Dissertation Chapters:

Aliens Through Space and Time: Anti-Immigrant Opinion in Longitudinal Perspective

This paper investigates anti-immigrant opinion from a longitudinal perspective. Using Stimson's (1999) "recursive dyadic dominance" algorithm, I develop a temporal measure of anti-immigrant opinion based on the common variance shared by nearly 30 years of U.S. polling data on immigration. With this measure in hand, I investigate the determinants of anti-immigrant opinion over time. The analysis yields two important findings. First, American opposition to immigration is underpinned by a common attitudinal dimension across time, such that, in the aggregate, the U.S. public makes no distinction between legal and illegal immigrants. This suggests that, as far as the macro-measurement of anti-immigrant opinion is concerned, the American public views illegal and legal immigration as one and the same. Importantly, this common attitudinal dimension appears to encompass a broad swath of the American public, which includes both native-born blacks and whites in the United States. Second, Americans' opposition to immigration responds to both immediate and long-term influences. In the short-run, opposition to immigration responds to the changing volume of news coverage about this issue, such that greater news coverage registers immediate increases in anti-immigrant opinion. In the long-run, however, Americans' position on immigration is gradually adjusted to comport with its economic outlook, such that greater economic pessimism about the future gradually increases opposition to foreigners. While underlining the relationship between increases in anti-immigrant opinion and subjective perceptions about the American polity, these findings provide new insight into the longitudinal mechanics of anti-immigrant opinion by demonstrating that it is not a whimsical, knee-jerk response, but rather, an evolving public stance which responds to immediate triggers and long-term views about the United States.



Working Papers:

Lost in Translation? Validity & Reliability in Bilingual Political Surveys

Are measurement instruments comparable across linguistic groups? As the largest minority in the U.S., Latinos are the focus of increasing attention from survey researchers interested in understanding their political behavior. Scholars have relied on a growing repository of survey data to measure a wide array of constructs among Latinos, including political identity, political sophistication, policy attitudes, and political values (e.g., Sanchez 2006; Nicholson et al. 2006; de la Garza et al. 1996; de la Garza 1992). However, because these constructs and their attendant measures were primarily developed and tested on English-speaking populations, it is reasonable to expect some discrepancies in the performance of these items among a bilingual population like Latinos. In particular, these constructs may not accurately capture political concepts as understood by Latinos (validity), just as the measures we use to gauge these concepts among this population may be of questionable quality (reliability) (Little et al. 1999; Millsap 1998; Reise et al. 1993; Byrne et al. 1989). This concern for validity and reliability is not insignificant. Lee (2001), for instance, has uncovered strong evidence of language-of-interview effects among Latino respondents in one of our discipline's primary sources of attitudinal data on this population. The persistence of these effects strongly suggests the need to take stock of our efforts to measure political phenomena in a multi-lingual, multi-ethnic context. To that end, I develop a series of multi-group measurement models that test the invariance of several political constructs across 1) English-speaking Latinos and 2) Spanish-speaking Latinos. Drawing on commonly used polling data, such as the 1989-1990 Latino National Political Survey, the 1999 Harvard/Kaiser/Washington Post Latino Survey, and the 2004 National Survey of Latinos, this investigation identifies a moderately weak level of validity and reliability in bilingual survey items. The paper concludes by detailing several strategies to enhance the performance of these items in applied analyses.

*This paper was presented at the 2007 Annual Meeting of the Midwest Political Science Association.


Framing Effects & Attitudes Toward Assimilation

In the annals of immigration, the question of assimilation is one of the most fundamental and controversial (Ngai 1999; Higham 1981). Yet insofar as opinion toward assimilation is concerned, scholars have generally examined it with respect to immigrants as a whole. This paper, in contrast, takes a group-centered approach. By many accounts, research demonstrates that Americans are strongly in favor of having immigrants assimilate to American culture writ large (e.g., Fetzer 2000; Fuchs 1992; Huddy and Sears 1995; Citrin et al. 1990; Higham 1981). We contend, however, that attitudes toward assimilation are conditional and highly susceptible to framing effects. That is, these attitudes are influenced by how the issue is formulated and which immigrant group is referenced in the presentation of the issue. Framing effects occur when different but logically equivalent words or phrases lead individuals to modify their decisions (e.g., Druckman 2002; Quattrone 1988; Tversky and Kahneman 1981; Kahneman and Tversky 1979). Extant research demonstrates that framing effects work by highlighting and attaching different weights to specific attitudes during the decision-making process (e.g., Nelson et al. 1997; Nelson et al. 1999). In particular, attitudes toward groups have been shown to be highly susceptible to framing effects, and very influential in decreasing support for policies associated with specific groups (e.g., affirmative action and blacks) (Nelson and Kinder 1996). Such group-centrism is considered to stem from a frame's emphasis on a group which is held in low esteem by the public. This paper extends this reasoning to investigate whether attitudes toward assimilation are entangled with negative evaluations of Latinos. Using the 1999 Harvard/Kaiser/Washington/Post Latino Survey, we test this proposition by: 1) varying the direction of assimilation (i.e., changing versus maintaining a group's culture); and 2) changing the immigrant group involved in assimilation.

*This paper was presented at the 2006 Annual Meeting of the Western Political Science Association.