CLASSIFICATIONS OF VARIABLES AND THEIR USES B. simply defines groups of subjects.

Here you may have more

than 2 categories of equivalent magnitude.
WHAT IS VARIABLE?
- A “characteristic that has two or more mutually exclusive For example, a basketball player’s number is used to distinguish
him from other players. It certainly does not follow that player 10 is better
values or properties” Sevilla and Other, 1988
than player 8. Other examples are blood type, hair color, and mode of
- Bernard (1994) defines a variable as something that can transportation.
take more than one value, and values can be words or C. from the name itself denotes that a variable is ranked in a certain order.
numbers. This variable can have a qualitative or quantitative attribute.
For example, a survey questionnaire may have a numerical rating
- A variable specifically refers to characteristics, or attributes
as choices like 1, 2, 3, 4, and 5 ranked accordingly (5=highest, 1=lowest) or
of an individual or an organization that can be measured or categorical rating like strongly agree, agree, neutral, disagree, and strongly
observed and that varies among the people or organization disagree. Other examples or ordinal variables: cancer stage (Stage I, Stage
being studied (Creswell, 2002). II, Stage III), Spotify Top 20 hits, academic honors (with highest, with high,
with honors).
- anything that has a quantity or quality that varies. The root
word of the word variable is “vary” or simply “can change”.
These variables are among the fundamental concepts of Other Classifications Of Variables:
research, alongside with measurement, validity, reliability,
cause and effect, and theory.  INDEPENDENT VARIABLES
 DEPENDENT VARIABLES
TWO MAJOR KINDS OF VARIABLES:  INTERVENING OR MEDLING VARIABLES
1. Quantitative / Numerical Variables  CONTROL VARIABLES
2. Qualitative / Categorical Variables  CONFOUNDING VARIABLES

Quantitative Variables  INDEPENDENT VARIABLES – Those that probably cause,

 Also called numerical variables are the type of variables influence, or affect outcomes. They are invariably called
used in quantitative research because they are numeric and treatment, manipulated, antecedent, or predictor variables.
can be measured. This is the cause variable or the one responsible for the
 Under this category are discrete and continuous variables. conditions that act on something else to bring about
changes.
Types of Quantitative Variables:
 DEPENDENT VARIABLES – those that depend on the
 Discrete variables are countable whole numbers. It does independent variables; they are the outcomes or results of
not take negative values or values between fixed points. For the influence of the independent variable. That is why it is
example: the number of students in a class, group size, and also called the outcome variable.
frequency. A discrete variable is a variable that takes on
distinct, countable values. In theory, you should always be  INTERVENING OR MEDLING VARIABLES – Variables that
able to count the values of a discrete variable. “stand between” the independent and dependent variables,
Examples: and they show the effects of the independent variable on the
 Years of schooling dependent variable.
 Number of goals made in a soccer match It helps to explain how or why the independent
 Number of red M&M’s in a candy jar
variable has an effect on the dependent variable. It clarifies
 Votes for a particular politician
 Number of times a coin lands on heads after ten coin tosses the underlying mechanism.
Goal in research - To identify and understand
 Continuous variables take fractional (non-whole number) intervening variables to gain a deeper and more nuanced
values that can either be positive or negative. Example: understanding of the causal pathway. This often involves
height, temperature. A continuous variable is a variable that statistical mediation analysis.
EXAMPLE:
can take on any value within a range. A continuous variable Even if farm production is good, if the attitude towards payment is
takes on an infinite number of possible values within a given negative, loan repayment would be low, whereas, if the attitude towards
range. repayment is positive or favorable, loan repayment would be high.
Examples:
 The time it takes sprinters to run 100 meters
 CONTROL VARIABLES – A special type of independent
 The size of real estate lots in a city
 The weight of baby elephants variable that is measured in the study because it potentially
 The body temperature of patients with the flu influences the dependent variable.
 The deployment altitude of skydivers Control variables are important in scientific
experiments because they help scientists understand the
Two levels of measurement relationship between different things they are testing.
 Intervals are quantitative variables where the interval or
differences between consecutive values are equal and CONFOUNDING VARIABLES – These variables are
meaningful but the numbers are arbitrary. unmeasured factors that can influence the relationship
For example, the difference between 36 degrees and 37 degrees between an independent variable and a dependent variable,
is the same as between 100 degrees and 101 degrees. The zero point does
not suggest the absence of a property being measured. Temperature at 0 making it seem like there is a correlation when there might
degree Celsius is assigned as the melting point of ice. not be.
A confounding variable is an "extra" variable that
 Ratio type of data is similar to interval, the only difference is affects both the independent variable and the dependent
the presence of a true zero value. The zero point in this variable, creating a spurious or misleading association
scale indicates the absence of the quantity being measured. between them. It obscures or distorts the true relationship
Examples are age, height, weight and distance between the IV and DV.

• It is related to the Independent Variable (IV).

Qualitative Variables • It causes or influences the Dependent Variable (DV).
 Also referred to as Categorical Variables are not expressed • It is not caused by the Independent Variable.
in numbers but are descriptions or categories. It can be
further divided into nominal, ordinal or dichotomous. Goal in research –
To identify and control for confounding variables to isolate
A. Dichotomous - Consisting of only two distinct categories or the true effect of the independent variable on the dependent
values. variable. This can be done through research design (e.g.,
For example, a response to a question either be a yes or no. randomization, matching) or statistical analysis (e.g.,
regression, stratification)
 Characteristics of Quantitative Research WEAKNESSES OF QUANTITATIVE RESEARCH
Quantitative research is commonly used in natural The disadvantages of quantitative research are as
sciences research problems because of the following follows:
characteristics: 1. Quantitative research requires a large number of
1. LARGE SAMPLE SIZE. To obtain more respondents. It is assumed that the larger the
meaningful statistical result, the data must come from sample is, the more statistically accurate the
a large sample size. findings are.
2. OBJECTIVE. Data gathering and analysis of 2. It is costly. Since, there are more respondents
results are done accurately, objectively, and are compared to qualitative research, the expenses
unaffected by the researcher’s intuition and personal will be greater in reaching out to these people and
guesses. in reproducing questionnaires.
3. VISUAL RESULT PRESENTATION. Data is 3. The information which is contextual factors to
numerical, which makes presentation through graphs, help interpret the results or to explain
charts, and tables possible and for better conveyance variations are usually ignored. It does not
and interpretation. consider the distinct capacity of the respondents to
4. FASTER DATA ANALYSIS. The use of a share and elaborate further information unlike the
statistical tools give way for a less time-consuming qualitative research.
data analysis. 4. Much information are difficult to gather using
5. GENERALIZED DATA. Data taken from a structured research instruments, specifically on
sample can be applied to the population if sampling is sensitive issues like pre-marital sex, domestic
done accordingly, i.e., sufficient size and random violence, among others.
samples were taken. 5. If not done seriously and correctly, data from
6. FAST DATA COLLECTION. Depending on questionnaires may be incomplete and
the type of data needed, collection can be quick and inaccurate. Researchers must be on the look-out
easy. Quantitative research uses standardized on respondents who are just guessing in
research instruments that allow the researcher to answering the instrument
collect data from a large sample size efficiently. For
instance, a single survey form can be administered  Descriptive design is used to describe a particular
simultaneously to collect various measurable phenomenon by observing it as it occurs in nature.
characteristics like age, gender, socio-economic There is no experimental manipulation and the
status, etc. researcher does not start with a hypothesis. The goal
7. RELIABLE DATA. Data is taken and of descriptive research is only to describe the person
analyzed objectively from a sample as a or object of the study. An example of descriptive
representative of the population, making it more research design is “the determination of the different
credible and reliable for policymaking and decision kinds of physical activities and how often high school
making. students do it during the quarantine period.”
8. REPLICATION. The Quantitative method can  The correlational design identifies the relationship
be repeated to verify findings enhancing its validity, between variables. Data is collected by observation
free from false or immature conclusions. since it does not consider the cause and effect for
example, the relationship between the amount of
STRENGTHS OF QUANTITATIVE RESEARCH physical activity done and student academic
The following are the strengths of quantitative achievement.
research.  Ex post facto design is used to investigate a
1. It is objective. The most reliable and valid way of possible relationship between previous events and
concluding results, giving way to a new present conditions. The term “Ex post facto”, means
hypothesis or to disproving it. Because of bigger after the fact, looks at the possible causes of an
number of the sample of a population, the results already occurring phenomenon. Just like the first two,
or generalizations are more reliable and valid. there is no experimental manipulation in this design.
Since it provides numerical data, it can’t be easily An example of this is “how does the parent’s
misinterpreted. academic achievement affect the children obesity?”
2. The use of statistical techniques facilitates  A quasi-experimental design is used to establish
sophisticated analyses and allows you to the cause and effect relationship of variables.
comprehend a huge amount of vital Although it resembles the experimental design, the
characteristics of data. quasi-experimental has lesser validity due to the
3. It is real and unbiased. If the research is absence of random selection and assignment of
properly designed, it filters out external factors, subjects. Here, the independent variable is identified
and so can be seen as real and unbiased. but not manipulated. The researcher does not modify
4. The numerical data can be analyzed in a quick pre-existing groups of subjects. The group exposed
and easy way. By employing statistically valid to treatment (experimental) is compared to the group
random models, findings can be generalized to unexposed to treatment (control): example, the
the population about which information is effects of unemployment on attitude towards following
necessary. safety protocol in ECQ declared areas.
5. Quantitative studies are replicable.  Experimental design like quasi- experimental is
Standardized approaches allow the study to be used to establish the cause and effect relationship of
replicated in different areas or over time with two or more variables. This design provides a more
formulation of comparable findings. conclusive result because it uses random assignment
6. Quantitative experiments are useful for testing of subjects and experimental manipulations. For
the results gained by a series of qualitative example, a comparison of the effects of various
experiments, leading to a final answer, and blended learning to the reading comprehension of
narrowing down of possible directions to follow. elementary pupils.