Key Design Features of the 2012 Latino Immigrant National Election Study

The Latino Immigrant National Election Study (LINES), conducted in two installments during the fall of 2012, is a nationally representative telephone survey of foreign-born adult residents of the United States who emigrated from one of the Spanish-speaking countries of Latin America. The Russell Sage Foundation, the Carnegie Corporation of New York, Purdue University, and Cornell University provided support for the study. Much of the instrumentation for LINES was adapted from the questionnaire of the 2012 American National Election Study (ANES), so that the political attitudes and behaviors of Latino immigrants can be systematically compared with other groups within the United States. Unlike the ANES, sampling for LINES was not conditional on civic status or voting eligibility.

How politically engaged are Latino immigrants vis-à-vis Latinos who were born and raised in the United States? Vis-à-vis African Americans or Anglos (whites)? Are immigrants without voting rights less inclined to take part in civic life? Are immigrants who remain involved in politics in their country of birth less likely to follow public affairs in the United States? Or does engagement in one national context complement involvement in the other? Such questions, among others, may be examined as never before through the 2012 LINES. The articles in this issue of RSF amply demonstrate the diverse scholarly literatures to which this survey contributes.

The first installment of the study was fielded at the height of the campaign season, between October 10 and November 5, 2012. In total, 853 immigrants took part in the preelection survey. Contact information for respondents was obtained from the marketing research firm Geoscape. Individuals identified on Geoscape's Hispanicity index as likely to have emigrated from Latin America were contacted at random and invited to take part in the investigation once it was confirmed that they fit the study profile. Both landline and cellular numbers were selected (AAPOR RR 4 = 0.31; Cooperation Rate = 0.93). Professional bilingual interviewers affiliated with the polling firm Latino Decisions conducted the surveys; nearly all (95 percent) were in Spanish.

Following the elections on November 6, 2012, we contacted as many immigrants as possible again for another round of interviewing. The fielding period for this installment lasted until December 20, 2012. Up to fifteen attempts were made to reach each respondent. A total of 435 participants from the preelection study took part in this second round, for a recontact rate of 51 percent. This rate is less than what is typically obtained in household panel surveys such as the ANES, but it is somewhat better than that in recent election-year telephone panel surveys of the Mexican-born population (see McCann, Cornelius, and Leal 2009; McCann and Nishikawa Chávez, forthcoming). As noted in the LINES codebook, panel attrition biases were relatively minor. Immigrants who were politically attentive were slightly more likely to take part in the second survey wave, as were those who primarily spoke Spanish at home. Age, level of formal education, family income, gender, naturalization status, the number of years spent in the United States, and frequency of church attendance were not significantly correlated with panel attrition.

While this second survey round was being administered, an additional 451 Latino immigrants were randomly selected and interviewed, so that the postelection N is comparable in size to that from the preelection wave. The full N for LINES is thus 1,304. Interviewing Services of America surveyed these fresh 451 postelection respondents, who were recruited for the study using procedures that were identical to those employed in the preelection round.

The distributions of socio-demographic variables in LINES were compared with those of Latino immigrants over eighteen in the 2011 American Community Survey (ACS). In most respects, the LINES sample conformed to the ACS, though significant discrepancies were found for education, age, and gender. A weighting variable was consequently calculated through iterative proportional fitting (that is, “raking”). When the LINES data are weighted, the distributions for educational group, age group, and gender match the ACS. Users wishing to pool LINES with the 2012 ANES may calculate additional weights as needed.

Table A1 provides a breakdown of selected social and demographic variables for LINES respondents. These variables have long been associated with orientations toward politics and participation: formal education, family income, gender, age, marital status, country of birth, years living in the United States, and civic status. For each of these items, variation is considerable. Researchers wishing to examine how socioeconomic resources, exposure to American society, or civic status, among other factors, shape democratic engagement among immigrants have much analytical leverage. At the same time, comparative benchmarks from the 2012 ANES indicate how foreign-born Latinos differ from the public at large. Most notably, Latino immigrants tend to be younger, less educated, and less affluent.

Nearly all of the telephone contact records for LINES respondents also included current street addresses. To protect anonymity, these addresses cannot be publicly archived. Users wishing to incorporate contextual geographical variables into multilevel analyses may contact the PIs for five-digit Federal Information Processing Standard (FIPS) codes, which identify the counties of residence for respondents. In the current version of the study, a number of politically relevant county and census tract-level variables have been incorporated, including age distributions, home values, education levels, percent receiving public assistance, and size of the noncitizen population. A full listing of these variables and the data sources is given in table A2. The 2012 LINES is archived for general use; citation instructions are given in the study codebook. Questions concerning the design and use of LINES data can be directed to the PIs.

• Copyright © 2016 by Russell Sage Foundation. All rights reserved. Printed in the United States of America. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the publisher. Reproduction by the United States Government in whole or in part is permitted for any purpose.