Cross-sectional vs longitudinal
MCAT glossary · MCAT
Definition
Cross-sectional: data collected at one point in time across a population (cheap and fast; cannot establish temporal sequence or causality). Longitudinal: same subjects measured repeatedly over time (allows analysis of within-subject change, supports temporal-sequence inferences). Longitudinal designs are NECESSARY for causal claims but not SUFFICIENT — without randomization or robust confounding control, observational longitudinal studies still face confounding, selection, and reverse-causation threats.
Cross-sectional vs longitudinal in plain English
These are two ways to run a study over time. Cross-sectional is a single snapshot — you measure a bunch of different people once, all at the same moment (fast and cheap, but you can't tell what came first, so no real cause-and-effect). Longitudinal follows the SAME people repeatedly over months or years, so you can actually watch individuals change and see which event happened before which. Here's the subtle part: following people over time is REQUIRED to argue causation (you need the cause to come before the effect), but it's not enough by itself — without careful controls, other explanations like confounders or 'the effect actually caused the cause' can still fool you.
Practice Cross-sectional vs longitudinal
Sign up free: all 171 MCAT terms as flashcards, plus practice questions with full explanations →
Related MCAT terms
MCAT practice questions · MCAT flashcards · MCAT study guide