The Fox News Poll is directed by Anderson Robbins Research (D) and Shaw & Company Research (R). The poll is conducted via telephone among a nationally representative sample of approximately 900 registered voters. Of the completed interviews, roughly 650 are conducted with respondents on landlines and 250 with respondents on cell phones.
The Fox News Poll uses a dual-frame sample design covering landline and cell phones. Both the sample components are designed using a probability proportionate to size method, which means that phone numbers for each state are proportional to the number of voters in each state.
The landline sample is developed using random digit dial procedures (RDD), in which all landline numbers (listed and unlisted) in the continental United States have an equal probability of being selected.
The cell phone sample is also developed using a probability selection method which involves: (1) identifying blocks of telephone numbers that are exclusively cell phones; and (2) randomly generating the last two-digits and then dialing these numbers within the block.
In addition, all respondents are screened to ensure he/she is registered to vote. In states without registration or same-day registration, respondents are asked about the regularity of their voting behavior. In the months prior to elections, interviewers sometimes screen for “likely voters” through a series of additional questions about past voting behavior and interest in voting in the current election.
Quotas are applied to ensure the sample mirrors the proportions of voters nationally. Specifically, the aim is for a gender split nationwide of 53 percent female and 47 percent male, as well as regional quotas.
All interviews for the Fox News Poll are conducted by live interviewers. Furthermore, all interviews are conducted by trained professionals from a central, monitored location. Interviews are conducted through a computer-assisted telephone interviewing (CATI) system, and both human supervisors and software monitor each step of the interviewing process.
All telephone numbers are dialed by hand so prospective respondents always find a live interviewer when they answer their phone. On average, six calls are placed to each unanswered number during the three day period when the poll is fielded.
Cell phone respondents are screened to ensure they are not driving or participating in an activity that requires their full attention. Cell phone respondents are also offered a $3 reimbursement payment to cover the cost of the minutes they use for the telephone call.
When necessary, minor weights are used to ensure the demographics of survey respondents are representative of the national registered voter population (as measured by the national exit poll). Generally, weights are only applied to age and race variables. The Fox News Poll is not weighted by political party.
For a sample of about 900 interviews, the error due to sampling is plus or minus 3.3 percentage points at the 95% confidence level. For example, when the survey says “47 percent of voters . . .” then the probability is very high that no less than 44 percent and no more than 50 percent of all voters would have responded the same way. Sampling error associated with subgroup results is higher.
In addition to sampling error, question wording and question order can influence poll results.