Use Difference-in-Differences without an A/B test
WHAT IT TESTS: causal inference when randomization is impossible. OUTLINE: give a scenario like a region-wide launch, apply Difference-in-Differences comparing treated vs control over time, and state the parallel-trends assumption.
WHAT IT TESTS: whether you can estimate causal impact when you cannot randomize, such as a feature rolled out to an entire market for legal or technical reasons. ANSWER OUTLINE: propose Difference-in-Differences, which compares the before-versus-after change in a treated group against the same change in an untreated control group, subtracting out shared trends. Its core assumption is parallel trends: absent the launch, treated and control groups would have moved in parallel.
Read the original → interview
- #causal-inference
- #difference-in-differences
- #quasi-experiment
- #parallel-trends
- #experimentation
Get five bites like this every day.
Tezvyn delivers a daily feed of 60-second tech bites with quizzes to lock in what you learn.