Distinguishing One- and Two-Sided Hypothesis Tests in Radiology Research

AJR review clarifies why distinguishing one- and two-sided hypothesis tests is essential for interpreting radiology research: two-sided tests are default; one-sided tests require prespecified directional hypotheses. Misuse inflates false-positive rates.

• Design: Educational review on statistical methodology (no original data).
• Key message: Two-sided tests should be used unless a directional difference is explicitly hypothesized before data collection; one-sided tests double the type I error rate if applied post hoc.
• Limitation: Not a clinical study; practical impact depends on reader awareness of statistical reporting.

RadPigeon summaries are original and for information only. They are not clinical advice.