Ideal Points and American Political Development: Beyond DW-NOMINATE

Abstract

This article aims to persuade historically oriented political scientists that ideal point techniques such as DW-NOMINATE can illuminate much about politics and lawmaking and be very useful to better understanding some of the key questions put forward by American political development (APD) scholars. We believe that there are many lines of inquiry of interest to APD scholars where ideal point measure could be useful, but which have been effectively foreclosed because of the assumptions undergirding DW-NOMINATE. In particular, we focus on three issues as particularly important: (1) the assumption of linear change; (2) the collapsing of distinct policy issue areas into a single “ideology” score; and (3) an agnosticism toward policy development, institutional context, and historical periodization. We go over these issues in detail and propose that many of these concerns can be addressed by taking seriously the proposition that policy substance, historical and political context, and the temporal dimension of political processes be integrated into the core of our measures and analyses. We also discuss a set of techniques for addressing these issues in order to answer specific questions of broad interest to both APD scholars and other Americanists.

Do 40-Year-Old Facts Still Matter? Long-Run Effects of Federal Oversight under the Voting Rights Act

“…Table B6 of the online Appendix shows that treatment effects are largely invariant to alternative area-time 37 While the second dimension of DW-NOMINATE historically tracked policy issues that cut across party lines, such as bimetallism and slavery, congressional voting since the 1960s has been virtually unidimensional (McCarty, Poole, and Rosenthal 1997). As Bateman and Lapinski (2016) notes, "this measure bears little relationship to patterns of voting on civil rights," and during the 1970s, second-dimension scores of southern Republicans were actually higher than those of northern Democrats but lower than those of southern Democrats. Given the difficulty of interpreting these scores, results from their analysis, which reveal no clear or consistent treatment effect, are left to the online Appendix.…”

Anatomy of Party Sorting: Partisan Polarization of Voters and Party Switching

“…However, after the passages of the 1964 Civil Rights Act and the 1965 Voting Rights Act, the second dimension slowly declined in importance and has been almost absent since the early 1980s. Analysts contend that the measurement of DW Score could be biased due to the differential agendas and smoothing method (Bateman and Lapinski ; Caughey and Schickler ; Clinton and Meirowitz ). Nonetheless, the first dimension scores alone explain more than 80 percent of the voting decisions in all the Congressional sessions since the mid‐nineteenth century (Poole ; Poole and Rosenthal ).…”

“…Analysts contend that the measurement of DW Score could be biased due to the differential agendas and smoothing method (Bateman and Lapinski 2016;Caughey and Schickler 2016;Clinton and Meirowitz 2001). Nonetheless, the first dimension scores alone explain more than 80 percent of the voting decisions in all the Congressional sessions since the mid-nineteenth century (Poole 2005;Poole and Rosenthal 1997).…”

Estimating Ideal Points from Roll-Call Data: Explore Principal Components Analysis, Especially for More Than One Dimension?

“…For further comments on this issue, see (e.g.) Bailey (2007), Bateman and Lapinski (2016), Cillizza (2014), and Sides (2011).…”

“…For other works that provide differing views about P-R or how it compares with CJR, see (e.g.) Krehbiel and Peskowitz (2015); Caughey and Schickler (2016); Bateman and Lapinski (2016) and McCarty (2016). PCA is based on the eigenvalues and eigenvectors of S. Let L k denote the k-th largest eigenvalue (k = 1, 2, ...) and g k (J × 1) the corresponding eigenvector.…”