV)/acquired immunodeficiency syndrome (AIDS). In this case, HIV/AIDS is the dependent variable. When another article about the relationship between HIV/AIDS and maternal parenting practices is considered, HIV/AIDS status is the independent variable. Whether a variable is independent or dependent is a func- tion of the role it plays in a particular study.
Population
The population (a well-defined set that has certain properties) is either specified or implied in the research question. If the scope of the question has been narrowed to a specific focus and the variables have been clearly identified, the nature of the population will be evident to the reader of a research report. For example, a research question may ask, “What is the effect of a psychoeducational intervention on quality of life (QOL) in breast cancer survivors in posttreatment survivorship?” This question suggests that the population under consideration includes breast cancer survivors who have completed treatment for breast cancer (e.g., surgery, adjuvant therapy, reconstruction). It is also implied that some of the breast cancer survivors were involved in a psychoeducational intervention (consisting of face-to-face sessions and monthly follow-up sessions by telephone and in person) in contrast to other survivors (who received four monthly control telephone calls). The researcher or reader will have an initial idea of the composition of the study population from the outset (see Chapter 10).
EVIDENCE-BASED PRACTICE TIP
Make sure that the population of interest and the setting have been clearly described so that if you were going to replicate the study, you would know exactly who the study population needed to be.
Testability
The research question must imply that it is testable, that is, measurable by either qualitative or quantitative methods. For example, the research question “Should postoperative patients control how much pain medication they receive?” is stated incorrectly for a variety of reasons. One reason is that it is not testable; it represents a value statement rather than a research question. A scientific research question must propose a relationship between an independent and a dependent variable and do this in such a way that it indicates that the variables of the relationship can somehow be measured. Many interesting and important clinical questions are not valid research questions because they are not amenable to testing. The question “Should postoperative patients control how much pain medication they receive?” could be revised from a philosophical question to a research question that implies testability. Two examples of the revised research question might be the following.
PROPERTY
were involved in a psychoeducational intervention (consisting of face-to-face sessions and
PROPERTY
were involved in a psychoeducational intervention (consisting of face-to-face sessions and monthly follow-up sessions by telephone and in person) in contrast to other survivors (who
PROPERTY
monthly follow-up sessions by telephone and in person) in contrast to other survivors (who received four monthly control telephone calls). The researcher or reader will have an initial
PROPERTY
received four monthly control telephone calls). The researcher or reader will have an initial idea of the composition of the study population from the outset (see Chapter 10).
PROPERTY
idea of the composition of the study population from the outset (see Chapter 10).
PROPERTYPROPERTYPROPERTYPROPERTY
EVIDENCE-BASED PRACTICE TIP
PROPERTY
EVIDENCE-BASED PRACTICE TIP
Make sure that the population of interest and the setting have been clearly described so that if you were
PROPERTY
Make sure that the population of interest and the setting have been clearly described so that if you were
PROPERTYPROPERTYPROPERTYPROPERTY
going to replicate the study, you would know exactly who the study population needed to be. PROPERTY
going to replicate the study, you would know exactly who the study population needed to be.
OF
includes breast cancer survivors who have completed treatment for breast cancer (e.g., surgery,
OF
includes breast cancer survivors who have completed treatment for breast cancer (e.g., surgery, adjuvant therapy, reconstruction). It is also implied that some of the breast cancer survivors
OF
adjuvant therapy, reconstruction). It is also implied that some of the breast cancer survivors were involved in a psychoeducational intervention (consisting of face-to-face sessions and OF
were involved in a psychoeducational intervention (consisting of face-to-face sessions and monthly follow-up sessions by telephone and in person) in contrast to other survivors (who monthly follow-up sessions by telephone and in person) in contrast to other survivors (who^ OF
ELSEVIER
(a well-defined set that has certain properties) is either specified or implied
ELSEVIER
(a well-defined set that has certain properties) is either specified or implied in the research question. If the scope of the question has been narrowed to a specific focus
ELSEVIER
in the research question. If the scope of the question has been narrowed to a specific focus and the variables have been clearly identified, the nature of the population will be evident to
ELSEVIER
and the variables have been clearly identified, the nature of the population will be evident to the reader of a research report. For example, a research question may ask, “What is the effect
ELSEVIER
the reader of a research report. For example, a research question may ask, “What is the effect of a psychoeducational intervention on quality of life (QOL) in breast cancer survivors in
ELSEVIER
of a psychoeducational intervention on quality of life (QOL) in breast cancer survivors in posttreatment survivorship?” This question suggests that the population under consideration
ELSEVIER
posttreatment survivorship?” This question suggests that the population under consideration includes breast cancer survivors who have completed treatment for breast cancer (e.g., surgery, ELSEVIER
includes breast cancer survivors who have completed treatment for breast cancer (e.g., surgery, adjuvant therapy, reconstruction). It is also implied that some of the breast cancer survivors adjuvant therapy, reconstruction). It is also implied that some of the breast cancer survivors^ ELSEVIER
SAMPLESAMPLESAMPLESAMPLESAMPLE
EVIDENCE-BASED PRACTICE TIP
SAMPLE
EVIDENCE-BASED PRACTICE TIP
Make sure that the population of interest and the setting have been clearly described so that if you were
SAMPLE
Make sure that the population of interest and the setting have been clearly described so that if you were
SAMPLESAMPLESAMPLESAMPLE
going to replicate the study, you would know exactly who the study population needed to be.
SAMPLE
going to replicate the study, you would know exactly who the study population needed to be.
CONTENT
posttreatment survivorship?” This question suggests that the population under consideration
CONTENT
posttreatment survivorship?” This question suggests that the population under consideration includes breast cancer survivors who have completed treatment for breast cancer (e.g., surgery,
CONTENT
includes breast cancer survivors who have completed treatment for breast cancer (e.g., surgery, adjuvant therapy, reconstruction). It is also implied that some of the breast cancer survivors
CONTENT
adjuvant therapy, reconstruction). It is also implied that some of the breast cancer survivors were involved in a psychoeducational intervention (consisting of face-to-face sessions and
CONTENT
were involved in a psychoeducational intervention (consisting of face-to-face sessions and monthly follow-up sessions by telephone and in person) in contrast to other survivors (who
CONTENT
monthly follow-up sessions by telephone and in person) in contrast to other survivors (who received four monthly control telephone calls). The researcher or reader will have an initial
CONTENT
received four monthly control telephone calls). The researcher or reader will have an initial idea of the composition of the study population from the outset (see Chapter 10). CONTENT
idea of the composition of the study population from the outset (see Chapter 10).
posttreatment survivorship?” This question suggests that the population under consideration
posttreatment survivorship?” This question suggests that the population under consideration includes breast cancer survivors who have completed treatment for breast cancer (e.g., surgery,includes breast cancer survivors who have completed treatment for breast cancer (e.g., surgery,^ -
NOT
in the research question. If the scope of the question has been narrowed to a specific focus
NOT
in the research question. If the scope of the question has been narrowed to a specific focus and the variables have been clearly identified, the nature of the population will be evident to
NOT
and the variables have been clearly identified, the nature of the population will be evident to the reader of a research report. For example, a research question may ask, “What is the effect
NOT
the reader of a research report. For example, a research question may ask, “What is the effect of a psychoeducational intervention on quality of life (QOL) in breast cancer survivors in NOT
of a psychoeducational intervention on quality of life (QOL) in breast cancer survivors in posttreatment survivorship?” This question suggests that the population under consideration posttreatment survivorship?” This question suggests that the population under consideration^ NOT
FINALFINAL
(a well-defined set that has certain properties) is either specified or implied
FINAL
(a well-defined set that has certain properties) is either specified or implied in the research question. If the scope of the question has been narrowed to a specific focus FINAL
in the research question. If the scope of the question has been narrowed to a specific focus and the variables have been clearly identified, the nature of the population will be evident to and the variables have been clearly identified, the nature of the population will be evident to^ FINAL
CHapteR 2 research Questions, hypotheses, and clinical Questions 37
- Is there a relationship between patient-controlled analgesia (PCA) versus nurse- administered analgesia and perception of postoperative pain?
- What is the effect of PCA on pain ratings by postoperative patients? These examples illustrate the relationship between the variables, identify the independent and dependent variables, and imply the testability of the research question. Now that the elements of the formal research question have been presented in greater detail, this information can be integrated by formulating a formal research question about whether self-reported pain (pain sites and pain intensity) and disability (physical and social functional limitations) differ between black and white older adults. This research question was originally derived from a general area of interest—pain experiences of older adults of different racial groups. The topic was more specifically defined by delineating a particular research question, self-reported pain and outcomes (e.g., physical and social functional limitations). The question crystallized further after a preliminary literature review (Horgas et al., 2008). Table 2-3 identi- fies the components of this research question as they relate to and are congruent with the three research question criteria.
taBLe 2-3 Components of the Research Question and Related Criteria Variables Population Testability Independent variable: Pain intensity Pain sites race health (number of limiting diagnoses) Dependent variable: Management effectiveness functional status
Black and white older adults differential effect of pain intensity and number of painful sites on functional disability (physical and social functioning)
HELPFUL HINT
- Remember that research questions are often not explicitly stated. The reader has to infer the research question from the title of the report, the abstract, the introduction, or the purpose.
- Using your focused question, search the literature for the best available answer to your clinical question.
STUDY PURPOSE, AIMS, OR OBJECTIVES
Once the research question is developed and the literature review is critiqued in terms of the level, strength, and quality of evidence available for the particular research question, the purpose, aims, or objectives of the study become focused so that the researcher can decide whether a hypothesis should be tested or a research question answered. The purpose of the study encompasses the aims or objectives the investigator hopes to achieve with the research, not the question to be answered. These three terms are synonymous with each other. For example, a nurse working with rehabilitation patients who have bladder dysfunction may be disturbed by the high incidence of urinary tract infections. The nurse may propose the following research question: “What is the optimum frequency of changing
PROPERTY
crystallized further after a preliminary literature review (Horgas et al., 2008). Table 2-3 identi-
PROPERTY
crystallized further after a preliminary literature review (Horgas et al., 2008). Table 2-3 identi- fies the components of this research question as they relate to and are congruent with the
PROPERTY
fies the components of this research question as they relate to and are congruent with the three research question criteria.
PROPERTY
three research question criteria.
PROPERTYPROPERTYPROPERTYPROPERTYPROPERTY
HELPFUL HINT
PROPERTY
HELPFUL HINT
PROPERTY
PROPERTY
that question from the PROPERTY
question from the
OF
groups. The topic was more specifically defined by delineating a particular research question,
OF
groups. The topic was more specifically defined by delineating a particular research question, self-reported pain and outcomes (e.g., physical and social functional limitations). The question
OF
self-reported pain and outcomes (e.g., physical and social functional limitations). The question crystallized further after a preliminary literature review (Horgas et al., 2008). Table 2-3 identi- OF
crystallized further after a preliminary literature review (Horgas et al., 2008). Table 2-3 identi- fies the components of this research question as they relate to and are congruent with the fies the components of this research question as they relate to and are congruent with the^ OF
ELSEVIER
and dependent variables, and imply the testability of the research question.
ELSEVIER
and dependent variables, and imply the testability of the research question. Now that the elements of the formal research question have been presented in greater detail,
ELSEVIER
Now that the elements of the formal research question have been presented in greater detail, this information can be integrated by formulating a formal research question about whether
ELSEVIER
this information can be integrated by formulating a formal research question about whether self-reported pain (pain sites and pain intensity) and disability (physical and social functional
ELSEVIER
self-reported pain (pain sites and pain intensity) and disability (physical and social functional limitations) differ between black and white older adults. This research question was originally
ELSEVIER
limitations) differ between black and white older adults. This research question was originally derived from a general area of interest—pain experiences of older adults of different racial
ELSEVIER
derived from a general area of interest—pain experiences of older adults of different racial groups. The topic was more specifically defined by delineating a particular research question, ELSEVIER
groups. The topic was more specifically defined by delineating a particular research question, self-reported pain and outcomes (e.g., physical and social functional limitations). The question self-reported pain and outcomes (e.g., physical and social functional limitations). The question^ ELSEVIER
SAMPLESAMPLESAMPLESAMPLESAMPLESAMPLESAMPLESAMPLE
that
SAMPLE
that research
SAMPLE
research question from the
SAMPLE
question from the title of the report, the abstract, the introduction, or the purpose.
SAMPLE
title of the report, the abstract, the introduction, or the purpose. your
SAMPLE
your focused
SAMPLE
focused
CONTENT
derived from a general area of interest—pain experiences of older adults of different racial
CONTENT
derived from a general area of interest—pain experiences of older adults of different racial groups. The topic was more specifically defined by delineating a particular research question,
CONTENT
groups. The topic was more specifically defined by delineating a particular research question, self-reported pain and outcomes (e.g., physical and social functional limitations). The question
CONTENT
self-reported pain and outcomes (e.g., physical and social functional limitations). The question crystallized further after a preliminary literature review (Horgas et al., 2008). Table 2-3 identi-
CONTENT
crystallized further after a preliminary literature review (Horgas et al., 2008). Table 2-3 identi- fies the components of this research question as they relate to and are congruent with the
CONTENT
fies the components of this research question as they relate to and are congruent with the three research question criteria.
CONTENT
three research question criteria.
CONTENTCONTENTCONTENT
derived from a general area of interest—pain experiences of older adults of different racial
derived from a general area of interest—pain experiences of older adults of different racial groups. The topic was more specifically defined by delineating a particular research question,groups. The topic was more specifically defined by delineating a particular research question,^ -
NOT
Now that the elements of the formal research question have been presented in greater detail,
NOT
Now that the elements of the formal research question have been presented in greater detail, this information can be integrated by formulating a formal research question about whether
NOT
this information can be integrated by formulating a formal research question about whether self-reported pain (pain sites and pain intensity) and disability (physical and social functional
NOT
self-reported pain (pain sites and pain intensity) and disability (physical and social functional limitations) differ between black and white older adults. This research question was originally NOT
limitations) differ between black and white older adults. This research question was originally derived from a general area of interest—pain experiences of older adults of different racial derived from a general area of interest—pain experiences of older adults of different racial^ NOT
FINAL
postoperative
FINAL
postoperative patients?
FINAL
patients? the relationship between the variables, identify the independent
FINAL
the relationship between the variables, identify the independent and dependent variables, and imply the testability of the research question.
FINAL
and dependent variables, and imply the testability of the research question. Now that the elements of the formal research question have been presented in greater detail, FINAL
Now that the elements of the formal research question have been presented in greater detail, this information can be integrated by formulating a formal research question about whether this information can be integrated by formulating a formal research question about whether^ FINAL
CHapteR 2 research Questions, hypotheses, and clinical Questions 39
the study by Meneses and colleagues (2007), the hypotheses are embedded in the Methods section of the article; you must interpret that the statement, “The overall hypotheses are to determine the effect of the breast cancer psychoeducational intervention (BCEI) on overall quality of life (QOL) and on the individual quality of life (QOL) and to examine whether the effects of the intervention were durable over time,” represents the hypotheses that test the effect of the BCEI on QOL in female breast cancer survivors. In light of that stylistic reality, it is important for you to be acquainted with the components of hypotheses, how they are developed, and the standards for writing and evaluating them. Hypotheses flow from the research question, literature review, and theoretical framework. Figure 2-2 illustrates this flow. A hypothesis is a statement about the relationship between two or more variables that suggests an answer to the research question. A hypothesis is a declarative statement that predicts an expected outcome. It explains or predicts the relationship or differences between two or more variables in terms of expected results or outcomes of a study. Hypotheses are formulated before the study is actually conducted because they provide direction for the collection, analysis, and interpretation of data.
Figure 2-2 Interrelationships of
research question, literature review,
theoretical framework, and
hypothesis.
Hypothesis Theoretical framework
Literature review
Problem statement
HELPFUL HINT
when hypotheses are not explicitly stated by the author at the end of the introduction section or just before the Methods section, they will be embedded or implied in the results or discussion section of a research article.
Characteristics
Nurses who are conducting research or critiquing published research studies must have a working knowledge about what constitutes a “good” hypothesis. Such knowledge will enable them to have a standard for evaluating their own work or the work of others. The following discussion about the characteristics of hypotheses presents criteria to be used when formulating or evaluating a hypothesis.
PROPERTY
it is important for you to be acquainted with the components of hypotheses, how they are
PROPERTY
it is important for you to be acquainted with the components of hypotheses, how they are developed, and the standards for writing and evaluating them.
PROPERTY
developed, and the standards for writing and evaluating them.
PROPERTY
Hypotheses flow from the research question, literature review, and theoretical framework.
PROPERTY
Hypotheses flow from the research question, literature review, and theoretical framework. Figure 2-2 illustrates this flow. A
PROPERTY
Figure 2-2 illustrates this flow. A two or more variables that suggests an answer to the research question. A hypothesis is a
PROPERTY
two or more variables that suggests an answer to the research question. A hypothesis is a declarative statement that predicts an expected outcome. It explains or predicts the relationship
PROPERTY
declarative statement that predicts an expected outcome. It explains or predicts the relationship
PROPERTY
or differences between two or more variables in terms of expected results or outcomes of a
PROPERTY
or differences between two or more variables in terms of expected results or outcomes of a study. Hypotheses are formulated before the study is actually conducted because they provide PROPERTY
study. Hypotheses are formulated before the study is actually conducted because they provide direction for the collection, analysis, and interpretation of data. direction for the collection, analysis, and interpretation of data.^ PROPERTY
OF
the effects of the intervention were durable over time,” represents the hypotheses that test the
OF
the effects of the intervention were durable over time,” represents the hypotheses that test the effect of the BCEI on QOL in female breast cancer survivors. In light of that stylistic reality,
OF
effect of the BCEI on QOL in female breast cancer survivors. In light of that stylistic reality, it is important for you to be acquainted with the components of hypotheses, how they are it is important for you to be acquainted with the components of hypotheses, how they are OF
ELSEVIER
the study by Meneses and colleagues (2007), the hypotheses are embedded in the Methods
ELSEVIER
the study by Meneses and colleagues (2007), the hypotheses are embedded in the Methods section of the article; you must interpret that the statement, “The overall hypotheses are to
ELSEVIER
section of the article; you must interpret that the statement, “The overall hypotheses are to determine the effect of the breast cancer psychoeducational intervention (BCEI) on overall
ELSEVIER
determine the effect of the breast cancer psychoeducational intervention (BCEI) on overall quality of life (QOL) and on the individual quality of life (QOL) and to examine whether
ELSEVIER
quality of life (QOL) and on the individual quality of life (QOL) and to examine whether the effects of the intervention were durable over time,” represents the hypotheses that test the the effects of the intervention were durable over time,” represents the hypotheses that test the ELSEVIER
SAMPLE
Figure 2-2 illustrates this flow. A
SAMPLE
Figure 2-2 illustrates this flow. A two or more variables that suggests an answer to the research question. A hypothesis is a
SAMPLE
two or more variables that suggests an answer to the research question. A hypothesis is a declarative statement that predicts an expected outcome. It explains or predicts the relationship
SAMPLE
declarative statement that predicts an expected outcome. It explains or predicts the relationship
SAMPLE
or differences between two or more variables in terms of expected results or outcomes of a
SAMPLE
or differences between two or more variables in terms of expected results or outcomes of a study. Hypotheses are formulated before the study is actually conducted because they provide
SAMPLE
study. Hypotheses are formulated before the study is actually conducted because they provide
SAMPLE
direction for the collection, analysis, and interpretation of data.
SAMPLE
direction for the collection, analysis, and interpretation of data.
CONTENT
quality of life (QOL) and on the individual quality of life (QOL) and to examine whether
CONTENT
quality of life (QOL) and on the individual quality of life (QOL) and to examine whether the effects of the intervention were durable over time,” represents the hypotheses that test the
CONTENT
the effects of the intervention were durable over time,” represents the hypotheses that test the effect of the BCEI on QOL in female breast cancer survivors. In light of that stylistic reality,
CONTENT
effect of the BCEI on QOL in female breast cancer survivors. In light of that stylistic reality, it is important for you to be acquainted with the components of hypotheses, how they are
CONTENT
it is important for you to be acquainted with the components of hypotheses, how they are developed, and the standards for writing and evaluating them.
CONTENT
developed, and the standards for writing and evaluating them.
CONTENT
Hypotheses flow from the research question, literature review, and theoretical framework.
CONTENT
Hypotheses flow from the research question, literature review, and theoretical framework. Figure 2-2 illustrates this flow. A CONTENT
Figure 2-2 illustrates this flow. A two or more variables that suggests an answer to the research question. A hypothesis is a two or more variables that suggests an answer to the research question. A hypothesis is a^ CONTENT
quality of life (QOL) and on the individual quality of life (QOL) and to examine whetherquality of life (QOL) and on the individual quality of life (QOL) and to examine whether - -
NOT
the study by Meneses and colleagues (2007), the hypotheses are embedded in the Methods
NOT
the study by Meneses and colleagues (2007), the hypotheses are embedded in the Methods section of the article; you must interpret that the statement, “The overall hypotheses are to
NOT
section of the article; you must interpret that the statement, “The overall hypotheses are to determine the effect of the breast cancer psychoeducational intervention (BCEI) on overall determine the effect of the breast cancer psychoeducational intervention (BCEI) on overall NOT
FINAL
40 paRt i OvErviEw Of rEsEarch and EvidEncE-BasEd PracticE
Relationship Statement
The first characteristic of a hypothesis is that it is a declarative statement that identifies the predicted relationship between two or more variables. This implies that there is a systematic relationship between an independent variable (X) and a dependent variable (Y). The direction of the predicted relationship is also specified in this statement. Phrases such as greater than; less than; positively, negatively, or curvilinearly related; and difference in connote the directional- ity that is proposed in the hypothesis. The following is an example of a directional hypothesis: “The rate of continuous smoking abstinence (dependent variable) at 6 months postpartum, based on self-report and biochemical validation, will be significantly higher in the treatment group (postpartum counseling intervention) than in the control group (independent vari- able).” The dependent and independent variables are explicitly identified, and the relational aspect of the prediction in the hypothesis is contained in the phrase significantly higher than. The nature of the relationship, either causal or associative, is also implied by the hypothesis. A causal relationship is one in which the researcher can predict that the indepen- dent variable (X) causes a change in the dependent variable (Y). In research, it is rare that one is in a firm enough position to take a definitive stand about a cause-and-effect relation- ship. For example, a researcher might hypothesize that blood pressure telemonitoring plus usual care would lead to a greater reduction in blood pressure than usual care alone from baseline over a 12-month follow-up (Artinian et al., 2007). It would be difficult for a researcher to predict a strong cause-and-effect relationship, however, because of the multiple intervening variables (e.g., age, medication, and lifestyle changes) that might also influence the subject’s health status. Variables are more commonly related in noncausal ways; that is, the variables are system- atically related but in an associative way. This means that the variables change in relation to each other. For example, there is strong evidence that asbestos exposure is related to lung cancer. It is tempting to state that there is a causal relationship between asbestos exposure and lung cancer. Do not overlook the fact, however, that not all of those who have been ex - posed to asbestos will have lung cancer and not all of those who have lung cancer have had asbestos exposure. Consequently, it would be scientifically unsound to take a position advocat- ing the presence of a causal relationship between these two variables. Rather, one can say only that there is an associative relationship between the variables of asbestos exposure and lung cancer, a relationship in which there is a strong systematic association between the two phenomena.
Testability
The second characteristic of a hypothesis is its testability. This means that the variables of the study must lend themselves to observation, measurement, and analysis. The hypothesis is either supported or not supported after the data have been collected and analyzed. The pre- dicted outcome proposed by the hypothesis will or will not be congruent with the actual outcome when the hypothesis is tested. Hypotheses advance scientific knowledge by confirm- ing or refuting theories. Hypotheses may fail to meet the criteria of testability because the researcher has not made a prediction about the anticipated outcome, the variables are not observable or measurable, or the hypothesis is couched in terms that are value-laden.
PROPERTY
but in an associative way. This means that the variables change in relation to
PROPERTY
but in an associative way. This means that the variables change in relation to each other. For example, there is strong evidence that asbestos exposure is related to lung
PROPERTY
each other. For example, there is strong evidence that asbestos exposure is related to lung cancer. It is tempting to state that there is a causal relationship between asbestos exposure
PROPERTY
cancer. It is tempting to state that there is a causal relationship between asbestos exposure and lung cancer. Do not overlook the fact, however, that not all of those who have been ex
PROPERTY
and lung cancer. Do not overlook the fact, however, that not all of those who have been ex posed to asbestos
PROPERTY
posed to asbestos
PROPERTY
will have lung cancer and not all of those who have lung cancer have had
PROPERTY
will have lung cancer and not all of those who have lung cancer have had asbestos exposure. Consequently, it would be scientifically unsound to take a position advocat-
PROPERTY
asbestos exposure. Consequently, it would be scientifically unsound to take a position advocat- ing the presence of a causal relationship between these two variables. Rather, one can say only
PROPERTY
ing the presence of a causal relationship between these two variables. Rather, one can say only that there is an associative relationship between the variables of asbestos exposure and PROPERTY
that there is an associative relationship between the variables of asbestos exposure and lung cancer, a relationship in which there is a strong systematic association between the two lung cancer, a relationship in which there is a strong systematic association between the two PROPERTY
OF
Variables are more commonly related in noncausal ways; that is, the variables are system
OF
Variables are more commonly related in noncausal ways; that is, the variables are system but in an associative way. This means that the variables change in relation to OF
but in an associative way. This means that the variables change in relation to each other. For example, there is strong evidence that asbestos exposure is related to lung each other. For example, there is strong evidence that asbestos exposure is related to lung^ OF
ELSEVIER
one is in a firm enough position to take a definitive stand about a cause-and-effect relation-
ELSEVIER
one is in a firm enough position to take a definitive stand about a cause-and-effect relation- ship. For example, a researcher might hypothesize that blood pressure telemonitoring plus
ELSEVIER
ship. For example, a researcher might hypothesize that blood pressure telemonitoring plus usual care would lead to a greater reduction in blood pressure than usual care alone from
ELSEVIER
usual care would lead to a greater reduction in blood pressure than usual care alone from baseline over a 12-month follow-up (Artinian et al., 2007). It would be difficult for a
ELSEVIER
baseline over a 12-month follow-up (Artinian et al., 2007). It would be difficult for a researcher to predict a strong cause-and-effect relationship, however, because of the multiple
ELSEVIER
researcher to predict a strong cause-and-effect relationship, however, because of the multiple intervening variables (e.g., age, medication, and lifestyle changes) that might also influence
ELSEVIER
intervening variables (e.g., age, medication, and lifestyle changes) that might also influence
Variables are more commonly related in noncausal ways; that is, the variables are system Variables are more commonly related in noncausal ways; that is, the variables are system^ ELSEVIER
SAMPLE
will have lung cancer and not all of those who have lung cancer have had
SAMPLE
will have lung cancer and not all of those who have lung cancer have had asbestos exposure. Consequently, it would be scientifically unsound to take a position advocat-
SAMPLE
asbestos exposure. Consequently, it would be scientifically unsound to take a position advocat- ing the presence of a causal relationship between these two variables. Rather, one can say only
SAMPLE
ing the presence of a causal relationship between these two variables. Rather, one can say only that there is an associative relationship between the variables of asbestos exposure and
SAMPLE
that there is an associative relationship between the variables of asbestos exposure and
SAMPLE
lung cancer, a relationship in which there is a strong systematic association between the two
SAMPLE
lung cancer, a relationship in which there is a strong systematic association between the two phenomena. SAMPLE
phenomena.
CONTENT
intervening variables (e.g., age, medication, and lifestyle changes) that might also influence
CONTENT
intervening variables (e.g., age, medication, and lifestyle changes) that might also influence
Variables are more commonly related in noncausal ways; that is, the variables are system
CONTENT
Variables are more commonly related in noncausal ways; that is, the variables are system but in an associative way. This means that the variables change in relation to
CONTENT
but in an associative way. This means that the variables change in relation to each other. For example, there is strong evidence that asbestos exposure is related to lung
CONTENT
each other. For example, there is strong evidence that asbestos exposure is related to lung cancer. It is tempting to state that there is a causal relationship between asbestos exposure
CONTENT
cancer. It is tempting to state that there is a causal relationship between asbestos exposure and lung cancer. Do not overlook the fact, however, that not all of those who have been ex CONTENT
and lung cancer. Do not overlook the fact, however, that not all of those who have been ex will have lung cancer and not all of those who have lung cancer have had will have lung cancer and not all of those who have lung cancer have had CONTENT
intervening variables (e.g., age, medication, and lifestyle changes) that might also influence
intervening variables (e.g., age, medication, and lifestyle changes) that might also influence^ NOT
ship. For example, a researcher might hypothesize that blood pressure telemonitoring plus
NOT
ship. For example, a researcher might hypothesize that blood pressure telemonitoring plus usual care would lead to a greater reduction in blood pressure than usual care alone from
NOT
usual care would lead to a greater reduction in blood pressure than usual care alone from baseline over a 12-month follow-up (Artinian et al., 2007). It would be difficult for a
NOT
baseline over a 12-month follow-up (Artinian et al., 2007). It would be difficult for a researcher to predict a strong cause-and-effect relationship, however, because of the multiple NOT
researcher to predict a strong cause-and-effect relationship, however, because of the multiple intervening variables (e.g., age, medication, and lifestyle changes) that might also influence intervening variables (e.g., age, medication, and lifestyle changes) that might also influence^ NOT
FINAL
The nature of the relationship, either causal or associative, is also implied by the
FINAL
The nature of the relationship, either causal or associative, is also implied by the hypothesis. A causal relationship is one in which the researcher can predict that the indepen-
FINAL
hypothesis. A causal relationship is one in which the researcher can predict that the indepen- (Y).
FINAL
(Y). In research, it is rare that
FINAL
In research, it is rare that one is in a firm enough position to take a definitive stand about a cause-and-effect relation-
FINAL
one is in a firm enough position to take a definitive stand about a cause-and-effect relation- ship. For example, a researcher might hypothesize that blood pressure telemonitoring plus FINAL
ship. For example, a researcher might hypothesize that blood pressure telemonitoring plus usual care would lead to a greater reduction in blood pressure than usual care alone from usual care would lead to a greater reduction in blood pressure than usual care alone from^ FINAL
42 paRt i OvErviEw Of rEsEarch and EvidEncE-BasEd PracticE
taBLe 2-4 Examples of How Hypotheses Are Worded
Variables * Hypothesis
Type of Design; Level of Evidence Suggested
1. THERE ARE SIGNIFICANT DIFFERENCES IN SELF-REPORTED CANCER PAIN, SYMPTOMS ACCOMPANYING PAIN, AND FUNCTIONAL STATUS ACCORDING TO SELF-REPORTED ETHNIC IDENTITY. iv: Ethnic identity dv: self-reported cancer pain dv: symptoms accompanying pain dv: functional status
nondirectional, research nonexperimental; Level iv
2. INDIVIDUALS WHO PARTICIPATE IN USUAL CARE (UC) PLUS BLOOD PRESSURE (BP) TELEMONITORING (TM) WILL HAVE A GREATER REDUCTION IN BP FROM BASELINE TO 12-MONTH FOLLOW-UP THAN WOULD INDIVIDUALS WHO RECEIVE UC ONLY. iv: telemonitoring (tM) iv: Usual care (Uc) dv: Blood pressure
directional, research Experimental; Level ii
3. THERE WILL BE A GREATER DECREASE IN STATE ANXIETY SCORES FOR PATIENTS RECEIVING STRUCTURED INFORMATIONAL VIDEOS BEFORE ABDOMINAL OR CHEST TUBE REMOVAL THAN FOR PATIENTS RECEIVING STANDARD INFORMATION. iv: Preprocedure structured videotape information iv: standard information dv: state anxiety
directional, research Experimental; Level ii
4. THE INCIDENCE AND DEGREE OF SEVERITY OF SUBJECT DISCOMFORT WILL BE LESS AFTER ADMINISTRATION OF MEDICATIONS BY THE Z-TRACK INTRAMUSCULAR INJECTION TECHNIQUE THAN AFTER ADMINISTRATION OF MEDICATIONS BY THE STANDARD INTRAMUSCULAR INJECTION TECHNIQUE. iv: Z-track intramuscular injection technique iv: standard intramuscular injection technique dv: subject discomfort
directional, research Experimental; Level ii
5. NURSES WITH HIGH SOCIAL SUPPORT FROM CO-WORKERS HAVE LOW PERCEIVED JOB STRESS. iv: social support dv: Perceived job stress
directional, research nonexperimental; Level iv
6. THERE WILL BE NO DIFFERENCE IN ANESTHETIC COMPLICATION RATES BETWEEN HOSPITALS THAT RELY PRIMARILY ON CERTIFIED REGISTERED NURSE ANESTHETIST (CRNA) OBSTETRICAL ANESTHESIA VERSUS THOSE THAT RELY PRIMARILY ON ANESTHESIOLOGISTS. iv: type of anesthesia provider (crna or Md) dv: anesthesia complication rates
nondirectional; null nonexperimental; Level iv
7. THERE WILL BE NO SIGNIFICANT DIFFERENCE IN THE DURATION OF PATENCY OF A 24-GAUGE INTRAVENOUS LOCK IN A NEONATAL PATIENT WHEN FLUSHED WITH 0.5 ml OF HEPARINIZED SALINE (2 U/ml), STANDARD PRACTICE, COMPARED WITH 0.5 ml OF 0.9% NORMAL SALINE. iv: heparinized saline iv: normal saline dv: duration of patency of intravenous lock
nondirectional; null Experimental; Level ii
*abbreviations: IV, independent variable; DV, dependent variable.
PROPERTY
HIGH SOCIAL SUPPORT FROM CO-WORKERS HAVE LOW PERCEIVED JOB STRESS.
PROPERTY
HIGH SOCIAL SUPPORT FROM CO-WORKERS HAVE LOW PERCEIVED JOB STRESS.
BE NO DIFFERENCE IN ANESTHETIC COMPLICATION RATES BETWEEN HOSPITALS THAT RELY PRIMARILY PROPERTY
BE NO DIFFERENCE IN ANESTHETIC COMPLICATION RATES BETWEEN HOSPITALS THAT RELY PRIMARILY ON CERTIFIED REGISTERED NURSE ANESTHETIST (CRNA) OBSTETRICAL ANESTHESIA VERSUS THOSE THAT RELY ON CERTIFIED REGISTERED NURSE ANESTHETIST (CRNA) OBSTETRICAL ANESTHESIA VERSUS THOSE THAT RELY PROPERTY
OF
MEDICATIONS BY THE Z-TRACK INTRAMUSCULAR INJECTION TECHNIQUE THAN AFTER ADMINISTRATION OF
OF
MEDICATIONS BY THE Z-TRACK INTRAMUSCULAR INJECTION TECHNIQUE THAN AFTER ADMINISTRATION OF MEDICATIONS BY THE STANDARD INTRAMUSCULAR INJECTION TECHNIQUE. OF
MEDICATIONS BY THE STANDARD INTRAMUSCULAR INJECTION TECHNIQUE.
ELSEVIERELSEVIER
BE A GREATER DECREASE IN STATE ANXIETY SCORES FOR PATIENTS RECEIVING STRUCTURED
ELSEVIER
BE A GREATER DECREASE IN STATE ANXIETY SCORES FOR PATIENTS RECEIVING STRUCTURED INFORMATIONAL VIDEOS BEFORE ABDOMINAL OR CHEST TUBE REMOVAL THAN FOR PATIENTS RECEIVING STANDARD
ELSEVIER
INFORMATIONAL VIDEOS BEFORE ABDOMINAL OR CHEST TUBE REMOVAL THAN FOR PATIENTS RECEIVING STANDARD
irectional, research
ELSEVIER
irectional, research
AND DEGREE OF SEVERITY OF SUBJECT DISCOMFORT WILL BE LESS AFTER ADMINISTRATION OF ELSEVIER
AND DEGREE OF SEVERITY OF SUBJECT DISCOMFORT WILL BE LESS AFTER ADMINISTRATION OF MEDICATIONS BY THE Z-TRACK INTRAMUSCULAR INJECTION TECHNIQUE THAN AFTER ADMINISTRATION OF MEDICATIONS BY THE Z-TRACK INTRAMUSCULAR INJECTION TECHNIQUE THAN AFTER ADMINISTRATION OF^ ELSEVIER
SAMPLE
HIGH SOCIAL SUPPORT FROM CO-WORKERS HAVE LOW PERCEIVED JOB STRESS.
SAMPLE
HIGH SOCIAL SUPPORT FROM CO-WORKERS HAVE LOW PERCEIVED JOB STRESS.
BE NO DIFFERENCE IN ANESTHETIC COMPLICATION RATES BETWEEN HOSPITALS THAT RELY PRIMARILY
SAMPLE
BE NO DIFFERENCE IN ANESTHETIC COMPLICATION RATES BETWEEN HOSPITALS THAT RELY PRIMARILY ON CERTIFIED REGISTERED NURSE ANESTHETIST (CRNA) OBSTETRICAL ANESTHESIA VERSUS THOSE THAT RELY
SAMPLE
ON CERTIFIED REGISTERED NURSE ANESTHETIST (CRNA) OBSTETRICAL ANESTHESIA VERSUS THOSE THAT RELY
SAMPLE
PRIMARILY ON ANESTHESIOLOGISTS. SAMPLE
PRIMARILY ON ANESTHESIOLOGISTS. crna crna^ SAMPLE
CONTENTCONTENT
AND DEGREE OF SEVERITY OF SUBJECT DISCOMFORT WILL BE LESS AFTER ADMINISTRATION OF
CONTENT
AND DEGREE OF SEVERITY OF SUBJECT DISCOMFORT WILL BE LESS AFTER ADMINISTRATION OF MEDICATIONS BY THE Z-TRACK INTRAMUSCULAR INJECTION TECHNIQUE THAN AFTER ADMINISTRATION OF
CONTENT
MEDICATIONS BY THE Z-TRACK INTRAMUSCULAR INJECTION TECHNIQUE THAN AFTER ADMINISTRATION OF MEDICATIONS BY THE STANDARD INTRAMUSCULAR INJECTION TECHNIQUE.
CONTENT
MEDICATIONS BY THE STANDARD INTRAMUSCULAR INJECTION TECHNIQUE. d
CONTENT
d
CONTENT
irectional, research
CONTENT
irectional, research
HIGH SOCIAL SUPPORT FROM CO-WORKERS HAVE LOW PERCEIVED JOB STRESS. HIGH SOCIAL SUPPORT FROM CO-WORKERS HAVE LOW PERCEIVED JOB STRESS. CONTENT
AND DEGREE OF SEVERITY OF SUBJECT DISCOMFORT WILL BE LESS AFTER ADMINISTRATION OFAND DEGREE OF SEVERITY OF SUBJECT DISCOMFORT WILL BE LESS AFTER ADMINISTRATION OF^ -^ -
NOT
irectional, research
NOT
irectional, research
FINALFINAL
BE A GREATER DECREASE IN STATE ANXIETY SCORES FOR PATIENTS RECEIVING STRUCTURED
FINAL
BE A GREATER DECREASE IN STATE ANXIETY SCORES FOR PATIENTS RECEIVING STRUCTURED INFORMATIONAL VIDEOS BEFORE ABDOMINAL OR CHEST TUBE REMOVAL THAN FOR PATIENTS RECEIVING STANDARD FINAL
INFORMATIONAL VIDEOS BEFORE ABDOMINAL OR CHEST TUBE REMOVAL THAN FOR PATIENTS RECEIVING STANDARD
CHapteR 2 research Questions, hypotheses, and clinical Questions 43
not statistically significant, the hypothesis was not supported, thereby indicating that the REST intervention did not significantly reduce parenting stress for parents with irritable infants. The examples in Table 2-4 represent research hypotheses. A statistical hypothesis, also known as a null hypothesis, states that there is no relation- ship between the independent and dependent variables. The examples in Table 2-5 illustrate statistical hypotheses. If, in the data analysis, a statistically significant relationship emerges between the variables at a specified level of significance, the null hypothesis is rejected. Rejec- tion of the statistical hypothesis is equivalent to acceptance of the research hypothesis. For example, a study by Simonson and colleagues (2007) that sought to identify differences in the rates of anesthetic complications in hospitals whose obstetrical anesthesia is provided solely by certified registered nurse anesthetists (CRNAs) compared to hospitals with only anesthe- siologists. The null hypothesis, that there would be no differences in anesthetic complication rates between the hospitals that rely primarily on CRNA obstetrical anesthesia versus those that rely primarily on anesthesiologists, was supported, thereby indicating that there were no differences in anesthesia-related complications according to the type of provider, nurse anes- thetist or physician. Because the difference in outcomes was not greater than expected by chance, the null hypothesis was accepted (see Chapter 18). Some researchers refer to the null hypothesis as a statistical contrivance that obscures a straightforward prediction of the outcome. Others state that it is more exact and conservative statistically, and that failure to reject the null hypothesis implies that there is insufficient evidence to support the idea of a real difference. You will note that when hypotheses are stated, research hypotheses are generally used more often than statistical hypotheses because they are more desirable to state the researcher’s expectation. Readers then have a more precise idea of the proposed outcome. In any study that involves statistical analysis, the underlying null hypothesis is usually assumed without being explicitly stated.
Directional versus Nondirectional Hypotheses
Hypotheses can be formulated directionally or nondirectionally. A directional hypothesis is one that specifies the expected direction of the relationship between the independent and
taBLe 2-5 Examples of Statistical Hypotheses
Hypothesis Variables * Type of Hypothesis
Type of Design Suggested Oxygen inhalation by nasal cannula of up to 6 L/min does not affect oral temperature measurement taken with an electronic thermometer.
iv: Oxygen inhalation by nasal cannula dv: Oral temperature
statistical; null Experimental
there will be no difference in the performance accuracy of adult nurse practitioners (anPs) and family nurse practitioners (fnPs) in formulating accurate diagnoses and acceptable interventions for suspected cases of domestic violence.
iv: nurse practitioner (anP or fnP) category dv: diagnosis and intervention performance accuracy
statistical; null nonexperimental
*abbreviations: IV, independent variable; DV, dependent variable.
PROPERTY
tion of the statistical hypothesis is equivalent to acceptance of the research hypothesis. For
PROPERTY
tion of the statistical hypothesis is equivalent to acceptance of the research hypothesis. For example, a study by Simonson and colleagues (2007) that sought to identify differences in
PROPERTY
example, a study by Simonson and colleagues (2007) that sought to identify differences in the rates of anesthetic complications in hospitals whose obstetrical anesthesia is provided solely
PROPERTY
the rates of anesthetic complications in hospitals whose obstetrical anesthesia is provided solely by certified registered nurse anesthetists (CRNAs) compared to hospitals with only anesthe-
PROPERTY
by certified registered nurse anesthetists (CRNAs) compared to hospitals with only anesthe- siologists. The null hypothesis, that there would be no differences in anesthetic complication
PROPERTY
siologists. The null hypothesis, that there would be no differences in anesthetic complication rates between the hospitals that rely primarily on CRNA obstetrical anesthesia versus those
PROPERTY
rates between the hospitals that rely primarily on CRNA obstetrical anesthesia versus those that rely primarily on anesthesiologists, was supported, thereby indicating that there were no
PROPERTY
that rely primarily on anesthesiologists, was supported, thereby indicating that there were no
PROPERTY
differences in anesthesia-related complications according to the type of provider, nurse anes- PROPERTY
differences in anesthesia-related complications according to the type of provider, nurse anes- thetist or physician. Because the difference in outcomes was not greater than expected by thetist or physician. Because the difference in outcomes was not greater than expected by PROPERTY
OF
f, in the data analysis, a statistically significant relationship emerges
OF
f, in the data analysis, a statistically significant relationship emerges between the variables at a specified level of significance, the null hypothesis is rejected. Rejec-
OF
between the variables at a specified level of significance, the null hypothesis is rejected. Rejec- tion of the statistical hypothesis is equivalent to acceptance of the research hypothesis. For OF
tion of the statistical hypothesis is equivalent to acceptance of the research hypothesis. For example, a study by Simonson and colleagues (2007) that sought to identify differences in example, a study by Simonson and colleagues (2007) that sought to identify differences in^ OF
ELSEVIER
not statistically significant, the hypothesis was not supported, thereby indicating that the
ELSEVIER
not statistically significant, the hypothesis was not supported, thereby indicating that the REST intervention did not significantly reduce parenting stress for parents with irritable
ELSEVIER
REST intervention did not significantly reduce parenting stress for parents with irritable infants. The examples in Table 2-4 represent research hypotheses.
ELSEVIER
infants. The examples in Table 2-4 represent research hypotheses. also known as a null hypothesis, states that there is no relation-
ELSEVIER
also known as a null hypothesis, states that there is no relation- and
ELSEVIER
and dependent
ELSEVIER
dependent f, in the data analysis, a statistically significant relationship emerges ELSEVIER
f, in the data analysis, a statistically significant relationship emerges between the variables at a specified level of significance, the null hypothesis is rejected. Rejec- between the variables at a specified level of significance, the null hypothesis is rejected. Rejec-^ ELSEVIER
SAMPLE
siologists. The null hypothesis, that there would be no differences in anesthetic complication
SAMPLE
siologists. The null hypothesis, that there would be no differences in anesthetic complication rates between the hospitals that rely primarily on CRNA obstetrical anesthesia versus those
SAMPLE
rates between the hospitals that rely primarily on CRNA obstetrical anesthesia versus those that rely primarily on anesthesiologists, was supported, thereby indicating that there were no
SAMPLE
that rely primarily on anesthesiologists, was supported, thereby indicating that there were no
SAMPLE
differences in anesthesia-related complications according to the type of provider, nurse anes-
SAMPLE
differences in anesthesia-related complications according to the type of provider, nurse anes- thetist or physician. Because the difference in outcomes was not greater than expected by
SAMPLE
thetist or physician. Because the difference in outcomes was not greater than expected by chance, the null hypothesis was accepted (see Chapter 18). SAMPLE
chance, the null hypothesis was accepted (see Chapter 18). Some researchers refer to the null hypothesis as a statistical contrivance that obscures a Some researchers refer to the null hypothesis as a statistical contrivance that obscures a^ SAMPLE
CONTENT
dependent
CONTENT
dependent f, in the data analysis, a statistically significant relationship emerges
CONTENT
f, in the data analysis, a statistically significant relationship emerges between the variables at a specified level of significance, the null hypothesis is rejected. Rejec-
CONTENT
between the variables at a specified level of significance, the null hypothesis is rejected. Rejec- tion of the statistical hypothesis is equivalent to acceptance of the research hypothesis. For
CONTENT
tion of the statistical hypothesis is equivalent to acceptance of the research hypothesis. For example, a study by Simonson and colleagues (2007) that sought to identify differences in
CONTENT
example, a study by Simonson and colleagues (2007) that sought to identify differences in the rates of anesthetic complications in hospitals whose obstetrical anesthesia is provided solely
CONTENT
the rates of anesthetic complications in hospitals whose obstetrical anesthesia is provided solely by certified registered nurse anesthetists (CRNAs) compared to hospitals with only anesthe- CONTENT
by certified registered nurse anesthetists (CRNAs) compared to hospitals with only anesthe- siologists. The null hypothesis, that there would be no differences in anesthetic complication siologists. The null hypothesis, that there would be no differences in anesthetic complication CONTENT
variables.
variables. f, in the data analysis, a statistically significant relationship emergesf, in the data analysis, a statistically significant relationship emerges^ -
NOT
not statistically significant, the hypothesis was not supported, thereby indicating that the
NOT
not statistically significant, the hypothesis was not supported, thereby indicating that the REST intervention did not significantly reduce parenting stress for parents with irritable
NOT
REST intervention did not significantly reduce parenting stress for parents with irritable infants. The examples in Table 2-4 represent research hypotheses.
NOT
infants. The examples in Table 2-4 represent research hypotheses. also known as a null hypothesis, states that there is no relation- NOT
also known as a null hypothesis, states that there is no relation- variables. variables.^ NOT
FINAL
not statistically significant, the hypothesis was not supported, thereby indicating that the FINAL
not statistically significant, the hypothesis was not supported, thereby indicating that the REST intervention did not significantly reduce parenting stress for parents with irritable REST intervention did not significantly reduce parenting stress for parents with irritable^ FINAL
CHapteR 2 research Questions, hypotheses, and clinical Questions 45
RELATIONSHIP BETWEEN THE HYPOTHESIS, THE
RESEARCH QUESTION, AND THE RESEARCH DESIGN
Regardless of whether the researcher uses a statistical or a research hypothesis, there is a sug- gested relationship between the hypothesis, the research design of the study, and the level of evidence provided by the results of the study. The type of design, experimental or nonexperi- mental (see Chapters 8 and 9), will influence the wording of the hypothesis. For example, when an experimental design is used, the research consumer would expect to see hypotheses that reflect relationship statements, such as the following:
- X 1 is more effective than X 2 on Y.
- The effect of X 1 on Y is greater than that of X 2 on Y.
- The incidence of Y will not differ in subjects receiving X 1 and X 2 treatments.
- The incidence of Y will be greater in subjects after X 1 than after X 2. Such hypotheses indicate that an experimental treatment (i.e., independent variable X ) will be used and that two groups of subjects, experimental and control groups, are being used to test whether the difference in the outcome (i.e., dependent variable Y ) predicted by the hypothesis actually exists. Hypotheses reflecting experimental designs also test the effect of the experimental treatment (i.e., independent variable X ) on the outcome (i.e., dependent variable Y ). This would suggest that the strength of the evidence provided by the results would be Level II (experimental design) or Level III (quasi-experimental design). In contrast, hypotheses related to nonexperimental designs reflect associative relationship statements, such as the following:
- X will be negatively related to Y.
- There will be a positive relationship between X and Y. This would suggest that the strength of the evidence provided by the results of a study that examined hypotheses with associative relationship statements would be at Level IV (nonex- perimental design). Table 2-6 provides additional examples of this concept. The Critical Thinking Decision Path shown in the following diagram will help you determine the type of hypothesis presented in a study, as well as the study’s readiness for a hypothesis-testing design.
taBLe 2-6 Elements of a Clinical Question Population Intervention Comparison Intervention Outcome People with advanced cancer Pain diaries no pain diaries increased pain control
EVIDENCE-BASED PRACTICE TIP
think about the relationship between the wording of the hypothesis, the type of research design suggested, and the level of evidence provided by the findings of a study using each kind of hypothesis. You may want to consider which type of hypothesis potentially will yield the strongest results applicable to practice.
PROPERTY
hypothesis actually exists. Hypotheses reflecting experimental designs also test the effect of
PROPERTY
hypothesis actually exists. Hypotheses reflecting experimental designs also test the effect of the experimental treatment (i.e., independent variable
PROPERTY
the experimental treatment (i.e., independent variable
PROPERTY
). This would suggest that the strength of the evidence provided by the results would
PROPERTY
). This would suggest that the strength of the evidence provided by the results would be Level II (experimental design) or Level III (quasi-experimental design).
PROPERTY
be Level II (experimental design) or Level III (quasi-experimental design). In contrast, hypotheses related to nonexperimental designs reflect associative relationship
PROPERTY
In contrast, hypotheses related to nonexperimental designs reflect associative relationship statements, such as the following:
PROPERTY
statements, such as the following: X PROPERTY
X will be negatively related to PROPERTY
XX will be negatively related towill be negatively related to PROPERTY
XX will be negatively related to PROPERTY
- There PROPERTY
- There This would suggest that the strength of the evidence provided by the results of a study that PROPERTY This would suggest that the strength of the evidence provided by the results of a study that
OF
be used and that two groups of subjects, experimental and control groups, are being used to
OF
be used and that two groups of subjects, experimental and control groups, are being used to test whether the difference in the outcome (i.e., dependent variable OF
test whether the difference in the outcome (i.e., dependent variable hypothesis actually exists. Hypotheses reflecting experimental designs also test the effect of hypothesis actually exists. Hypotheses reflecting experimental designs also test the effect of OF
ELSEVIER
when an experimental design is used, the research consumer would expect to see hypotheses
ELSEVIER
when an experimental design is used, the research consumer would expect to see hypotheses that reflect relationship statements, such as the following:
ELSEVIER
that reflect relationship statements, such as the following:
is greater than that of
ELSEVIER
is greater than that of not differ in subjects receiving
ELSEVIER
not differ in subjects receiving be greater in subjects after
ELSEVIER
be greater in subjects after Such hypotheses indicate that an experimental treatment (i.e., independent variable ELSEVIER
Such hypotheses indicate that an experimental treatment (i.e., independent variable be used and that two groups of subjects, experimental and control groups, are being used to be used and that two groups of subjects, experimental and control groups, are being used to^ ELSEVIER
SAMPLE
be Level II (experimental design) or Level III (quasi-experimental design).
SAMPLE
be Level II (experimental design) or Level III (quasi-experimental design). In contrast, hypotheses related to nonexperimental designs reflect associative relationship
SAMPLE
In contrast, hypotheses related to nonexperimental designs reflect associative relationship statements, such as the following:
SAMPLE
statements, such as the following: will be negatively related to
SAMPLE
will be negatively related to
SAMPLE
SAMPLE
SAMPLE
will be
SAMPLE
be This would suggest that the strength of the evidence provided by the results of a study that SAMPLE
This would suggest that the strength of the evidence provided by the results of a study that examined hypotheses with associative relationship statements would be at Level IV (nonex- examined hypotheses with associative relationship statements would be at Level IV (nonex- SAMPLE
CONTENT
Such hypotheses indicate that an experimental treatment (i.e., independent variable
CONTENT
Such hypotheses indicate that an experimental treatment (i.e., independent variable be used and that two groups of subjects, experimental and control groups, are being used to
CONTENT
be used and that two groups of subjects, experimental and control groups, are being used to test whether the difference in the outcome (i.e., dependent variable
CONTENT
test whether the difference in the outcome (i.e., dependent variable hypothesis actually exists. Hypotheses reflecting experimental designs also test the effect of
CONTENT
hypothesis actually exists. Hypotheses reflecting experimental designs also test the effect of the experimental treatment (i.e., independent variable
CONTENT
the experimental treatment (i.e., independent variable
CONTENT
). This would suggest that the strength of the evidence provided by the results would CONTENT
). This would suggest that the strength of the evidence provided by the results would be Level II (experimental design) or Level III (quasi-experimental design). CONTENT be Level II (experimental design) or Level III (quasi-experimental design). In contrast, hypotheses related to nonexperimental designs reflect associative relationship CONTENT In contrast, hypotheses related to nonexperimental designs reflect associative relationship
be greater in subjects after
be greater in subjects after Such hypotheses indicate that an experimental treatment (i.e., independent variableSuch hypotheses indicate that an experimental treatment (i.e., independent variable^ -
NOT
X
NOT
X 2
NOT
XX 22
NOT
XX 2 on
NOT
on Y.
NOT
Y.
not differ in subjects receiving NOT
not differ in subjects receiving be greater in subjects after be greater in subjects after NOT
FINAL
evidence provided by the results of the study. The type of design, experimental or nonexperi-
FINAL
evidence provided by the results of the study. The type of design, experimental or nonexperi- mental (see Chapters 8 and 9), will influence the wording of the hypothesis. For example,
FINAL
mental (see Chapters 8 and 9), will influence the wording of the hypothesis. For example, when an experimental design is used, the research consumer would expect to see hypotheses
FINAL
when an experimental design is used, the research consumer would expect to see hypotheses
46 paRt i OvErviEw Of rEsEarch and EvidEncE-BasEd PracticE
CRITICAL THINKING DECISION PATH Determining the Type of Hypothesis or Readiness for
Hypothesis Testing
Examine the literature review and theoretical framework
Assess the primary purpose of the study and the explicit or implied research problem
Research or scientific hypothesis(es)
Directional hypothesis
Nondirectional hypothesis
Statistical or null hypothesis(es)
Is the primary purpose exploratory, descriptive, or hypothesis-generating?
Yes
Formulate research question(s)
No
Formulate hypothesis(es)
Yes No
Assess for presence of IV and DV, a predictive statement, testability, and appropriate theory base
Is the primary purpose testing causal or associative relationships?
PROPERTYPROPERTYPROPERTYPROPERTY
OF
ELSEVIERELSEVIERELSEVIERELSEVIERELSEVIERELSEVIER
SAMPLE
CONTENTCONTENTCONTENTCONTENTCONTENTCONTENTCONTENT
Research
CONTENT
Research or scientific
CONTENT
or scientific hypothesis(es) CONTENT
hypothesis(es) CONTENT
NOTNOTNOTNOTNOT
Yes Yes NOT
FINALFINALFINALFINALFINALFINALFINALFINALFINAL
Is the primary purpose
FINAL
Is the primary purpose testing causal or associative
FINAL
testing causal or associative relationships?
FINAL
relationships?
