Publication Type: Conference Paper/Unpublished Manuscript
Abstract: Standard models of collegial (multi-member) courts ignore the most distinctive and characteristic features of judicial policy-making: legal cases and legal rules. Meanwhile, traditional legal perspectives ignore collegiality and the problems inherent in majoritarian choice. In short, we lack a theory of the collegial politics of legal doctrine. I present the "Case-Space Model," which puts cases and rules at the analytic center. I study how preferences that take the form of legal rules can be aggregated. The main result is the Median Rule Theorem, demonstrating the existence and stability of an implicit median rule, even where there is no median judge. I discuss implications for the development of law (in terms of the authenticity, coherence, and consistency of collective choice), for recurring debates between law and politics, and for the study of collegial courts. I also sketch supplemental applications of the case-space model to the separation-of-powers game, and to the choice between rules and standards as instruments of judicial policy-making.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: ABSTRACT. Comparative policy research
has created problems for traditional justifications of democracy as
being uniquely responsive to popular opinion. There seems little party
convergence on majority opinion in practice. Few governments rest on
spontaneous popular majorities.
‘Consensus democracy’ relies on party elites making policy by excluding
popular passions. Can a necessary connection between public opinion and
public policy only be made by direct policy voting under direct
democracy? We suggest an alternative mechanism – a ‘median mandate’ –
and check evidence for its operation in 16 nations from 1972-91 in the
important area of central government size, a key component of
left-right differences.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: This paper tests hypotheses derived from the debate between proponents of the conditional party government theory (Aldrich 1995; Aldrich and Rohde 2000; Rohde 1991) and those who argue that a preference based model accounts for outcomes as well as the more complex conditional party government theory (Krehbiel 1998). Analyzing data from the 87th (1961-62) through the 107th (2001-02) Congresses, we find that: (1) the median House member is significantly more likely to vote with the majority party median than with the minority party median; (2) the significant increase in the probability of joining a majority party coalition is independent of the ideological distance between the chamber median and the party medians; and (3) the asymmetric pull toward the majority party increases when the conditions of the conditional party government theory are met.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: Public interest in aggregate marital status data continues to exist despite the fact that in 1989, the National Center for Health Statistics (NCHS) ended its publication of annual marriage and divorce reports which provided detailed marital status characteristics based on marriage certificates. Demographers, the public, and policy makers relied on NCHS data, and need a replacement.
The U.S. Census Bureau has made up for some of the lack by reporting marriage data based on the Current Population Survey (CPS). The CPS is able to closely replicate NCHS data on median age at first marriage at the national level and has been reporting these results for years, using an indirect method of producing the estimated median age at first marriage based on the proportion of people who were never married for 5-year age groups ranging from 15 to 54. However, the CPS does not have the sample size to be able to report data for geographies below the national level (which used to be reported by NCHS) or to report race and ethnic differences in marriage patterns.
This poster will suggest how median age at first marriage can be estimated in the future, using the American Community Survey (ACS). The results are based on a comparison of NCHS, CPS, Census 2000 and ACS marital status data. We will show that the ACS provides the best continuous data available at the state level as well as greater detail for characteristics such as race and Hispanic origin groups.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: Are legislators' party affiliations or are constituency preferences the greatest predictor of legislators' positions in support of the president? Spatial theories imply that a legislator will locate at the district median. However, empirical evidence examining roll-call support of the president's agenda in Congress indicates the opposite, showing that party and ideology cause a legislator to deviate from the median (and that constituency preferences have a muted effect on presidential support). We explain this paradox by showing that legislators' platforms, upon which spatial election theories are based, are different from legislators' votes on roll-calls. Our findings show that constituency preferences are the only significant predictor of legislative support for the president when analyzing legislators' platforms, and that party has no effect. When analyzing roll-call votes, our analysis suggests that party and ideology are the key predictors. We analyze legislators' platforms with a unique dataset of mass mailings sent by members of Congress. The results are in contrast to recent empirical work testing the spatial model in the electoral context, and suggest that party affects legislator behavior in the legislature but not in the electorate.