PROBLEM
AND PURPOSE STATEMENTS.
RESEARCH QUESTIONS AND
HYPOTHESES
Overview Of Research Problems
Studies begin as problems that researchers want to solve or as questions
they want to answer. This chapter discusses the formulation and development of
research problems. We begin by clarifying some relevant terms.
Basic Terminology. At the most general level,
a researcher selects a topic or a phenomenon on which to focus.
Examples of research topics are adolescent smoking, patient compliance,
coping with disability, and pain management. Within each of these broad topics
are many potential research problems. In this section, we illustrate various
terms using the topic side effects of chemotherapy.
A research problem is an enigmatic, perplexing, or troubling condition.
Both qualitative and quantitative researchers identify a research problem
within a broad topic area of interest. The purpose of research is to “solve”
the problem — or to contribute to its solution — by accumulating relevant
information. A problem statement articulates the problem to be addressed and
indicates the need for a study. Table 4-1 presents a problem statement related
to the topic of side effects of chemotherapy.
Research questions are the specific queries researchers want to
answer in addressing the research problem. Research questions guide the types
of data to be collected in a study. Researchers who make specific
predictions regarding answers to the research question pose hypotheses that are
tested empirically.
Many reports include a statement of purpose (or purpose statement), which
is the researcher’s summary of the overall goal of a study. A researcher might
also identify several research aims or objectives—the specific
accomplishments the researcher hopes to achieve by conducting the study. The
objectives include obtaining answers to research questions or testing research
hypotheses but may also encompass some broader aims (e.g., developing
recommendations for changes to nursing practice based on the study results).
These terms are not always consistently defined in research methods
textbooks, and differences between the terms are often subtle. Table 4-1
illustrates the interrelationships among terms as we define them.
Research Problems and Paradigms. Some research problems are
better suited for studies using qualitative versus quantitative methods.
Quantitative studies usually involve concepts that are fairly well
developed, about which there is an existing body of literature, and for which
reliable methods of measurement have been developed. For example, a
quantitative study might be undertaken to determine if postpartum depression is
higher among women who are employed 6 months after delivery than among those
who stay home with their babies. There are relatively accurate measures of
postpartum depression that would yield quantitative information about the level
of depression in a sample of employed and nonemployed
postpartum women.
Qualitative studies are often undertaken because some aspect of a
phenomenon is poorly understood, and the researcher wants to develop a rich,
comprehensive, and context-bound understanding of it. Qualitative studies are
usually initiated to heighten awareness and create a dialogue about a
phenomenon. In the example of postpartum depression, qualitative methods would
not be well suited to comparing levels of depression among the two groups of
women, but they would be ideal for exploring, for example, the meaning of
postpartum depression among new mothers. Thus, the nature of the research
question is closely allied to paradigms and research traditions within
paradigms.
Sources Of
Research Problems
Students are sometimes puzzled about the origins of research
problems. Where do ideas for research problems come from? How do esearchers select
topic areas and develop research questions? At the most basic level, research
topics originate with researchers’ interests. Because research is a
time-consuming enterprise, curiosity about and interest in a topic are essential
to a project’s success.
Explicit sources that might fuel researchers’ curiosity include
experience, the nursing literature, social issues, theories, and ideas from
others.
Experience and Clinical Fieldwork. The nurse’s everyday
clinical experience is a rich source of ideas for research problems. As you are
performing your nursing functions, you are bound to find a wealth of
research ideas if you are curious about why things are the way they are or
about how things could be improved if something were to change. You may be well
along the way to developing a research idea if you have ever asked the
following kinds of questions:Why are things done this
way? What information would help to solve this problem? What is the process by
which this situation arose? What would happen if ... ? For beginning
researchers in particular, clinical experience (or clinical coursework) is
often the most compelling source for topics. Immediate problems that need a
solution or that excite the curiosity are relevant and interesting and, thus,
may generate more enthusiasm than abstract and distant problems inferred from a
theory. Clinical fieldwork before a study may also help to identify
clinical problems.
TIP: Personal experiences in clinical settings are a provocative source
of research ideas. Here are some hints on how to proceed:
• Watch for recurring problems and see if you can discern a pattern in
situations that lead to the problem.
Example: Why do many patients complain of being tired after being
transferred from a coronary care unit to a progressive care unit?
• Think about aspects of your work that are irksome, frustrating, or do
not result in the intended outcome — then try to identify factors contributing
to the problem that could be changed.
Example: Why is suppertime so frustrating in a nursing home?
• Critically examine some decisions you make in performing your
functions. Are these decisions based on tradition, or are they based on
systematic evidence that supports their efficacy? Many practices in
nursing that have become custom might be challenged.
Example: What would happen if visiting hours in the intensive care unit
were changed from 10 minutes every hour to the regularly scheduled hours
existing in the rest of the hospital?
Nursing Literature. Ideas for research
projects often come from reading the nursing literature. Beginning nurse
researchers can profit from regularly reading nursing journals, either
clinical specialty journals or research journals such as Nursing Research or
the Western Journal of Nursing Research. Nonresearch
articles can be helpful in alerting researchers to clinical trends and issues
of importance in clinical settings. Published research reports may suggest
problem areas indirectly by stimulating the imagination and directly by
specifying further areas in need of investigation.
Example of a direct suggestion for further research:
Stranahan (2001) studied the relationship
between nurse practitioners’ attitudes about spiritual care and their spiritual
care practices. She made several recommendations for further research in her
report, such as the following: “Studies should be conducted to determine
reasons why nurse practitioners do not practice spiritual care in the primary
care setting” (p. 87).
Inconsistencies in the findings reported in nursing literature
sometimes generate ideas for studies. For example, there are inconsistencies
regarding which type of tactile stimulation or touch (e.g., gentle touch,
stroking, rubbing) has the most beneficial physiologic and behavioral
effects on preterm infants. Such discrepancies can lead to the design of a
study to resolve the matter.
Researchers may also wonder whether a study similar to one reported in a
journal article would yield comparable results if applied in a different
setting or with a different population. Replications are needed to establish
the validity and generalizability of previous findings.
In summary, a familiarity with existing research, or with problematic and
controversial nursing issues that have yet to be understood and investigated
systematically, is an important route to developing a research topic. Students
who are actively seeking a problem to study will find it useful to read
widely in areas of interest.
TIP: In a pinch, do not hesitate to replicate a study that is reported in
the research literature. Replications are a valuable learning experience and
can make important contributions if they corroborate (or even if they
challenge) earlier findings.
Social Issues. Sometimes, topics are suggested by
more global contemporary social or political issues of relevance to the health
care community. For example, the feminist movement has raised questions about
such topics as sexual harassment, domestic violence, and gender equity in
health care and in research. The civil rights movement has led to research on
minority health problems, access to health care, and culturally sensitive
interventions. Thus, an idea for a study may stem from a familiarity with
social concerns or controversial social problems.
Theory. The fourth major source of research
problems lies in the theories and conceptual schemes that have been developed
in nursing and related disciplines. To be useful in nursing practice, theories
must be tested through research for their applicability to hospital units,
clinics, classrooms, and other nursing environments.
When researchers decide to base a study on an existing theory, deductions
from the theory must be developed. Essentially, researchers must ask the following
questions: If this theory is correct, what kind of behavior would I expect to
find in certain situations or under certain conditions? What kind of evidence
would support this theory? This process would eventually result in a specific
problem that could be subjected to systematic investigation.
Ideas From External Sources. External sources can sometimes
provide the impetus for a research idea. In some cases, a research topic may be
given as a direct suggestion. For example, a faculty member may give students a
list of topics from which to choose or may actually assign a specific
topic to be studied. Organizations that sponsor funded research, such as
government agencies, often identify topics on which research proposals are
encouraged. Ideas for research are also being noted on various websites on the
internet (see, for example, Duffy, 2001).
Research ideas sometimes represent a response to priorities that are
established within the nursing profession. Priorities for nursing research have
been established by many nursing specialty practices. Priority lists can often
serve as a useful starting point for exploring research topics.
Often, ideas for studies emerge as a result of a brainstorming session.
By discussing possible research topics with peers, advisers or mentors, or
researchers with advanced skills, ideas often become clarified and
sharpened or enriched and more fully developed. Professional conferences often
provide an excellent opportunity for such discussions.
Development And
Refinement Of Research Problems
Unless a research problem is developed on the basis of theory or an explicit
suggestion from an external source, the actual procedures for developing a
research topic are difficult to describe. The process is rarely a smooth
and orderly one; there are likely to be false starts, inspirations, and
setbacks in the process of developing a research problem statement. The few
suggestions offered here are not intended to imply that there are techniques
for making this first step easy but rather to encourage beginning
researchers to persevere in the absence of instant success.
Selecting a Topic. The development of a
research problem is a creative process that depends on imagination and
ingenuity. In the early stages, when research ideas are being generated, it is
wise not to be critical of them immediately. It is better to begin by relaxing
and jotting down general areas of interest as they come to mind. At this point,
it matters little if the terms used to
remind you of your ideas are abstract or concrete, broad or specific,
technical, or colloquial—the important point is to put some ideas on paper.
Examples of some broad topics that may come to mind include nurse—patient
communication, pain in patients with cancer, and postoperative loss of
orientation.
After this first step, the ideas can be sorted in terms of
interest, knowledge about the topics, and the perceived feasibility of turning
the topics into a research project. When the most fruitful idea has been
selected, the rest of the list should not be discarded; it may be necessary to
return to it.
Narrowing the Topic. Once researchers have
identified a topic of interest, they need to ask questions that lead to a
researchable problem. Examples of question stems that may help to focus an inquiry
include the following:
• What is going on with ...?
• What is the process by which ...?
• What is the meaning of ...?
• Why do ...?
• When do ...?
• How do ...?
• What can be done to solve ...?
• What is the extent of ...?
• How intense are ...?
• What influences ...?
• What causes ...?
• What characteristics are associated with ...?
• What differences exist between ...?
• What are the consequences of ...?
• What is the relationship between ...?
• What factors contribute to ...?
• What conditions prevail before ...?
• How effective is ...?
Here again, early criticism of ideas is often counterproductive. Try not
to jump to the conclusion that an idea sounds trivial or uninspired without
giving it more careful consideration or without exploring it with advisers or
colleagues.
Beginning researchers often develop problems that are too broad in scope
or too complex and unwieldy for their level of methodologic
expertise. The transformation of the general topic into a workable problem is typically
accomplished in a number of uneven steps, involving a series of successive
approximations. Each step should result in progress toward the goals of
narrowing the scope of the problem and sharpening and defining the
concepts.
As researchers move from general topics to more specific
researchable problems, more than one potential problem area can emerge. Let us
consider the following example. Suppose you were working on a medical unit and
were puzzled by that fact that some patients always complained about having to
wait for pain medication when certain nurses were assigned to them and, yet,
these same patients offered no complaints with other nurses. The general
problem area is discrepancy in complaints from patients regarding pain
medications administered by different nurses. You might ask the following: What
accounts for this discrepancy? How can I improve the situation? Such questions
are not actual research questions; they are too broad and vague. They may,
however, lead you to ask other questions, such as the following: How do the two
groups of nurses differ? What characteristics are unique to each group of
nurses? What characteristics do the group of complaining patients share? At
this point, you may observe that the ethnic background of the patients and
nurses appears to be a relevant factor. This may direct you to a review of the
literature for studies concerning ethnicity in relation to nursing care, or it
may provoke you to discuss the observations with others. The result of these
efforts may be several researchable questions, such as the following:
• What is the essence of patient complaints among patients of different
ethnic backgrounds?
• What is the patient’s experience of waiting for pain medication?
• How do complaints by patients of different ethnic backgrounds get
expressed by patients and perceived by nurses?
• Is the ethnic background of nurses related to the frequency with which they dispense pain
medication?
• Is the ethnic background of patients related to the frequency and
intensity of complaints when waiting for pain medication?
• Does the number of patient complaints increase when patients are of
dissimilar ethnic backgrounds as opposed to when they are of the same ethnic
background as nurses?
• Do nurses’ dispensing behaviors change as a function of the similarity
between their own ethnic background and that of patients?
All these questions stem from the same general problem, yet each would be
studied differently—for example, some suggest a qualitative approach and others
suggest a quantitative one. A quantitative researcher might become curious
about nurses’ dispensing behaviors, based on some interesting evidence in the
literature regarding ethnic differences.
Both ethnicity and nurses’ dispensing behaviors are variables that can be
measured in a straightforward and reliable manner. A qualitative researcher who
noticed differences in patient complaints would likely be more interested in
understanding the essence of the complaints, the patients’ experience of
frustration, the process by which the problem got resolved, or the full nature of the nurse—patient interactions
regarding the dispensing of medications. These are aspects of the research
problem that would be difficult to quantify.
Researchers choose the final problem to be studied based on several
factors, including its inherent interest to them and its compatibility with a
paradigm of preference. In addition, tentative problems usually vary in their
feasibility and worth. It is at this point that a critical evaluation of ideas
is appropriate.
Evaluating Research Problems. There are no rules for
making a final selection of a research problem. Some criteria, however,
should be kept in mind in the decision-making process. The four most important
considerations are the significance,
researchability, and feasibility of the problem, and
its interest to the researcher.
Significance of the Problem. A crucial factor in selecting
a problem to be studied is its significance to nursing—especially to
nursing practice. Evidence from the study should have the potential of
contributing meaningfully to nursing knowledge. Researchers should pose the
following kinds of questions: Is the problem an important one? Will patients,
nurses, or the broader health care community or society benefit from the
evidence that will be produced? Will the results lead to practical
applications? Will the results have theoretical relevance? Will the findings
challenge (or lend support to) untested assumptions? Will the study help to
formulate or alter nursing practices or policies? If the answer to all these
questions is “no,” then the problem should be abandoned.
Researchability of the Problem. Not all problems are
amenable to study through scientific investigation. Problems or questions
of a moral or ethical nature, although provocative, are incapable of being
researched. Take, for example, the following: Should assisted suicide be
legalized?
The answer to such a question is based on a person’s values. There are no
right or wrong answers, only points of view. The problem is suitable to debate,
not to research. To be sure, it is possible to ask related questions that could
be researched. For instance, each of the following questions could be
investigated in a research project:
• What are nurses’ attitudes toward assisted suicide?
• Do oncology nurses hold more favorable opinions of assisted suicide
than other nurses?
• What moral dilemmas are perceived by nurses who might be involved in
assisted suicide?
• What are the attitudes of terminally ill patients toward assisted
suicide?
• Do terminally ill patients living with a high level of pain hold more favorable
attitudes toward assisted suicide than those with less pain?
• How do family members experience the loss of a loved one through
assisted suicide?
The findings from these hypothetical projects would have no
bearing, of course, on whether assisted suicide should be legalized, but the
information could be useful in developing a better understanding of the issues.
In quantitative studies, researchable problems are ones involving
variables that can be precisely defined and measured. For example,
suppose a researcher is trying to determine what effect early discharge has on
patient well-being. Well-being is too vague a concept for a study. The
researcher would have to sharpen and define the concept so that it could
be observed and measured. That is, the researcher would have to establish
criteria against which patients’ progress toward well-being could be assessed.
When a new area of inquiry is being pursued, however, it may be
impossible to define the concepts of interest in precise terms. In such
cases, it may be appropriate to address the problem using in-depth qualitative
research. The problem may then be stated in fairly broad terms to permit full
exploration of the concept of interest.
Feasibility of
Addressing the Problem. A problem that is both significant and researchable may
still be inappropriate if a study designed to address it is not feasible. The
issue of feasibility encompasses various considerations. Not all of the
following factors are relevant for every problem, but they should be kept in
mind in making a final decision.
Time and Timing. Most studies have
deadlines or at least goals for completion. Therefore, the problem must be one
that can be adequately studied within the time allotted. This means that the
scope of the problem should be sufficiently restricted that enough time
will be available for the various steps and activities reviewed in Chapter 3.
It is wise to be conservative in estimating time for various tasks because
research activities often require more time to accomplish than anticipated.
Qualitative studies may be especially time-consuming.
A related consideration is the timing of the project. Some of the research
steps — especially data collection — may be more readily performed at certain
times of the day, week, or year than at other times. For example, if the
problem focused on patients with peptic ulcers, the research might be more
easily conducted in the fall and spring because of the increase in the number
of patients with peptic ulcers during these seasons. When the timing
requirements of the tasks do not match the time available for their
performance, the feasibility of the project may be jeopardized.
Availability of Study Participants. In any study involving humans, researchers
need to consider whether individuals with the desired characteristics will be
available and willing to cooperate. Securing people’s cooperation may in some
cases be easy (e.g., getting nursing students to complete a questionnaire in a
classroom), but other situations may pose more difficulties. Some people
may not have the time, others may have no interest in a study that has little
personal benefit, and others may not feel well enough to participate.
Fortunately, people usually are willing to cooperate if research demands are
minimal. Researchers may need to exert extra effort in recruiting
participants—or may have to offer a monetary incentive—if the research is
time-consuming or demanding. An additional problem may be that of identifying
and locating people with needed characteristics.
For example, if we were interested in studying the coping strategies of
people who had lost a family member through suicide, we would have to develop a
plan for identifying prospective participants from this distinct and
inconspicuous population.
Cooperation of Others. Often, it is insufficient to obtain the
cooperation of prospective study participants alone. If the sample includes
children, mentally incompetent people, or senile individuals, it would be
necessary to secure the permission of parents or guardians. In institutional or
organizational settings (e.g., hospitals), access to clients, members,
personnel, or records usually requires administrative authorization. Many
health care facilities require that any project be presented to a panel of
reviewers for approval. Critical requirement in many qualitative studies is
gaining entrée into an appropriate community, setting, or group, and developing
the trust of gatekeepers.
Facilities and Equipment. All studies have resource requirements,
although in some cases, needs may be modest. It is prudent to consider what
facilities and equipment will be needed and whether they will be available
before embarking on a project to avoid disappointment and frustration. The
following is a partial list of considerations:
• Will assistants be needed, and are such assistants available?
• If technical equipment and apparatus are needed, can they be secured,
and are they functioning properly? Will audiotaping or videotaping equipment be
required, and is it of sufficient sensitivity for the research
conditions? Will laboratory facilities be required, and are they available?
• Will space be required, and can it be obtained?
• Will telephones, office equipment, or other supplies be required?
• Are duplicating or printing services available, and are they reliable?
• Will transportation needs pose any difficulties? Money. Monetary requirements for research projects
vary widely, ranging from $10 to $20 for small student projects to hundreds of
thousands (or even millions) of dollars for large-scale, government-sponsored
research. The investigator on a limited budget should think carefully about
projected expenses before making the final selecion
of a problem. Some major categories of research-related expenditures are the
following:
• Literature costs—computerized literature search and retrieval service
charges, Internet access charges, reproduction costs, index cards, books and
journals
• Personnel costs—payments to individuals hired to help with the data
collection (e.g., for conducting interviews, coding, data entry, transcribing,
word processing)
• Study participant costs—payment to participants as an incentive for
their cooperation or to offset their own expenses (e.g., transportation or
baby-sitting costs)
• Supplies—paper, envelopes, computer disks, postage, audiotapes, and so
forth
• Printing and duplication costs—expenditures for printing forms,
questionnaires, participant recruitment notices, and so on
• Equipment—laboratory apparatus, audio- or video-recorders, calculators,
and the like
• Computer-related expenses (e.g., purchasing software)
• Laboratory fees for the analysis of biophysiologic
data
• Transportation costs
Experience of the Researcher. The problem should be
chosen from a field about which investigators have some prior knowledge
or experience.
Researchers have difficulty adequately developing a study on a
topic that is totally new and unfamiliar—although clinical fieldwork
before launching the study may make up for certain deficiencies. In addiion to substantive knowledge, the issue of technicalexpertise should not be overlooked. Beginning
researchers with limited methodologic skills should
avoid research problems that might require the development of sophisticated
measuring instruments or that involve complex data analyses.
Ethical Considerations. A research problem may not
be feasible because the investigation of the problem would pose unfair or
unethical demands on participants. The ethical responsibilities of researchers
should not be taken lightly. People engaged in research activities should be
thoroughly knowledgeable about the rights of human or animal subjects.
Interest to the Researcher. Even if the tentative
problem is researchable, significant, and feasible, there is one more
criterion: the researcher’s own interest in the problem. Genuine interest in
and curiosity about the chosen research problem are critical prerequisites to a
successful study. A great deal of time and energy are expended in a study;
there is little sense devoting these personal resources to a project that does
not generate enthusiasm.
TIP: Beginning researchers often seek out suggestions on topic areas, and
such assistance may be helpful in getting started. Nevertheless, it is rarely
wise to be talked into a topic toward which you are not personally inclined. If
you do not find a problem attractive or stimulating during the beginning
phases of a study—when opportunities for creativity and intellectual enjoyment
are at their highest—then you are bound to regret your choice later.
Communicating
The Research Problem
It is clear that a study cannot progress without the choice of a problem;
it is less clear, but nonetheless true, that the problem and research questions
should be carefully stated in writing before proceeding with the design of the
study or with field work. Putting one’s ideas in writing is often sufficient
to illuminate ambiguities and uncertainties.
This section discusses the wording of problem statements, statements of
purpose, and research questions, and the following major section discusses
hypotheses.
Problem Statements. A problem statement is an
expression of the dilemma or disturbing situation that needs investigation for
the purposes of providing understanding and direction. A problem statement
identifies the nature of the problem that is being addressed in the study
and, typically, its context and significance.
In general, the problem statement should be broad enough to include
central concerns, but narrow enough in scope to serve as a guide to study
design.
Example of a problem statement from a quantitative study:
Women account for an increasing percentage of adults with human immunodeficiency
virus (HIV).... Most of these HIV-infected women are in their childbearing
years. As a result, approximately 7,000 infants are exposed prenatally each
year.... All infants exposed to HIV prenatally are at risk for developmental
problems.... Little is known about the quality of parental caregiving for
infants of mothers with HIV, because few studies have examined parenting in
this vulnerable group.... The purpose of this report is to describe the
development of infants of mothers with HIV and to determine the extent to which
child characteristics, parental caregiver characteristics, family
characteristics, and parenting quality influence development (Holditch-Davis, Miles, Burchinal,
O’Donnell, McKinney, & Lim, 2001, pp. 5–6).
In this example, the general topic could be described as infant
development among at-risk children. The investigators’ more specific
focus is on four factors that influence
infant development among children exposed to HIV prenatally. The problem
statement asserts the nature of the problem (these children are at risk of
developmental problems) and indicates its breadth (7000 infants annually). It
also provides a justification for conducting a new study: the dearth of
existing studies on parenting in this population.
The problem statement for a qualitative study similarly expresses the
nature of the problem, its context, and its significance, as in the
following example:
Example of a problem statement from a qualitative study:
Members of cultural minority groups may find themselves surrounded
by people whose values, beliefs, and interpretations differ from their own
during hospitalization. This is often the case for
As in the previous example, these qualitative researchers clearly
articulated the nature of the problem and the justification for a new
study.
Qualitative studies that are embedded in a particular research tradition
usually incorporate terms and concepts in their problem statements that
foreshadow their tradition of inquiry (Creswell, 1998). For example, the
problem statement in a grounded theory study might refer to the need to
generate a theory relating to social processes. A problem statement for a
phenomenological study might note the need to know more about people’s
experiences (as in the preceding example) or the meanings they attribute to
those experiences. And an ethnographer might indicate the desire to describe
how cultural forces affect people’s behavior.
Problem statements usually appear early in a research report and are
often interwoven with a review of the literature, which provides context by
documenting knowledge gaps.
Statements of Purpose. Many researchers first
articulate their research goals formally as a statement of purpose, worded in
the declarative form. The statement captures—usually in one or two clear
sentences—the essence of the study.
The purpose statement establishes the general direction of the inquiry.
The words purpose or goal usually appear in a purpose statement (e.g., The
purpose of this study was..., or, The goal of this study was...), but sometimes
the words intent, aim,or objective are used instead.
Unfortunately, some research reports leave the statement of purpose implicit,
placing an unnecessary burden on readers to make inferences about the goals.
In a quantitative study, a statement of purpose identifies the key
study variables and their possible interrelationships, as well as the nature of
the population of interest.
Example of a statement of purpose from a quantitative study:
“The purpose of this study was to determine whether viewing a video of an
actual pediatric inhalation induction would reduce the level of parental
anxiety” (Zuwala & Barber, 2001, p. 21).
This statement identifies the population of interest (parents whose
child required inhalation induction), the independent variable (viewing a video
of such an induction, versus not viewing the video), and the dependent variable
(parental anxiety).
In qualitative studies, the statement of purpose indicates the nature of
the inquiry, the key concept or phenomenon, and the group, community, or
setting under study.
Example of a statement of purpose from a qualitative study:
Gallagher and Pierce (2002) designed their qualitative study for the
following two purposes: “to gain the family caregivers’ perspective of dealing
with UI [urinary incontinence] for the care recipient who lives in a home
setting, and to gain care recipients’ perspective on the UI care given by
family caregivers in the home setting” (p. 25).
This statement indicates that the central phenomenon of interest is
perspectives on caregiving and that the groups under study are UI patients in
home settings and the family caregivers caring for them.
Often, the statement of purpose specifically mentions the
underlying research tradition, if this is relevant.
Example of a statement of purpose from a grounded theory study:
The purpose is “to generate a grounded substantive theory of the process
of forgiveness in patients with cancer” (Mickley and
Cowles, 2001, p. 31).
The statement of purpose communicates more than just the nature of the
problem. Through researchers’ selection of verbs, a statement of purpose suggests
the manner in which they sought to solve the problem, or the state of knowledge
on the topic. That is, a study whose purpose is to explore or describe some
phenomenon is likely to be an investigation of a little-researched topic, often
involving a qualitative approach such as a phenomenology or ethnography. A
statement of purpose for a qualitative study—especially a grounded theory
study—may also use verbs such as understand, discover, develop, or generate.
Creswell (1998) notes that the statements of purpose in qualitative studies
often “encode” the tradition of inquiry not only through the researcher’s
choice of verbs but also through the use of certain terms or “buzz words”
associated with those traditions, as follows:
• Grounded theory: Processes; social structures; social interactions
• Phenomenological studies: Experience; lived experience; meaning;
essence
• Ethnographic studies: Culture; roles; myths; cultural behavior
Quantitative researchers also suggest the nature of the inquiry through
their selection of verbs. A purpose statement indicating that the study purpose
is to test or determine or evaluate the effectiveness of an intervention
suggests an experimental design, for example. A study whose purpose is to
examine or assess the relationship between two variables is more likely to
refer to a nonexperimental quantitative design. In
some cases, the verb is ambiguous: a purpose statement indicating that the
researcher’s intent is to compare could be referring to a comparison of
alternative treatments (using an experimental approach) or a comparison of two
preexisting groups (using a nonexperimental
approach). In any event, verbs such as
test, evaluate, and compare suggest an existing knowledge base, quantifiable
variables, and designs with tight scientific controls.
Note that the choice of verbs in a statement of purpose should connote
objectivity. A statement of purpose indicating that the intent of the study was
to prove, demonstrate,or showsomething
suggests a bias.
TIP: In wording your research questions or statement of purpose, look at
published research reports for models. You may find, however, that some
reports fail to state unambiguously the study purpose or specific
research questions. Thus, in some studies, you may have to infer the research
problem from several sources, such as the title of the report. In other
reports, the purpose or questions are clearly stated but may be difficult
to find. Researchers most often state their purpose or questions at the
end of the introductory section of the report.
Research Questions. Research questions are, in
some cases, direct rewordings of statements of purpose, phrased interrogatively
rather than declaratively, as in the following example:
• The purpose of this study is to assess the relationship between the
dependency level of renal transplant recipients and their rate of recovery.
• What is the relationship between the dependency level of renal
transplant recipients and their rate of recovery?
The question form has the advantage of simplicity and directness.
Questions invite an answer and help to focus attention on the kinds of data
that would have to be collected to provide that answer. Some research reports
thus omit a statement of purpose and state only research questions. Other
researchers use a set of research questions to clarify or lend greater specificity
to the purpose statement.
Example of research questions clarifying a statement of purpose:
Statement of Purpose: The purpose of this study was to explore the
relationship between method of pain management during labor and specific
labor and birth outcomes.
Research Questions: (1) Are nonepidural and
epidural methods of pain relief associated with augmentation during the first
stage of labor? (2) Is the length of second stage labor associated with
epidural and nonepidural methods of pain relief? (3)
Are newborn Apgar scores at 1 minute and 5 minutes associated with method of
pain relief? (4) Does epidural anesthesia affect maternal temperature? (Walker
& O’Brien, 1999)
In this example, the statement of purpose provides a global message about
the researchers’ goal to explore relationships among several variables.
The research questions identified the two methods of pain
management (the independent variable) and the specific labor and birth
outcomes of interest (the dependent variables).
Research Questions in Quantitative Studies. In quantitative studies,
research questions identify the key variables (especially the independent and
dependent variables), the relationships among them, and the population under
study. The variables are all measurable concepts, and the questions suggest
quantification. For example, a descriptive research question might ask about
the frequency or prevalence of variables, or their average values (What
percentage of women breastfeed their infants? or What is the average
interstitial fluid volume at 60 minutes after intravenous infiltration
following treatment with cold applications?).
Most quantitative studies, however, ask questions about relationships
between variables. These can be illustrated with an example of women’s
emotional responses to miscarriage:
1. Existence of relationship: Is there a relationship between miscarriage
and depression—that is, are there differences in depression levels of pregnant
women who miscarry compared with those who do not?
2. Direction of relationship: Do women who miscarry exhibit higher (or
lower) levels of depression than pregnant women who do not?
3. Strength of relationship: How strong is the risk of depression among
women who miscarry?
4. Nature of relationship: Does having a miscarriage contribute to
depression? Does depression contribute to a miscarriage? Or does some other
factor influence both?
5. Moderated relationship: Are levels of depression among women who
miscarry moderated by whether the woman has previously given birth? (i.e., Is
the relationship between depression and miscarriage different for primiparas and multiparas?)
6. Mediated relationship: Does a miscarriage directly affect depression
or does depression occur because the miscarriage had a negative effect on
marital relations?
The last two research questions involve mediator and moderator variables,
which are variables of interest to the researcher (i.e., that are not extraneous)
and that affect the relationship between the independent and dependent
variables. A moderator variable is a variable that affects the strength or
direction of an association between the independent and dependent variable. The
independent variable is said to interact with the moderator: the independent
variable’s relationship with the dependent variable is stronger or weaker for
different values of the moderator variable (Bennett, 2000). In the preceding
example, it might be that the risk of depression after a miscarriage is low
among women who had previously given birth (i.e., when the moderating variable
parity is greater than 0), but high among women who do not have children (i.e.,
when parity equals 0). When all women are considered together without taking
parity into account, the relationship between experiencing a miscarriage (the
independent variable) and levels of depression (the dependent variable) might
appear moderate. Therefore, identifying parity as a key moderator is important
in understanding when to expect a relationship between miscarriage and
depression, and this understanding has clinical relevance.
Research questions that involve mediator variables concern the
identification of causal pathways. A mediator variable is a variable that
intervenes between the independent and dependent variable and helps to explain
why the relationship exists. In our hypothetical example, we are asking whether
depression levels among women who have experienced a miscarriage are influenced
by the negative effect of the miscarriage on marital relations. In research
questions involving mediators, researchers are typically more interested in the
mediators than in the independent variable, because the mediators are key
explanatory mechanisms.
In summary, except for questions of a purely descriptive nature, research
questions in quantitative research focus on unraveling relationships among
variables.
Example of a research question from a quantitative study:
Watt-Watson, Garfinkel, Gallop, Stevens, and Streiner (2000) conducted a study about acute care nurses’
empathy and its effects on patients. Their primary research question was about
the existence and direction of a relationship:
Do nurses with greater empathy have patients experiencing less pain and
receiving adequate analgesia than those with less empathy?
Research Questions in Qualitative Studies. Researchers in the
various qualitative traditions vary in their conceptualization of what types of
questions are important. Grounded theory researchers are likely to ask process questions, phenomenologists tend to
ask meaning questions, and ethnographers generally ask descriptive questions about cultures.
The terms associated with the various traditions, discussed previously in
connection with purpose statements, are likely to be incorporated into the
research questions.
Example of a research question from a phenomenological study:
What is the lived experience of caring for a family member with
Alzheimer’s disease at home? (Butcher, Holkup, & Buckwalter, 2001)
It is important to note, however, that not all qualitative studies are
rooted in a specific research tradition. Many researchers use
naturalistic methods to describe or explore phenomena without focusing on
cultures, meaning, or social processes.
Example of a research question from a qualitative study:
Wilson and Williams (2000) undertook a qualitative study that explored
the potential effects of visualism (a prejudice in
favor of the seen) on the perceived legitimacy of telephone work in community
nursing. Among the specific research questions that guided their in-depth
interviews with community nurses were the following:
Is telephone consultation considered real work? Is it considered real
communication? Can telephone consultation bring the community and its nursing
services into close relationship?
In qualitative studies, research questions sometimes evolve over the
course of the study. The researcher begins with a focus that defines the
general boundaries of the inquiry. However, the boundaries are not cast in
stone; the boundaries “can be altered and, in the typical naturalistic inquiry,
will be” (Lincoln & Guba, 1985, p. 228). The
naturalist begins with a research question that provides a general starting
point but does not prohibit discovery; qualitative researchers are often sufficiently
flexible that the question can be modified as new information makes
it relevant to do so.
Research
Hypotheses
A hypothesis is a prediction about the relationship between two or more
variables. A hypothesis thus translates a quantitative research question into a
precise prediction of expected outcomes. In qualitative studies, researchers do
not begin with a hypothesis, in part because there is usually too little known
about the topic to justify a hypothesis, and in part because qualitative
researchers want the inquiry to be guided by participants’ viewpoints rather
than by their own. Thus, this discussion focuses on hypotheses used to guide
quantitative inquiries (some of which are generated within qualitative
studies).
Function of Hypotheses in Quantitative Research. Research questions, as we
have seen, are usually queries about relationships between variables.
Hypotheses are proposed solutions or answers to these queries. For
instance, the research question might ask: Does history of sexual abuse in
childhood affect the development of irritable bowel syndrome in women? The
researcher might predict the following: Women who were sexually abused in
childhood have a higher incidence of irritable bowel syndrome than women who
were not.
Hypotheses sometimes follow directly from a theoretical framework.
Scientists reason from theories to hypotheses and test those hypotheses in the
real world. The validity of a theory is never examined directly. Rather, it is
through hypothesis testing that the worth of a theory can be evaluated. Let us
take as an example the theory of reinforcement. This theory maintains that
behavior that is positively reinforced (rewarded) tends to be learned or
repeated.
The theory itself is too abstract to be put to an empirical test, but if
the theory is valid, it should be possible to make predictions about certain
kinds of behavior. For example, the following hypotheses have been deduced from
reinforcement theory:
• Elderly patients who are praised (reinforced) by nursing personnel for
self-feeding require less assistance in feeding than patients who are not
praised.
• Pediatric patients who are given a reward (e.g., a balloon or
permission to watch television) when they cooperate during nursing procedures
tend to be more obliging during those procedures than nonrewarded
peers.
Both of these propositions can be put to a test in the real world. The
theory gains support if the hypotheses are confirmed.
Not all hypotheses are derived from theory. Even in the absence of a
theory, well-conceived hypotheses offer direction and suggest explanations.
Perhaps an example will clarify this point.
Suppose we hypothesized that nurses who have received a baccalaureate
education are more likely to experience stress in their first nursing job
than are nurses with a diploma-school education. We could justify our
speculation based on theory (e.g., role conflict theory, cognitive
dissonance theory), earlier studies, personal observations, or on the basis of
some combination of these.
The development of predictions in and of itself forces researchers to
think logically, to exercise critical judgment, and to tie together earlier
research findings.
Now let us suppose the preceding hypothesis is not confirmed by the
evidence collected; that is, we find that baccalaureate and diploma
nurses demonstrate comparable stress in their first job.
The failure of data to support a prediction forces researchers to analyze
theory or previous research critically, to carefully review the limitations of
the study’s methods, and to explore alternative explanations for the findings.
The use of hypotheses in quantitative studies tends to induce critical
thinking and to facilitate understanding and interpretation of the data.
To illustrate further the utility of hypotheses, suppose we conducted the
study guided only by the research question, Is there a relationship between
nurses’ basic preparation and the degree of stress experienced on the first
job? The investigator without a hypothesis is, apparently, prepared to accept
any results. The problem is that it is almost always possible to explain
something superficially after the fact, no matter what the findings
are. Hypotheses guard against superficiality and minimize the possibility
that spurious results will be misconstrued.
Characteristics of Testable Hypotheses. Testable research
hypotheses state expected relationships between the independent variable (the
presumed cause or antecedent) and the dependent variable (the presumed effect
or outcome) within a population.
Example of a research hypothesis:
Cardiac patients receiving an intervention involving “vicarious
experience” through support from former patients have (1) less anxiety; (2)
higher self-efficacy expectation; and (3) higher self-reported activity
than other patients (Parent & Fortin, 2000).
In this example, the independent variable is receipt versus nonreceipt of the intervention, and the dependent variables
are anxiety, self-efficacy expectation, and activity. The hypothesis
predicts better outcomes among patients who receive the intervention.
Unfortunately, researchers occasionally present hypotheses that fail to
make a relational statement. For example, the following prediction is not an
acceptable research hypothesis:
Pregnant women who receive prenatal instruction regarding postpartum
experiences are not likely to experience postpartum depression.
This statement expresses no anticipated relationship; in fact, there is
only one variable (postpartum depression), and a relationship by definition
requires at least two variables.
When a prediction does not express an anticipated relationship, it cannot
be tested. In our example, how would we know whether the hypothesis was
supported—what absolute standard could be used to decide whether to accept or
reject the hypothesis? To illustrate the problem more concretely, suppose we
asked a group of mothers who had been given instruction on postpartum
experiences the following question 1 month after delivery: On the whole, how
depressed have you been since you gave birth? Would you say (1) extremely
depressed, (2) moderately depressed, (3) somewhat depressed, or (4) not at all
depressed?
Based on responses to this question, how could we compare the actual
outcome with the predicted outcome? Would
all the women have to say they were “not at all depressed?” Would the
prediction be supported if 51% of the women said they were “not at all
depressed” or “somewhat depressed?” There is no adequate way of testing the
accuracy of the prediction.
A test is simple, however, if we modify the prediction to the following:
Pregnant women who receive prenatal instruction are less likely to experience
postpartum depression than those with no prenatal instruction. Here, the
dependent variable is the women’s depression, and the independent variable is
their receipt versus nonreceipt of prenatal
instruction. The relational aspect of the prediction is embodied in the phrase
less than. If a hypothesis lacks a phrase such as more than, less than, greater
than, different from, related to, associated with, or something similar, it is
not amenable to testing in a quantitative study. To test this revised
hypothesis, we could ask two groups of women with different prenatal
instruction experiences to respond to the question on depression and then
compare the groups’ responses. The absolute degree of depression of either
group would not be at issue.
Hypotheses, ideally, should be based on sound, justifiable
rationales. The most defensible hypotheses follow from previous research findings
or are deduced from a theory. When a relatively new area is being investigated,
the researcher may have to turn to logical reasoning or personal experience to
justify the predictions. There are, however, few problems for which research
evidence is totally lacking.
The Derivation of Hypotheses. Many students ask the
question, How do I go about developing hypotheses? Two basic
processes—induction and deduction—constitute the intellectual machinery
involved in deriving hypotheses.
An inductive hypothesis is a generalization based on observed
relationships. Researchers observe certain patterns, trends, or associations
among phenomena and then use the observations as a basis for predictions.
Related literature should be examined to learn what is already known on a
topic, but an important source for inductive hypotheses is personal
experiences, combined with intuition and critical analysis. For example, a
nurse might notice that presurgical patients who ask
a lot of questions relating to pain or who express many pain-related
apprehensions have a more difficult time in learning appropriate
postoperative procedures. The nurse could then formulate a hypothesis, such as
the following, that could be tested through more rigorous procedures: Patients
who are stressed by fears of pain will have more difficulty in deep
breathing and coughing after their surgery than patients who are not stressed.
Qualitative studies are an important source of inspiration for inductive
hypotheses.
Example of deriving an inductive hypothesis:
In Beck’s (1998) qualitative study of postpartum-onset panic disorder,
one of her findings was a theme relating to self-esteem: “As a result of
recurring panic attacks, negative changes in women’s lifestyles ensued—lowering
their self-esteem and leaving them to bear the burden of disappointing not only
themselves but also their families” (p. 134). A hypothesis that can be derived
from this qualitative finding might be as follows:
Women who experience postpartum onset panic disorder have lower
self-esteem than women who do not experience this disorder.
The other mechanism for deriving hypotheses is through deduction.
Theories of how phenomena behave and interrelate cannot be tested directly.
Through deductive reasoning, a researcher can develop hypotheses based on
general theoretical principles. Inductive hypotheses begin with specific
observations and move toward generalizations; deductive hypotheses have as a
starting point theories that are applied to particular situations. The
following syllogism illustrates the reasoning process involved:
• All human beings have red and white blood cells.
• John Doe is a human being.
• Therefore, John Doe has red and white blood cells.
In this simple example, the hypothesis is that John Doe does, in fact,
have red and white blood cells, a deduction that could be verified.
Theories thus can serve as a valuable point of departure for hypothesis
development. Researchers must ask: If this theory is valid, what are the
implications for a phenomenon of interest? In other words, researchers deduce
that if the general theory is true, then certain outcomes or consequences can
be expected. Specific predictions derived from general principles must
then be subjected to testing through the collection of empirical data. If these
data are congruent with hypothesized outcomes, then the theory is strengthened.
The advancement of nursing knowledge depends on both inductive and
deductive hypotheses. Ideally, a cyclical process is set in motion wherein
observations are made (e.g., in a qualitative study); inductive hypotheses are
formulated; systematic and controlled observations are made to test the
hypotheses; theoretical systems are developed on the basis of the results;
deductive hypotheses are formulated from the theory; new data are gathered;
theories are modified, and so forth.
Researchers need to be organizers of concepts (think inductively),
logicians (think deductively), and, above all, critics and skeptics of
resulting formulations, constantly demanding evidence.
Wording of Hypotheses. A good hypothesis is
worded in simple, clear, and concise language. Although it is cumbersome to
include conceptual or operational definitions of terms directly in the
hypothesis statement, it should be specific enough so that readers
understand what the variables are and whom researchers will be studying.
Simple Versus Complex Hypotheses. For the purpose of this
book, we define a simple hypothesis as a hypothesis that expresses an
expected relationship between one independent and one dependent variable. A
complex hypothesis is a prediction of a relationship between two (or more)
independent variables and/or two (or more) dependent variables. Complex
hypotheses sometimes are referred to as multivariate hypotheses because they
involve multiple variables.
We give some concrete examples of both types of hypotheses, but let us first
explain the differences in abstract terms. Simple hypotheses state a
relationship between a single independent variable, which we will call X, and a single dependent variable, which we
will label Y. Y is the predicted effect, outcome, or consequence of X, which is
the presumed cause, antecedent, or precondition. The nature of this
relationship is presented in Figure 4-1A. The hatched area of the circles,
which represent variables X and Y, signifies the strength of the
relationship between them. If there were a one-to-one correspondence between
variables X and Y, the two circles would completely overlap, and the entire
area would be hatched. If the variables were totally unrelated, the circles
would not overlap at all.
Example of a simple hypothesis:
Patients receiving a warmed solution for body cavity irrigation during
surgical procedures [X] will maintain a higher core body temperature [Y] than
patients receiving a room temperature solution (Kelly, Doughty, Hasselbeck, & Vacchiano,
2000).
Most phenomena are the result not of one variable but of a complex array
of variables. A person’s weight, for example, is affected simultaneously by
such factors as the person’s height, diet, bone structure, activity level, and
metabolism. If Y in Figure 4-1A was weight, and
X was a person’s caloric intake, we would not be able to explain or
understand individual variation in weight completely.
For example, knowing that Dave Harper’s daily caloric intake averaged
2500 calories would not allow us a perfect prediction of his weight.
Knowledge of other factors, such as his height, would improve the
accuracy with which his weight could be predicted.
Figure 4-1B presents a schematic representation of the effect of two
independent variables on one dependent variable. The complex hypothesis would
state the nature of the relationship between Y on the one hand and X1 and X2 on
the other. To pursue the preceding example, the hypothesis might be: Taller
people (X1) and people with higher caloric intake (X2) weigh more (Y) than
shorter people and those with lower caloric intake. As the figure shows,
a larger proportion of the area of Y is hatched when there are two independent
variables than when there is only one. This means that caloric intake and
height do a better job in helping us explaining variations in weight (Y) than
caloric intake alone. Complex hypotheses have the advantage of allowing
researchers to capture some of the complexity of the real world. It is not
always possible to design a study with complex hypotheses.
Practical considerations (e.g., researchers’ technical skills and
resources) may make it difficult to test complex hypotheses. An important
goal of research, however, is to explain the dependent variable as thoroughly
as possible, and two or more independent variables are typically more
successful than one alone.
Example of a complex hypothesis—multiple independent variables:
Among breast cancer survivors, emotional well-being [Y] is influenced
by the women’s self-esteem [X1], their resourcefulness [X2] and their degree of
social support [X3] (Dirksen, 2000).
Just as a phenomenon can result from more than one independent variable,
so a single independent variable can have an effect on, or be antecedent to,
more than one phenomenon. Figure 4-
Example of a complex hypothesis—multiple dependent variables:
The implementation of an evidence-based protocol for urinary incontinence
[X] will result in decreased frequency of urinary incontinence episodes (Y1), decreased urine loss per episode [Y2],
and decreased avoidance of activities [Y3] among women in ambulatory care
settings (Sampselle
et al., 2000).
Finally, a more complex type of hypothesis, which links two or more
independent variables to two or more dependent variables, is shown in Figure
4-1D. An example might be a hypothesis that smoking and the consumption of
alcohol during pregnancy might lead to lower birth weights and lower Apgar scores
in infants.
Hypotheses are also complex if mediator or moderator variables are
included in the prediction. For example, it might be hypothesized that the
effect of caloric intake (X) on weight (Y) is moderated by gender (Z)—that is,
the relationship between height and weight is different for men and women.
Example of a complex hypothesis with mediator:
The quality of life of a family [Y] during the survivor phase after
cancer diagnosis is affected by family resources [X1] and illness survival
stressors such as fear of recurrence [X2], through the mediating variable, the
family meaning of the illness [Z] (Mellon & Northouse,
2001).
In general, hypotheses should be worded in the present tense. Researchers
make predictions about relationships that exist in the population, and not just
about a relationship that will be revealed in a particular sample. Hypotheses
can be stated in various ways as long as the researcher specifies or
implies the relationship to be tested. Here are examples:
1. Older patients are more at risk of experiencing a fall than younger
patients.
2. There is a relationship between the age of a patient and the risk of
falling.
3. The older the patient, the greater the risk that she or he will fall.
4. Older patients differ from younger ones with respect to their risk of
falling.
5. Younger patients tend to be less at risk of a fall than older
patients.
6. The risk of falling increases with the age of the patient.
Other variations are also possible. The important point to remember is
that the hypothesis must specify the independent variable (here, patients’ age)
and the dependent variables (here, risk of falling) and the anticipated
relationship between them.
Directional Versus Nondirectional
Hypotheses. Sometimes hypotheses are described as being either directional or nondirectional. A directional hypothesis is one that specifies
not only the existence but the expected direction of the relationship between
variables. In the six versions of the hypothesis in the preceding list,
versions 1, 3, 5, and 6 are directional because there is an explicit prediction
that older patients are at greater risk of falling than younger ones.
A nondirectional hypothesis, by contrast, does
not stipulate the direction of the relationship. Versions 2 and
Hypotheses derived from theory are almost always directional because
theories explain phenomena, thus providing a rationale for expecting variables
to be related in certain ways. Existing studies also offer a basis for
directional hypotheses. When there is no theory or related research, when the findings
of related studies are contradictory, or when researchers’ own experience leads
to ambivalence, nondirectional hypotheses may be
appropriate. Some people argue, in fact, that nondirectional
hypotheses are preferable because they connote a degree of impartiality.
Directional hypotheses, it is said, imply that researchers are intellectually
committed to certain outcomes, and such a commitment might lead to bias. This
argument fails to recognize that researchers typically do have hunches about
outcomes, whether they state those expectations explicitly or not. We prefer
directional hypotheses—when there is a reasonable basis for them—because they
clarify the study’s framework and demonstrate that researchers have thought
critically about the phenomena under study. Directional hypotheses may also
permit a more sensitive statistical test through the use of a one-tailed test.
Research Versus Null Hypotheses. Hypotheses are sometimes
classified as being either research hypotheses or null hypotheses.
Research hypotheses (also referred to as substantive, declarative,or
scientific hypotheses) are statements of expected relationships between
variables. All the hypotheses presented thus far are research hypotheses that
indicate researchers’ actual expectations.
The logic of statistical inference operates on principles that are
somewhat confusing to many beginning students. This logic requires that
hypotheses be expressed such that no relationship is expected. Null hypotheses (or statistical hypotheses)
state that there is no relationship between the independent and dependent
variables.
The null form of the hypothesis used in our preceding examples would be a
statement such as:
“Patients’ age is unrelated to their risk of falling” or “Older patients
are just as likely as younger patients to fall.” The null hypothesis might be
compared with the assumption of innocence of an accused criminal in our system
of justice: the variables are assumed to be “innocent” of any relationship
until they can be shown “guilty” through appropriate statistical procedures.
The null hypothesis represents the formal statement of this assumption of
innocence.
TIP: If you formulate hypotheses, avoid stating them in null form. When
statistical tests are performed, the underlying null hypothesis is assumed
without being explicitly stated. Stating hypotheses in the null form gives an
amateurish impression.
Hypothesis Testing. Hypotheses are formally
tested through statistical procedures; researchers seek to determine through
statistics whether their hypotheses have a high probability of being correct.
However, hypotheses are never proved through hypothesis testing; rather, they
are accepted or supported. Findings are always tentative. Certainly, if the
same results are replicated in numerous investigations, then greater confidence
can be placed in the conclusions. Hypotheses
come to be increasingly supported with mounting evidence.
Let us look more closely at why this is so.
Suppose we hypothesized that height and weight are related. We predict
that, on average, tall people weigh more than short people. We then obtain
height and weight measurements from a sample and analyze the data. Now suppose
we happened by chance to choose a sample that consisted of short, heavy people,
and tall, thin people. Our results might indicate that there is no relationship
between a person’s height and weight. Would we then be justified in
stating that this study 
proved or demonstrated that height and weight in humans are
unrelated?
As another example, suppose we hypothesized that tall nurses are more
effective than short ones.
This hypothesis is used here only to illustrate a point because, in
reality, we would expect no relationship between height and a nurse’s job
performance. Now suppose that, by chance again, we drew a sample of nurses in
which tall nurses received better job evaluations than short ones.
Could we conclude definitively that height is related to a nurse’s
performance? These two examples illustrate the difficulty of using
observations from a sample to generalize to a population. Other issues, such as
the accuracy of the measures, the effects of uncontrolled extraneous variables,
and the validity of underlying assumptions prevent researchers from concluding
with finality that hypotheses are proved.
TIP: If a researcher uses any statistical tests (as is true in most
quantitative studies), it means that there are underlying hypotheses—regardless
of whether the researcher explicitly states them—because statistical tests are
designed to test hypotheses. In planning a quantitative study of your own, do
not be afraid to make a prediction, that is, to state a hypothesis.
Research
Examples
This section describes how the research problem and research questions
were communicated in two nursing studies, one quantitative and one qualitative.
Research Example of a Quantitative Study. Van Servellen,
Aguirre, Sarna, and Brecht (2002) studied emotional
distress in HIV-infected men and women. The researchers noted that, despite the
fact that AIDS rates have been dropping for men but increasing for women, few
studies have described the health experiences of HIV-infected women or compared
them with those of men. This situation was viewed as especially troubling
because of certain evidence indicating that, once HIV infected, women may be at
greater risk than men for illness-related morbidity and adverse outcomes.
As stated by the researchers, the purpose of their study was “to describe
and compare patterns of emotional distress in men and women with symptomatic
HIV seeking care in community-based treatment centers” (p. 50). The researchers
went on to note that understanding gender differences and similarities in
relation to sociodemographic characteristics, health
status, and stress-resistant resources could “provide important information in
designing gender-specific programs to improve quality of life and reduce
emotional distress in clients affected by HIV” (p. 50).
The conceptual framework for the study was attribution theory, which
offers explanations of links between life stressors and emotional distress.
This framework guided the development of the four study hypotheses, which were
as follows:
Hypothesis 1: Sociodemographic vulnerability
(less than high school education, etc.) will be associated with emotional
distress in both men and women.
Hypothesis 2: Poor physical and functional health status will be
associated with emotional distress in both men and women.
Hypothesis 3: Optimism and social support will be associated with
positive mental health outcomes ... in both men and women.
Hypothesis 4: Women will have higher levels of emotional distress than
men (pp. 53–54).
Data for the study were collected from 82 men and 44 women with HIV
disease in
However, the first hypothesis was not supported in this low-income
sample: there were no significant relationships between any sociodemographic vulnerability indicators and the subjects’
level of anxiety or depression.
Research Example of a Qualitative Study. Beery, Sommers,
and Hall (2002) studied the experiences of women with permanent cardiac
pacemakers.
The researchers stated that biotechnical devices such as pacemakers are
increasingly being implanted into people to manage an array of disorders, yet
relatively little research has examined the emotional impact of such an
experience. They further noted that women may have distinctive responses to
implanted devices because of cultural messages about the masculinity of
technology, but little was known about women’s unique responses to permanent
cardiac pacemakers.
The purpose of Beery and colleagues’ study was to explore women’s
responses to pacemaker implementation, using in-depth interviews to solicit the
women’s life stories. The researchers identified two specific
research questions for their study: “What is the experience of women living
with permanent cardiac pacemakers?” and “How do women incorporate permanent
cardiac pacemakers into their lives and bodies?” (p. 8).
A sample of 11 women who were patients at the cardiology service of a
large hospital participated in the study. During interviews, the women were
asked a series of questions regarding life events that led up to, and occurred
during and after, their pacemaker’s implantation. Each woman participated in
two interviews. An example of the questions asked in the initial interview is:
“What has living with a pacemaker been like for you?” (p. 12). In the follow-up
interviews, more specific questions were asked, such as, “How often do
you think about the pacemaker?” and “When might you be reminded of it?” (p.
12).
The researchers’ analysis revealed eight themes that emerged from the
interview data: relinquishing care, owning the pacemaker, experiencing fears
and resistance, imaging their body, normalizing, positioning as caregivers, finding
resilience, and sensing omnipotence.
SUMMARY POINTS
•A research problem is a perplexing or enigmatic situation that a
researcher wants to address through disciplined inquiry.
• Researchers usually identify a broad topic, narrow the scope of the problem,
and then identify questions consistent with a paradigm of choice.
• The most common sources of ideas for nursing research problems are
experience, relevant literature, social issues, theory, and external sources.
• Various criteria should be considered in assessing the value of a
research problem. The problem should be clinically significant;
researchable (questions of a moral or ethical nature are inappropriate);
feasible; and of personal interest.
• Feasibility involves the issues of time, cooperation of study
participants and other people, availability of facilities and equipment,
researcher experience, and ethical considerations.
• Researchers communicate their aims in research reports as problem statements,
statements of purpose, research questions, or hypotheses. The problem statement
articulates the nature, context, and significance of a problem to be
studied.
•A statement of purpose summarizes the overall study goal; in both
qualitative and quantitative studies, the purpose statement identifies
the key concepts (variables) and the study group or population.
• Purpose statements often communicate, through the use of verbs and
other key terms, the underlying research tradition of qualitative studies, or
whether study is experimental or nonexperimental in
quantitative ones.
•A research question is the specific query researchers want to
answer in addressing the research problem. In quantitative studies, research
questions usually are about the existence, nature, strength, and direction of
relationships.
• Some research questions are about moderating variables that affect the
strength or direction of a relationship between the independent and dependent
variables; others are about mediating variables that intervene between the
independent and dependent variable and help to explain why the relationship
exists.
• In quantitative studies, a hypothesis is a statement of predicted
relationships between two or more variables. A testable hypothesis states the
anticipated association between one or more independent and one or more
dependent variables.
• Simple hypotheses express a predicted relationship between one
independent variable and one dependent variable, whereas complex hypotheses
state an anticipated relationship between two or more independent variables and
two or more dependent variables (or state predictions about mediating or
moderating variables).
• Directional hypotheses predict the direction of a relationship; nondirectional hypotheses predict the existence of
relationships, not their direction.
• Research hypotheses predict the existence of relationships; statistical
or null hypotheses express the absence of a relationship.
• Hypotheses are never proved or disproved in an ultimate sense—they are
accepted or rejected, supported or not supported by the data.
DEVELOPING HYPOTHESIS AND RESEARCH QUESTIONS
Introduction
■
Processes involved before formulating the
hypotheses.
■
Definition
■
Nature of Hypothesis
■
Types
■
How to formulate a Hypotheses in
Quantitative Research
Qualitative Research
■
Testing and Errors in Hypotheses
■
Summary
The research structure helps us create research that is:
Quantifiable Verifiable Replicable Defensible
Corollaries among the model common sense & paper format
|
Model |
Common
Sense |
Paper
Format |
|
Model |
Why |
Intro |
|
Research
Question |
Your
Answer |
Intro
|
|
Develop
a Theory |
How
|
Method
|
|
Identify
Variables (if applicable) |
Expectations
|
Method
|
|
Identify
hypotheses |
Collect/Analyze
data |
Results
|
|
Test
the hypotheses |
What
it Means |
Conclusion
|
|
Evaluate
the Results |
What
it doesn’t Mean |
Conclusion |
|
Critical
Review |
Why
|
Intro
|
