October 29, 2014

Latino Support for Democrats Falls, but Democratic Advantage Remains

Appendix B: National Survey of Latinos Methodology

Results for this study are based on telephone interviews conducted by SSRS, an independent research company, among a nationally representative sample of 1,520 Latino respondents ages 18 and older. It was conducted on cellular and landline telephones from September 11 through October 9, 2014.

For the full sample, a total of 682 respondents were native born (including Puerto Rico), and 838 were foreign born (excluding Puerto Rico). Of the foreign born, 359 were U.S. citizens, 215 were legal residents, and 248 were neither U.S. citizens nor legal residents. For results based on the total sample, one can say with 95% confidence that the error attributable to sampling is plus or minus 3.2 percentage points.

2014 National Survey of Latinos details

For this survey, SSRS used a staff of Spanish-speaking interviewers who, when contacting a household, were able to offer respondents the option of completing the survey in Spanish or English. A total of 826 respondents (54%) were surveyed in Spanish, and 694 respondents (46%) were interviewed in English. Any person ages 18 or older of who said they were of Latino origin or descent was eligible to complete the survey.

To ensure the highest possible coverage of the eligible population, the study employed a dual-frame landline/cellular telephone design. The sample consisted of a landline sampling frame (yielding 481 completed interviews) and a cell phone sampling frame (1,039 interviews).8 Both the landline and cell phone sampling frames used a stratified sampling design, oversampling areas with higher densities of Latino residents. Overall the study employed six strata.

For the landline sampling frame, the sample was compared with InfoUSA and Experian landline household databases, and phone numbers associated with households that included persons with known Latino surnames were subdivided into a Surname stratum. The remaining, unmatched and unlisted landline sample was used to generate a stratum with a high incidence of Latinos, based upon the share of Latinos in the sample telephone exchange.

Marketing System Group’s (MSG) GENESYS sample generation system was used to generate cell phone sample, which was divided into High and Medium strata, based upon the share of Latinos in the sample telephone area code.

Samples for the low-incidence landline and low-incidence cell strata were drawn from previously interviewed respondents in SSRS’s weekly dual-frame Excel omnibus survey. Respondents who indicated they were Latino on the omnibus survey were eligible to be re-contacted for the present survey. Altogether, a total of 315 previously-interviewed respondents were included in this sample.

It is important to note that the existence of a surname stratum does not mean the survey was a surname sample design. The sample is RDD, with the randomly selected telephone numbers divided by whether or not they were found to be associated with a Spanish surname. This was done simply to increase the number of strata and to ease administration by allowing for more effective assignment of interviewers and labor hours.

Interviews by Strata

A multi-stage weighting procedure was used to ensure an accurate representation of the national Hispanic population.

Finally, the data were put through a post-stratification sample balancing routine. The post-stratification weighting utilized estimates of the U.S. adult Hispanic population based on the 2013 U.S. Census Bureau’s Current Population Survey, March Supplement, on gender, age, education, Census region, heritage, years in the U.S., phone status (i.e., cell phone only, cell phone mostly, mixed/landline only/landline mostly) and density of the Latino population.

  1. According to calculations by the National Center for Health Statistics National Health Interview Survey (NHIS), from July to December 2013, 53% of Hispanic adults were living in wireless-only households and 17% were in wireless-mostly households (Blumberg and Luke, 2014).