280000 bills proposed in the US House and Senate from 1973 to 2004 93rd108th Congresses recorded in Thomas over 21 million cosponsorship signatures partitioned by chamber and Congress to create 32 separate cosponsorship networks ID: 330298
Download Presentation The PPT/PDF document "Cosponsorship Data" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Cosponsorship Data
280,000
“
bills
”
proposed in the U.S. House and Senate from 1973 to 2004 (93rd-108th Congresses) recorded in Thomas
over 2.1 million cosponsorship signatures
partitioned by chamber and Congress to create 32 separate cosponsorship networks
http://jhfowler.ucsd.edu/cosponsorship.htm
Slide2
Cosponsorship in the HouseSlide3
Cosponsorship in the SenateSlide4
Mutual Cosponsorship RelationsSlide5
Connectedness: An Alternative Measure
Traditional measures of centrality generate plausible names
None takes advantage of information about the
strength
of social relationships
Total number of cosponsors on each bill
Legislators recruit first those legislators to whom they are most closely connected.
More cosponsors = lower probability of direct connection
Bills with fewer total cosponsors more reliable
Strength of the connection between
i and j = 1/cijTotal number of bills sponsored by j and cosponsored by iMore bills in common = stronger relationshipWeighted cosponsorship distance Slide6
Weighted cosponsorship distanceSlide7
Legislative connectedness
Suppose
direct
distance from legislator
j
to legislator
i
is simple inverse of the cosponsorship weight
Then use Dijkstra’s algorithm (Cormen et al. 2001)
Starting with legislator
j, identify from a list of all other legislators the closest legislator iReplace each of the distances with Remove legislator i from the list and repeat until there are no more legislators on the list. Connectedness is the inverse of the average of these distances from all other legislators to legislator j.Slide8
Results for the HouseSlide9
Results for the SenateSlide10
Quality of Strongest Weighted Relationships
Institutional Ties
House committee chairs and ranking members
Senate majority and minority leaders
Regional Ties
From the same state
In the House they are often from contiguous districts
Issue Ties
Rep. Jim DeMint and Sue Myrick -- Republican Study Committee
Sen. George Mitchell and Jim Sasser -- Federal Housing Reform
Sen. Kay Bailey Hutchinson and Sam Brownback -- marriage penalty relief and bankruptcy reformPersonal TiesSenator John McCain chaired Senator Phil Gramm’s 1996 Presidential campaignMcCain has told the media that they have been friends since 1982 when they served together in the House (McGrory 1995)Slide11
108th House Top 20Slide12
108th Senate Top 20Slide13
External Validity: Legislative Influence
Widely used measure of legislative influence is number of successful floor amendments
Hall 1992; Sinclair 1989; Smith 1989; Weingast 1991
1 SD increase in connectedness increases successful floor amendments
53% in House
65% in SenateSlide14
External Validity: Roll Call Votes
Model roll call votes as in Poole and Rosenthal, adding connectedness score of sponsor
1 SD increase in connectedness of sponsor increases votes for bill by
5.2 in House
8.2 in Senate
2 SD increase would change 16% of House votes and 20% of Senate votesSlide15
Landmark Legislation:
An Alternative to MayhewSlide16Slide17
Modularity
(Newman and Girvan 2004)
Define modularity to be
Q = (number of edges within groups) – (expected number within groups).
Actual Number of Edges between
i
and j is
Expected Number of Edges between
i
and j isSlide18
Modularity Matrix
So Q is a sum of
over pairs (i, j) that are in the same group
Or we can write in matrix form as
Where B is a new characteristic matrix, the
modularity marix
,
(s
i
, s
j) Slide19
Modularity Matrix
Algorithm
Calculate the leading eigenvector of the modularity matrix
Divide the vertices according to the signs of the elements
Note that there is no need to forbid the solution with all the vertices in a single group.Slide20
ExampleSlide21
Example Applications
Books on politics
The vertices represent 105 recent books sold from Amazon.com
Divide the books according to their political alignment
Liberal / Conservative / Centrist Slide22
ExampleSlide23
Polarization in the 108th SenateSlide24
108th House by PartySlide25
108th House by IdeologySlide26
108th House by StateSlide27
108th House by CommitteeSlide28
Polarization Over Time in the HouseSlide29
Polarization Over Time in the SenateSlide30
Black Legislators in the 103rd HouseSlide31
Black Legislators in the 104th HouseSlide32
Black Legislators in the 108th HouseSlide33
Poor Districts in the 108th House