Chattanooga Times Free Press
Gardenhire questions poll’s use of ‘illegal immigrant’
NASHVILLE — A poll commissioned by The Tennessee Star, a conservative political news website, says 84 percent of the 1,007 “likely” Republican voters surveyed oppose state legislation “providing in-state college tuition to illegal immigrant students.”
The sponsor of that legislation, Sen. Todd Gardenhire, R-Chattanooga, says he doesn’t find that surprising, given the phrasing.
“The question was worded in such a way to get the result that they wanted,” said Gardenhire, who passed the bill two years ago in the Senate only to see it fail in the House by a single vote. Both chambers are solidly in GOP control.
The bill failed this year in a House panel on a tie vote.
Last month, Vanderbilt University released its semi-annual poll that included a similar question. But pollsters used a different word, describing the students as “undocumented.”
The Vanderbilt poll of 1,004 registered voters including Republicans, Democrats and independents.
“Now thinking about children of
undocumented immigrants who are brought to this country when they are young,” the Vanderbilt poll question began. “If these children attend Tennessee public schools, graduate from a Tennessee high school and are accepted at one of Tennessee’s public colleges and universities, do you think they should be eligible for the in-state tuition rate, or shouldn’t they?”
Sixty-six percent of respondents said yes, including 55 percent of self-described Republicans.
In the Tennessee Star poll, an automated landline telephone survey conducted by Triton Communications, the question was: “In 2018, the Tennessee General Assembly is expected to reconsider a bill to provide in-state college tuition to illegal immigrant students. Do you support or oppose providing in-state college tuition to illegal immigrant students?”
Gardenhire said that’s misleading. He said “we’re not giving them anything” because the students live in Tennessee and would pay the same tuition as other students who live here.
In an interview last week, Steve Gill, a former conservative radio talk show host with ties to The Tennessee Star, defended the question’s wording. That’s how it will play out in a Republican primary election, said Gill, who runs an independent political expenditure group.
Gill added he knows that firsthand, having used the issue in a direct mail attack last year against Sen. Brian Kelsey, R-Germantown, who voted for Gardenhire’s bill in 2015. Kelsey was running in Tennessee’s 8th Congressional District Republican primary.
“That issue destroyed him,” said Gill, whose mailer said Kelsey voted for a bill “Requiring Tennessee Taxpayers Pay Some Of The Costs Of College Tuition For Illegal Aliens.”
The margin of error in The Tennessee Star poll was plus or minus 3.1 percent. The Vanderbilt poll’s margin of error was plus or minus 3.3 percent overall, with larger margins in subcategories such as the GOP voter breakouts.
Vanderbilt uses live interviewers to do its telephone polling and apportions calls to both landline and cellphone users. Triton uses automated polling, which under federal law excludes cellphones for a variety of reasons.
Dr. John Geer, a Vanderbilt political science professor and co-director of the university’s poll, questioned the Triton poll’s accuracy, saying, “Automated polls are simply not reliable. … I would take little stock in the validity of the result.”
Moreover, Geer said, younger voters tend to use cellphones, “and so really big problems crop up with that group.”
After the Vanderbilt poll was released, The Tennessee Star had a story in which various Republicans criticized its findings.
Gardenhire agreed with Geer’s point about automated surveys. He noted he has been involved in various polls since the 1971 Chattanooga fire and police commissioner race between Gene Roberts, whom Gardenhire backed, and James E. “Bookie” Turner, in which live-interviewer polling correctly predicted Roberts’ victory.
The senator said he has for years used a Louisiana polling firm, which serves both Republican and Democratic clients, and that it was one of the few firms to predict Republican Donald Trump’s presidential victory.
Tennessee-based Republican strategist Ward Baker, who during the 2016 campaign cycle was executive director of the National Republican Senatorial Committee, questioned the Triton survey, which covered a number of issues and political races, in a memorandum to “interested parties.”
Baker’s memorandum took issue with the poll’s look at declared or potential GOP candidates in the 2018 Republican primary contest for governor.
In the past, Baker has worked for Tennessee Republican congresswomen Diane Black and Marsha Blackburn, and would likely be involved in Black’s campaign should she run for governor.
The Triton poll showed Black at No. 1 with 9.9 percent support, slightly ahead of GOP businessman Randy Boyd, who came in at 8.1 percent. Six other candidates or potential candidates had lesser support. The poll said 60.9 percent of those surveyed were undecided.
Baker argued, “We should take these results with a grain of salt. This was an automated Interactive Voice Response [IVR] telephone survey.”
That, he noted, meant there were no live interviewers asking questions but an automated voice with a recorded prompt. Secondly, Baker wrote, “this means they could not legally call cellphones under this methodology. Any survey in 2017 that does not have a significant cellphone sample should be seriously questioned.”