Review Article IJRRMS 2013;3(4)

Variables in research
Kaur SP

INRODUCTION Variables can be defined in terms of measurable

factors through a process of operationalization. It
“Variable” is a term frequently used in research will convert difficult concepts into easily
projects. It is pertinent to define and identify the understandable concepts which then can be
variables while designing quantitative research measured, empirically. “It is essential to define the
projects. A variable incites excitement in any term as variables so that they can be quantified
research than constants. It is therefore critical for and measured. That is, the variable have to be able
beginners in research to have clarity about this to work for you to operate, or becomes
term and the related concepts. operational”.4
There are different types of variables and having
Variable, to put in layman statement is something their influence differently in a study viz.
that can change and or can have more than one Independent & dependent variables, Active and
value. ''A variable, as the name implies, is
1” attribute variables, Continuous, discrete and
something that varies”. It may be weight, height,
categorical variable, Extraneous variables and
anxiety levels, income, body temperature and so
Demographic variables.
on. Each of these properties varies from one person
to another and also has different values along a
continuum. It could be demographic, physical or INDEPENDENT & DEPENDENT VARIABLE
social and include religion, income, occupation,
temperature, humidity, language, food, fashion, The independent variable is the antecedent while
etc. Some variables can be quite concrete and clear, the dependent variable is the consequent. If the
such as gender, birth order, types of blood group etc independent variable is an active variable then we
while others can be considerably more abstract and manipulate the values of the variable to study its
vague. affect on another variable. In the above example,
we alter anxiety level to see if responsiveness to
“Variable is a property that takes on different pain reduction medication is enhanced. Anxiety
2 3
values''. It is also a logical grouping of attributes. level is the active independent variable.
Attributes are characteristics or qualities that
describe an object. For example if gender is a
Dependent variable is the variable that is affected
variable then male and female are the attributes. If
by the independent variable. Responsiveness to
residence is the variable then urban, semi urban,
pain reduction medication is the dependent
rural become the attributes. So attributes here
describe the residence of an individual. variable in the above example. The dependent
variable is dependent on the independent variable
It is pertinent for a researcher to know as how
certain variables within a study are related to each ACTIVE AND ATTRIBUTE VARIABLES
other. It is thus important to define the variables to
facilitate accurate explanation of the relationship Variables are often characteristics of research
between the variables. There is no limit to the subjects, such as their age, health beliefs, or
number of variables that can be measured, weight etc. ''Variables which cannot be
although the more variables, the more complex the manipulated are attribute variables and the
study and the more complex the statistical analysis. variables that the researcher creates are the active
Moreover the longer the list of variables, the longer variables''. 2 Active variables can also be
the time required for data collection. independent variables. E.g. effectiveness of
IJRRMS | VOL-3 | No.4 | OCT - DEC | 2013 36
Kaur SP. Variables in research IJRRMS 2013;3(4)

communication board in meeting the needs of the “When categorical variables takes on only two
intubated patients. Communication board is an values, they are sometimes referred to as
1
'active independent variable' as it can be modified dichotomous variables”. Some examples of
according to the needs of the patients or according dichotomous variables and multiple variables are:
to the requirement in the study and it is EXTRANEOUS VARIABLES
researcher's concept. It is also the cause i.e.
independent variable. . It happens sometimes that after completion of the
study we wonder that the actual result is not what
Attribute variable is a variable where we do not we expected. In spite of taking all the possible
alter the variable during the study. It can also be the measures the outcome is unexpected. It is because
independent variable, but it has limitations. Some of extraneous variables. Variables that may affect
attribute variables are age, gender, blood group, research outcomes but have not been adequately
color of eyes, etc. We might want to study the effect considered in the study are termed as extraneous
of age on weight. We cannot change a person's age, variables. Extraneous variables exist in all studies
but we can study people of different ages and and can affect the measurement of study variables
weights. “An active variable in one study could be an and the relationship among these variables.
attribute variable in another study''. “Extraneous variables that are not recognized until
CONTINOUS, DISCRETE AND CATEGORIAL the study is in process, or are recognized before
VARIABLES the study is initiated but cannot be controlled, are
referred to as confounding variables”.5 Certain
Sometimes variables take on a wide range of values external variables may influence the relationship
on a continuum. ''A continuous variable can assume between the research variables, even though
1
an infinite number of values between two points''. researcher cannot see it. These variables are called
If we consider the continuous variable weight: intervening variables. For example, girl's
between 1 and 2 Kg, the number of values is knowledge and practices helps in maintaining
limitless: 1.005, 1.7, 1.33333, and so on. menstrual hygiene. Here, motivation, mother and
Continuous measures in actual use are contained in friends, mass media, are some intervening
a range each individual obtains a score within the variables which may also help in maintaining
1
range. On the other hand, a discrete variable is one menstrual hygiene. Thus, if these two factors are
that has a finite number of values between any two not controlled it would be impossible to know
points, representing discrete quantities. what the underlying cause really is.

Categorial variables, belongs to a kind of DEMOGRAPHIC VARIABLES:

measurement called nominal. In nominal
measurements there are two or more subsets of “Demographic variables are characteristics or
the set objects being measured. “They have a attributes of subjects that are collected to
simple requirement that all the members of the describe the sample”.5 They are also called sample
subset are considered the same and all are assigned characteristics. It means these variables describe
the same name (nominal) and the same numeral”.
2 study sample and determine if samples are
That is, they can be measured only in terms of representative of the population of interest.
whether the individual items belong to certain Although demographic variables cannot be
distinct categories, but we cannot quantify or even m a n i p u l a t e d , re s e a rc h e rs c a n ex p l a i n
rank order the categories. The variable gender, for relationships between demographic variables and
example, has only two values (male and female). dependent variables. Some common
Variables that take on only a handful of discrete non demographic variables are age, gender,
quantitative values are categorical variables. occupation, marital status, income etc.

37 IJRRMS | VOL-3 | No.4 | OCT - DEC | 2013

Kaur SP. Variables in research IJRRMS 2013;3(4)

VARIABLES EXAMPLES variables are categorized that can be ranked. Eg.

Dichotomous Gender: Male and female
Levels of pain mild, moderate or severe. Interval
Variables Type of property: Commercial and residential scale measures equal numerical distances
Pregnant and non pregnant between the intervals. It can be categorized, as
Alive and dead
HIV positive and HIV negative well as ranked. Eg. difference between 70 degree
Education: Literate and illiterate and 80 degree will be same as the difference
Trichotomous Residence: Urban, semi urban and rural between 30 degree and 40 degree. So the interval
Variables Religion: Hindu, muslim, and Christianity. is of 10 degree in both the categories. Ratio scale
Multiple Blood groups: A,B,AB and O
measures variables which can be categorized,
Variables ranked, have equal intervals and can represent a
continuum of values.
MEASURING VARIABLES
Thus, the most difficult part of planning a research
In order to do an analysis, the variables have to be
study is identifying the research variables and
quantified; this means measuring giving values
research design. Considerable time and thought
and scale. Sometimes identification of variables
needs to be given to this step. Once the key
and determining how to measure them looks quite
variables have been identified, then the research
simple, but due to vaguely defined variables
study can be developed.
measuring can be difficult too. There are four levels
of measurements on a continuum of discrete and
AUTHOR NOTE
continuous: nominal-scale, ordinal-scale, interval-
scale and ratio-scale. Nominal scale is used when
Simer Preet Kaur, Principal;email:
the variables can be categorized but cannot be
[email protected]
ranked. E.g. gender, marital status, race, diagnosis,
Samarpan Institute of Nursing Sciences
blood group etc. Ordinal scale is one were the

REFERENCES
1. Polit Denise F., Beck CT. Nursing Research, Principles and Methods. 7th edition. Pliladephia: LippincottWilliams & Wilkins;
2004. p 29-32,37-38.
2. Kerlinger Fred N. Foundation of Behavioral Research. 2nd edition. U.S.A: Holt, Rinehart and Winston, Inc; 1983. p 29-40.
3. Earl Babbie. The Practice of Social Research. 3rd edition. Belmont:Wadsworth publishing Company; 1983. p 20-21.
4. Aparna Bhaduri, Marie Farrell. Health Research: A community - based approach. WHO: Regional office for South East Asia.
Series No 12 .p 70
5. Nancy Burns, Grove Susan.K. Understanding Nursing Research. 4th edition. Missouri: Elsevier Publication; 2007. p 125-129.

IJRRMS | VOL-3 | No.4 | OCT - DEC | 2013

38