Search TESS:


Local and National Influences on Inter-group Attitudes

Download data and study materials


Principal Investigator(s):

Daniel Hopkins
Georgetown University
Home Page:

Van C. Tran
Columbia University
Home Page:

Abigail Fisher Williamson
Trinity College
Home Page:

Sample size: 566
Field period: 2/23/2008 - 2/28/2008


Are Americans? responses to immigration cues similar to their responses to racial cues? This paper explores how implicit and explicit cues influence immigration attitudes. To explain explicit cues? varying impacts, it reconceptualizes the implicit/explicit distinction as continuous rather than binary. The paper hypothesizes that the effect of cues will be moderated by personal experience with immigrants. It tests these claims using four survey experiments with more than 2,500 respondents. Seeing Spanish text acts as a potent implicit cue. Even brief exposures to written Spanish can undermine support for immigration among key subgroups, including Democratic voters and those who hear Spanish frequently. Yet explicit anti-immigration appeals backfire, particularly among those who frequently interact with Latino immigrants. Thus the distinction between implicit and explicit cues proves critical in understanding attitudes on immigration. Contact with immigrants can increase the effect of cues, suggesting that they are most influential when they resonate with personal experience.


* Framing effects on immigration would be more potent for respondents who live near many immigrants.

* Concern about immigrants would be higher when local issues were primed, so long as respondents lived near large immigrant populations.

* Information about the share of immigrants locally would dampen concern about immigration for those living in areas with few immigrants.

Experimental Manipulation:

The TESS module included several experimental manipulations. One exposed respondents to a welcome note in Spanish, to identify if brief uses of foreign languages primed anti-immigrant sentiment. Another exposed respondents to varying pro- and anti-immigration frames, to see under what conditions we might observe social desirability or framing effects in response patterns. Still another experiment varied whether immigrants were described as legal, illegal, or neither, to determine if the word "immigrants" itself elicits responses that are closer to those for explicitly legal or explicitly illegal immigrants. We also examined whether providing information about the number of immigrants in one's local area influenced respondent attitudes.

Key Dependent Variables:

The outcome variables constitute a wide variety of measures of immigration-related attitudes and threat. From Sniderman et al. (2004), we drew one generic measure of threat. We also replicated the National Election Study question about the number of foreigners who should be permitted to come to the U.S. to live, as well as other questions from the General Social Survey tapping concerns about immigrants raising taxes, increasing crime, and taking jobs.

Summary of Findings:

Two findings stand out as especially noteworthy, and both have spawned follow-up research. First, we learned that respondents who commonly encounter Spanish in their day-to-day lives have negative reactions to it, whereas those who do not commonly see Spanish in their day-to-day lives have positive or neutral reactions to it. Second, we identified strong social desirability effects, or other contrast effects. Respondents expressed lower levels of threat when exposed to anti-immigration arguments.


Hopkins, Daniel J., Van C. Tran, and Abigail Fisher Williamson. 2014. "See No Spanish: Language, local context, and attitudes toward immigration." Politics, Groups, and Identities.

Copyright © 2014, TESS