Applied Statistics Workshop (Gov 3009)


Wednesday, December 4, 2013, 12:00pm to 1:30pm


Room K354, CGIS Knafel. 1737 Cambridge Street, Cambridge MA 02138

This methodological workshop offers a tour of Harvard's statistical innovations and applications with weekly stops in different disciplines. It presents opportunities for faculty, visitors, and graduate students to present ongoing research. For additional details, please click here

Graduate Student Coordinator: Tess Wise

This lecture meets on the 3rd floor of CGIS Knafel. Lunch will be served. 

Presentation Details

Presenter: Jason Anastasopoulos

Title: Not in My Backyard: Understanding Local Opposition to Undocumented Immigration Using a National Survey Experiment

Abstract: Understanding what motivates animus toward undocumented immigrants presents several empirical challenges.  Estimates of undocumented immigrants at the state and metropolitan area level are unreliable, rendering studies which base conclusions on them questionable at best. Furthermore, because undocumented immigrants mostly come from Latin American countries, assessing the role that undocumented immigrant characteristics such as race or ethnic background play in shaping policy attitudes related to undocumented immigration is practically impossible using observational data.  To overcome these challenges, I designed a survey experiment in which undocumented immigrant skin tone and perceived proximity to the respondent are manipulated using images and respondent Internet Protocol address, respectively. I find that perceived undocumented immigrant proximity polarizes responses on several immigration policy questions while the interaction between dark skin tone and close proximity causes increases in levels of support for state laws which seek to identify and detain undocumented immigrants. These findings have two important implications. First, they support the claim that surveys and opinion polls which ask broad questions about immigration policy will tend elicit ``top of the head'' responses rather than true preferences since responses have no real life consequences (Bertand and Mullainathan 2001; Taylor and Fiske 1978). Second, they suggest that racial threat plays an important role in motivating opposition to undocumented immigration.