University of georgia Congress Project


Since 2010, the University of Georgia Congress Project has sought to expose undergraduate students to data collection and political science research on congressional politics. Working in conjunction with Professor Michael Lynch, we teach a small course (5-20 students) each semester related to congressional politics. The course is integrated with a broader research project that attempts to model the roll call generating process by examining amendments dispensed with on the floor of the United States Congress from 1877-2014.  In accordance with the policies established by the University of Georgia's Center for Undergraduate Research (“CURO”), students are asked to spend five hours a week collecting data on the congressional amending process. Students are assigned a set of important pieces of congressional legislation and is tasked with reading through congressional debates over the measure through an online version of the Congressional Record.  They then enter information about all amendments offered on the floor to these bills (included whether they received a recorded or unrecorded vote) into an excel spreadsheet in a shared Dropbox folder. 

The sponsor of perhaps the best known amendment in American history, David Wilmot (D-PA).  The Wilmot Proviso was adopted in the House via an unrecorded teller vote.

The sponsor of perhaps the best known amendment in American history, David Wilmot (D-PA).  The Wilmot Proviso was adopted in the House via an unrecorded teller vote.

The project has been serviced by 108 separate undergraduates, 17 graduate students and three high school students. To date, students have collected data on 150,415 amendments to 2,322 landmark enactments across 69 congresses. Data collected by students in the project has led to 17 separate conference paper presentations—seven co-authored with undergraduate students and five co-authored with graduate students—one published paper co-authored with an undergraduate, one forthcoming book chapter co-authored with an undergraduate and two published papers co-authored with graduate students. Undergraduate students participating in the project have also been active with independent undergraduate research, winning ten CURO research awards and making 14 presentations at the UGA CURO symposium. Preliminary data from the project has been published in the U.S. News and World Report, the Washington Post’s Monkey Cage blog and online in conjunction with the R Street Institute.

In addition to coding data on the lawmaking process, students are required to complete a paper analyzing the passage of a “landmark piece” of American legislation. It is our hope that this paper will eventually become part of broader project on American lawmaking and published on a website accessible to the general public. In recent years, scholars of congressional politics have made great strides in generating important new datasets on congressional activity. Consistent with efforts to make congressional politics more transparent, these scholars have made their data available online. However, data alone is likely insufficient for members of the general public interested in congressional lawmaking. Most casual observers of politics are unlikely to download a dataset and analyze it. Making readable landmark bill histories available, however, should aide both scholars and journalists as well as enrich the knowledge of the general public. To date, students and faculty have completed over 40 such landmark bill histories. This will be published on a project website in the near future. An example bill history on the National Firearms Act of 1934 is available online.

During the Constitutional Convention, Delegate Nathaniel Gorham expressed concern about allowing for members to call for recorded votes in Congress, suggesting it would lead to “frivolous” votes to “mislead the people, who never know the reasons determining the votes.”

During the Constitutional Convention, Delegate Nathaniel Gorham expressed concern about allowing for members to call for recorded votes in Congress, suggesting it would lead to “frivolous” votes to “mislead the people, who never know the reasons determining the votes.”

Students have also reported that a number of internships and jobs have inquired about whether they had undergraduate research skills.  In this regard, the practical training we provide them with in terms of creating and managing large spreadsheets, mastering Microsoft Excel, writing short reports summarizing their activities and engaging in independent research has proved valuable.  Many former students have participated in the University of Georgia's Washington Semester Program. Additionally, they have accepted positions with Governor Andrew Cuomo’s (D-NY) Washington D.C. office, the House Rules Committee, various House and Senate offices, the Podesta Group lobbying firm, Capitol Hill Consulting Group, Ballotpedia and the Georgia State House of Representatives, as well as moved on to top law school programs including Yale, Duke, NYU, Cal-Berkeley, Virginia, Chicago and UGA, Political Science PhD programs including Vanderbilt, UNC, Georgetown and UGA, several Public Administration graduate programs and one became a white-water rafting guide in Montana. 

The data

From a research standpoint, the project is the first systematic effort to model the roll call generating process.  Using roll call votes, political scientists have demonstrated fairly convincingly that the two political parties are more polarized than they have been since the years leading up to the Civil War.  This polarization is commonly treated by the media as being solely driven by ideology.  Similarly, campaigns and interest groups routinely use roll call votes as latent or true measures of politicians’ attitudes on issues. This suggests that the solution to solving the problem of gridlock is to “vote the bums out” and replace them with less-ideological members. 

Figure 1: Important Enactments per Congress

Figure 1: Important Enactments per Congress

In contrast, we argue that the observed high levels of polarization are driven in part by unobserved changes in the roll call generating process. Roll call votes need to be formally requested by a member and supported by a sufficient second of one-fifth of a quorum, as specified by the Constitution. The framers debated the roll call voting clause and viewed the sufficient second as a comprise that balanced the need for transparency in government with a fear, as Nathaniel Gorham of Massachusetts put it, of stuffing the record with “frivolous” votes to “mislead the people, who never know the reasons determining the votes.”

The University of Georgia Congress project sought to evaluate how the roll call generating process has changed by collecting not only roll call vote data, but data on measures that did not generate roll calls. Our first step involved constructing a dataset of “important” legislation from 1877 to 2015. This combined several lists of “landmark legislation” compiled by scholars with several routine appropriation bills considered in each Congress. The decision to focus on important enactments was motivated by several factors, which are discussed in greater detail in Appendix A. Altogether, this left us with a list of 2,322 enactments across 69 congresses.  Figure 1 plots the number of important enactments per Congress. As the Figure demonstrates, data collection is complete for the 59th Congress (1905-1906) to the 113th Congress (2013-2014).

Figure 2: Amendments Filed per Congress by Chamber, 1905-2014

Figure 2: Amendments Filed per Congress by Chamber, 1905-2014

We then tasked teams of coders to read through the Congressional Record and collect data on all amendments to those bills. This left us with data on 150, 415 amendments. Discussed in Appendix B, our primary dependent variable is whether or not the amendment received a recorded vote. Because data collection is complete from the 59th Congress (1905-1906) to the 113th Congress (2013-2014), we focused on the 143,255 amendments filed during this period. Of those amendments, 90,706 were offered on the floor of either the House or Senate. Figure 2 plots the amendments filed per Congress by chamber from the 59th Congress (1905-1906) to the 113th Congress (2013-2014).

Consistent with our expectations, amendments that are offered on the floor appear to be substantially more likely to receive roll call votes in modern congresses. From the 59th Congress to the 79th Congress, 7.39% of the 31,707 amendments received roll call votes. From the 80th to 113th Congress, 27.76% of the 56,470 amendments received roll call votes.  Figure 3 plots the percentage of amendments considered on the floor with the percentage that received roll call votes. The figure also includes a simple lowess smoothing line of recorded votes per Congress to indicate the general trend in the data and plots the percentage of amendments considered on the floor that received roll calls on a second y axis.

Figure 3: Roll Call Votes per Amendments Considered by Congress

Figure 3: Roll Call Votes per Amendments Considered by Congress

Preliminary evidence suggests that much of this increase is attributable to amendments sponsored by more extreme members for electoral purposes. This, we argue, suggests that much of the observed increase in polarization is not due to changes in ideology, rather shifting electoral incentives and procedural changes. While this does not alter the primary products of polarization—crippling gridlock on salient issues and anemic legislative productivity—it does suggest alternative means of reform.  Specifically, if polarization is partially driven by electoral and institutional factors, then solving the problem requires a more nuanced set of solutions then simply replacing legislators.  In the absence of specific procedural reforms and increased public education about how Congress operates, “voting the bums out” will only led to the creation of new “bums.” 

Research Team

We would like to thank all the students that have made this project possible: Haidi Al-Shabrawey, Nathaniel Ament-Stone, Rain Ammons, Whitney Arp, Matthew Baker, Alice Barker, Becca Bennett, Ethan Boldt, Allison Brill, Jason Byers, Maitri Chittidi, Lauren Corbett Bryant, Kasey Clark, Aaron Cooperman, Ananda Costa, Shellea Crochet, Amanda Delaperriere, Michael Evans, Jason Fern, James Floyd, Matthew Fowler, Ryan Freeman, Jacob Frenkel, Catherine Funk, Vinita Gandhi, David Gelman,  Sophie Giberga, Kunal Goel, Braden Goodgame, Katherine Graham, Hannah Greenberg, Casey Grippando, Cody Hall, Leyall Harb, Spencer Hardin, Jacquelyn Harms, Sharne Haywood, Daniel Helmick, Cameron Henderson, Rory Hibbler, Kyle Hollimon, Eileen Hong, Nick Howard, Eric Howell, Elise Hynd, Dory Ille, Taylor Johnston, Sydney Juliano, Da Hae Kim, Cody Knapp, Haley Lattke, Maggie Little, Jill Maloney, Caleb Masten, Megan Mayfield, Jordan McKissick, Hayden McRee, Annabel McSpadden, Kayce Mobley, Amber Morgan, Erin Munger, Wes Nichols, Katie Nisbet, Charlotte Norsworthy, Rob Oldham, Landon O’Neal, Tiernan O’Neill, Katie Opacity, Mark Owens, Kevin Parker, Caroline Pearson, Colin Phillips, Justin Pinkerman, Nathan Pinnell, Elaina Polson, Alex Porter, Maribeth Portier, Nathaniel Rice, Scott Riley, Matthew Roberts, Dustin Sammons, Adrien Sandercock, Laine Shay, Melissa Siegel, Joel Sievert, Veselin Simonov, Rob Smalley, Darrian Stacy, Chase Stell, Melissa Strickland, Jessica Suh, Rachel Surminsky, Zach Taylor, Andrew Teal, Kelsey Thomas, Javier Trejo, Katie Umholtz, Abbe Van Gorder, Adam Veale, Michael Watson, Hannah Weiss, Ryan Williamson, Simon Williamson, Andrew Wills, Adam Wittenstein and Sarah Young.

Research from the Project

We have already written several papers using data from the collection project. The first paper listed below gives an overview of the project and our most complete analysis of the amendment data to date.

  • Michael S. Lynch and Anthony Madonna. 2017. "Broken Record: Causes and Consequences of the Changing Roll Call Voting Record in the U.S. Congress." Unpublished Manuscript. (link)
  • Keith L. Dougherty, Michael S. Lynch, and Anthony J. Madonna. 2014. "Partisan Agenda Control and the Dimensionality of Congress." American Politics Research. 42: 600-627. (link)
  • Michael S. Lynch and Anthony J. Madonna. 2013. "Viva Voce: Implications of the Disappearing Voice Vote, 1865-1996." Social Science Quarterly. 94: 530-550. (Wiley Online).
  • Jaime L. Carson, Anthony J. Madonna, Mark E. Owens, and Joel Sievert. 2012. "Building a Record: Amending Activity, Position Taking, and the Seventeenth Amendment." Unpublished Manuscript. (link).
  • David Gelman, Michael S. Lynch, Anthony J. Madonna, and Mark E. Owens. 2012. "Majority Party Influence in an Open Rule Setting: Examining Amending Activity in the 45th Congress, 1877-1879." Unpublished Manuscript. (link).
Out of over 100 photos, somehow this was the least awkward.

Out of over 100 photos, somehow this was the least awkward.


Additional supplementary materials can be found here:

Amendment Codebook

Important Enactment List

Questions and Comments

If you have questions about the project, interest in the undergraduate research course or would like to learn more about it, please contact us at either or Additionally, quotes from the Congressional Record are often posted on the project's twitter account Congressional Quotes.