Difference-in-Differences

Definition

A causal-inference design comparing before/after changes in a treatment group to before/after changes in a control group.

Example

The Card-Krueger minimum-wage study compared NJ fast-food employment to Pennsylvania.

How the lesson puts it

Difference-in-differences is the workhorse method for natural experiments. You take the change over time in the treatment group, then subtract the change over time in a control group. Subtracting removes anything that hit both groups equally, like a nationwide trend, leaving the policy's own effect.

Where you learn it

← All terms

Download on theApp Store