Jump to ContentJump to Main Navigation
How Policy Shapes PoliticsRights, Courts, Litigation, and the Struggle Over Injury Compensation$

Jeb Barnes and Thomas F. Burke

Print publication date: 2015

Print ISBN-13: 9780199756117

Published to Oxford Scholarship Online: December 2014

DOI: 10.1093/acprof:oso/9780199756117.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 25 February 2017

(p.209) Appendix II Model of Hearing Participation

(p.209) Appendix II Model of Hearing Participation

Source:
How Policy Shapes Politics
Publisher:
Oxford University Press

There is no accepted theory of which groups are selected for hearing participation in Congress. However, a number of factors may plausibly affect interest group participation. Analytically, these variables can be divided into the following categories: (1) hearing type variables (2) committee-level variables, (3) Congress-level variables, and (4) congressional session/time. Each is discussed below.

Hearing type variables. Although there is little theory on why specific witnesses are called, it makes sense that different hearing attributes might affect the types of witnesses called. For example, not all hearings are substantively equivalent. As noted in chapter 2, we focused on three different types of hearings. We coded for hearings on specific reform proposals, which we expected to generate more participation, as one of the few regularities in politics is that people are more energetic about preserving benefits that they already have than seeking new ones. We also coded for hearings on yearly budgetary reviews and routine appropriation hearings, which we expected to generate less interest because the politics of policy change seem more likely to generate interest and conflict than the politics of policy maintenance (see generally Pierson 1994). Finally, we coded for oversight hearings in which Congress sought to gather information about problems related to court and agency practices. Common topics include the reasons for delays in processing lawsuits or claims, or the unintended policy consequences of the rising costs of claim adjudication. We controlled for these categories by coding each hearing as a dummy variable. (See Table AII.1) In the model, budgetary hearings were the omitted category.

Table AII.1 Hearing Type Variables

Variable

Measure

Referral

o if the hearing does not consider a specific bill or reform proposal concerning who pays, how much, to whom, or who decides; 1 if it does

Oversight

o if the hearing considers a specific bill or is a routine budget hearing; 1 if the hearing focus on gathering information about court or agency practices related to injury compensation policy, such as issues related to the cost of resolving claims, delays in claim processing, problems in administration

Budgetary

o if the hearing considers a specific bill or gathers information about agency or court practices; 1 if the hearing is on yearly appropriations

Committee-level variables. It also makes sense that a variety of committee-level variables would potentially affect hearings, including (a) whether the committee is a House or Senate committee, as each chamber is organized differently and represents different types of constituencies; (b) the type and specific identity of the committee, on the theory that money committees that are responsible for balancing fiscal concerns might conduct different types of hearings than policy committees and that specific committees may have their own practices; (c) the party, seniority, and ideology of the (p.210) committee chair, on the theory that liberal and conservative chairs are likely to call different types of witnesses; (d) the party and ideology of the ranking member, on similar grounds; (e) the ideological distance of the chair from key players in the process, including the ranking member, the majority leader, the median voter in the chair’s chamber, and the president, given that a chair who is ideologically alienated from the leadership and median voter of their chamber might conduct different types of hearings than those whose preferences align with those key players; and (f) the committee chair party’s risk losing its majority (see generally Binder 1997; Farhang 2008), as we might expect that committee chairs whose majority is threatened might conduct hearings differently than those who anticipate retaining their positions. Table AII.2 summarizes these variables.

Table AII.2 Committee-level Variables

Variable

Measure

House Committee

o for Senate committees; 1 for House committees

Committee Type

o for policy committees; 1 for money committees

Committee Name

Dummy variables for each committee in the sample

Chair Risk of Loss

Gains or losses as a proportion of total seats in the next election by the majority party in the relevant chamber

Chair Political Party

o for GOP chair; 1 for Democratic chair

Chair Seniority

Cumulative number of years served as chair for a specific committee

Chair Ideology

Poole and Rosenthal’s Common Space NOMINATE first dimension

Ranking Member Political Party

Common Space NOMINATE first dimension

Chair-Ranking Member Ideological Distance

Absolute value of the difference between Common Space NOMINATE first dimension scores of Chair and Ranking Member

Chair-Median Voter Distance

Absolute value of the difference between Common Space NOMINATE first dimension scores of Chair and Median Voter of relevant chamber

Chair-Chamber Leader Distance

Absolute value of the difference between Common Space NOMINATE first dimension scores of Chair and majority leader of relevant chamber

Chair-President Distance

Absolute value of the difference between Common Space NOMINATE first dimension scores of Chair and President

Note: All common space NOMINATE data were downloaded on July 25, 2011.

Congress-level Variables. Variables at the level of Congress might also potentially influence the conduct of hearings. Because this is not well theorized, we used a control for each session of Congress in our sample on the theory that this would represent the broadest possible control for variation in Congress. (It also doubled as a control for time.) In alternative models, we used more refined variables, including whether the Congress involved divided government, ideological distance between chambers, the distance between the chambers and the president, and the ratio of seats in Congress controlled by the party opposing the president. These variables are listed in Table AII.3.

Table AII.3 Congress-level Variables

Variable

Measure

Congressional Session

Dummy variable for each congressional session in the sample

Divided Government

o for undivided government; 1 for divided government

House-Senate Ideological Distance

Absolute value of the difference between CSN-1 scores of median voter of each chamber of Congress

Interbranch Conflict

Absolute value of the greatest difference between CSN-1 scores of median voter of each chamber of Congress and the President

Opposition Seat Ratio

Proportion of seats held by the party opposite the President averaged across chambers

Time Control Variables. A final issue concerns how to control for time, as the data span a period of 40 years. We used a variety of alternative measure to control for time: (a) congressional session; (b) year; (c) administration; (d) pre- and post-1994; and (e) the number of the session (1 through 40), this number squared, and then cubed Table AII.4.

Table AII.4 Time Control Variables

Variable

Measure

Congressional Session

Dummy variable for each congressional session in the sample

Year

Dummy variable for each year in the sample

Administration

Dummy variable for each presidential administration in the sample

Session Number, Session Number Squared and Cubed

Session number (1 through 40) and these numbers squared and cubed

(p.211) As noted in chapter 2, using these variables, we ran a number of alternative models using different combinations of controls. Our results with respect to the relationship between hearings of adversarial legalism and the number of group types were robust across the models and alternative specifications (see Table 2.5). (p.212)