Quantitative Research & Variable Classifications

Lesson 1: Quantitative Research

Definition & Core Idea

• Quantitative research = “explaining phenomena by collecting numerical data that are analyzed using mathematically-based methods” (Aliaga & Gunderson, 2000).
• Designed for problems that demand measurement and numerical evidence.
• Employs deductive reasoning ⇒ formulates hypotheses/predictions, then tests them with empirical data.
• Central activity: attach numbers to variables, run statistical analyses (descriptive or inferential) to answer the research question.

Phenomenon (What We Study)

• Any observable event/experience that invites investigation.
• Measurable in quantity, rate, or proportion.
• Researchers seek to uncover causes, effects, or relationships through measurement.
• Illustrative examples:
– Increase in sales.
– Changes in employee turnover rates.
– Drop-out rates shifting.
– Fewer failing students in Mathematics.
– Percentage shift of dengue patients.
– Rise in youth drug addiction rate.
– Decline in juvenile crime in rural areas.

Characteristics of Quantitative Research

  1. Reliable & objective.

  2. Uses statistics to generalize findings to the wider population.

  3. Reduces/restructures complex problems into a limited set of variables.

  4. Examines variable connections; establishes cause–effect under highly controlled conditions.

  5. Tests theories or hypotheses.

  6. Assumes the sample represents the population.

  7. Methodological subjectivity = secondary concern.

  8. Deals with detailed aspects of the subject.

Advantages

  1. Enables objective measurement & analysis ⇒ objective answers.

  2. Large samples → generally reliable outcomes.

  3. Standardized instruments, sampling, & statistics → study is replicable.

  4. Minimal personal bias (little/no direct interaction).

  5. Procedures are simplified, systematic, and easy to follow.

  6. Statistical treatment condenses results, permitting concise interpretation.

Disadvantages

  1. Often ignores natural context; study environment may be artificial.

  2. Large samples → more resources required.

  3. Findings limited to numerical analysis; lacks rich narrative detail.

  4. Provides less elaborate accounts of human perception.

  5. Experimental control (e.g., lab settings) may not mirror real-world conditions.

  6. Pre-set answer options may not reflect true participant views.

  7. Researcher perspective seldom influences findings (participants often anonymous) → may overlook nuance.

Importance Across Fields

Education

• Apply quantifiable best practices.
• Measure student & teacher performance levels.
• Evaluate instructional methods, programs, stakeholder satisfaction.

Business

• Supply market size, demographic, & preference data.
• Inform overall marketing strategy & product/service decisions.
• Capture consumer opinions to boost productivity.
• Guide product development & targeted campaigns.

Medicine & Health-Allied Services

• Patient-satisfaction data & profiles inform health-fund distribution.
• Recovery rates, illness frequency, medication efficacy → basis for treatment & intervention.
• Experimental studies identify effective medicines/vaccines.

Science & Technology

• Drives accountability & responsibility through measurable data.
• Assesses device processing rates, procedure duration, efficiency tests.
• Records data during experiments on new devices, inventions, & innovations.

Lesson 2: Classifications of Variables

What Is a Research Variable?

• Also called a data item.
• Any factor/property a researcher measures, controls, and/or manipulates.
• Exhibits differing amounts/types (i.e., a changing quantity).
• A logical set of attributes/values, countable or measurable.

Main Classification Schemes

1. According to Numbers

a. Numerical (Quantitative) Variables

• Answer “how many/how much.”
– Continuous/Interval: can assume any value on a real-number scale.
• Examples: time, age, temperature, height, weight.
– Discrete: take only whole numbers within limits.
• Examples: number of cars, business locations, children in a family, student population.

b. Categorical (Qualitative) Variables

• Describe quality/characteristics – answer “what type/which category.”
– Ordinal: logically ordered/ranked.
• Examples: grades (A,B,C), clothing sizes, Likert attitudes (strongly agree → strongly disagree).
– Nominal: no intrinsic order.
• Examples: business type, eye color, religion, language, learner type.
– Dichotomous: exactly two categories.
• Examples: gender (male/female), yes/no, true/false.
– Polychotomous: multiple categories.
• Examples: educational attainment levels, performance tiers.

2. Experimental Role

• Variables manipulated or controlled in experiments.
– Independent / Manipulated / Exploratory: intentionally varied to observe effect.
• Example: amount of sleep in cognition study.
– Dependent / Response / Predicted: observed outcome influenced by independent variable.
• Example: cognitive test score.
– Extraneous / Mediating / Intervening / Covariate: existing variables that could influence results.
• Example: participant stress levels.
– Controlled: held constant to eliminate confounding influence.
• Example: room temperature during testing.

Example 1
• Research title: “An Experiment on the Methods of Teaching and Language Achievement Among Elementary Pupils.”
– Independent V: teaching method.
– Dependent V: language achievement.
– Mediating V: ventilation facilities, physical ambiance.

Example 2
• Research title: “Use of Gardening Tools and Types of Fertilizer: Effects on the Amount of Harvest.”
– Independent V: gardening tools, fertilizer types.
– Dependent V: amount of harvest.
– Mediating V: humidity, seed/plant type.

3. Non-Experimental Role

• Variables studied without manipulation; useful when manipulation is unethical/impossible.
– Predictor variable: presumed cause.
– Criterion variable: presumed effect.

Example 1
• Title: “Competencies of Teachers and Students’ Behavior in Selected Private Schools.”
– Predictor: teacher competencies.
– Criterion: student behavior.

Example 2
• Title: “Conduct of Guidance Counseling Programs and Degree of Absenteeism & Drop-Out Rate Among Grade 8 Classes.”
– Predictor: guidance counseling program conduct.
– Criterion: absenteeism & drop-out rate.

4. According to Number Studied

• Univariate study: one variable.
• Bivariate study: two variables.
• Polyvariate study: more than two variables.

Key Take-Aways & Practical Tips

• Always match variable type to measurement scale; improper scale selection can compromise validity.
• In quantitative designs, strive for clear operational definitions so variables can be reliably measured.
• Maintain control of extraneous variables (experimental) or carefully record them (non-experimental) to support causal inference.
• When ethical or practical constraints prevent manipulation, opt for predictor/criterion framing while acknowledging limits on causality.

Ethical & Practical Considerations

• Large-scale quantitative work demands resources; plan budgeting for sampling & data collection.
• Use fair, unbiased instruments; preset categories must not marginalize participants or distort data.
• Remember that heavy control (lab context) can reduce ecological validity; balance precision with real-world applicability.

Numerical/Statistical Notation

• Percentages, proportions, means, standard deviations, and inferential tests (e.g., t-test, \chi^{2}, ANOVA, r) underpin quantitative analysis.
• Generalization to population relies on probability sampling & parameter estimation (e.g., 95% confidence interval \pm1.96\times SE).
• Always report effect sizes (e.g., d, \eta^{2}) alongside p-values.

Reference Backbone (for further study)

• Cristobal & De La Cruz-Cristobal (2022).
• Dalisay & Ramos (2022).
• Microsoft Copilot (GPT-4) (2025).