Nursing Research (FBNG) Spring 2012 Dr. Susan Norwood
Oct 08, 2014
Nursing Research (FBNG)
Spring 2012
Dr. Susan Norwood
Overview of content What variables are
Why this information is important
Types of variables
Defining variables
Research purposes and variables
Theoretical frameworks and variables
Variables and doing a literature review
Practice!
What are we talking about? A “variable” is something that can be different among a
study’s participants or that can change as a part of a study.
All studies have variables.
There are different types of variables –
each is managed differently or has different
roles in the research process.
Why this is important – Different types of variables are managed in different
ways -- you need to know what to do with the different variables in your study.
Implications for research design (study organization)
Implications for data collection (what you will measure)
Implications for enhancing study credibility (research control strategies)
Implications for needed data analysis procedures
Types of variables • Independent variables
• Dependent variables
• Extraneous variables
• Demographic variables
Independent variables The cause in a presumed cause and effect
relationship.
• The study intervention or treatment
• The situation that is expected to result in a certain outcome.
• May be introduced and manipulated by the researcher (experimental variable)
• May be naturally occurring.
• May be under the control of the study’s participants.
Dependent variables The effect in a cause and effect relationship.
• The results or outcome of a treatment or intervention.
• Differences in amounts or qualities of the dependent variable depend on differences in the independent variable.
X Y
Extraneous variables Variables in a research situation that exert an
unintended effect on the dependent variable.
Extraneous variables are not a part of a planned intervention or experiment
Function as rival explanations for a study outcome.
Can be considered nuisance or “noise” factors.
z z z z z
X Y
Extraneous variables, con’t
A confounding variable (or confounder) is a special type of extraneous variable.
Confounding occurs when the association between an independent and dependent variable (cause and effect relationship) is either partly or completely due to a third factor that is not a part of the experiment.
Confounders are related to the independent variable and have a known independent causal effect on the dependent variable.
Extraneous variables, con’t
An example of confounding –
Caffeine consumption during pregnancy
Cigarette smoking
Low birthweight infant
Extraneous variables, con’t
The Independent variable is the real cause –
Confounder present
Confounder absent
Independent variable present
++++++ ++++++
Independent variable absent
++
Extraneous variables, con’t
The confounder is the real cause –
Confounder present
Confounder absent
Independent variable present
++++++ ++
Independent variable absent
++++++
Demographic variables Characteristics of a study’s participants.
• Often function as extraneous variables.
• Sometimes seem to function as an “independent” variable.
Important demographic variables – age, sex, education, socioeconomic status, race/ethnicity, health status
Researcher roles with different variables -- Independent variables – must be consistently applied
Dependent variables – must be carefully measured
Extraneous variables – must be recognized and controlled, if possible
Demographic variables – must be documented; sometimes are controlled.
Defining variables All variables need to have both a conceptual definition
and an operational definition –
Conceptual definition – the “thinking” definition of a variable. A description of the concept that is the foundation of the variable; involves using other concepts. The “mental picture” of the variable that the researcher is using.
Example – “Pain is conceptualized as a
subjective feeling of discomfort.”
Defining variables, con’t
Operational definition – the “measurement definition” of a variable. A description of how the quality or amount of a variable will be determined for the purposes of the study.
Example – “Pain will be operationalized as the participant’s self-report of amount of pain, as reported on a 0 to 10 point scale, where 0 indicates no pain at all and 10 indicates the worst pain imaginable.”
Defining variables, con’t
Discrete versus continuous variables –
Discrete variables – variables that differ in terms of quality or characteristics.
• When “measured,” these variables are placed into categories.
• Categories represent differences in quality (nominal-level data) or in relative amount (ordinal-level data).
Examples – sex, mild/moderate/severe pain
Defining variables, con’t
Continuous variables – variables that differ in amount.
• Measurement of these variables reflects how much of the variable is present, in absolute terms.
• The measurement values fall along a continuum.
• These variables constitute interval or ratio-level data.
Examples – pulse rate, temperature, length of hospitalization, number of falls.
Research purposes and variables -- • All research has extraneous variables!
• Cause and effectiveness research has independent and dependent variables.
• Descriptive research focuses on single “research variables.”
• Correlational research focuses on linkages between pairs (or sets) of variables.
• Predictive research has “predictor” and “outcome” variables.
Research purposes and variables, con’t
Three conditions for detecting a cause and effect relationship –
Temporality – the independent variable (or how it changes) must take place before the dependent variable (and its changes).
Influence – the effect of the independent variable on the dependent variable can be detected with statistics, and the probability that the relationship was caused by chance is small.
Specificity – rival explanations for the observed relationship between the independent and dependent variable can be ruled out.
Theoretical frameworks and variables -- A theoretical framework is an explanation of how
different concepts are related to one another.
• The ideas to be tested in a study
• The hunches and assumptions that guide a study
• A “map” of how the different concepts in the research problem are linked to each other
Concepts are too abstract to be measured directly – so they are translated into variables.
Variables and doing a literature review -- Variables can be useful as search terms for doing a
literature review !
• Intervention of interest (independent variable)
• Outcome of interest (dependent variable)
• Population of interest (demographic variables)
Let’s practice --
Question #1 -- Identify the independent variable in the following
research question: “How does effectiveness of smoking cessation education for teenage boys differ when it is provided by nurses compared to physicians?”
a. Smoking cessation education
b. Teenage boys
c. Type of provider (nurse or physician)
Question #2 -- Identify the dependent variable in the following research
question: “How does gender of health care provider affect compliance with iron supplementation among pregnant women?”
a. Compliance
b. Gender of provider
c. Pregnant women
Question #3 -- Identify the independent variable in the following
research question: “Does average hours of sleep per night differ for second term MSN students who are married versus not married?”
a. Hours of sleep
b. Marital status
c. Term in MSN program
Question #4 -- “The purpose of this study is to explore the relationship
between teen age girls’ participation in sports and self esteem.” What is the independent variable in this study?
a. Participant gender
b. Participation in sports
c. Self esteem
Question #5 -- Which of the following would be operational definition
for obesity?
a. A body mass Index (BMI) of > 30
b. A condition characterized by excess body fat.
c. A risk factor for a variety of health conditions.
Question #6 -- Which of the following statements is an example of a
conceptual definition?
a. Depression is defined as a subjective sensation of pervasive sadness.
b. Depression is differentiated as situational or nonspecific.
c. Depression is determined by score on
the Beck Depression Inventory.
Question Question #7 -- A research report states, “Heart rate was recorded as
beats per minute counted by palpating the radial pulse for 15 seconds and multiplying by 4.” This statement is an example of a(an)
a. conceptual definition
b. intervention or research protocol
c. operational definition.