Answers are grounded in this paper and cite their sources.
This article shows when a triple difference strategy using an imperfect control category improves on the double difference strategy for estimating an average treatment effect. For example, a product is treated in one place and not another leading to a double difference strategy. When does comparison with an untreated product in triple difference strategy improve accuracy?
No takes yet. Share an insight, caveat, or question.
Berck et al. (2015) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: