1. Note that some proponents of the attitudinal model believe solicitors general influence the Court, making the organization the only “non-attitudinal influence” on the Court (Segal and Spaeth 1993, 237–38).
2. See Lindquist and Cross 2009 for a discussion of the origins of judicial restraint and its opposite, judicial activism.
3. Baum (1997, 59) notes that judicial restraint is not a legal doctrine per se but rather a doctrine about the appropriate role of the Court vis-à-vis the other branches of government.
4. Separation-of-powers models focus on the influence of the other branches of government. There are other possible external influences on justices. Maltzman, Spriggs, and Wahlbeck (2000) and Epstein and Knight (1998) demonstrate that justices take positions that reflect their understanding of the need to build a majority coalition. Numerous scholars (e.g., Caldeira and Wright  and Kearney and Merrill ) argue that interest groups influence the justices. Claims about interest group influence are explicitly rejected by Segal and Spaeth (1993, 241). There is also a large literature on the influence of public opinion on the Court (e.g., Mishler and Sheehan 1993). For a critical critique of the claim that justices take into account the preferences of those who are not on the bench, see Brenner and Whitmeyer 2009.
INSTITUTIONS AND TIME
1. The approach for measuring ideological preferences across institutions was originally detailed in Bailey 2007. The models presented throughout this book employ a similar approach for measuring ideology. They also employ a data set that is expanded beyond Bailey 2007 to include the Court appointments of John Roberts and Samuel Alito. Preference estimates and a methodological appendix are available at http://press.princeton.edu/titles/9598.html.
2. An exception is Epstein, Martin, Segal, and Westerland 2007, which re-scales Poole-Rosenthal Common Space scores and Martin and Quinn (2002) scores. Because their approach uses both Poole-Rosenthal and Martin-Quinn scores, it inherits the concerns we discuss below about both methods.
3. Segal’s use of the approach is understandable given the fact his work appeared prior to more recent breakthroughs in ideal point estimation.
4. This approach to inter-institutional preference measurement is not endorsed by Poole, Rosenthal, Martin, or Quinn; we present it simply as an example of assuming direct comparability across institutions.