This poll was conducted over a two-week period (10/15 – 10/28), one week prior to the start of early voting in Texas and during the first full-week of early voting.
The project serves as the first statewide poll for The University of Texas at Tyler. The poll also represents the only statewide university poll in Texas that contacts residents by phone. The blend of phone and internet surveys reflects the strongest methodology and creates an exceptional opportunity for student research and community engagement, two of the four pillars of The University of Texas at Tyler’s new strategic plan. The College of Arts and Sciences, Department of Political Science and History, and the Department of Social Sciences were essential in establishing this new polling center to expand opportunities for civic engagement.
The UT Tyler-Texas Opinion Survey poll was conducted by telephone using live callers October 15 to 28, 2018 with a scientifically selected random target sample (provided by the firm Survey Sampling International) of 6000 Texas registered voters, 18 or older. During a two-week period, forty-students called these registered voters on landlines and cell phones. The overall response rate was 8%, with a cooperation rate of 44%. This allowed us to generate a sample of 400 registered voters by phone.
The UT Tyler-Texas Opinion Survey poll was also conducted using a Survey Sampling International panel of registered voters that opt-in to take surveys. This is known as Aristotle. The online panel generated a sample of 633 registered voters.
The data were weighted to be representative of Texas adults. The weighting balanced sample demographics to population parameters. The sample is balanced to match parameters for gender, age, race/ethnicity, and education using through an iterated process known as raking. These parameters were derived from 2016 Current Population Survey Voting and Registration Tables. The use of these weights in statistical analysis ensures that the characteristics of the sample closely reflect the characteristics of registered voters in Texas.
In this poll, the sampling error for 1033 registered voters in Texas is +/- 3.03 percentage points at a 95 percent confidence interval.