Study Variables

Dr. Tekeste
Biostatistics-II
Sep, 2024
 Study Variables
• A variable is a measurable characteristic that
varies.
• It may change from group to group, person to
person, or even within one person over time.
• Variables are things that we measure, control, or
manipulate in research.
E.g. time students spent on social media and their
academic performance.
Outcome – dependent – Predictand (Y)
Exposure – independent – Predictors (X)
• The objective is to assess the relationship
between some factor of interest (the
independent variable), which we will
sometimes call exposure, and an outcome
variable (the dependent variable).
 What are the different types of variables used in
research?

There are six common variable types:

1. Continuous and categorical variables
2. Independent Vs Dependent Variables
3. INTERVENING VARIABLES.
4. MODERATOR VARIABLES.
5. CONTROL VARIABLES.
6. EXTRANEOUS VARIABLES.
• Continuous and categorical variables
• Independent Vs Dependent Variables
 1. Continuous and categorical variables
 Continuous variables: are variables that can be
represented along a continuum e.g. age of humans
can range from 0 to cover 100 years. Weight is a
continuous variable that can take on values that are
not whole numbers.
 Categorical variables: are those that take on a much
smaller range of values representing discrete
categories rather than incremental placement along
a continuum. (e.g the gender of a person has only
two values – male or female.
 Blood types (A, AB, and O), being pregnant/not pregnant,
smoker/nonsmoker).
 2. Independent Vs Dependent Variables:
 Independent variables: are experimental variables
over which the investigator has control. You can be
able to manipulate or vary the variables.
 Dependent variables: are those that are associated
with the changes introduced by varying the
independent variables.
– Example: Assume you are interesting in answering the
question:” Does smoking cause lung cancer?
– The presumed cause (smoking) is referred to as
independent variable as it can be varied.
– The presumed effects is referred to as the dependent
variable.
 Independent Vs Dependent Variables
A dependent variable is a factor whose value
depends on the level of another factor, which is
termed an independent variable.
 In the example of cigarette smoking and lung cancer
mortality, duration and/or number of cigarettes
smoked are independent variables upon which the
lung cancer mortality depends (thus, lung cancer
mortality is the dependent variable).
 IVs and DVs
• Independent variables (IVs) are experimental
variables over which the investigator has control.
• You can be able to manipulate or vary the
variables.
• Dependent variables (DVs) are those that are
associated with or related to the changes
introduced by varying the independent variables.
The observational studies are distinguished by
the point in time when measurements are made
on the dependent and independent variables, as
illustrated below.
 In cross-sectional studies, both the dependent and
independent (outcome and exposure) variables are
measured at the same time, in the present.
 In case-control studies, the outcome is measured
now and exposure is estimated from the past.
 In prospective studies, exposure (the independent
variable) is measured now and the outcome is
measured in the future.
 observational studies
Time of measurement
Past Present Future
Cross-sectional exposure &
outcome
Case-Control: exposure outcome

Prospective: exposure outcome

Cross-Sectional Versus Longitudinal Looks at Data

Prospective studies are sometimes also known as longitudinal studies, since
people are followed longitudinally, over time. Examination of longitudinal data
may lead to quite different inferences than those to be obtained from cross-
sectional looks at data. For example, consider age and blood pressure.
Variable
 Independent variable – In an
experiment, the treatment or
condition manipulated by the
experimenter.
 Dependent variable – In an
experiment, any aspect of a subject's
behaviour that is measured after the
administration of a treatment; the
expected effect of a treatment.
 INDEPENDENT (EXPOSURE) VARIABLES
Are those that the researcher has control over.
This "control" may involve manipulating existing
variables (e.g., reducing salt intake).
Whatever the case may be, the researcher
expects that the independent variable(s) will
have some effect on (or relationship with) the
dependent variables (eg. BP).
 DEPENDENT VARIABLES
• Show the effect of manipulating or introducing
the independent variables. For example, if the
independent variable is reducing salt intake, then
the dependent variable might be level of BP.

• In other words, the variation in the dependent

variable depends on the variation in the
independent variable.
 Dependent variable
• A dependent variable is a factor whose value
depends on the level of another factor, which is
the independent variable.

• In the example of cigarette smoking and lung

cancer mortality, duration and/or number of
cigarettes smoked are independent variables
upon which the lung cancer mortality depends
(thus, lung cancer mortality is the dependent
variable).
• Example: Assume you are interesting in
answering the question:” Does smoking cause
lung cancer?
• The presumed cause (smoking) is referred to
as independent variable as it can be varied.
• The presumed effects (lung cancer) is referred
to as the dependent variable.
 EXTRANEOUS VARIABLES
• Are those factors in the research environment
which may have an effect on the dependent
variable(s) but which are not controlled.
• Extraneous variables are dangerous. They may
damage a study's validity, making it impossible
to know whether the effects were caused by the
independent or some extraneous factor.
• If they cannot be controlled, extraneous
variables must at least be taken into
consideration when interpreting results.
• A confounding variable is an outside influence
that changes the effect of a dependent and
independent variable.
• This extraneous influence is used to influence
the outcome of an experimental design. ...
• Confounding variables can ruin an experiment
and produce useless results
 CONTROL VARIABLES

• It is not possible to consider every variable in a

single study. Therefore, the variables that are
not measured in a particular study must be held
constant, neutralized/balanced, or eliminated,
so they will not have a biasing effect on the
other variables.
• Variables that have been controlled in this way
are called control variables
 Intervening/mediating variable
• A variable that explains a relation or provides a
causal link between other variables.
• Example: The statistical association between income
and longevity needs to be explained because just
having money does not make one live longer. Other
variables intervene between money and long life.
People with high incomes tend to have better
medical care than those with low incomes. Medical
care is an intervening variable.
• It mediates the relation between income and
longevity.
 Moderator variable
• A moderator variable is a third variable that affects
the strength of the relationship between a
dependent and independent variable.
• In correlation, a moderator is a third variable that
affects the correlation of two variables.
• In a causal relationship, if x is the predictor variable
and y is an outcome variable, then z is the moderator
variable that affects the casual relationship of x and
y.
• Most of the moderator variables measure causal
relationship using regression coefficient. The
moderator variable, if found to be significant, can
cause an amplifying or weakening effect between x
and y.
• A mediating variable explains the relation
between the predictor (independent) and the
criterion (dependent). It is often depicted as the
following figure where MV is the mediator.
• A mediator can be a potential mechanism by
which an independent variable can produce
changes on a dependent variable.
• When you remove the effect of the mediator, the
relation between independent and dependent
variables may go away.
A moderator is a variable that affects the
strength of the relation between the predictor
and criterion variable. Moderators specify when
a relation will hold. It can be qualitative (e.g.,
sex, race, class…) or quantitative (e.g., dosage or
level of reward).
A moderator variable alters the effect that an
independent variable has on a dependent variable, on
the basis of the moderator’s value. The moderator thus
changes the effect component of the cause-effect
relationship between the two variables. This
moderation is also referred to as the interaction effect.
 Study designs and Variables
• The observational studies are distinguished by the
point in time when measurements are made on the
dependent and independent variables, as illustrated
below.
 In cross-sectional studies, both the dependent and
independent (outcome and exposure) variables are
measured at the same time, in the present.
 In case-control studies, the outcome is measured
now and exposure is estimated from the past.
 In prospective studies, exposure (the independent
variable) is measured now and the outcome is
measured in the future.
 Measures of Relative Risk:
• Risk is the likelihood that a particular event will
occur within a particular population.
• The relative risk compares the likelihood that a
disease or outcome will occur among individuals
who have a particular characteristic, exposure,
or risk factor with the likelihood that the
outcome will occur in individuals who do not
have it.
 Relative risk and Study Design
• For prospective studies, the calculation of
relative risk is straight forward e.g. eating carrots
and poor eyesight, smoking and developing lung
cancer
Case-control studies.
• Starts with patients already having poor vision,
sorts them according to who consumes carrots,
and makes the comparison with an
independently selected control group.
 Measures of Relative Risk:
• In epidemiologic studies, we are often interested
in knowing how much more likely an individual is
to develop a disease if he/she is exposed a
particular factor than the individual who is not
so exposed.
• A simple measure of such likelihood is called
relative risk (RR).
 Relative Risk
• Relative Risk is the ratio of two incidence rates:
the rate of development of the disease for
people with the exposure factor, divided by the
rate of development of the disease for people
without the exposure factor.
• Suppose we wish to determine the effect of high
blood pressure (hypertension) on the
development of cardiovascular disease (CVD). To
obtain the relative risk, we need to calculate the
incidence rates. We can use the data from a
classic prospective study, the Framingham Heart
Study.
• Among the most important predictive factors
identified in the Framingham study were
elevated blood pressure, elevated serum
cholesterol and cigarette smoking.
• Elevated blood glucose and abnormal resting
electrocardiogram findings are also predictive
of future cardiovascular disease.
 Relative risk can be determined by the following
calculation:
• Incidence rate of cardiovascular disease (new
cases) over a specified period of time among
people free of CVD at beginning of the study
period who have the risk factor in question (e.g.,
high blood pressure).
OVER
• Incidence rate of CVD in the given time period
among people free of CVD initially, who do not
have the risk factor in question (normal blood
pressure).
Table 1: Relative Risk in Population-Based (Follow-up)
studies
Variables Present Absent Total
Disease Disease

Factor present A=20 B = 80 A+B = 100

High BP (100) CVD
Factor absent C= 5 D = 80 C+D = 100
No BP (100) CVD
20/100/5/100 = 4
Interpretation: the likelihood/chance of developing CVDs is 4 times
higher for people with HTN than people without HTN.
Table 1: Relative Risk in Population-Based (Follow-up)
studies
Variables Disease Disease Total
Present Absent
Factor present A=20 B = 80 A+B = 100
High BP (100) CVD
Factor absent C= 5 D = 95 C+D = 100
No BP (100) CVD

20/100/5/100 = 4
Interpretation: the likelihood/chance of developing CVDs is 4 times
higher for people with HTN than people without HTN.
From the Framingham data we calculate for men in the study the
RR of CVD within 18 years after first exam =
• 353/10,000 persons at risk with definite hypertension 353.2 = 2.87
• 123/10,000 persons at risk with no hypertension 123.9

• This means that a man with definite hypertension is

2.87 times more likely to develop CVD in an 18-year
period than a man who does not have hypertension.
• For women the relative risk is 187 = 3.28
57
 Calculation of Relative Risk from
Prospective Studies:
• Calculation of Relative Risk from Prospective
Studies:
• Relative risk can be determined directly from
prospective studies by constructing a 2 × 2
table as follows:
 DISEASE (developing in the specified period)

RISK Present a = 90 b = 403 a + b = 493

FACTOR (high BP) (persons with
(determined factor)
at beginning
of study Absent c = 70 d = 1201 c + d = 1271
period) (normal BP) (persons
without factor)
 Relative risk is

• Incidence of disease among those with high BP

• Incidence of disease among those with normal BP
• a/(a+b) = 90/493 = 3.31
• c/(c+d) 70/1271

• Interpretation?