Downloads
![Image representing the file: IZA paper MB_TC_RJ.pdf](/sites/default/files/output_url_files/IZA%2520paper%2520MB_TC_RJ.pdf_0.jpg)
IZA paper MB_TC_RJ.pdf
PDF | 404.48 KB
A growing literature on inference in difference-in-differences (DiD) designs with grouped errors has been pessimistic about obtaining hypothesis tests of the correctsize, particularly with few groups. We provide Monte Carlo evidence for three points: (i) it is possible to obtain tests of the correct size even with few groups, and in many settings very straightforward methods will achieve this; (ii) the main problem in DiD designs with grouped errors is instead low power to detect real effects; and (iii) feasible GLS estimation combined with robust inference can increase power considerably whilst maintaining correct test size – again, even with few groups
Authors
![Mike Brewer](/sites/default/files/styles/square_desktop/public/2022-07/Mike%20Brewer.jpg?itok=3JdYlHV1)
Mike Brewer
![Thomas Crossley](/sites/default/files/styles/square_desktop/public/2022-07/Tom%20Crossley.jpg?itok=8BNVWAtq)
Research Fellow University of Michigan
Tom is a Research Fellow at IFS, a Research Professor for the Institute for Social Research at the University of Michigan.
![Robert Joyce](/sites/default/files/styles/square_desktop/public/2022-06/Rob_Joyce.jpg?itok=4UaBJrrb)
Deputy Director
Robert is a Deputy Director. His work focuses on primarily on the labour market, income and wealth inequality, and the design of the welfare system.
Report details
- Publisher
- IFS
Suggested citation
M, Brewer and T, Crossley and R, Joyce. (2013). Inference with Difference-in-Differences Revisited. London: IFS. Available at: https://ifs.org.uk/publications/inference-difference-differences-revisited-1 (accessed: 1 July 2024).
More from IFS
Understand this issue
![Microphone](/sites/default/files/styles/square_desktop/public/2024-06/Microphone.jpg?itok=soM7Wvbz)
Election Special: Your questions answered
27 June 2024
![Tom Waters](/sites/default/files/styles/square_desktop/public/2024-06/tom%20waters.jpg?itok=faDtwS3F)
What is the two-child limit in benefits?
27 June 2024
![Interview](/sites/default/files/styles/square_desktop/public/2024-06/1.jpg?itok=mjderngL)
Election Special: The big issues politicians haven't spoken about
25 June 2024
Policy analysis
![Welsh town with mountains behind](/sites/default/files/styles/square_desktop/public/2024-06/Welsh-town-feature.jpg?itok=5axb5wZn)
How would the parties’ tax and spending plans affect Scotland and Wales?
28 June 2024
![Polling station](/sites/default/files/styles/square_desktop/public/2024-06/Polling-station-school.jpg?itok=uS-l6b-1)
What are the parties’ plans for benefits and taxes?
24 June 2024
![Shopping street](/sites/default/files/styles/square_desktop/public/2024-06/Street-scene-.jpg?itok=R39cR6Xp)
How do the last five years measure up on levelling up?
19 June 2024
Academic research
![Working paper cover](/sites/default/files/styles/portrait/public/2024-06/WP202428-The-impact-of%20labour-demand-shocks-when-occupational-labour-supplies-are-heterogeneous.jpg?itok=Erq9-V9O)
The impact of labour demand shocks when occupational labour supplies are heterogeneous
28 June 2024
![Working Paper Cover](/sites/default/files/styles/portrait/public/2024-04/WP202413-Labour-market-inequality-and-the-changing-life-cycle-profile-of-male-and-female-wages_Page_001.jpg?itok=FGFp0iFj)
Labour market inequality and the changing life cycle profile of male and female wages
15 April 2024
![Working Paper Cover](/sites/default/files/styles/portrait/public/2024-04/WP202414-Interpreting-cohort-profiles-of-lifecycle-earnings-volatility_Page_01.jpg?itok=PhiPLJpT)
Interpreting cohort profiles of lifecycle earnings volatility
15 April 2024