This page has links to datasets used in my research, and supplementary materials like computer code to help you manipulate and analyze the data.  If you have any questions or problems with the data, send me an email.  If you use these data, please cite them as:

1. •Carey, John M.  [year.]  Carey data archive.  www.dartmouth.edu/~jcarey
   

Transparency

This project has two parts.  One part uses laboratory experiments to address the question of whether representatives behave differently when their actions are observed by citizens versus when they are not observed.  Intuition suggests transparency should matter, but non-transparent legislative processes in many countries present obstacles to conventional empirical research.  The data here were produced by preliminary experiments conducted during the summer of 2010, using Dartmouth students as participants.  The experiments involve Legislators proposing and voting on a budget that can be divided among the Legislators and the Public, followed by the Public deciding whether to reelect each Legislator for the next period.  The degree of transparency varies across different treatments.  

1. • Transparency experiments data in Stata format (August 2010)
   

2. • Transparency experiments analysis Stata do-file (August 2010)
   

3. • Transparency experiment codebook (August 2010)
   

A second part of this project is collecting data from the websites of national legislatures around the world to document what kind of information they make available, and in what format.  That is, my research assistants and I are trying to create a map of the world’s parliaments in terms of how visible legislators’ actions are, and how easy it is for citizens, journalists, activists, or academics to monitor those actions -- at least via the web.  We are particularly interested in how visible legislative vote records are, although we have collected more information than that.  This is a work in progress.  We have pretty well covered the Americas, and we have collected information on many other countries outside the Western Hemisphere, too, but much of the world of legislative transparency remains un-mapped, at least by us.  We will update our data as we collect it.  For now, you can see a map of the world, and see the reports on legislative transparency for the countries we’ve documented, here.

Legislative Voting Project

This project involves the collection of recorded votes from 21 legislative chambers in 19 countries, as well as interviews with 55 legislators and political party leaders in 8 countries.  The data are analyzed in various articles, as well as the book Legislative Voting & Accountability.  There are two types of material associated with this project:

1. •Datasets of recorded votes from various legislatures, and Stata do-files (computer programs written in the Stata statistical software package that allow users to generate from the raw datasets various indices of voting unity, as well as other statistics ).
   

2. •Transcripts of interviews with legislators and other politicians
   

Electoral System Design Project

This project, collaborative with Simon Hix (London School of Economics) examines all elections between 1945-2006 in democracies with populations over one million -- 610 elections in all, from 81 countries.  The dataset contains information on constitutional design and electoral rules (including mean and median magnitudes of districts in each election, legal thresholds, mixed-member formats), on disproportionality and party system fragmentation, on various indicators of political economy (growth rates, government spending, surpluses/deficits, inequality, human development index), and on demographic and historical information.  There are two files, the data and a codebook:

1. • Carey & Hix “Electoral Sweet Spot” data (March 2010)
   

2. • Carey&Hix “Electoral Sweet Spot” Stata do-file (March 2010)
   

3. • Carey & Hix Codebook (September 2008)
   

Primary Elections in Latin America Project

This project, some parts of which are collaborative with John Polga-Hecimovich (Dartmouth ‘05), examines whether candidates for executive office nominated by primary are stronger or weaker in general election competition than candidates nominated by other methods.  There are two datasets,  from:

1. •Presidential Primaries Data (revised and updated, October 2007)
   

2. •Primaries Replication Memo (October 2007)
   

3. •Mexican gubernatorial elections, 1998-2003
   

Constitutional Creation Project

This project examines ‘constitutional moments’ worldwide, from 1990-2005 to determine whether the way a constitution is created (e.g. democratic election of constitutional designers, inclusiveness of deliberations, use of referenda) affects whether new constitutions subsequently deliver stability and democracy.  A lot of the data for this project are adapted from data generously shared by Professors Zachary Elkins and Thomas Ginsberg (University of Illinois) from their Comparative Constitutions Project, and by Professor Jennifer Widner (Princeton) from her Constitution Writing and Conflict Resolution Project.  The two datasets available here adapt some of the variables from these projects and add some further information.

1. •Carey CCP adapted data
   

2. •Carey CWCR adapted data
   

3. •Carey Constitutions Codebook
   

Candidate Selection in Chile Project

This project collaborative with Peter Siavelis (Wake Forest University), examines the career trajectories of successful and unsuccessful candidates for the Chilean Congress to identify the strategies that party and coalition leaders use in order to recruit high-quality candidates to undertake risky electoral campaigns.  There are two datasets:

1. •an Excel file that includes all the descriptive data on legislative candidates
   

2. •a Stata file that includes just the variables from our Latin American Politics & Society article
   

U.S. State Legislature Term Limits Project

For data from this project, please contact Lynda Powell, at the University of Rochester.