**Welcome to the MozFest replication workshop!** The goal of the workshop is to give you a hands-on introduction to replicating an academic paper in political science or economics using publicly available data and code and freely available open source tools. **Papers:** Several economics and political science journals have repositories of data and code associated with the papers they publish. Most link to the data from the article's page on the journal site, but some use Harvard's Dataverse for their repository. - [*Review of Economics and Statistics*][9] (on Dataverse) - [*Review of Economic Studies*][1] - [*American Economic Journal: Applied*][2] - [*American Economic Review*][3] - [*American Journal of Political Science*][8] (on Dataverse) - [*Quarterly Journal of Political Science*][4] - [Institution for Social and Policy Studies][10] (Yale) [1]: http://restud.oxfordjournals.org/ [2]: https://www.aeaweb.org/aej-applied/ [3]: https://www.aeaweb.org/aer/index.php [4]: http://www.nowpublishers.com/QJPS [8]: https://dataverse.harvard.edu/dataverse/ajps [9]: https://dataverse.harvard.edu/dataverse/restat [10]: http://isps.yale.edu/research/data#.Vj3TSytB_Ms Though you can try and replicate any paper you'd like, we're prepared to help you with two: - Young, Joseph K. "State Capacity, Democracy, and the Violation of Personal Integrity Rights" *Journal of Human Rights* 2009:8(4), 283-300. - Avery, Robert B. and Kenneth P. Brevoort. "The Subprime Crisis: Is Government Housing Policy to Blame?" *The Review of Economics and Statistics* 2015:97(2):352-363. Search to find the data and code for one of these papers. **Tools:** We'll use the open source statistical program R to reproduce the results of two papers. Download R [here][6], as well as R Studio, a helpful GUI for R, [here][7]. [6]: http://www.r-project.org [7]: https://www.rstudio.com/products/RStudio/ The vast majority of economics papers are written in the proprietary software package [Stata][5]. Script files (called '.do files' in Stata) can be opened with any text editor, and Stata data files (.dta) can be opened in R with the 'foreign' package (but only through version 12). [5]: http://www.stata.com **Replication**: Use R to attempt to re-run the analysis provided by the authors. Can you get their code to run? Do you get the same results that are in the published paper? **Sharing**: Once you modify the code, share it publicly using the Open Science Framework([OSF][11]). If you prefer, you can also share using [Dataverse][12] or [GitHub][13], which can be linked to your OSF project. (Dataverse creates a DOI, so publishing there cannot be undone.) [11]: http://www.osf.io [12]: http://dataverse.harvard.edu [13]: http://www.github.com