Reproducibility in research – Ensuring the transparency and credibility of your work

Authors

Lars Vilhuber

Marie Connolly

Published

May 28, 2026

Location: TBD

Registration required, see [registration link] (coming).

Sponsored by Alfred P. Sloan Foundation grant to Cornell University (Lars Vilhuber) and by the Research Chair in Intergenerational Economics, ESG UQAM (Marie Connolly). Part of the Canadian Economics Association meetings pre-program.

ILR Cornell logo

UQAM logo

9:00 Welcome
9:05 Walkthrough
9:15 Goals
9:30 Technical setup, possible team formation
9:45 🔒Hands-on Exercise: A very imperfect example
10:00 Discussion of the “Very imperfect example”
10:30 Break (15 minutes)
10:45 Day 1:
- Setting yourself up for reproducibility
- How to correctly document a replication package (and why!)

12:30 How to run Stata! or R! (reproducibly)
12:45 Extra: How to install Stata packages
1:00 Topic A (see Survey)
1:45 Topic B (see Survey)
2:30 Hands-on: Improving the replication package (very imperfect -> a lot better)
2:50 Hands-on: Testing it all
2:55 Wrap up

For Part 2, we have a choice:

To identify what we should be speaking about in the afternoon, please fill out this other survey:

No data yet. Check back later.

Discussions on concerns about ethics in data dissemination (privacy, scraping, copyright, etc.)
Reproducibility for LLM and AI

Some additional guidance can be found on the website of the Social Science Data Editors (URLs subject to change):

With confidential data

https://doi.org/10.3886/E154241V2 not only code, but faces the problem that IRS data cannot have variables revealed. Their workaround is not the same one as in this tutorial.
https://doi.org/10.3886/E162581V1

Using containers:

Coming.