Correlational Design

Correlational Design

  • Overview: Focus on relationships between variables rather than differences.

  • Basics of Correlation: Measures how two variables relate to each other.

Scatterplots

  • Definition: Graphs depicting pairs of observations as dots.

  • Types: Can show negative or positive correlations.

Correlation Coefficient (Pearson's r)

  • Ranges from -1 to +1, indicating the strength and direction of a relationship.

  • Formula:
    r=COVXYsxsyr = \frac{COV_{XY}}{s_x s_y}

  • Strength indications:

    • 0 = no relationship

    • +1 = perfect positive

    • -1 = perfect negative

Coefficient of Determination

  • r2r^2: Proportion of variability in one variable attributed to another.

  • Example: If r=0.56r = 0.56, then r2=0.31r^2 = 0.31 indicates 31% of the variability in Y (e.g., GPA) is explained by X (e.g., IQ).

Regression Analysis

  • Simple Regression: Predicts Y from a single X using Y=(b)(X)+aY = (b)(X) + a where b is slope and a is the intercept.

  • Multiple Regression: Involves multiple predictors to analyze relationships among several variables.

  • Example: Predicting frequency of earworms includes factors like anxiety, liking of the song, and gender.

Directionality & Third Variables

  • Important to distinguish if A affects B or vice versa, and to rule out third variables that may influence the relationship.

Practical Applications of Correlational Studies

  • Useful for exploring hard-to-manipulate variables (e.g., effects of screen time on ADHD).

  • Strengths: Allow assessment of real-world relationships without manipulation.

Reliability and Validity in Research

  • Reliability Types: Test-retest, split-half, inter-rater.

  • Validity Types: Content, criterion, construct, convergent, discriminant.

Criteria for Causality

  • Association: Significant relationship exists.

  • Time Priority: Establish which variable occurred first.

  • Non-Spurious Relationship: No third variable explains the correlation (high internal validity).