Following Lives Over Time: Cohort Studies and Crime

In 1972, researchers in Dunedin, New Zealand, began tracking 1,037 babies born in a single year. They have followed those same people ever since, measuring their health, choices, and run-ins with the law across five decades. One striking finding: a small group of roughly 5 percent of the cohort accounted for a large share of the group's total criminal convictions. You cannot uncover a pattern like that from a one-time survey. You need a longitudinal cohort study.

Criminology has a founding question: why do some people commit crime while most do not, and why do most people who offend as teenagers stop as adults? Following the same individuals over many years is one of the most powerful ways to answer it.

By the end of this lesson you will be able to: - Define a longitudinal cohort study and contrast it with a cross-sectional snapshot. - Explain the age-crime curve and what cohort data revealed about it. - Identify the strengths and the serious limitations of this method (attrition, cost, correlation vs. causation). - Interpret a real finding about chronic offenders.

A cross-sectional study measures many people at ONE moment, like a single photograph of a crowd. It can tell you who is offending now, but not whether the same people keep offending, or what came first. A longitudinal cohort study follows the SAME group (the cohort) repeatedly over time, like a time-lapse film. Because it observes the same individuals as they age, it can establish TIME ORDER: it can show that something (say, harsh childhood adversity) came BEFORE later offending, which is a necessary condition for arguing one might cause the other.

In a famous 1972 study, Marvin Wolfgang and colleagues followed 9,945 boys born in Philadelphia in 1945 up to age 18. They found that a small fraction, about 6 percent of the boys, were 'chronic offenders' responsible for more than half of all the offenses committed by the entire cohort, including a large majority of the serious violent crimes. This '6 percent' finding reshaped policy debates. If a tiny group commits most serious crime, some argued, resources should focus on identifying and helping that group early. Critics warned this risks unfair labeling of children based on predictions that are often wrong.