Videos
View the MattWebbEcon channel on YouTube.
Good Controls Gone Bad: Difference-in-Differences with Covariates
Why covariate assumptions matter for difference-in-differences, and how integrated parallel trends and DID-INT can help.
Randomization Inference for Accounting Researchers
How placebo treatment effects support randomization-inference tests in small- and large-sample accounting research.
Pitfalls when Estimating Treatment Effects Using Clustered Data
Why treatment-effect inference can be unreliable with few treated clusters or unequal cluster sizes, even when the total number of clusters is large.
Fast and Wild: Bootstrap Inference In Stata Using Boottest
Fast wild-cluster-bootstrap tests and confidence sets in Stata using boottest, especially when conventional large-sample inference is unreliable.
Fast and Reliable Jackknife and Bootstrap Methods for Cluster-Robust Inference
Computationally efficient jackknife and wild-cluster-bootstrap methods for reliable inference with few clusters or substantially unequal cluster sizes.
Leverage, Influence, & Jackknife in Clustered Regression Models: Reliable Inference Using summclust
How summclust diagnoses cluster-level leverage and influence and computes jackknife standard errors for clustered regression models.
Cluster-Robust Jackknife and Bootstrap Inference for Logistic Regression Models
Reliable inference for logistic regression with clustered data using linearized cluster-jackknife and wild-cluster-bootstrap methods implemented in logitjack.
Which Policy Works, and Where? State-Level Treatment Effects in Difference-in-Differences
How treatment effects and inference should match policy questions at the jurisdiction, cohort, period, or policy-type level.
The Translation Assumption Unmasking Representation Risk
How pretrained image encoders can yield different predictions from the same images, and how LookAgain-ML helps researchers assess and select representations.
The Wild Bootstrap for Few (Treated) Clusters
Subcluster wild-bootstrap methods for inference with very few treated clusters, including when ordinary wild-bootstrap procedures can work well.
Cluster-robust inference: A guide to empirical practice
Practical guidance on when cluster-robust inference is needed and how to approach common design and implementation choices.
Testing for the Appropriate Level of Clustering in Linear Regression Models
An overview of score-variance tests and the statistical trade-offs involved in choosing a clustering level.
Jackknife Inference with Two-Way Clustering
Jackknife methods designed to improve inference in linear regression models with clustering in two dimensions.
Good Guys With Guns?
A research overview of Canadian administrative data on legal firearm ownership, deaths, and crime.
Wild Bootstrap Inference for Wildly Different Cluster Sizes
Why conventional cluster-robust inference can fail with highly unequal cluster sizes and how wild-bootstrap methods can help.
Randomization Inference for Difference-in-Differences with Few Treated Clusters
Randomization-inference procedures for difference-in-differences designs with few treated clusters.
Improved Inference for CSDID Using the Cluster Jackknife
How the cluster jackknife can provide more reliable inference for the Callaway-Sant’Anna difference-in-differences estimator.
Reworking Wild Bootstrap-Based Inference for Clustered Errors
A concise account of refinements to wild-cluster-bootstrap inference for settings with few clusters.
Wild Bootstrap Randomization Inference for Few Treated Clusters
An introduction to wild-bootstrap randomization inference for clustered designs with a small treated group count.
Impact of a Peer Comparison Intervention on Seasonal Influenza Vaccine Uptake in Community Pharmacy
A short overview of a national cluster-randomized study of peer comparison and influenza vaccination in community pharmacies.
A Simple, Graphical Procedure for Comparing Multiple Treatment Effects
A graphical method for comparing multiple treatment effects against a control and against one another.
Immigrant Parents’ Category of Admission and Outcomes of the Children: How Far Does the Apple Fall?
How immigrant admission categories in Canada relate to the long-run outcomes of children.
Targeting Tax Relief at Youth Employment
Evidence on the employment effects of Canada’s Youth Hires payroll-tax-relief program.
Finish It and It’s Free: An Evaluation of College Graduation Subsidies
An evaluation of Canadian graduate tax credits, university completion, and interprovincial mobility.
Choosing Portfolios of Rival Risky Options: Most People Violate the No Safety Schools Theorem
Experimental evidence on how people choose portfolios when options are risky and rival.