Theory and Variables
Review the differences in descriptive, explanatory, predictive, and prescriptive studies and select one of the predictive or prescriptive articles that you find. What are the independent variables (IV), dependent variables (DV), and mediating variables (MVs) that you find in the article? How do determining IVs, DVs, and MVs help you discern the theory tested in the research? In your particular article, what risk factors are identified, or what recommendations are provided?
SOURCE LESS THAN 5 YEARS
CLASS LESSON:
The introduction of a research article lays the groundwork for the study. The introduction typically describes the problem investigated in the study, along with statistical evidence of the extent of the problem. The research problem arises out of clinical, educational, or administrative practice. After identifying the problem, the researcher outlines the purpose of the research and specifies the questions or hypotheses to be posed in the study.
For example, a nurse identifies a problem of depression in cancer patients who have completed chemotherapy and obtains statistical data from the U.S. National Cancer Institute that describes the extent of the problem. After reading that recreational exercise may relieve depression in cancer patients who have completed chemotherapy, the nurse decides to conduct a study to confirm or disconfirm whether a program featuring recreational exercise relieves depression. The research question is, “Does a recreational exercise program relieve depression in cancer patients?” The problem, purpose, and research question may be diagrammed as the following.
Purposes of Research
Research purposes fall into four broad categories (Polit & Beck, 2011).
Descriptive
Explanatory
Predictive
Prescriptive
Descriptive Research
The purpose of this study is to describe X and Y in ABC population.
A descriptive study does what is implied: The researcher collects data that describe or explicate the characteristics of a population or the nature of a phenomenon using descriptive statistics or qualitative methods, such as interviewing (Polit & Beck, 2011).
Descriptive research is used to elucidate the characteristics of a population, typically such common demographic variables as age, gender, marital status, educational level, religion, and socioeconomic status. The most commonly reported statistics used in descriptive studies are means, standard deviations, frequency counts, and percentages.
These descriptive statistics are also used to report the nature of a phenomenon under study. For example, researchers could administer a survey on exercise to a group of research participants resulting in a description of the extent to which participants exercise on a normal day or to describe participants’ recreational exercise. Likewise, physiological parameters, such as vital signs, are often pertinent variables to describe in a medical study.
Descriptive statistics are often used to control for extraneous or mediating variables that could affect the results of a study. For example, religion might be an important variable to include in a study of spirituality.
Qualitative studies are always descriptive studies, but data are gathered in the form of words rather than numbers. In qualitative studies, participants are describing a situation or a feeling, the researcher is writing field notes about something that is happening in a culture or environment, or a historian is examining original documents. The transcripts from interviews, field notes, or historical documents are then analyzed to determine common themes.
Explanatory Research
The purpose of this study is to determine the correlation between X and Y in ABC population or the purpose of this study is to determine whether Variable X differs between Group A and Group B.
Explanatory studies are those in which researchers calculate correlations between variables or differences in a particular variable among groups.
Descriptive Research
The Question
What is an example of a descriptive study?
Your Answer
Compare Answers
Explanatory Research
The Question
What is an example of an explanatory study?
Your Answer
Compare Answers
Predictive Research
The purpose of this study is to determine what combination of variables best predicts Variable X.
Predictive studies are important in healthcare, because they address questions of risk. Thus, they can produce a theory: A + B + C à D. For example, a patient’s education level + lack of patient stress + reinforcement with two methods of teaching à increased retention of information on self-care.
Prescriptive Research
The purpose of this study is to determine whether applying Intervention A leads to Outcome B.
Prescriptive studies are experimental studies. They also yield a theory that treatment X results in a cure. When several studies confirm the cure, the theory becomes an accepted truth.
Predictive Research
The Question
What is an example of a predictive study?
Your Answer
Compare Answers
Prescriptive Research
The Question
What is an example of a prescriptive study?
Your Answer
Compare Answers
RCTs are the gold standard for intervention research (Melnyk & Fineout-Overholt, 2011). Meta-analyses and systematic reviews of RCTs are the highest forms of evidence on the levels-of-evidence pyramid. RCTs are tightly controlled studies designed to exclude the effects of mediating variables on the dependent variable after an intervention, or independent, variable is applied. Forms of control include comparing the intervention group with a control group and applying a pretest, as well as a posttest, to ensure that the groups are comparable at the beginning of the study. A cohort study comparing groups is used to demonstrate prospectively whether the intervention is effective. Quasiexperimental studies are used when random assignment and control groups and/or pretests are impractical.
Systematic reviews of research can be conducted with all types of research, but for those seeking answers to the best types of interventions, systematic reviews of prescriptive or intervention research are necessary.
Test your understanding of descriptive, explanatory, predictive, and prescriptive research with this exercise.
Variables
Variables are characteristics that differ from one individual to another (Polit & Beck, 2011). A variable could be an age, a score on a test, a blood pressure, a mean value on an opinion survey, or the presence or absence of a medical complication.
Three types of variables are seen in most studies: independent, dependent, and mediating. Simply stated, independent variables (IVs) are the presumed cause of a phenomenon, dependent variables (DVs) are the presumed effects or outcomes, and mediating (extraneous, intervening) variables are intervening factors or alternate explanations for dependent variables. The following diagram illustrates the differences among the three types of variables.
One way to remember the difference between independent and dependent variables is to think of the dependent variable as the variable that depends upon the independent variable; the independent variable comes first. Independent variables are the variables that are manipulated in a study, such as a medical intervention. Dependent variables are those that are measured at the end of a study to see if an intervention worked. Likewise, the mediating variable comes in the middle of the independent and dependent variables. Mediating variables are typically descriptive and include factors such as age, gender, and comorbidities.
In predictive studies, independent variables are the predictive factors that are tested. They are the risks that are being studied. For example, does obesity (the iV) lead to heart disease (the DV)? Predictive studies often use several independent variables as predictors.
One way to remember the difference between independent and dependent variables is to think of the dependent variable as the variable that depends upon the independent variable; the independent variable comes first. Independent variables are the variables that are manipulated in a study, such as a medical intervention. Dependent variables are those that are measured at the end of a study to see if an intervention worked. Likewise, the mediating variable comes in the middle of the independent and dependent variables. Mediating variables are typically descriptive and include factors such as age, gender, and comorbidities.
In predictive studies, independent variables are the predictive factors that are tested. They are the risks that are being studied. For example, does obesity (the iV) lead to heart disease (the DV)? Predictive studies often use several independent variables as predictors.
One way to remember the difference between independent and dependent variables is to think of the dependent variable as the variable that depends upon the independent variable; the independent variable comes first. Independent variables are the variables that are manipulated in a study, such as a medical intervention. Dependent variables are those that are measured at the end of a study to see if an intervention worked. Likewise, the mediating variable comes in the middle of the independent and dependent variables. Mediating variables are typically descriptive and include factors such as age, gender, and comorbidities.
In predictive studies, independent variables are the predictive factors that are tested. They are the risks that are being studied. For example, does obesity (the iV) lead to heart disease (the DV)? Predictive studies often use several independent variables as predictors.
One way to remember the difference between independent and dependent variables is to think of the dependent variable as the variable that depends upon the independent variable; the independent variable comes first. Independent variables are the variables that are manipulated in a study, such as a medical intervention. Dependent variables are those that are measured at the end of a study to see if an intervention worked. Likewise, the mediating variable comes in the middle of the independent and dependent variables. Mediating variables are typically descriptive and include factors such as age, gender, and comorbidities.
In predictive studies, independent variables are the predictive factors that are tested. They are the risks that are being studied. For example, does obesity (the iV) lead to heart disease (the DV)? Predictive studies often use several independent variables as predictors.
Research Questions, Variables, and Theory
Recall that nursing theories are statements of proposed relationships among variables. The research question, “Does the grade on the medical-surgical nursing course predict whether the nursing graduate passes the nurse licensing examination, controlling for nurse IQ?” is another way of making a simple theoretical statement: “The grade in a medical-surgical nursing course predicts whether a graduate nurse will pass the nurse licensing examination, depending upon the nurse’s IQ.”
Research questions should include the variables and population of interest of a study. From the earlier examples, identify the independent, mediating, and dependent variables in these theoretical statements:
The undergraduate GPA of a nursing student is predicted by his or her high school GPA and SAT score, controlling for IQ.
Controlling for the number of deliveries, the use of a doula on a labor and delivery unit increases patient satisfaction.
Pressure sores can be alleviated by a planned turning schedule, considering the patient’s nutritional status.
Hypotheses
Hypotheses are simply statements about relationships among variables. To that end, they reflect theories. Hypotheses are tested with inferential statistics, which are statistics designed to estimate an answer to a research question (Polit & Beck, 2011).
Recall from earlier courses that a null hypothesis states that there is no relationship between variables, and a research hypothesis states that there is a relationship. Null hypotheses are harder to understand than research hypotheses. Why would a researcher state that there is no relationship among variables? A lack of relationship only eliminates one possibility. A relationship allows for decision making based on that finding. However, the choice of hypotheses is up to researchers and may be influenced by the preference of scholars in their fields of study.
Hypotheses will not be seen in descriptive studies, because they merely outline what is. Explanatory studies may pose a research question or a hypothesis. Using the earlier example on fall rates,
the research question would be, “Are age and gender related to fall rates?”;
the null hypothesis would be, “Age and gender are not related to fall rates.”; and
the research hypothesis would be, “Age and gender are related to fall rates.”
Using the prescriptive study example concerning the provision of health services to pregnant high-school students,
the research question would be, “Does providing health services to pregnant high-school students prevent dropping out?”;
the null hypothesis would be, “Providing health services to pregnant high-school students does not prevent dropping out.”; and
the research hypothesis would be, “Providing health services to pregnant high-school students prevents dropping out.”
Relationships Among Types of Studies and the Levels-of-Evidence Pyramid
The relationships among the types of studies described in this lesson and the levels-of-evidence pyramid are illustrated below. The research design of an article is described in the methodology statement when evaluating studies and comparing them to the levels-of-evidence pyramid. Note the type of study and research questions appropriate for each level of evidence. Research from the top of the pyramid provides the best evidence for a change of practice, but patient concerns and expert opinions are also valuable.
Summary
The description of research purposes, questions, hypotheses, and variables in this lesson and the discussion of the relationship of research design to the levels-of-evidence pyramid lay the groundwork for the next lesson. Week 3 will outline the critique of various studies with an emphasis on internal and external validity, trustworthiness in qualitative research, and the critique of quantitative and qualitative research. Practicing the skill of critique is essential for developing a working knowledge of evidence-based practice.
References
Melnyk, B. M., & Fineout-Overholt, E. (2011). Evidence-based practice in nursing and healthcare: A guide to best practice (2nd ed.). Philadelphia, PA: Wolters Kluwer/Lippincott, Williams, & Wilkins.
Polit, D. F., & Beck, C. T. (2011). Nursing research: Generating and assessing evidence for nursing practice (8th ed.). Philadelphia, PA: Lippincott, Williams, & Wilkins.