Survey Methods

The Brandman University State of Orange County Survey was co-directed by Fred Smoller, Ph.D., Michael Moodian, Ed.D., Gregory Robinson, Ph.D. and Sarah Catz, J.D.  The random digit dial (RDD) telephone survey was administered to 675 Orange County adult residents between April 15 and April 27, 2010.  Random digit dialing begins with the random selection of “working blocks” allotted to residential and cell telephones, including the area code, three-digit prefix and the first two digits of the final four-digit sequence.  Appending two digits generated at random by computer to the random selection of working blocks results in listed and unlisted telephone numbers, those established for many years, and numbers connected as recently as three months ago.

Interviewing was conducted on weekend days and weekday nights. Each interview included 90 questions and required an average of 19 minutes and 20 seconds to complete. Interviews were administered in English and Spanish, as needed.  Field work was conducted by WestGroup Research of Phoenix, Arizona.
 
Calculated conservatively, the sampling error for this survey is +/- 3.85% at the 95% confidence level. This means that 95 times out of 100, the results reported will fall within a range 3.85 percentage points above and below what they would be if an adult in every Orange County household with a telephone was interviewed. The sampling error for any subgroup of the survey sample is larger. Sampling error is just one type of error that may affect survey results. Results may also be affected by question wording, ordering and survey timing.

The sample is compared in the tables below to estimates of the Orange County population produced in 2007 by the State of California, Department of Finance, in a table titled, “Race/Ethnic Population with Age and Sex Detail, 2000–2050,” and to a report generated by the County of Orange Registrar of Voters in April 2010.  Population statistics in the first two tables below include only adults 18 years of age and older, since that is the population of inference.  Persons younger than 18 years of age were not interviewed. 

Data was collected to over-represent high-propensity voters*.  As a consequence, Whites are over-represented and other key groups are under-represented in the survey sample.  Similarly, younger adults are under-represented, the proportion of adults 35 to 54 years of age is equivalent in the sample and in the County population, and adults 55 and older are over-represented in the survey sample.

The Orange County population 18 years of age and older is 50.9% female, compared to 55% in the survey sample, and 49.1% male, compared to 45% in the survey sample.

Figure 2 illustrates the fact that Republicans are slightly over-represented, and persons registered as “Decline to state” combined with those of other party affiliations are slightly under-represented in the survey sample.

^*See the Public Policy Institute of California Report (September, 2009) “Just the Facts: California’s Likely Voters” at http://www.ppic.org/content/pubs/jtf/JTF_LikelyVotersJTF.pdf