Jim Himes | Democrat for Congress | Chris Shays: Partisan When it Counts

Statistical analysis of Chris Shays� votes reveals that when it counts, Shays has been a loyal servant of the Republican Party. On the closest votes�when his vote has the most power to affect the outcome, when the Republican Party most needs his support, and when 4th District residents most need him to make the right choice�Shays has lined up behind his Party�s leadership 89% of the time, or nine out of ten times.

This finding�based on analysis of the 1,752 votes cast in the 110th Congress�shows that Shays is not �purple� as he claims, but is, in fact, a true Republican partisan. Real political courage doesn�t mean crossing the aisle when it�s safe and easy; it means exercising independent judgment on the close votes where your decision can really make a difference.

Chris Shays: Partisan When It Counts

Moreover, the narrow votes where Shays has been a reliable supporter of the Republican agenda include key measures such as:

- Voting against tax cuts for the middle class, including voting against extending credits for college tuition and voting against exempting middle class families from the Alternative Minimum Tax (HR 3996)
- Voting against a timetable for withdrawal from Iraq (HR 1591)
- Voting to allow torture by CIA interrogators (HR 2082)

For the purposes of this analysis, a vote was deemed �close� if it was in the closest 1/3 of all the votes taken so far in the 110th Congress (see below for methodology.)

Dr. Arthur Kover, a sociologist and Emeritus Professor at Fordham University, made the following statement regarding this research:

�I have reviewed this work, and there is no subjective interpretation of the data here � just a straightforward report of the facts of Congressman Shays� voting record. This is a commonly accepted method of analysis, using publicly available data to make an objective statement of fact. Anybody who knows how to use Excel could replicate the results using publicly available data on their home computer.

�The finding rests on dividing all the votes in the 110th Congress so far into two categories: close votes and safe votes. The closest one-third of all the votes are put into the �close votes� category, and the rest are put into the �safe votes� category. Then it is just a matter of calculating how often Congressman Shays votes with the Republican majority in each set of votes. It is a numerical fact that Congressman Shays sides with the Republican majority on about nine out of ten close votes.�

As Chris Shays tries to distance himself from George W. Bush and the Republican Party this fall, the facts will tell a very different story. Simply put, Chris Shays has voted with the Republicans nine out of ten times when it counts.

Speak out about Chris Shays' partisan voting record:

For a technical statement of the research methodology, click here.

A record of all Roll Call votes for the 110th Congress through Roll No. 566 of the 2nd Session on 8/1/2008, obtained in the standard format of the Inter-University Consortium for Political and Social Research at the University of Michigan, was used as the dataset for this analysis. The datafile itself was downloaded from the website of Dr. Keith Poole at the University of California, San Diego, where it is freely available (see http://voteview.com/dwnl.htm). The file was updated for consistency with the Library of Congress� THOMAS database, and the following procedure was then performed:

For each vote, the total number of Yea votes and Nay votes was determined.
For each vote, the vote margin was calculated as the absolute value of the difference between the number of Yea votes and the number of Nay votes.
For each vote, the numbers of Democrats voting Yea and Nay, and the numbers of Republicans voting Yea and Nay, were determined.
The majority of Democrats were determined to support a vote whenever the number of Democrats voting Yea exceeded the number of Democrats voting Nay; otherwise, the majority of Democrats were determined to oppose the vote. The same determinations were made for Republicans.
Rep. Shays was determined to have voted with the Republicans whenever he voted Yea and the majority of Republicans voted Yea, or he voted Nay and the majority of Republicans voted Nay. The same process was used to determine when Rep. Shays voted with the Democrats.
Votes which the majority of Democrats and the majority of Republicans either supported or opposed were excluded from the analysis. This ensures that the rate at which Rep. Shays votes with the Republicans is not artificially inflated by votes where he voted with the Republicans, but also with the Democrats.
Votes where Rep. Shays did not vote either Yea or Nay were excluded from the analysis.
Within the remaining dataset, which encompasses 980 votes, the vote margins ranged from 0 to 224. When these 980 votes are ordered by the size of their vote margins, the 33rd percentile encompasses votes with margins of 31 votes or less. These 333 votes, the closest 1/3 of the total set, were taken to comprise the set of "close votes."
The percentages of close votes (N=333) and other votes (N=647) where Rep. Shays voted with the Republicans were computed separately. The percentage for close votes comes to 89.2%.

The data necessary to replicate this procedure and the findings above are all publicly available. For replication purposes, note that the dataset does not include the non-binding votes of delegates from Guam, Virgin Islands, Puerto Rico, American Samoa, and the District of Columbia (including these would not have any material effect on the findings).