discussion

profilenurse444
ch24.docx

Chapter 2 &4

Chapter 2: The Research Process and Ways of Knowing

CHAPTER OBJECTIVES

The study of this chapter will help the learner to

· Discuss the philosophical orientations that influence the choice of a research design.

· Contrast the characteristics of quantitative and qualitative research.

· Review the steps involved in the research process.

· Determine the way that a design is linked to the research question.

· Classify research based on characteristics related to intent, type, and time.

· Evaluate which kind of evidence is best provided by quantitative and qualitative research.

KEY TERMS

1. Applied research

2. Basic research

3. Cross-sectional methods

4. Experimental research

5. Longitudinal studies

6. Mixed methods

7. Paradigm

8. Prospective studies

9. Qualitative research

10. Quantitative research

11. Quasi-experimental studies

12. Retrospective studies

Introduction

What is the nature of truth? It is hard to think of a more difficult question to answer. This fundamental question must be considered, however, to ensure that the research process is successful in providing evidence for practice. Research is about the search for truth. There are, however, multiple approaches to determining and describing truth. The successful researcher understands which approach is effective for the particular problem to be solved. The key is to consider assumptions about the nature of the world, the question to be answered, and the intent of the researcher.

The most fundamental questions to be answered in the beginning of a research process are philosophical but necessary ones: What constitutes knowledge? What is the nature of the world, and how can this research reflect that nature? The researcher should carefully consider these issues before proceeding with the design of the inquiry. It is a mistake to jump straight from research question to design without considering the philosophical foundation on which the study will be built.

These philosophical considerations must represent more than the researcher’s view of the world. That is, they must be carefully matched to a design that will address the specific nature of the research question. The goal is to produce knowledge that is relevant and applicable to the body of nursing knowledge and that becomes evidence for practice.

VOICES FROM THE FIELD

When I started my doctorate, I was sure I wanted to do a straightforward quantitative experiment. I like numbers and statistics, so this kind of study seemed to be a natural extension of my interests. My subject, however, was a bit novel: I was trying to build a comprehensive model to measure inpatient nurse workload. I had always worked in hospitals and used patient acuity systems (systems used to measure the intensity of a patient’s care needs) to assess the nursing workload, but a nurse said something that intrigued me: “If all I had to do was take care of my patients, I’d be fine.” I set out to find out what all those other demands were, and how they affected the nurse’s perception of workload.

I found out just how novel this topic was when I tried to do a literature review and discovered that I could not find any relevant literature. There were lots of opinion articles about measuring workload, and plenty of published quantitative studies focused on patient acuity, but none tried to look at workload holistically. Reluctantly, I concluded that I needed to utilize a mixed-methods design—that is, I needed first to figure out what the forces affecting the nurse’s workload were, and then to measure how much impact they had on the nurse’s day.

I conducted a series of focus groups with nurses, observed them during their regular workdays, and interviewed quite a few individually. I found that I could describe many nonpatient demands—equipment needed repair, supplies were missing, and other therapists and technicians interrupted patient care. In addition, there were some macro issues at play: Nurses said that strong teams were able to accomplish more work, but weak teams actually created more pressure. All of the nurses mentioned the effects of good leadership on recruitment and retention, and subsequently on the stability of the nursing staff, which helped build teams.

After theme analysis and triangulating the data from my focus groups, observations, and interviews, I developed a model of the demands on a nurse’s time. This preparation seemed to take forever, but when I finally began to test the model quantitatively, the work went quickly. I was able to determine the elements that directly affected workload and those that had an indirect effect. I figured out that teamwork, leadership, and retention were central to efficient unit operations. Demonstrating caring, communicating with team members, and entering information into the health record also consumed a lot of time. I discovered that “hunting for things” is a legitimate time drain.

This study was a classic case in which answering the research question required both quantitative and qualitative methods. The qualitative phase helped me determine the fundamental things that frustrate a nurse, and the quantitative phase let me demonstrate whether those influences were real and strong.

Janet Houser, PhD, RN

The Research Process

Regardless of the philosophical assumptions made in a specific study, some characteristics are universal to all research studies. Research by its very nature is systematic and rigorous; it is about a disciplined search for truth. “Systematic” implies that decisions are carefully considered, options weighed, and a rational basis documented to support the choices that are made. Those decisions and choices help form the foundation for and build a research study. They also make up phases of study that are more or less completed in sequence. These phases are depicted in  FIGURE 2.1 :

· Define a research problem: Identify a gap in the knowledge of nursing practice that can be effectively addressed with evidence.

· Scan the literature: Complete a systematic review of the literature to determine basic knowledge about the problem, so as to identify relevant evidence and a potential theoretical framework.

· Determine an appropriate design: Select a design that is appropriate for the philosophical assumption, the nature of the question, the intent of the researcher, and the time dimension.

· Define a sampling strategy: Design a sampling plan that details both how subjects will be recruited and assigned to groups, if appropriate, and how many subjects will be needed.

· Collect data: Gather the data using appropriate data collection protocols and reliable, valid methods.

· Analyze data: Apply analytic techniques that are appropriate for the type of data collected and that will answer the question.

· Communicate the findings: Disseminate the findings to the appropriate audiences through conferences and publication.

· Use the findings to support practice: Promote the uptake of the research by linking it to specific guidelines for nursing practice.

/var/folders/6f/zkh6yd_s3mx3pby73s6g7y580000gn/T/com.microsoft.Word/WebArchiveCopyPasteTempFiles/ch02_fig01.jpg

FIGURE 2.1 The Research Process: Building a Study

These phases may look as if they make up steps, with the end of one phase leading directly to the beginning of another. It is, however, misleading to call the research process a series of steps. Such a description implies that the tasks are done in a particular sequence and that the components are distinct and mutually exclusive. In reality, the design of a research study is a fluid process, one that may be considered a work in progress until the final plan is complete. The process may resemble an elaborate game of Chutes and Ladders more than anything else. In this game, progress is made until the player reaches a chute, which will take the player back to a lower level. In research, several things may happen more or less at the same time—for example, the search for a theoretical framework, the literature review, and construction of the research question. Although the researcher may complete most of these tasks and move on to the design of the study, occasionally a situation will arise that prompts the researcher to reconsider the phrasing of the question, or new literature may be published. As a consequence, the phases may be conducted out of sequence, or the researcher may go back and forth between phases. The phases may overlap, or some phases may not be visited at all. So many varieties of research are possible that any depiction of the research process must come with the caveat that it is a general guide that is adapted to the particular situation at hand.

In quantitative research, decisions are usually finalized before data collection begins, although emergent issues may, even then, require adaptation of the research plan. In contrast, in qualitative research, the research plan is adapted based on both the data generated by the respondents and the nature of those data. Qualitative design decisions may not be completed until the final report is written.

In general, the way the research process emerges and the particular phases that are implemented in a research study are based on many characteristics of both the research problem and the researcher. These characteristics and assumptions lend themselves to several general classifications of research. The choice of an overall research classification is the first step in determining the specifics of a research design.

Classification of Research by Philosophical Assumptions About the Nature of the World

The philosophical assumptions that drive the design of a study are rooted in the paradigms of those who are doing the studying. A paradigm is an overall belief system, a view of the world that strives to make sense of the nature of reality and the basis of knowledge. The disciplined study of nursing phenomena is rooted in two broad paradigms, both of which are relevant for nursing research. These two broad paradigms reflect methods that are primarily quantitative (based on the measurement of observable phenomena) or qualitative (based on the analysis of the meaning of events as depicted in the words and actions of others).

Paradigm: An overall belief system or way of viewing the nature of reality and the basis of knowledge.

Quantitative Research

Quantitative research is the traditional approach to scientific research. It is rooted in the philosophical assumptions of positivism and determinism. Positivism assumes that features of the environment have an objective reality; the world is viewed as something available for study in a more or less unchanging form. A related assumption underlying the scientific method is determinism: a belief that events are not random, but rather have antecedent causes. In the face of these beliefs—the existence of an objective reality, in which events can be linked to an associated cause—the researcher’s challenge is to understand the relationships among human phenomena. The task of positivist scientific inquiry, then, is to make unbiased observations of the natural and social world.

Quantitative research: A traditional approach to research in which variables are identified and measured in a reliable and valid way.

Quantitative research involves identifying the variables that represent characteristics of interest and then measuring them in a reliable, valid way. This type of research is characterized by a tightly controlled context that enables the researcher to rule out extraneous effects. Both the way subjects are selected and the protocols for the study are designed to eliminate bias. Statistical analysis is used to establish the level of confidence in the results and to rule out the effects of random error. These conclusions, then, constitute the contribution to scientific knowledge.

There is no doubt that the scientific study of cause and effect in nursing practice is necessary and important for evidence-based practice; quantitative approaches are particularly well suited for answering questions about the nursing actions that can influence outcomes. These studies produce some of the strongest evidence for the benefits of an intervention. Nevertheless, nurses pose many questions that are not adequately addressed by a strict adherence to measurement of an objective reality. In turn, the single adherence to a positivist view has drawn considerable criticism from nurse researchers, and many of these criticisms are legitimate. The nature of nursing care involves helping others attain their health goals, many of which are defined by the individual, not the nurse. Perceptions of quality of life, the meaning of a life event, and the willingness to endure side effects for a therapeutic result are all based on the patient’s construction of reality, not the nurse’s perceptions. In turn, many related questions are better addressed with a process of naturalistic inquiry.

Qualitative Research

Qualitative research is based on a naturalistic paradigm. This belief system is represented by a view of reality that is constructed by the individual, not the researcher. In the naturalistic view, reality is not a fixed entity, but rather exists in the context of what the research participant believes it to be. Qualitative researchers believe that many different views of reality are possible, and all of them are right. An associated belief for the naturalistic researcher is relativism, or the belief that there are always multiple interpretations of reality, and that these interpretations can exist only within an individual. The qualitative researcher, then, believes there is no process in which the ultimate basis for a singular truth can be identified.

Qualitative research: A naturalistic approach to research in which the focus is on understanding the meaning of an experience from the individual’s perspective.

Qualitative methods focus on an understanding of the meaning of an experience from the individual’s perspective. Extended observation of participants, in-depth interviews or focus groups, case studies, and studies of social interaction are examples of qualitative methods. The inquiry process focuses on verbal descriptions and observable behaviors as a basis for analysis and conclusions.

Qualitative methods are appropriate for addressing questions in which the meaning of the patient’s experience is central to understanding the best therapeutic approach. Issues of behavior change, motivation, compliance with a regimen, and tolerance of a treatment are all examples of topics in which the patient’s perception is central to assisting the patient to a healthy state. The analysis of themes that describe the meaning of the experience for the patient is based on words and observations, rather than on measurable phenomena. The researcher establishes a relationship with the subject, and bias is considered an inherent part of the research process. The findings from qualitative studies are used to enhance evidence-based practice by incorporating the patient’s preferences and values into guides for nursing practice.

The differences in philosophy, roles, and methods between quantitative and qualitative research are depicted in  Table 2.1 . These contrasts are made to help the student understand the variations between these two overall approaches. In reality, both types of research have many characteristics in common:

· A disciplined, rigorous approach based on external evidence

· Methods that require samples and the cooperation of individuals

· A focus on the rights of human subjects and ethical guidelines

· An ultimate aim of discovering new knowledge that can be used to improve nursing practice

Table 2.1 Quantitative Versus Qualitative Characteristics

Element

Quantitative

Qualitative

View of reality

Reality is objective and can be seen and measured.

Reality is constructed by the individual.

View of time

Reality is relatively constant.

Reality is continuously constructed.

Context

Reality can be separated from its context.

Reality is embedded in its context.

Researcher approach

Objective, detached.

Personally involved.

Populations studied

Samples that represent overall populations as subjects.

Individual cases, represented as informants.

Measures

Human behavior or other observable phenomena.

Study the meanings that individuals create.

Observations

Analyze reality as definable variables.

Make holistic observations of the total context.

Design

Preconceived and highly controlled.

Emergent and fluid, adaptable to informants’ views.

Analysis

Descriptive and inferential statistics.

Analytic induction to determine meaning.

Generalization

Use inference to generalize from a sample to a defined population.

Transfer knowledge from case analysis to similar cases.

Reports

Objective, impersonal reports in which the researcher’s opinions are undetectable.

Interpretive reports that reflect the researcher’s reconstruction of the meaning of the data.

Many nurse researchers assume they must select only one approach and carry out the study in a pure and inflexible way. In fact, it is the rare study that relies on just one approach or the other. The choices made in research design are probably less about a solely qualitative approach versus a solely quantitative approach, and more about selection from a continuum of choices that may overlap from one approach to the other. Many quantitative studies involve asking the subjects to respond to questions or give opinions in which the participants’ words are later analyzed to enhance the statistical findings. Experimental researchers may rate subject behaviors using scales that contain subjective elements, or they may record their own observations of behaviors. Conversely, many qualitative studies use measurement to determine the reliability of multiple raters in determining themes and to verify the trustworthiness of conclusions. A basic qualitative validation method is triangulation, or the search for multiple sources to confirm the same finding, in which numbers are often retrieved to confirm verbal data. There are many situations in which a blend of methods is appropriate, and these mixed methods designs are becoming more common.

Mixed Methods

Mixed methods are becoming an important tool in nursing research, particularly in evaluation research. Evaluation research is the application of research methods to the study of p rograms, projects, or phenomena. Increasingly, the question is not whether mixed methods are appropriate, but rather how they should be used.

Mixed methods: A research approach that combines quantitative and qualitative elements; it involves the description of the measurable state of a phenomenon and the individual’s subjective response to it.

Mixed-method designs can provide pragmatic advantages when exploring novel or complex nursing problems ( McCusker & Gunaydin, 2015 ). The qualitative data provide a deep understanding of the human experience, while the quantitative data enable the researcher to identify and measure relationships. Research that draws on the strengths of both paradigms is increasingly recognized as essential in all fields—including in medicine, where it is needed to support effective patient care guidelines.

Mixed methods are often applied in an ad hoc way, meaning the researcher initiates the study by using a primarily quantitative or qualitative method, and then integrates elements of the alternative approach as an afterthought. The most effective use of mixed methods, however, occurs when they are employed in a systematic way ( Kettles, Creswell & Zhange, 2011 ). Mixed methods are commonly used in descriptive studies, where they may be used to describe both the measurable state of a phenomenon and the individual responses to it. For example, mixed methods might be used for the following purposes:

· Describe the rate of hand washing on a nursing unit (quantitative) as well as the nurses’ perceptions about the importance of hand washing (qualitative)

· Measure the presence of bacteria on a nurse’s hands after washing (quantitative) and observe the hand-washing steps the nurse used (qualitative)

· Count the number of times a nurse washed his or her hands between patients (quantitative) and record the nurse’s report on the convenience of hand-washing facilities (qualitative)

Choosing a Design

Many considerations go into the choice of a general approach to research design. The philosophical orientation of the researcher is just one element. The nature of the research question, the skills and abilities of the researcher, and access to resources and samples all are important elements to consider prior to choosing the research methodology.

Of primary importance to the selection of an approach is the nature of the research question. Research questions that focus on the effectiveness of an intervention require a scientific approach (assuming effectiveness is defined as an objectively measured outcome). For example, the effectiveness of a skin-care regimen in preventing pressure ulcers is best studied by applying the proposed regimen to one group of patients, applying a standard regimen to another group of patients, and then measuring the rate of pressure ulcer development in both groups. If the regimen is effective, then the subjects getting the new regimen will have a lower pressure ulcer rate than those with the standard regimen. This is the traditional experiment, and it is still one of the most common research designs in health care.

In contrast, research questions that focus on the acceptability of an intervention may require a qualitative approach. The new regimen may be effective, but it may be painful, have an unpleasant smell, or consist of a cream that sticks to clothing. Assessment of these attributes, which will almost certainly affect whether a patient complies with the skin-care regimen, requires asking the patients about their preferences for the treatment and whether the outcome outweighs the unpleasant side effects.

Some of the considerations when choosing an approach are researcher driven. Many researchers have a personal preference for one approach over another. When the research question may be answered in several different ways, or when various aspects of a phenomenon require study before evidence can be deduced, then the researcher’s personal preference may drive the selection of an approach. The skills that are required for quantitative research include the capacity to define variables, recruit subjects, use random assignment methods, create reliable and valid measurements, and analyze results with statistical techniques. The skills that are required for qualitative researchers are quite different. They include the ability to find and select those subjects who can best inform the question, observe and record actions and interactions in detail, skillfully interview subjects or focus groups, and distill meaning from large amounts of word-based data. Both skill sets can require years to develop and hone. It is natural, then, that most researchers find themselves specializing in one approach or the other.

GRAY MATTER

Consider the following elements prior to choosing a research design:

· Philosophical orientation of the researcher

· Nature of the research question or problem

· Skills, abilities, and preferences of the researcher

· Resources and sample access

GRAY MATTER

The following skills are required for quantitative research:

· Defining variables

· Recruiting subjects

· Using random assignment methods

· Creating reliable and valid measurements

· Analyzing results with statistical techniques

GRAY MATTER

The following skills are required for qualitative research:

· Finding and selecting subjects appropriate for the question

· Observing and recording actions and interactions in detail

· Interviewing subjects skillfully

· Distilling meaning from large amounts of word-based data

A host of practical considerations must be addressed when selecting an approach. Quantitative methods require measurement tools, subjects who are willing to undergo experimental treatments (or the risk of no treatment), statistical software, and access to individuals knowledgeable in statistical analysis and interpretation. Qualitative methods need less in the way of tools and software, but they require informants who are willing to be observed or interviewed, often for extended periods of time. The particular individuals who are accessible as well as the material resources required may drive the selection of a feasible research approach.

Theoretical considerations may also influence the selection of a specific design. The researcher may start the design process by deciding which concepts are of interest. The way these concepts interact with each other and create a framework is called a theory. Theoretical models are commonly tested with both quantitative and qualitative designs, and they provide a roadmap for future research. Theoretical and conceptual frameworks are the necessary backbone of a research study. Using a well-founded and well-referenced framework lends credence to the study, but, even more importantly, allows for comparisons across studies as well as building from or between studies. Basing a study on a sound theoretical framework is one way to ensure the research will be systematically designed.

A Note About Using Theoretical Frameworks in Nursing Research

Theory is an attempt to explain the world around us. Nurses become part of the world of health care through an understanding of theories about nursing, which attempt to explain why nurses do what they do. Nursing care is a complex process, and explanations of human actions and interactions can be complicated and difficult to understand without a road map. Theory is a method of mapping these complex processes of human action and interaction that affect patient care and understanding their interrelationship.

The word theory comes from the Greek theoria, which means “vision.” Nurse scientists use theories to explain their visions of reality. Theories are not facts; instead, they are methods of posing what might be reality. Just as there are many visions of reality, so there are many theories that attempt to explain that reality. These theories often form the basis for research studies, in that many aspects of a conceptual model might potentially be the subject of study.

Nurse researchers also use theory as a framework for their studies. Developing a conceptual foundation involves a series of steps that focus on the selection and definition of concepts, concept analysis, relational statements, and conceptual models of action and interaction. In this way, theoretical frameworks form the backbone of a research study. Using a strong framework lends credence to the study’s results, but more importantly allows for the replication of the study and the synthesis of its outcomes into guidelines.

Myriad thoughtful decisions must be made to choose the right approach for a particular research problem. The key word is thoughtful. These decisions should be based on a sound rationale, and the researcher should be able to articulate the basis for these decisions.

Classifications of Research by the Intent of the Researcher

Research is classified by the basic belief system that drives its design features, but it must also reflect the intent of the researcher. There are two kinds of goals for research: (1) to provide new knowledge for the foundation of nursing and (2) to provide knowledge that can be immediately applied to the practice of nursing. The first of these is referred to as basic research; the latter is termed applied research.

Basic research is commonly referred to as theoretical, pure, fundamental, or bench research. One might think of the work done by scientists in laboratories as basic research. It is used to test theories and to build the body of knowledge that forms the foundation for practice, but it does not directly apply to the practice setting. Examples of basic research include measuring neuromuscular responses to stimuli and studying the effects of circulatory volume on neonatal cardiac function.

Basic research: Theoretical, pure, fundamental, or bench research done to advance knowledge in a given subject area.

Applied research is undertaken with the single goal of improving nursing practice. The findings from such research are intended to contribute in some way to a modification of nursing practice. Examples of applied research include investigating the effects of topical drugs on phlebitis and determining the efficacy of specific counseling techniques after the death of a spouse.

Applied research: Research conducted to gain knowledge that has a practical application and contributes in some way to a modification of practice.

Both basic and applied research may be conducted by using quantitative, qualitative, or mixed methods. Most clinical nursing research is considered applied research, and the findings that are generated as evidence for practice are exclusively of an applied nature. This is not to imply that basic research is not valuable. Indeed, one must have a clear understanding of the underlying theoretical and physiological basis for a given nursing practice to understand its mechanisms of effect.

Classifications of Research by the Nature of the Design

Another classification of research is associated with the nature of the design. Experimental researchrefers to studies of cause and effect, which are usually undertaken to determine the effectiveness of an intervention. In an experimental design, some type of randomization method is employed to select subjects or assign them to groups according to how well they represent the population of interest. The researcher manipulates some aspect of the patient’s treatment in a highly controlled setting and compares the outcomes to those for a group that has received no treatment or a standard treatment. If the outcomes are different, the researcher assumes the difference is a result of the treatment because all other variables have been controlled. Experimental designs are characterized by highly structured protocols for sample selection and assignment, intervention, measurement, and analysis. Such designs aim to eliminate bias and control for rival explanations for the outcome.

Experimental research: Highly structured studies of cause and effect, usually applied to determine the effectiveness of an intervention. Subjects are selected and randomly assigned to groups to represent the population of interest.

Nonexperimental designs cover a broad range of studies that do not share these characteristics and, therefore, cannot test cause and effect. Quasi-experimental studies mimic experimental designs in most ways except for the selection and assignment of subjects. Such studies often use convenience samples or existing groups to test interventions. For example, a quasi-experimental study might test an intervention by selecting populations in two different nursing homes, where one group gets the treatment and the other does not. However, subjects are not assigned to the nursing homes randomly.

Quasi-experimental studies: Studies of cause and effect similar to experimental design but using convenience samples or existing groups to test interventions.

Other nonexperimental designs include descriptive research, correlation research, and predictive research. Descriptive research involves the study of a particular situation or event that already exists. The researcher does not manipulate any variables, although the study itself is systematic and thorough. Correlation research focuses on the existing relationships between variables. Such a study might, for example, search for a relationship between a single variable in two populations (e.g., do teens with mothers who had teen pregnancies have a higher teen pregnancy rate themselves?). Correlation studies might also search for relationships between two variables in the same sample (e.g., do overweight teens have higher pregnancy rates?). Predictive research takes the correlation aspect one step further, searching for relationships in which the values of one variable can be used to predict the values of another (e.g., do certain family characteristics predict the risk of a teen pregnancy?). Predictive research is particularly helpful in public health studies and research involving the determination of whether a risk factor will lead to a particular health condition.

Chapter 4: Finding Problems and Writing Questions

CHAPTER OBJECTIVES

The study of this chapter will help the learner to

· Discuss strategies for identifying evidence-based practice problems.

· Describe the process for translating a practice problem into a researchable question.

· Define and contrast problem statements and purpose statements.

· Develop and articulate problem statements and purpose statements.

· Perform a critical analysis of the question, problem statement, and purpose statement from a research article.

KEY TERMS

1. Concepts

2. Deductive

3. Directional hypothesis

4. Hypothesis

5. Inductive

6. Nondirectional hypothesis

7. Null hypothesis

8. Problem statements

9. Purpose statements

10. Replication study

11. Research question

Introduction

The best research starts with two words: “I wonder.” A sense of curiosity is all that is needed to begin the research process. Observations about a practice problem become questions, and these questions lead to nursing research that provides evidence to solve the problem.

Finding and developing significant problems for nursing research is critical to improving outcomes for patients, nurses, organizations, and communities. The evolution of a research problem from a general topic of interest to the articulation of problem and purpose statements serves to narrow the focus to a researchable question. This progression moves the research problem from the conceptual (abstract concepts) to the operational (measurable concepts or variables).  FIGURE 4.1  depicts how the individual steps in translating a problem into a researchable question follow this path.

/var/folders/6f/zkh6yd_s3mx3pby73s6g7y580000gn/T/com.microsoft.Word/WebArchiveCopyPasteTempFiles/ch04_fig01.jpg

FIGURE 4.1 Traditional Evolution of the Research Process

VOICES FROM THE FIELD

The idea for this research study actually had its beginnings in a class on critically reading research that I took. I work on a medical–surgical unit, and my team decided that we would like to try to do a small project. We decided to do some observation and find a question to study, even though it is a really busy unit.

About that time, we had a physician who began doing more bariatric surgery. The standing orders for these patients were to have physical and occupational therapy personnel evaluate the patients and get them up and walking. We had always interpreted that order as being implemented on the next morning, because the therapists were generally not available in the evenings when these patients were in shape to start moving. We had patients who wanted to get up and start walking the first evening, though, so we would help them walk. We noticed that these patients seemed to get less nausea. Limiting nausea and retching is important in these patients because we cannot get their intravenous line (IV) out until they are not vomiting, and retching is very painful for them. They get their pain medications through the IV, and if they are vomiting we cannot switch to oral medications and pull their IV lines. If the patients had to wait until the second day to walk, it seemed they had more nausea and vomiting, and those reactions just backlogged everything. It usually meant that their discharge was delayed until the evening of the second day. We wondered if perhaps the earlier walking was helping with the nausea.

We started with a literature search. Originally, we planned to find a study and replicate it; we never thought we could do a study of our own. We just wanted to duplicate what someone else had done. But there were no studies to be found. We found lots of studies of the effects of ambulation in the postoperative period, but nothing dealing with this specific group of patients, and none of the studies measured nausea as the outcome measure. So we thought, “Maybe we need to do a study.” We were going to do something very simple—not even go through the institutional review board (IRB), but instead undertake an investigation that was more like a quality study. We were nervous about having to go to the IRB; we thought that would be far too deep for us. We thought a little study would be a good way—a really simple way—for the staff to become involved in research, and we thought it was doable.

We had an opportunity to consult a researcher through our evidence-based practice council. The researcher told us, “This is a good study; this is publishable,” and that was a turning point for us. We realized that this topic was as important as what other nurse researchers studied, and we recognized that we had an opportunity to make a contribution to practice.

What started as a simple little question—Does walking affect nausea?—has evolved into something more complicated. Our research question is now based on time—in other words, how soon does the patient have to walk to get a benefit? The process forced us to produce criteria for when a patient is ready to walk, which became a conversation in which the whole staff participated. We introduced another element after consulting with physical therapy personnel. We now have one group in which patients use a bedside pedaler and a second group in which patients walk, and we will see if one approach offers an advantage. That would be helpful to know, because when we have really chaotic days, we may not have a lot of time to stop and help someone walk. If we find that the pedaler does help the patient’s gut “wake up” faster, then we can use this device instead of providing walking assistance, because it takes much less time.

When the people on the unit realized that our goal was publication, they got on board with our research. We have learned to appreciate the nurses with whom we work who have stayed in medical– surgical nursing. One of the driving forces behind this effort was our quest to gain some respect for the fact that we are a highly qualified group of nurses who care deeply about patient care and doing the right thing for patients. Taking our nursing practice to the next level through publication of original research could be a real source of pride for the staff—and we think that is why they are so solidly behind it.

Our study has encouraged us to look at our whole nursing practice and realize it is not insignificant—that our research is something someone would want to read. Now that we have finished the IRB process, we have realized, yeah, we really can do that.

Maureen Wenzel, FNP-BC

The traditional method for finding and developing research problems suggests a deductive, sequential process moving from a general interest to the development of a research question.  FIGURE 4.2  illustrates how the individual steps might look in the development of a specific researchable question.

Deductive: A process of reasoning from the general to the specific.

In truth, the process of finding and developing research problems can be as chaotic as a busy parking lot. Some motorists drive their cars headfirst into the spaces, some motorists back their cars into the spaces, and still other motorists drive their cars into and out of the spaces until their cars are properly positioned. Research question development likewise is amenable to many different approaches. Some researchers do, indeed, use a sequential set of steps to arrive at a specific and well-articulated question. Many other nurse researchers use nontraditional methods for finding and developing research problems. These processes may be more inductive, in which specific observations serve as the starting points, leading to a general focus or interest. This approach is more common in evidence-based practice studies, as specific problems generally spark interest in finding an overall solution via research.  FIGURE 4.3  demonstrates a nontraditional example of finding and developing a research problem. Still other methods may begin somewhere in the middle of the traditional process by recognizing a gap, then proceed to identify the big picture as well as the specific research question.

/var/folders/6f/zkh6yd_s3mx3pby73s6g7y580000gn/T/com.microsoft.Word/WebArchiveCopyPasteTempFiles/ch04_fig02.jpg

FIGURE 4.2 Example of a Traditional Evolution of the Problem

Inductive: A process of reasoning from specific observations to broader generalizations.

/var/folders/6f/zkh6yd_s3mx3pby73s6g7y580000gn/T/com.microsoft.Word/WebArchiveCopyPasteTempFiles/ch04_fig03.jpg

FIGURE 4.3 Nontraditional Example of the Evolution of a Research Problem

Regardless of the approach—deductive, inductive, or somewhere in between— finding and developing research problems may be a process best characterized as a work in progress. The goal is always the same, though: to narrow the focus of the research problem so that a feasible research question emerges. The importance of narrowing the focus of the research cannot be overstressed. Research problems that have not been narrowed generate too many concepts and relationships to test.

Concepts: Abstract ideas or topics of interest that must be narrowed to researchable questions to be investigated.

The initial phase of implementing evidence-based practice is identifying the need for change that will improve patient care ( Facchiano, Snyder & Nunez, 2011 ). The need for change may become evident to nurses because of emerging problems or new knowledge ( Boswell & Cannon, 2012 ). During this phase, the topic is selected and preliminary, problem-based questions are considered. Key words in problem-based questions drive the search and retrieval of literature and set the stage for research design. This narrowing process is used both in the research process and in evidence-based practice ( Boswell & Cannon, 2012 ).

Subjects or topics that are overly broad are problematic for researchers because they mean that methodological complexities increase, expert methodologists are required, and resource demands (for example, money, people, and time) increase. With every additional concept or associated relationship examined, the feasibility of the study may be affected.

The primary objective of nursing research is to increase knowledge so as to improve nursing practice, but, of course, it can be accomplished only if the research is actually completed. Some researchers spend a lifetime studying a single concept; others spend their careers completing multiple, small studies. There is no shame in starting, or staying, small. Narrow questions are far easier for the novice researcher to address and may help the nurse learn skills that can then be applied to larger, more complex studies.

As the research problem moves from a broad topic of interest to a narrower, researchable question, measurable variables and outcomes should become evident. It is helpful to consider the type of concept under investigation when approaching this determination. Concepts may represent many things, but they generally fall into three categories: patient sensitive, staff member sensitive, or organizationally sensitive. Concepts studied by nurses, especially novice nurse researchers, should be limited to those within the nursing span of control. For example, a nurse interested in the development and severity of hematomas at arterial access sites after diagnostic or interventional arteriography would be better suited to asking research questions about positioning and turning the patient, rather than addressing questions about the method of arterial access used for the procedure. The first research question identifies nurse-sensitive concepts (repositioning and turning), whereas the second research question identifies medically sensitive concepts (arterial access methods).  Table 4.1 provides examples of each of these kinds of concepts.

GRAY MATTER

Nursing research concepts can be organized into three categories:

· Patient sensitive

· Staff member sensitive

· Organizationally sensitive

Finding and Developing Research Problems

Often, research problems find the nurse, rather than the other way around. Frustrations with ineffective procedures, the search for a “better way,” or the need to help a single patient may motivate the nurse to seek research-based evidence to improve patient care. The search for research problems is one of the easiest parts of the research process; researchable problems surround the contemporary nurse in practice.

Sources of Research Problems

Researchable problems can come from an almost unlimited number of sources ( American Association of Colleges of Nursing [AACN], 2013 Boswell & Cannon, 2012 Schmidt & Brown, 2011 ). The following are some sources for researchable problems:

· Clinical practice observations

· Educational processes and experiences

· Consumer/customer feedback and personal experience

· Theoretical models and frameworks

· Professional literature

· Performance improvement studies

· Research reports and priorities

· Social issues

Clinical Practice

Research problems may be generated from active, passive, or other organizational activities. Active methods of problem discovery include experiences with direct patient care and discussion with other members of the healthcare team through formal or informal communications. Patient problems, ineffective clinical procedures, and changes in protocols all present opportunities for research. Passive methods for identifying problems include medical record review and observation. Data collection activities, such as those performed for quality improvement or risk management purposes, may also bring problems to light.

Table 4.1 Examples of Patient-, Nursing-, and Organizationally Sensitive Concepts

Patient-Sensitive Concepts

Nurse-Sensitive Concepts

Organizationally Sensitive Concepts

Anxiety

Burnout

Cost

Skin integrity

Lower back injuries

Length of stay

Functional independence

Medication errors

Readmission

Blood glucose

Pain management

Resource utilization

Blood pressure

Patient falls

Satisfaction with nursing care

Quality of life

Restraint prevalence

Satisfaction with nursing care

The need for confirmed evidence-based practice generates numerous research questions. Nurses are in a unique position to change and confirm clinical practices. Through structured decision-making processes, nurses, other members of the healthcare team, and patients and families together have the ability to improve healthcare-related processes, outcomes, safety, satisfaction, and costs ( Boswell & Cannon, 2012 ). Nursing practice comprises a mixture of evidence and tradition. It is essential that nurses continually ask themselves and one another the crucial questions: What is the evidence for this approach to nursing care? Can I solve this problem with evidence?

Educational Processes and Experiences

Nursing students who are taking research courses are required to develop problem, purpose, and research statements from required and/or self-determined topics. Research-focused educational institutions may also support institution-driven research activities examining interests such as age-specific care, the effects of caring, or nursing shortage outcomes. Within the assignments and disciplined inquiry that occur during the educational experience, particularly graduate study, nursing students and their mentors may generate many researchable problems.

Consumer Feedback or Personal Experience

Research problems may be generated from consumer or personal experience. Such problems are distinguished from social issues; that is, consumers of health care have the opportunity to share ideas about and perceptions of care and treatment, and to suggest research questions that can potentially promote health throughout the continuum of care, improve patient outcomes and safety, and influence delivery models and cost-effectiveness ( Centers for Medicare & Medicaid Services [CMS], 2013 Hospital Care Quality Information from the Consumer Perspective [HCAHPS], 2013 Patient-Centered Outcomes Research Institute [PCORI], 2013 ). Nurses may also have experiences as family members or as patients themselves. This unique perspective, in which nurses view nursing care from the other side of the bed, gives rise to opportunities to transform that care. Research problems generated from personal experience may expose clinical, consumer, social, and other opportunities for improvement and drive the research priorities of both individuals and organizations.

Research problems generated from the results of activities aimed at soliciting consumer and/or patient feedback may be generated from the following sources:

· Patients and customers of the institution, both external and internal

· Leaders who represent the interests of specific services (e.g., cardiac care)

· Departments (e.g., coronary care unit) and service lines (e.g., cardiac services) within an organizational structure

· Members of general or specialty professional organizations (e.g., American Nurses Association, American Association of Critical-Care Nurses)

· Advisory boards and other consumer/customer input organizations

· Accrediting, regulatory, and safety organizations

The feedback garnered from these groups may be used to generate problem statements, purpose statements, research questions, and testable hypotheses, as well as to set priorities for performance improvement or other research activities. Feedback may be solicited through surveys, during formal and informal meetings, and at conferences and workshops, or it may be received electronically.

Professional Literature

Research problems may be generated from the results of professional literature reviews. Sources of professional literature reviews include clinical and nonclinical works, databases, and letters and opinions. Clinical works include books and journals addressing nursing and medical topics, both general and specialty. Nonclinical works include books and journals presenting non-nursing and nonmedical topics from other fields of study that may be generalized into an appropriate, researchable problem to expand nursing knowledge. Many databases, both clinical and nonclinical, are capable of provoking inquiry. Examples of these databases include those that hold data from previous studies (clinical) and census data (nonclinical). Many research problems have been developed by using the data collected by other researchers and taking a unique approach to the analysis.

Published letters and opinions are another interesting source of research problems. Letters and opinions written by nurses and other medical professionals often express concern, as well as issue directives, about researchable problems, gaps in current knowledge, limitations of available research, and recommendations for future research.

Performance Improvement Activities

Performance improvement activities, also known as quality improvement activities, are used to improve processes and outcomes and to meet regulatory requirements. Tools and techniques specific to performance improvement activities do not meet the requirements of traditional research methods. In particular, performance improvement studies are often characterized by methodological limitations, a lack of control over extraneous variables, violation of assumptions for statistical testing, and small sample sizes in a single setting; all of these factors affect the generalizability of their findings. Nevertheless, the results of performance improvement activities may be used as a springboard to engage in formal research activities. Researchable problems may start as performance improvement activities and expand into formal research projects with alterations in methodological approach, sampling strategy, and informed consent procedures.

Research Reports and Priorities

Research problems may also be generated from the outcomes of other research studies and evidence-based practice reviews. Previous research may influence the generation of research problems either directly or indirectly. By convention, most research reports include a section on “suggestions for future research,” which outlines ways to extend and expand on the currently available research. Researchers may directly influence the generation of subsequent research problems by explicitly stating remaining problems, gaps, and questions. One means of directly influencing subsequent studies is the type of research known as a replication study. Replication studies may be used to validate findings and knowledge, increase their generalizability (population and setting), or eliminate or minimize limitations (methodology). Replication studies are good exercises because they increase the knowledge of inexperienced researchers. Research also may indirectly influence the generation of a problem when the reader identifies a problem with the written report (e.g., a discrepancy, gap, inconsistency, or unidentified limitation) or disagrees with the methodology and/or results of the original investigator.

Replication study: A study generated from previous research studies in which the research is reproduced to validate findings, increase generalizability, or eliminate or minimize limitations.

The directives and recommendations of individuals and organizations may also serve as a source of research problems. In developing these directives and recommendations, individuals (educators and researchers) and organizations (clinical, educational, funding, and regulatory) apply their expertise to identify problems and gaps in current knowledge. Some of these experts may have developed problem statements and research questions to prioritize future research.

Social Issues

Research problems may be generated from social issues. Social issues include, but are not limited to, the effects of age, culture, education, gender, income, race, religion, and sexual preference. Social issues may be examined in the context of current events, the environment, and health policy. They may also be addressed in relation to local, state, national, or international populations. For example, obesity may be considered in terms of age, gender, race, and other social classifications. Notably, the “obesity epidemic” has given rise to equally epidemic rates of diabetes and heart disease. Additionally, obesity and its associated comorbidities affect national issues such as healthcare expenditures (percentage of the gross national product) and the ability of the government to maintain military readiness. Given this condition’s widespread impacts, it should not be surprising that a large number of organizations are examining obesity-related research questions. Collaborative, multidisciplinary studies addressing research questions that are clinically and socially relevant and that could potentially improve outcomes in local, national, and international communities are more likely to receive support and to receive funding ( White, 2012 ).

This list of sources for research problems is not exhaustive. Research problems are often the product of both internal and external driving forces; they are seldom generated from a single source. Any process or outcome associated with patient care, staff members’ work environment, or organizational success may become the basis for study. The potential nurse researcher can scrutinize the current practices and ask the following questions:

· Why are we doing it this way?

· Is there a better way of doing it?

· Should we be doing it at all?

All of these questions may give rise to researchable problems, whose solutions may add valuable evidence to the effective practice of nursing.

Articulation of Research Problem Statements

Research problem statements are declarations of disparity: the difference (gap) between what is known and what needs to be known about a topic. They articulate a discrepancy that is to be addressed by the research process. The disparity, whether a small gap or a large chasm, defines the area(s) of concern and focuses the research methods ( Boswell & Cannon, 2012 Schmidt & Brown, 2011 ). Most problem statements are explicitly stated; however, some problem statements may be inferred. The inferred research problem statement may describe the importance or potential consequences of the disparity as it pertains to clinical practice.

Problem statements: Statements of the disparity between what is known and what needs to be known and addressed by the research.

Problem statements, explicitly stated or inferred, are usually found at the beginning of a research report, in the introduction, or in the review of literature, and they may be repeated throughout the written report. The idea of a single problem statement is misleading; problem statements may resemble problem paragraphs and often can be several sentences long. Problem statements may be written as either questions or statements, and well-written ones contain clear, concise, and well-defined components.

Two examples of well-written problem statements from published articles follow:

· “The background research identified eight articles that looked at the gendered experience of being a nurse, six from the male perspective and two from the female, but none comparing these experiences” ( Rowlinson, 2013 , p. 218).

· “The majority of studies were quasi-experimental, and all examined disposable infant diapers under conditions of high humidity and/or radiant heat sources. One study found that no significant changes occurred after six hours; all others found that changes in diaper weights occurred. The question remained unanswered as to what changes might occur in diaper weights over time in open-air, open bed patient environments” ( Carlisle et al., 2012 , p. 224).

Development of Research Purpose Statements

Whereas research problem statements identify a gap in knowledge that requires disciplined study, research purpose statements are declarations of intent. Purpose statements indicate the general goal of the study and often describe the direction of inquiry ( Boswell & Cannon, 2012 Schmidt & Brown, 2011 ). The purpose of the research should be clearly stated.

Purpose statements: Declarative and objective statements that indicate the general goal of the study and often describe the direction of the inquiry.

Purpose statements are written as objective statements. They are easily identified in reports by their inclusion of words such as aim, goal, intent, objective, or purpose. They contain clear, concise, and well-defined components including key variables to be studied, possible interrelationships, and the nature of the population of interest.

Two examples of well-written purpose statements from published articles follow:

· “The objective was to explore, analyze, and compare and contrast the gender differences with the existing literature” ( Rowlinson, 2013, p. 219 ).

· “The purpose of this study was to examine changes in disposable infant diaper weights at selected intervals post-wetting” ( Carlisle et al., 2012 , p. 224).

Researcher bias is a limitation that may cause readers to question the validity of research processes and outcomes. When developing purpose statements, researchers should use unbiased verbs such as compare, describe, develop, discover, explore, test, and understand and avoid biased verbs such as demonstrate, prove, and show.

· Example of using an unbiased verb: The purpose of the study was to explore the effects of music therapy on speech recovery in adult stroke patients in a rehabilitation facility.

· Example of using a biased verb: The purpose of the study was to prove that music therapy improves speech recovery in adult stroke patients in a rehabilitation facility.