Unit 3

What is a Variable?

A variable is an image, perception, or concept that can be measured and takes on different values. It refers
to any factor, trait, or condition that can exist in different amounts or types. Variables are essential in
research because they allow for the quantification of concepts and enable the use of statistical methods to
explore relationships and test hypotheses.

Key Definitions of a Variable:

• Kerlinger (1986) defines a variable as "a property that takes on different values." It is something
that varies and can be assigned numerals or values.
• Black and Champion (1976) define a variable as "rational units of analysis that can assume any one
of a number of designated sets of values."

Examples of Variables:

1. Program Effectiveness: If someone says, "This program is effective," they are making a judgment
based on some criteria. To measure this, one might use specific indicators like performance
outcomes or satisfaction levels.
2. Quality of Service: A statement like "We provide a quality service" can be turned into a variable by
identifying measurable factors, such as customer satisfaction, service delivery time, or error rates.
3. Discrimination in an Institution: If someone claims, "In this institution, women are discriminated
against," this can be investigated by measuring variables like pay disparities, promotion rates, or job
assignments between men and women.

Variables in Scientific Measurement

Some argue that feelings, preferences, and values cannot be scientifically measured. However, these can
often be measured indirectly using indicators, as they are based on observable behaviors in real life.

The Difference Between a Concept and a Variable

• Concepts: Concepts are mental images or perceptions and are subjective by nature. For example,
beauty, quality, and effectiveness are concepts because they can mean different things to different
people.
• Variables: Variables are measurable representations of concepts. They can be measured on different
scales (nominal, ordinal, interval, ratio) and allow researchers to gather data that can be analyzed and
compared.

The key difference between a concept and a variable is measurability:

• Concepts: These are not directly measurable because they are abstract ideas or impressions.
• Variables: These are measurable and can take different values, thus allowing comparison and
classification.

Why Variables Matter

In research, using variables enables the application of quantitative techniques to explore and test
relationships, causality, and the effectiveness of interventions. By converting concepts into measurable
variables, researchers ensure their findings are accurate, comparable, and reproducible.
Examples of Concepts and Variables:

Concept Variable
Happiness Satisfaction scale
Intelligence IQ score
Program effectiveness Percentage of goal achieved
Service quality Customer satisfaction rating

Variables, therefore, provide a uniform and precise method for comparing, classifying, and analyzing data
in research.

Converting Concepts into Variables

Operationalization is the process of converting abstract concepts into measurable variables that can be
quantified and analyzed. This involves identifying indicators—specific criteria that reflect the concept—
which can then be turned into measurable variables. The choice of indicators should logically connect to the
concept being studied.

Steps to Operationalize a Concept:

1. Identify the Concept: Determine the abstract concept you want to measure. For example, if you are
interested in the concept of wealth, it is crucial to clarify what "wealth" means in the context of your
research.
2. Determine Indicators: Select indicators that reflect the concept. For the concept of wealth, you
might choose:
o Income: This can be measured directly in monetary terms (e.g., annual salary).
o Assets: This includes property, investments, vehicles, etc.
3. Convert Indicators into Variables:
o For Income: It is already measurable, so it functions as a variable directly.
o For Assets: Identify specific types of assets (e.g., house, car, investments) and assign
monetary values to each. Sum these values to get a total asset value.
4. Establish Classification Levels: Define thresholds based on relevant data, such as community
income distribution, to classify individuals. For instance:
o An individual with a total asset value above a certain dollar amount and an annual income
above a certain threshold may be classified as "rich."
5. Data Analysis: Analyze the collected data on income and total asset value to draw conclusions about
wealth classification.

Example of Operationalization

Concept Indicators Variables

Wealth Income, Assets Income (in dollars), Total Assets (in dollars)
Effectiveness Outcomes, Satisfaction Program outcome measures, Satisfaction rating (e.g., Likert scale)

Challenges in Operationalization

Some concepts, such as effectiveness or impact, can be more challenging to operationalize because they
may involve multiple dimensions or subjective interpretations. For example:

• Effectiveness of a Program: This might require establishing specific outcome measures (e.g.,
percentage improvement in participant skills) and subjective measures (e.g., participant satisfaction).
Differences Between Quantitative and Qualitative Research

In quantitative research, the goal is often to create uniformity in measurement across respondents, leading to
consistent and comparable data. This involves rigorous operationalization of concepts into measurable
variables.

Conversely, qualitative research focuses on understanding perceptions, beliefs, and feelings without
necessarily establishing uniformity across participants. Qualitative methods may use more flexible indicators
that capture the richness of individual experiences without requiring strict measurement.

Types of Variables in Research Methodology

Understanding variables is crucial for designing and conducting research. Variables can be classified based
on three main criteria: causal relationships, study design, and unit of measurement.

1. Causal Relationship

In studies that aim to explore causal relationships or associations, variables can be categorized as follows:

a. Independent Variables (Change Variables)

• Definition: These variables are manipulated or controlled to observe their effect on other variables.
They are considered the "cause" in the cause-and-effect relationship.
• Characteristics:
o Typically manipulated in experimental designs.
o Responsible for bringing about change.
• Example: In a study examining the effects of smoking on cancer, the independent variable would be
the extent of smoking (e.g., number of cigarettes smoked).

b. Dependent Variables (Outcome Variables)

• Definition: These variables are measured to assess the effects of the independent variable. They are
the outcomes or consequences of the independent variable.
• Characteristics:
o Reflect the changes caused by the independent variable.
o Dependent on the independent variable.
• Example: In the smoking study, the incidence of cancer would be the dependent variable.

c. Extraneous Variables

• Definition: These are unmeasured variables that can affect the relationship between independent and
dependent variables. They may introduce bias or confounding effects in the study.
• Characteristics:
o Can either enhance or diminish the observed relationship between independent and dependent
variables.
• Example: Factors such as age, dietary habits, and physical activity can all impact the relationship
between smoking and cancer.
d. Intervening Variables (Connecting or Linking Variables)

• Definition: These variables mediate the relationship between the independent and dependent
variables. They are essential for understanding how the independent variable affects the dependent
variable.
• Characteristics:
o Often seen in causal chains where another variable must be present for the effect to occur.
• Example: If biological conditions mediate the relationship between smoking and cancer, these
biological conditions would be the intervening variables.

Summary of Causal Variables

Type Definition Example

Independent Amount of smoking (e.g., number of
The cause that brings about change
Variable cigarettes)
Dependent The outcome or change caused by the
Incidence of cancer
Variable independent variable
Extraneous
Other factors affecting the dependent variable Age, diet, exercise
Variable
Intervening
Links independent and dependent variables Biological conditions affecting cancer
Variable

2. Study Design

The classification of variables can also depend on the study design:

a. Active Variables

• Definition: These are variables that the researcher can manipulate or control in an experimental
setting.
• Example: Different teaching methods used in a study evaluating educational effectiveness.

b. Attribute Variables

• Definition: These are characteristics of the study population that cannot be manipulated. They reflect
inherent traits or demographics.
• Example: Characteristics such as age, gender, or education level of participants.

3. Unit of Measurement

Variables can also be classified based on the unit of measurement:

a. Categorical Variables

• Definition: These variables represent categories or groups and can be further classified into:
o Constant Variables: Have only one category (e.g., water, tree).
o Dichotomous Variables: Have two categories (e.g., yes/no, male/female).
o Polytomous Variables: Have more than two categories (e.g., political affiliation, religious
beliefs).
b. Continuous Variables

• Definition: These variables can take any value within a range and are measured on a continuum.
• Example: Variables like age, income, and temperature can take a wide range of values.

Categorical/Continuous and Quantitative/Qualitative Variables

Variable
Description Examples
Type
Represents distinct categories or groups.
Categorical
Can be divided into three sub-types:
- Constant Only one category. - Water
- Tree
- Taxi
-
Has only two categories. - Yes/No
Dichotomous
- Male/Female
- Good/Bad
- Rich/Poor
- Political Parties (e.g., Labor, Liberal,
- Polytomous Has more than two categories.
Democrat)
- Attitudes (e.g., Strongly Favourable,
Favourable, Uncertain, Unfavourable, Strongly
Unfavourable)
- Day/Night
- Age Groups (e.g., Child, Young, Middle-Aged,
Old)
Measured on a scale and can take any
Continuous
value within a range.
-
Represents numerical values. - Income ($)
Quantitative
- Age (years, can also be measured in
months/days)
- Weight (kg)
- Temperature (°C or °F)
Based on categorical measurements but
Qualitative can also be expressed in terms of
quantities.
- Gender (e.g., Male, Female)
- Educational Level (e.g., High School,
Bachelor's Degree, Master's Degree)
- Income categories (e.g., Low, Middle, High)
- Attitudes classified qualitatively (e.g., Strongly
Favourable, Unfavourable)