(Joint House and Senate Scaling)
Updated 26 June 2016 (Updates only occur when Congress is in Session)
This is the twenty-eigth release of WeeklyCommon Space DW-NOMINATE scores for the House and Senate and the eighth for the second session
of the 114th Congress. The House and Senate were scaled as if they were one legislature using
the 650 Legislators who served in both the House and Senate as "glue" (bridge observations).
That is, we estimated a single ideal point for each member of Congress based upon
his/her entire record of service in Congress. In the Poole-Rosenthal framework we used the Constant model so that each unique legislator has the same ideal point throughout his or her
career.
In order to easily update the Common Space DW-NOMINATE scores when new roll calls
are cast in Congress we had to write a new DW-NOMINATE program that required as input
only the roll call matrix from Congresses 1 to 114 and the previous Legislator and
Roll Call output files for Congresses 1 - 113 from the former program. Jeff Lewis
wrote a batch file that combines PERL and Python scripts to combine all the roll call
vote matrices together and then run the program. When we have everything completed
these scores will be posted at UCLA and the links below will go there with updated
numbers of roll calls and legislators.
The New DW-NOMINATE program uses LBFGS to simultaneously estimate the roll call paraments
and to simultaneously estimate the legislator parameters. Beta and the 2nd dimension
weight are estimated using the Brent local minimization algorithm (Brent, Richard.
2002. Algorithms for Minimization Without Derivatives. New York: Dover). Legislators and the Roll Call Midpoints are constrained to lie
in the unit circle.
As of 26 June 2016 there were a total of 104,302 roll calls of which 93,456 were
scalable. The number of unique legislators is 12,044 (this counts Warren Davidson
(R-OH) who was elected to replace former Speaker John Boehner) producing a total of
17,392,378 choices. The second dimension weight is 0.4158 and Beta is 7.7793. The
correct classification is 87.37 percent with an APRE of 0.6277 and a geometric mean
probability of 0.7561.
In order to calculate distances from these Common Space DW-NOMINATE scores you must
multiply the second dimension by the weight parameter. To calculate the choice probabilities
you must apply both the second dimension weight and the Beta parameter. Use the Yea and Nay outcome coordinates with considerable caution because, as we explain
in Congress: A Political Economic History of Roll Call Voting, they are poorly identified. However, the cutting line is identified and can be used
safely.
Please note that these files contain scores for most Presidents. For Presidents prior
to Eisenhower these are based on roll calls corresponding to Presidential requests.
These roll calls were compiled by an NSF project headed by Elaine Swift ( Study No. 3371, Database of Congressional Historical Statistics, 1789-1989). Many of these scores are based upon a small number of roll calls so use them with caution!
In the files below the House Coordinates for each Congress are stacked on top of the
Senate coordinates. If you have questions or need help with these files please send us e-mail at jblewis_at_ucla.edu
(Jeff Lewis) or ktpoole_at_uga.edu (Keith Poole).Updated 26 June 2016 (Updates only occur when Congress is in Session)
Please note that at the end of each Congress we will post a final set of coordinates
with bootstrapped standard errors on our Common Space DW-NOMINATE download page.
The format of the legislator files is:
1. Congress Number 2. ICPSR ID Number: 5 digit code assigned by the ICPSR as corrected by Howard Rosenthal and myself. 3. State Code: 2 digit ICPSR State Code. 4. Congressional District Number (0 if Senate or President) 5. State Name 6. Party Code: 100 = Dem., 200 = Repub. (See PARTY3.DAT) 7. Name 8. 1st Dimension Coordinate 9. 2nd Dimension Coordinate 10. Log-Likelihood 11. Number of Votes 12. Number of Classification Errors 13. Geometric Mean Probability The format of the roll call files is: 1. Congress Number 2. Roll Call Number 3. Spread on 1st Dimension -- if the roll call was not scaled, there 4. Midpoint on 1st Dimension -- are 0.000's in all four fields 5. Spread on 2nd Dimension -- 6. Midpoint on 2nd Dimension --
Legislator Estimates 1st to 114th Houses and Senates (Text File, 47,042 lines, 26 June 2016)
Legislator Estimates 1st to 114th Houses and Senates (STATA 14 File, 47,042 lines, 26 June 2016)
Legislator Estimates 1st to 114th Houses and Senates (STATA 12 File, 47,042 lines, 26 June 2016)
Legislator Estimates 1st to 114th Houses and Senates (STATA 9 File, 47,042 lines, 26 June 2016)
Legislator Estimates 1st to 114th Houses and Senates (EVIEWS 9 File, 47,042 lines, 26 June 2016)
Roll Call Estimates 1st to 114th Houses and Senates (Text File, 104,302 lines, 26 June 2016)
Roll Call Estimates 1st to 114th Houses and Senates (STATA 14 File, 104,302 lines, 26 June 2016)
Roll Call Estimates 1st to 114th Houses and Senates (STATA 12 File, 104,302 lines, 26 June 2016)
Roll Call Estimates 1st to 114th Houses and Senates (STATA 9 File, 104,302 lines, 26 June 2016)
Roll Call Estimates 1st to 114th Houses and Senates (EVIEWS 9 File, 104,302 lines, 26 June 2016)
Below is STATA output showing regressions of these new coordinates onto the old coordinates for Congresses 1 - 113. All the r-squares are greater than 0.96 so that the new program is producing essentially
the same coordinates as the old program. However, note that as roll calls are added (1,496 -- 2015-16) that will slightly change
the scores for Represenatives/Senators who served in the 114th and previous Congresses.
------------------------------------------------------------------------------- . regress dwnom1new dwnom1 Source | SS df MS Number of obs = 46,506 -------------+---------------------------------- F(1, 46504) > 99999.00 Model | 6345.70352 1 6345.70352 Prob > F = 0.0000 Residual | 14.8732055 46,504 .000319826 R-squared = 0.9977 -------------+---------------------------------- Adj R-squared = 0.9977 Total | 6360.57672 46,505 .13677189 Root MSE = .01788 ------------------------------------------------------------------------------ dwnom1new | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- dwnom1 | .976272 .0002192 4454.33 0.000 .9758425 .9767016 _cons | -.0019001 .0000829 -22.91 0.000 -.0020626 -.0017375 ------------------------------------------------------------------------------ . regress dwnom2new dwnom2 Source | SS df MS Number of obs = 46,506 -------------+---------------------------------- F(1, 46504) > 99999.00 Model | 10043.9766 1 10043.9766 Prob > F = 0.0000 Residual | 252.298837 46,504 .005425315 R-squared = 0.9755 -------------+---------------------------------- Adj R-squared = 0.9755 Total | 10296.2755 46,505 .221401473 Root MSE = .07366 ------------------------------------------------------------------------------ dwnom2new | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- dwnom2 | 1.007128 .0007402 1360.63 0.000 1.005678 1.008579 _cons | .0105437 .0003416 30.86 0.000 .0098741 .0112133 ------------------------------------------------------------------------------ . regress spread1new spread1 if (vardum==1) Source | SS df MS Number of obs = 92,182 -------------+---------------------------------- F(1, 92180) > 99999.00 Model | 10377.4062 1 10377.4062 Prob > F = 0.0000 Residual | 46.1624947 92,180 .000500786 R-squared = 0.9956 -------------+---------------------------------- Adj R-squared = 0.9956 Total | 10423.5687 92,181 .113077193 Root MSE = .02238 ------------------------------------------------------------------------------ spread1new | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- spread1 | 1.022429 .0002246 4552.17 0.000 1.021988 1.022869 _cons | -.0000352 .0000738 -0.48 0.634 -.0001797 .0001094 ------------------------------------------------------------------------------ . regress mid1new mid1 if (vardum==1) Source | SS df MS Number of obs = 92,182 -------------+---------------------------------- F(1, 92180) > 99999.00 Model | 11907.1705 1 11907.1705 Prob > F = 0.0000 Residual | 116.576043 92,180 .001264657 R-squared = 0.9903 -------------+---------------------------------- Adj R-squared = 0.9903 Total | 12023.7465 92,181 .130436278 Root MSE = .03556 ------------------------------------------------------------------------------ mid1new | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- mid1 | .9840311 .0003207 3068.44 0.000 .9834025 .9846596 _cons | -.0000832 .0001171 -0.71 0.478 -.0003128 .0001464 ------------------------------------------------------------------------------ . regress spread2new spread2 if (vardum==1) Source | SS df MS Number of obs = 92,182 -------------+---------------------------------- F(1, 92180) > 99999.00 Model | 23216.6893 1 23216.6893 Prob > F = 0.0000 Residual | 951.684925 92,180 .010324202 R-squared = 0.9606 -------------+---------------------------------- Adj R-squared = 0.9606 Total | 24168.3743 92,181 .262183902 Root MSE = .10161 ------------------------------------------------------------------------------ spread2new | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- spread2 | 1.046978 .0006982 1499.59 0.000 1.045609 1.048346 _cons | -.0005859 .0003347 -1.75 0.080 -.0012419 .0000701 ------------------------------------------------------------------------------ . regress mid2new mid2 if (vardum==1) Source | SS df MS Number of obs = 92,182 -------------+---------------------------------- F(1, 92180) > 99999.00 Model | 29231.7209 1 29231.7209 Prob > F = 0.0000 Residual | 265.905226 92,180 .00288463 R-squared = 0.9910 -------------+---------------------------------- Adj R-squared = 0.9910 Total | 29497.6261 92,181 .319996811 Root MSE = .05371 ------------------------------------------------------------------------------ mid2new | Coef. Std. Err. t P>|t| [95% Conf. Interval] -------------+---------------------------------------------------------------- mid2 | .9956775 .0003128 3183.33 0.000 .9950644 .9962905 _cons | -.0001382 .0001769 -0.78 0.435 -.000485 .0002086 ------------------------------------------------------------------------------
VOTEVIEW Blog
NOMINATE Data, Roll Call Data, and Software
Course Web Pages: University of Georgia (2010 - )
Course Web Pages: UC San Diego (2004 - 2010)
University of San Diego Law School (2005)
Course Web Pages: University of Houston (2000 - 2005)
Course Web Pages: Carnegie-Mellon University (1997 - 2000)
Analyzing Spatial Models of Choice and Judgment with R
Spatial Models of Parliamentary Voting
Recent Working Papers
Analyses of Recent Politics
About This Website
K7MOA Log Books: 1960 - 2016
Bio of Keith T. Poole
Related Links