Common Misconceptions
Statistical Pitfalls
Four misconceptions that appear frequently on exams — and in published papers. Click a pitfall to expand it, then test yourself with the scenarios below.
Evaluate these conclusions
Two researchers write up the same result differently. Which statement is more appropriate?
A researcher runs a simple linear regression of exercise frequency (hours/week) on resting heart rate in 120 university students. The result is significant: b = −2.4 bpm per hour/week, P < 0.001, r² = 0.38.
A clinical trial of a new antibiotic compares recovery time in n = 12 patients (antibiotic) vs. n = 12 (placebo). Welch's t-test gives t₂₂ = 1.8, P = 0.085. The mean difference is 1.4 days (95% CI: −0.2 to 3.0 days).
A genetics study tests whether a point mutation affects enzyme activity. An ANOVA comparing three genotype groups gives F₂,₈₇ = 6.4, P = 0.002.