Why

Chapter 5. Observation

How can we learn about causes just by watching how things work?

One day on my commute I saw an ad on the New York City subway that read, “If you finish high school, get a job, and get married before having children, you have a 98% chance of not being in poverty.” This ad is meant to discourage teen pregnancy, but it is not clear how to interpret such a statistic. It seems to imply that if a teenager does all of these things, she will have a 98% chance of non-poverty. But is that true? And does this mean she won’t be in poverty at the current time or will never fall into poverty? The number comes from a study examining poverty rates among people with various characteristics such as marital status, age, and education level, and calculating what fraction of that population were in poverty.¹ Yet the resulting statistic is based solely on observational data.

No one enacted policies (individual or societal) to make teenagers get pregnant or not, or force them into poverty or not. This means that the statistic only describes a characteristic observed in a population: 98% of people studied who finished high school, got jobs, and got married before having children did not end up in poverty. If an individual finishes high school, gets a job, and gets married before having children, their personal odds of being in poverty may differ. Thinking back to Chapter 1, this is similar to the distinction between the odds of any family being affected by SIDS, and an individual ...