Correlational & descriptive research

Ice cream sales and drowning deaths rise together every summer — so does ice cream cause drowning? Of course not (hot weather drives both). That's the whole point of correlational research, and the trap examiners love: correlation does not equal causation.

Correlational research measures whether two variables relate to and vary together, without manipulating anything. The correlation coefficient (r) runs from −1.00 to +1.00. The sign tells direction: positive = variables move the same way (both up or both down); negative = they move in opposite directions (one up, the other down). The absolute value tells strength: closer to 1.00 (either sign) = stronger; near 0 = weak or no linear relationship. A strong negative like −.85 is stronger than a weak positive like +.20 — sign is not strength.

Because nothing is manipulated and groups aren't randomly assigned, correlational data cannot establish cause. Two classic reasons: directionality (does A cause B, or B cause A?) and the third-variable problem (an unmeasured factor — like summer heat — drives both). Correlation is excellent for prediction and for studying variables you can't ethically manipulate, but not for causal claims.

Descriptive research aims to describe — not explain — characteristics of a group: what exists, how often, in what pattern. Survey methods are the workhorse here, gathering self-report data via questionnaires or interviews from a sample. Key survey concerns: sampling (is it representative?), response rate, and wording bias (leading questions skew answers). Two survey structures worth knowing: cross-sectional (many groups measured once, at one time point) versus longitudinal (the same group followed over time).

Clinical Pearl: The number-one testable line — correlation ≠ causation; a relationship between two variables never proves one causes the other (watch for the third variable). And remember sign = direction, magnitude = strength: r = −.85 is a stronger relationship than r = +.20.