W-NOMINATE Files With Parametric Bootstrapped Standard Errors
by
Jeff Lewis and Keith Poole
(Updated 24 October 2006)
1 1059990999 0USA 10000CLINTON -0.8407 0.5414 -0.7995 0.5040 0.0928 0.1560 0.3781 2 1054970041 0ALABAMA 20001SESSIONS 0.8796 -0.3991 0.9206 -0.3642 0.0524 0.0829 0.7469 3 1059465941 0ALABAMA 20001SHELBY 0.6532 -0.0546 0.7302 -0.0772 0.0892 0.1210 0.0299 4 1051490781 0ALASKA 20001MURKOWSKI 0.5494 0.3198 0.6009 0.3218 0.0696 0.1322 -0.0433 5 1051210981 0ALASKA 20001STEVENS 0.3120 0.6300 0.3827 0.6287 0.0832 0.1257 0.0180 etc etc etc The first column is an integer counter The column in red is the Congress number The column in blue is the ICPSR ID Number The column in Magenta is the ICPSR State Code The column of zeroes will show the two digit Congressional District Number for House Roll Call Matrices The column in Coral is the first seven letters of the State Name The column in Midnightblue is the party code -- see the Party Codes Page for a complete listing The column in Firebrick are the ICPSR Occupancy and Election Codes Name of Senator/Representative The columns in Turquoise are the W-NOMINATE Coordinates. The columns in Purple are the Mean Coordinate Values For the 1000 Trials The columns in Green are the Parametric Bootstrapped Standard Errors The last column in Deeppink is the Pearson Correlation Between the First and Second Dimension Coordinate estimates across the 1000 Bootstrap Trials This can be used to obtain the corresponding covariance term etc etc etc 97 1051492256 0WEST VI 10001ROCKEFELLER -0.7299 0.6440 -0.7095 0.5983 0.0544 0.1049 -0.0280 98 1054930925 0WISCONS 10001FEINGOLD -0.7723 -0.6353 -0.7796 -0.6066 0.0465 0.0685 -0.7627 99 1051570325 0WISCONS 10001KOHL -0.6280 -0.3566 -0.6327 -0.4450 0.0492 0.1445 -0.0130 100 1054970668 0WYOMING 20001ENZI 0.7526 -0.1695 0.8195 -0.2170 0.0805 0.1246 0.0087 101 1051563368 0WYOMING 20001THOMAS 0.6881 -0.0141 0.7503 -0.0363 0.0758 0.1184 0.0068The file containing the 1000 bootstrapped sets of coordinates for the 105th Senate looks like this:
11059990999 0USA 10000CLINTON 42 6 1 30 0.831 -0.841 0.541 0.117 0.910 21054970041 0ALABAMA 20001SESSIONS 278 20 13 169 0.859 0.880 -0.399 0.051 0.137 31059465941 0ALABAMA 20001SHELBY 293 30 20 141 0.749 0.653 -0.055 0.036 0.112 41051490781 0ALASKA 20001MURKOWSKI 296 20 13 139 0.835 0.549 0.320 0.042 0.130 51051210981 0ALASKA 20001STEVENS 313 17 35 119 0.780 0.312 0.630 0.041 0.120 etc etc etc The fields through the name of the Senator/Representative are the same as above The cross classification is in Deeppink. The second and third columns are classification errors -- predicted yes voted no and predicted no voted yes. So for Clinton correct classification = (42+30)/(42+30+6+1) = 91.1% The column in Black is the Geometric Mean Probability The two columns in Turquoise are the Legislator coordinates The last two columns in Green are the Conditional Standard Errors etc etc etcThe format of this file is exactly the same as that for NOM31.DAT and it is discussed in detail on the W-NOMINATE Page. This file has 1001 sets of coordinates. The first set is identical to the ones contained in the file with the Standard Errors and the following 1000 sets are the 1000 parametric bootstrap trials.