FACULTY OF EDUCATION

PEDAGOGY FOR NATIONAL AND FOREIGN

LANGUAGES ONLINE

TOPIC:

QUANTITATIVE METHODS AND PROCESSES FOR APPLIED

LINGUISTICS RESEARCH

AUTHORS:
BURGOS GORDILLO VICTORIA NARCISA

ROMERO ANZOATEGUI LUIS ROBERTO

SALVADOR BAEZ SABRINA MAYERLIE

SHINGRE TADAY NAYDELIN DAYANA

ZAMORA MALDONADO RAUL HERNAN

COURSE:
SIXTH SEMESTER – C3

PROFESSOR:
BRIONES ZAMBRANO MARIANA MARISOL

SUBMISSION DATE:
28-04-2025

INTAKE:
August 2024 -December 2024

MILAGRO-ECUADOR
 Different Variables in Quantitative Research

A variable is an entity that takes on different values. In theory, it can change the
interaction within the theory, and anything that affects or changes the results of a
study can change. Therefore, we must keep in mind that anything that changes can be
considered a variable. It is necessary to understand the differences in all types of
studies, since a variable has several forms, such as age, gender, lifestyle, musical
tastes, colors, and others.

Variable Classifications

Variables are classified as numerical, experimental, categorical, and non-experimental.

Their classification helps us determine the roles each one plays depending on the type
of study being conducted. It is fundamental in statistics and research, as it allows data
to be organized and analyzed effectively.

Numeric Variable

Describes a measurable numerical quantity and is considered quantitative data. These

numerical variables are divided into two types: continuous or interval variables and
discrete variables.

Continuous variables ensure that any value falls within a set of numbers depending on
the scale used. We can use this variable for time, weight, height, and age, since this
variable, even with its simplest description, can recognize values between integers.

Discrete variables handle any integer value within the limits of the variables presented.
This variable serves as an example for class attendance, house or apartment numbers,
and family size, since this variable only focuses on an integer value because we cannot
use half of a number.
Discrete Variables

They are variables that can only take specific whole number values within a range
defined by the researchers. These values are counted, and there is a clear gap between
one value and another. For example, the number of subjects passed (1, 2, 3, 4, 5, 6, 7,
8). You can have two or three subjects approved, but not two and a half.

"Measurement" does not always mean using a metric or numerical expression. To

measure means placing the unit of analysis into one of the possible values of the
variable. Variables are classified based on what actions can be performed with their
values.

Examples of discrete variables:

1. Age of graduates from a course.

2. Number of pets owned by people in a city.

3. Number of animals in different ecological reserves of a country.

4. Number of sales made by each employee in a store.

5. Number of points scored by basketball players.

6. Number of loaves of bread baked daily in a bakery.

7. Number of properties owned by each citizen of a country.

Categorical Variables

Categorical variables are very common in any research field. However, it is important
to understand that whether a variable is categorical depends more on the
measurement system used than on the variable itself.

For instance, if I study light color, I can treat it as a qualitative variable with categories
like blue, green, red, or yellow. Alternatively, I could measure it by its wavelength or
frequency, giving it a numerical value.

Thus, categorical variables are a way to describe something when there is no numeric
measurement available.

Because categorical data is not mathematical, its direct analysis is limited. In

descriptive statistics, the only measure that can be found for a categorical variable is
the mode. Other analyses like correlation, regression, or probability distributions
require numeric data with mathematical meaning.

Examples of categorical variables:

● Payment methods (Cash or Credit)

● Machine settings (Low, Medium, High)

● Product types (Wood, Plastic, Metal)

Nominal Variables

A nominal variable is a type of categorical data that has no numerical value, order, or
ranking. Categories are separate and can be identified by unique labels. Nominal data
is mainly used to group or classify information.
Simply put, a nominal variable labels or names different things without giving them a
number or order. For example, a list of fruits (apple, orange, banana) is just a set of
categories without any ranking.

Nominal data is collected through surveys, questionnaires, observations, or existing

records. The questions are usually multiple-choice, yes/no, or open-ended.

Quantitative data analysis, instead, deals with numeric and measurable data, using
mathematical methods and visualizations. For example, recording temperature
readings over a week is quantitative data analysis.

Nominal and ordinal data are considered qualitative, while interval and ratio data are
considered quantitative. Nominal data gives the least amount of information, while
interval and ratio data give the most detail.

Examples of nominal variables:

● Hair color: Blonde, Brown, Black, etc.

● Marital status: Single, Married, Divorced, Widowed, etc.

● Nationality: Spanish, French, Italian, etc.

● Gender: Male, Female, Other

● Religion: Catholic, Protestant, Jewish, Muslim, etc.

● Type of transport: Car, Bicycle, Bus, Train, etc.

● Car brand: Toyota, Honda, Ford, etc.

● Product type: Milk, Yogurt, Cheese, etc.

 ● Person's name: John, Mary, Peter, etc.

● Identification number: ID, Passport, etc.

● Field of study: Medicine, Engineering, Law, etc.

Definition: Variables that represent categories with no natural order or ranking.

Examples:
- Gender (male, female, non-binary)
- Blood type (A, B, AB, O)
- Nationality (Brazilian, Japanese, Canadian)

Categories are different, but none is 'higher' or 'lower' than another.

Ordinal Variables

Definition: Variables that represent categories with a meaningful order, but intervals
between categories are not necessarily equal.
Examples:
- Education level (high school, bachelor's degree, master's degree)
- Customer satisfaction (very dissatisfied, dissatisfied, neutral, satisfied, very satisfied)

There is a clear ranking, but we cannot measure exactly how much 'better' one
category is compared to another.
Nominal vs Ordinal Variables

Feature Nominal Variables Ordinal Variables

Order between values ❌ No order ✅ Order exists

Examples Colors, blood types, gender Education level, satisfaction

Dichotomous vs Polytomous Variables

Dichotomous Variables:
- Definition: A variable with only two possible categories.
- Examples:
- Yes / No
- True / False
- Pass / Fail

It’s binary — just two choices.

Polytomous Variables:
- Definition: A variable with more than two possible categories.
- Examples:
- Eye color (brown, blue, green)
- Political party (Democrat, Republican, Independent, Other)

Experimental Variables

Experimental variables are the things (factors or items) that the researcher changes or
measures to see if they will affect the outcome of the experiment. It must be noted
that the variables above are crucial for determining such cause–results.
Examples:

● Drug Dose: In a clinical study, investigators may measure the response time for
patients to recover after administering different doses of a new drug (e.g., low,
medium, high).
● Temperature Conditions: In a study on plant growth, investigators may have
subjects growing in different temperatures (e.g., 20°C, 25°C etc.).

Independent Variables

Independent variables are those that researchers change or control in an experiment.

They are presumed to influence the dependent variable.

Examples:

● Study Method: In a study on learning effectiveness, the independent variable

could be the type of study method used (e.g., group study vs. individual study).
● Exercise Frequency: In a health study, researchers might manipulate the
frequency of exercise (e.g., 2 days a week, 4 days a week) to see how it impacts
weight loss.

Dependent Variables

The dependent variables are the responses or outcomes that researchers observe in
an experiment. They are presumed to react to the IV.

Examples:

● Test Scores: In an educational study, the dependent measure might be the

students’ scores on an exam and subjects are tested to determine how their
performance is affected by various study methods.
● Blood Pressure Levels: In a health survey related to diet, the outcome variable
would be the blood pressure measurements for a subject after a particular
compound gained.
Control Variables

Constants, or control variables, are any factors in an experiment that are kept constant
in an experiment so that the independent variable can have the greatest effect.

Examples:

● Age - If examining the impact of a new educational scheme, then the age of
students might be kept as the same for all participants but then we have
eliminated age as a confounding variable.
● Environmental Conditions In a plant growth experiment, scientists could
manipulate variables such as type of soil, amount of light, and watering
frequency to ensure they do not inadvertently affect the results.

Moderator Variables

Moderators are variables (what I sometimes call "driftwood") that can impact how
strong or weak (or in what direction) the relationship between independent and
dependent variables are. They are useful to analyze under which conditions
[associated] effects occur.

Examples:

● Sex: If you are conducting a research on the effect of stress on performance,

then sex could be a moderating variable if male and female respond to stress
differently. Scholars could find that stress affects the performance of men
more than it does women.
● Socioeconomic Status: In a study on educational achievement, socioeconomic
status could serve as a moderator variable if it is found to influence how well
students perform under varying levels of parental involvement.
Examples about non-experimental variables

1. Influence of Management Styles on Employee contentment

Predictor Variable: Management Styles
Criterion Variable: Employee Satisfaction

2. This study focuses on the effect of a counselling programme on dropout rates

among 8th grade students. So the main objective is to analyse the effectiveness
of a counselling programme,specifically in a given school district. The aim is to
find out whether this programme really contributes to lower dropout rates and
whether its impact is positive for all pupils in the district. To better understand
Student Absenteeism, it is important to define it in a clearer way. This term
refers to the percentage of absences a student has during a specific period.
It is generally used to evaluate attendance in the educational environment.
Counselling is not assigned or denied on purpose.
3. Predictor Variable:The time students prefer to study, whether in the morning,
afternoon or evening. Also considered is the level of support they receive from
their parents at home, which is measured by the frequency with which they
receive help or encouragement. Additionally, the type of device they use to
study is taken into account, whether it is a tablet, laptop or mobile phone.
Criterion Variable:The overall academic performance of students, which can be
assessed through their grade point average or final grades. Another aspect to
consider is the homework completion rate, which is expressed as the
percentage of homework assignments that have been handed in. In addition,
students' motivation levels are measured through surveys.

Reflection
Identifying non-experimental variables and their classification, either as
predictor or criterion, provides researchers with a deeper and more direct
understanding of how these variables are related without the need to
manipulate them.
In addition, it allows interpretation of the natural relationships that exist
between different factors, which is essential for proper analysis.

However, this also helps to avoid common errors that are generated by
assuming causal relationships without adequate experimental control.

Moreover, by designing studies where it is not possible to manipulate the

variable, such as in work situations where someone cannot be assigned a
management type for research purposes alone, more ethical and practical
approaches can be created.

Finally, this broader understanding encourages a richer discussion of the

results, taking into account the limitations that arise from not fully controlling
for all conditions.