Study Notes on Discrete and Continuous Variables

Discrete and Continuous Variables

Introduction to Variables

  • Variables in a study can be characterized by the types of values they can assume.

  • The type of values influences the statistical procedures for summarizing or making inferences.

Discrete Variables

  • Definition: Discrete variables consist of separate, indivisible categories.

  • Characteristics:

    • No intermediate values exist between two adjacent categories.

    • Example: Number of questions answered correctly on a 10-item multiple-choice quiz.

    • Between values 7 (seven correct) and 8 (eight correct), no intermediate values can be observed.

  • Common Characteristics:

    • Typically represented by whole, countable numbers (integers).

    • Examples include:

    • Number of children in a family.

    • Number of students attending a class (e.g., 18 students one day, 19 the next).

      • Cannot observe a value between 18 and 19 students.

  • Qualitative Observations:

    • Discrete variables can also include qualitative distinctions.

    • Examples:

      • Birth order (first-born vs. later-born).

      • Occupations (nurse, teacher, lawyer, etc.).

      • Academic major (art, biology, chemistry, etc.).

Continuous Variables

  • Definition: Continuous variables are not limited to a fixed set of separate categories.

  • Characteristics:

    • Continuous variables can be divided into an infinite number of fractional parts.

    • Example: Measurement of time (hours, minutes, seconds, fractions of seconds).

  • Detailed Explanation:

    • Continuous variables can represent an infinite number of values between any two observed values.

    • Example: Measuring weights in a diet study.

    • Visual representation: Continuous line with infinite possible points, no gaps between neighboring points.

    • For any two distinct points on a continuous variable line, a third value can always be found between them.

  • Weight Measurement Example:

    • A measurement of 150.5 does not strictly belong to 150 or 151 but resides at the boundary.

    • Placement depends on rounding rules:

    • Round up to 151 or round down to 150.

Considerations for Continuous Variables

  1. Unique Measurements:

    • Identical measurements between different individuals should be rare, indicating true continuity.

    • If tied scores emerge, either:

      • The variable may not be truly continuous.

      • The measurement procedure might be imprecise (restricted to discrete values).

  2. Measurement Categories:

    • Researchers must define measurement categories on the measurement scale.

    • Example: Weight measured to the nearest pound (categories: 149, 150, etc.).

    • Each measurement category is an interval defined by boundaries (real limits).

      • For a score of 150 pounds:

      • Lower real limit: 149.5

      • Upper real limit: 150.5

    • Individuals with weight between these limits (e.g., 149.6 and 150.3) are both assigned a score of 150.

Real Limits

  • Definition: Real limits are boundaries for intervals of scores on a continuous number line.

  • Characteristics:

    • Each score has two real limits:

    • Upper Real Limit: Top of the interval.

    • Lower Real Limit: Bottom of the interval.

  • Application:

    • Real limits apply to any measurement of continuous variables.

    • Example: Measuring time to the nearest tenth of a second.

    • Categories: 31.0, 31.1, 31.2, etc.

    • A score of X = 31.1 seconds corresponds to an interval defined by:

      • Lower real limit: 31.05

      • Upper real limit: 31.15

    • Real limits are essential for constructing graphs and performing calculations with continuous scales.

Important Distinction

  • Continuous vs. Discrete Measurement:

    • Terms ‘continuous’ and ‘discrete’ refer to the underlying variables and not the scores obtained from measurements.

    • Example: Height measured to the nearest inch produces discrete scores (60, 61, 62, etc.) but is fundamentally a continuous variable.

    • Key to determining variable type:

    • Continuous variables can be divided into any number of fractional parts (e.g., height can be measured to the nearest inch, 0.5 inch, or 0.1 inch).

    • Various measurement approaches:

    • Pass/fail system classifying students' knowledge is discrete.

    • A 10-point quiz: 11 categories (0 to 10) is discrete.

    • A 100-point exam: 101 categories (0 to 100) remains discrete.

    • Allowing choice in precision or number of categories indicates a continuous variable.

Conclusion

  • Understanding the differences between discrete and continuous variables is crucial for choosing appropriate statistical methods and interpreting study results.