Email campaign in one country: the diff-in-diff

Question

A company runs an email campaign to users in Country A but not in Country B, which serves as a control. Revenue per user before and after, in euros:

Country A (treated): 50 before, 58 after.
Country B (control): 50 before, 52 after.

Compute the difference-in-differences estimate of the campaign's effect per user.

Accepted Answer

Isolate the campaign from the trend. Country A's revenue per user rose 8 euros (from 50 to 58). But Country B, which got no campaign, rose 2 euros (from 50 to 52) over the same window, reflecting a shared trend such as a general uptick in spending or a seasonal effect. Attributing the full 8 euros to the campaign would over-credit it by the 2 euros that would have appeared anyway. Difference-in-differences. Subtract the control's change from the treated group's change: 	ext{DiD} = (58 - 50) - (52 - 50) = 8 - 2 = 6. The estimated campaign effect is 6 euros per user. Because both countries sta…

Email campaign in one country: the diff-in-diff

Your answer

Discussion

Learn the concepts

Related questions