About this poll

This survey was conducted for the Los Angeles Times and the University of Southern California College of Letters, Arts and Sciences by Greenberg Quinlan Rosner in conjunction with American Viewpoint, both based in Washington, DC.

The findings are based on a random-sample survey of 1,689 registered voters in the state of California conducted from November 3-14, 2010. A subset of 1,457 respondents were voters in the 2010 election.

Interviews were conducted by telephone using live interviewers from Interviewing Services of America. Voters were randomly selected from a list of registered voters statewide and reached on landline or cellphone depending on the number they designated on their voter registration. The percentage of cellphone respondents in this study matches the percentage of those who list their cellphone on the voter file. Up to five attempts were made to reach and interview each randomly selected voter. In order to include a wider range of questions in this study, some batteries of related questions were split into random half-samples, with one-half of 844 registered voters and the other half of 845 registered voters.

A separate oversample of 350 Latino registered voters were interviewed by telephone by Latino Decisions, a joint effort between Pacific Market Research and professors Dr. Gary Segura from Stanford University and Dr. Matt Barreto of the University of Washington. An additional 70 interviews were conducted with respondents who identified as Latino in the base sample for a total of 420 interviews with Latinos. Funding for the Latino oversample was provided by The California Community Foundation.

All interviews from the Latino sample were carried out by bilingual Latino interviewers, and conducted in the preferred language of the survey respondent, English or Spanish. Overall, 37% of interviews with the Latino sample were conducted in Spanish and 63% in English. The technique of using fully bilingual interviewers was used because it does not require terminating calls with Spanish-language households and arranging for a callback, which can be difficult to schedule with language barriers. As a result, the LAT-USC oversample of Latinos has higher response and cooperation rates than other surveys which limit the use of bilingual interviewers.

A separate oversample of 338 Asian Americans was included in the survey. Respondents in the Asian sample were offered the option to take the survey in Mandarin, Cantonese, Korean, Vietnamese or Tagalog. Nearly 50% of interviews were conducted in an Asian language. An additional 64 interviews were conducted with respondents who identified as Asian in the base sample for a total of 402 interviews with Asian Americans. Funding for the Asian American oversample and methodological guidance was provided in a grant to USC by the Asian Pacific American Legal Center.

Upon completion of all interviewing, the results were weighted to bring the Latino and Asian oversample populations into line with the racial and ethnic composition of registered voters in California. The data were weighted to reflect the total population of registered voters throughout the state, balancing on regional and demographic characteristics for gender, age, race and education according to known census estimates and voter file projections. The vote-choice questions for the governor and Senate races were weighted to the actual results according to the California secretary of state.

The maximum sampling error for the overall sample of 1,689 registered voters is +/- 2.4 percentage points at the 95% confidence level. The margin of error for the 1,442 2010 voters is +/- 2.6 percentage points. The margin of error for the 420 Latino respondents is +/- 4.8 percentage points. The margin of error for the 402 Asian American respondents is +/- 4.9 percentage points.