Difference-in-differences
A measurement method for when you cannot randomize but have a comparison group: instead of comparing levels after launch, you compare how much each group changed over the same window, subtracting out the shared background trend. It holds only if the groups would have moved in parallel without the AI, which must be stated as an assumption, not proven.
Defined in 2 GAGE programs, which carry 2 distinct definitions of it. The wording above is taught in Business AI Transformation.
How each discipline defines it
The same term does different work depending on who is using it. These are the definitions as each program teaches them, unedited.
A measurement method for when you cannot randomize but have a comparison group: instead of comparing levels after launch, you compare how much each group changed over the same window, subtracting out the shared background trend. It holds only if the groups would have moved in parallel without the AI, which must be stated as an assumption, not proven.
A comparison method that measures how much a treated group's metric changed relative to how much a comparable untreated group's metric changed over the same period, isolating the treatment's own contribution from factors that affected both groups equally.
Where it is taught
The exact lessons this term appears in. The first 7 topics of every program are free with a free account.
- The AI Value Office and Experimentation Discipline · Implementation from Pilot to Scale, Business AI Transformation
Terms it appears with
Not an alphabetical neighbourhood: these are the terms taught in the same lessons, ranked by how often they appear together.