Poll Methodology Statement
Restoration PAC commissioned these surveys. Polling was conducted in partnership with Hodas & Associates of Springfield, Illinois.
The poll was conducted by telephone from a representative sample of 600 registered voters that are likely to vote in the 2020 General Elections. Of the completed interviews, roughly 70 percent came from respondents on landlines, and 30 percent came from respondents on cellphones.
SAMPLING
INTERVIEWING
MARGIN OF ERROR
SAMPLING
Surveys use a mixed mode sample design, encompassing landlines and cellphones. Both sample components are designed using a probability proportionate to size method, which means phone numbers are selected proportional to the number of registered voters by region, gender, and age in the state, based on known demographics provided by the sample provider, Aristotle, Inc.
The voter sample is scrubbed to identify landlines and cellphone / VoIP / Mobile Ported numbers.
The landline sample is randomized and all households in the target area have an equal probability of being selected. The cellphone sample is manually dialed in conformity with Telephone Consumer Protection Act (TCPA) requirements. All cellphones identified in the target area have an equal probability of being selected.
All telephone respondents are screened to ensure they are registered to vote in the target area. Interviewers also screen respondents for their likelihood to vote in the upcoming election cycle.
Quotas are applied to ensure the sample mirrors age, gender, and regional proportions in the targeted area. Because the poll mirrors the voting population, no weighting has been applied to raw data.
INTERVIEWING
All interviews are conducted by live interviewers. No automated calls or online surveys are included in the data collection.
Interviews are conducted by experienced and trained professionals from a monitored call center. Interviews are conducted through a computer-assisted telephone interviewing (CATI) system, and both human supervisors and software monitor each step of the interviewing process.
For landline calls, the predictive dialing algorithm is set to no more than a 2:1 line ratio. This is designed to ensure that each respondent is routed to a live interviewer with a less than 1% chance of encountering “dead air” when answering their phone. On average, four call attempts are dialed to each unanswered number during the period when the poll is fielded.
Cellphone respondents are screened to ensure they are in a safe place (not driving or working).
MARGIN OF ERROR
Surveys have a margin of error of +/- 3% at the 95% confidence interval. That means that in 95 out of 100 samples of the same size and type, the results obtained would vary by no more than plus or minus 3 percentage points from the result we would obtain if we could interview every member of the population. The margin of error is higher among the various sub-groups.