Lesson 3 Notes — Variables of Research

Characteristics of Quantitative Research

  • Uses Numerical Data
    • Data collected are measurable and expressed in numbers (e.g., test scores, percentages, frequencies).
    • Enables precision and the ability to apply statistical tools.
  • Objective and Systematic
    • Follows a structured process—surveys, experiments, standardized instruments.
    • Designed to minimize personal bias and maximize replicability.
  • Uses Statistical Analysis
    • Employs mathematical or statistical techniques to detect patterns, relationships or trends.
    • Examples: correlation coefficients, tt-tests, ANOVA, regression.
  • Focuses on Measurable Variables
    • Investigates clearly defined variables that can be quantified and compared (reading level, age, income).

Central Premises About Quantitative Research

  • Numerical values can describe phenomena and infer relationships.
  • Preferred in scientific settings because hypotheses can be empirically tested.
  • Variables: attributes that give quantitative meaning to an object, phenomenon, or group. They are the foundation for hypothesis development and evaluation.

Key Measurement Questions Raised in the Lesson

  • • Is the attribute measurable at all times?
    • Do the values change over time or context?
    • Is the variable suitable for descriptive, correlational, ex-post-facto, quasi-experimental, or experimental designs?

Learning Objectives of the Lesson

  • IDENTIFY variables in research.
  • APPLY appropriate variables within specific quantitative research methods.

What Is a Variable?

  • "Anything that has a quantity or quality that varies."
  • Without variance, a concept cannot function as a variable.

Observation & Measurement in Research

  • Researchers answer inquiries by observing and measuring the quality or quantity of the study object.

Running Example: Tomato-Plant Study

  • Independent Variables (IVs): water, sunlight, nutrients in soil, kinds of soil.
  • Dependent Variables (DVs): how fast tomato seedlings grow, number of fruits produced.
  • Shows causal logic—IVs presumed cause; DVs presumed effect.

Dependent vs. Independent Variables

  • Independent Variable
    • Actively manipulated or modified by researcher (especially in experiments).
  • Dependent Variable
    • Depends on the IV; observed & measured outcome.
  • Application across designs:
    • Descriptive, correlational, ex-post-facto (no manipulation) vs. experimental (active manipulation).

Extraneous & Confounding Variables

  • Extraneous Variables
    • Factors that may influence the DV but are not manipulated by the researcher.
    • Examples: room temperature during a test, participant mood.
  • Confounding Variables
    • When an extraneous variable is uncontrolled and exerts a substantial effect on the DV, threatening internal validity.

Quantitative (Numerical) Variables

  • Numeric, measurable—core of quantitative research.

Discrete Variables

  • Take countable, whole-number values; no fractions or negatives.
  • Examples:
    • Number of children (0,1,2,…)
    • Cars owned
    • Students in class
    • Doctor visits per year
    • Books read per month
    • Defective products
    • Goals scored in football match

Continuous Variables

  • Can assume any value within a range, including fractions & decimals.
  • Examples:
    • Height (165.2 cm)
    • Weight (65.8 kg)
    • Temperature (22.4 °C)
    • Time (15.3 s)
    • Distance (120.5 km)
    • Blood pressure (120.6 mmHg)
    • Precisely measured income ($25,346.75)
    • Speed (55.8 km/h)

Levels of Measurement for Numerical Data

Interval Data

  • Ordered with equal, meaningful intervals, no true zero.
  • Permits addition/subtraction; ratios meaningless.
  • Examples & rationale:
    • Temperature in °C/°F (0 °C ≠ "no temperature")
    • IQ (120 ≠ "twice" 60)
    • Calendar years
    • Clock time of day.

Ratio Data

  • Ordered, equal intervals with a true zero: enables all arithmetic (+,−,×,÷).
  • Example logic: 80 kg40 kg=2\frac{80\,\text{kg}}{40\,\text{kg}} = 2 ➔ 80 kg is twice 40 kg.
  • Examples: height, weight, age, income, elapsed time.

Illustrative Numerical-Data Table (from transcript)

  • Age: 23.5 years (Continuous)
  • Test score: 88.6/100 (Continuous)
  • Number of siblings: 2 (Discrete)
  • Monthly income: $2,450.75 (Continuous)
  • Hospital visits per year: 3 (Discrete)
  • Patient weight: 68.2 kg (Continuous)
  • Products sold: 120 (Discrete)

Qualitative (Categorical) Variables

  • Not numeric; describe category membership. Sub-types: nominal, ordinal, dichotomous.

Nominal Variables

  • No inherent order; mutually exclusive categories.
  • Examples: gender identities, blood type, marital status, car type, religion.

Ordinal Variables

  • Categories possess a natural rank, but interval size unknown.
  • Examples:
    • Education: High-school < Bachelor’s < Master’s < PhD
    • Socio-economic status: Low < Middle < High
    • Customer satisfaction (Very unsatisfied → Very satisfied)
    • Pain scale (None, Mild, Moderate, Severe)
    • Cancer stage, academic honors, Spotify Top charts.

Dichotomous Variables (Binary)

  • Special case of nominal with exactly two categories; often coded 0/1.
  • Examples: binary gender (male/female), smoker status, test result (positive/negative), employment (yes/no).

Connections & Practical Implications

  • Research Design Alignment: Selecting variable types dictates statistical tests (e.g., chi-square for nominal, tt-test for interval/ratio).
  • Ethical Dimension: Misclassification (e.g., reducing gender to binary) can marginalize groups; researchers must justify variable operationalization.
  • Validity & Reliability: Controlling extraneous/confounding variables ensures accurate causal claims.
  • Real-World Application: From tomato growth to clinical trials, clear IV/DV structures enable actionable insights.

Blank Slides

  • Pages 12 & 13 contained no additional instructional content—likely placeholders.