Summary of key findings

Health care No. 1 state problem. Health care was named by 22% as the state’s No. 1 problem in an open-ended question – down from 29% last spring. But a year ago, just 19% mentioned health care. Education placed second. (Bob Wyatt, 615-477-8389)

Attitudes toward health paradoxical. A majority (51%) say the federal government has a responsibility to guarantee health care for all, but 53% also prefer the current system of private health insurance over a government-run plan. (Bob Wyatt, 615-4778389)

State barometer still looking up. The reading remains 62 out of 100, similar to last spring. The measure is up markedly from 51 in spring 2006. The barometer incorporates approval of the governor, and perceptions of the economy and the state’s direction. (Bob Wyatt, 615-477-8389)

Bredesen rating high but down. Approval of the governor is 59%, down somewhat from 67% last spring. This may likely be a post-election bounce up and down: His approval before last fall’s election was 57%. (Bob Wyatt, 615-477-8389)

Majority rate state economy fair or poor. Fully 53% are satisfied with the direction Tennessee is going, but just 43% rate the state’s economy as excellent or good. A racial divide between black and white perceptions is apparent. (Bob Wyatt, 615-4778389)

Race makes a big difference on many issues. These include taxes, crime, welfare, perceptions of the economy, legislative approval, approval of the governor and president, and race relations. (Bob Wyatt, 615-477-8389)

To whom are you willing to speak your mind? You find yourself at odds with an opinion someone is expressing about the war in Iraq. Would you keep quiet, or speak out? And would being among strangers, friends, or family make a difference? (Ken Blake, 615-210-6187)

Health care continues as No. 1 state problem.

(Contact Bob Wyatt, 615-477-8389)

Health care once again trumped all other state issues, with 22% naming health insurance and health services as the No. 1 problem in an open-ended question – down from 29% last spring. However, a year ago, just 19% mentioned health care. This year, 18% mentioned education, down from 21% last spring. Crime rates 10%, up from 7% in the spring. Only 5% mentioned immigration, similar to last fall’s 6%. All shifts except in health care are not statistically significant.

Once again, race proved the best predictor of the No. 1 problem. For blacks, 22% named crime first compared with only 8% of whites. But only 8% of blacks named health, compared with 25% of whites.

Tennesseans’ attitudes toward health care system paradoxical

(Contact Bob Wyatt, 615-477-8389)

A majority (51%) of Tennesseans say the federal government has a responsibility to guarantee health care for all – compared to 69% in a recent nationwide Gallup survey. But a majority (53%) of Tennesseans also prefer the current system of private health insurance over a government-run plan – results close to Gallup’s nationwide findings on this issue. There are, however, wide differences among Tennesseans according to political orientation and income level.

Liberals and those on the left (80%) and middle-of-the-roaders (59%) are far more likely to favor federal health guarantees than conservatives (24%). A majority of Democrats (53%) favor government-run insurance over a private insurance, compared to only one-third of independents (33%) and 14 percent of Republicans.

Thus, it seems that most of us like the current system of mixed private control of health care buffered by government programs for those who cannot currently get or afford coverage.

Tennesseans are quite satisfied with their current health care, with 35% rating it excellent and 35% good. Just 15% chose fair and 12% poor.

Income level is intimately related to assessments of one’s health care quality. About half (52%) of those with family incomes of $15,000 or less rate their care as poor or fair, and only 20% say excellent. By contrast, nearly half (49%) of those making $75,000 and above score their care excellent, while only 11% opt for poor or fair.

State barometer still looking up

(Contact Bob Wyatt, 615-477-8389)

Things are still looking up for the state, with our Tennessee barometer standing at 62 out of 100, a finding similar to last spring. The barometer is up markedly from 51 in spring 2006.

The state barometer is based on an index computed from the rating of the governor, the perception of the state’s economy, and the direction the state seems to be going in. Higher barometer scores indicate higher satisfaction with how things are going in the state.

Education makes a big difference in one’s barometer score. For high school graduates or less, the barometer stands at 50 – while for those with some college or trade school, it is a full 10 points higher. And for college graduates or above, it rises to 74.

Bredesen still riding high but back to pre-election rating

(Contact Bob Wyatt, 615-477-8389)

Approval of Gov. Phil Bredesen remains a high 59%, down somewhat from 67% last spring. The decline may likely be a post-election result. His approval before last fall’s election was 57%. Then came the bounce up to 67% – before settling back into the high 50s after the election.

Bredesen’s approval is higher among older voters. While 51% of those 18-34 approve of the job he is doing, that figure rises to 63% among those 35 and older.

Approval of the state’s Legislative leaders has followed a similar course, bouncing from 40% last fall to 50% this past spring, and now sliding back to 40%. Proportionally fewer minorities (24%) than whites (44%) approve of the state’s Legislative leaders. Among whites, males are more likely than females to disapprove, and females are more likely than males to offer no opinion.

Majority rate state economic conditions fair or poor.

(Contact Bob Wyatt, 615-477-8389)

Fully 53% are satisfied with the direction Tennessee is going, but just 43% rate the state’s economy as excellent or good. By contrast, 54% rate it only fair or poor. Race is an important predictor of economic rating. Among blacks, 58% rate the economy as only fair or poor, compared to 53% of whites. These results are even more dramatic when one looks at the percentages rating the economy poor: 31% for blacks and 12% for whites.

Race makes big difference in how several issues are perceived

(Contact Bob Wyatt, 615-477-8389)

As so many of our polling results make clear, there are huge differences between black and white Tennesseans in how they perceive the health of the economy, other issues, and the candidates. Here is a brief summary:

Crime is the No. 1 state problem for blacks, while whites name health care first. A healthy majority (56%) of whites are satisfied with how things are going in the state, compared to 39% of blacks. Blacks rate the state’s economy far lower than whites. Legislative approval stands at 44% for whites but 25% for blacks; Bredesen approval splits 63-42%; Bush approval splits 41%11%.

In the presidential race, 44% of blacks and only 21% of whites opt for Clinton, while 33% of blacks and only 4% of whites opt for Obama. Thompson’s support is 38% white and 8% black.

On the issues, blacks and whites are about equally affected by energy and the environment, terrorism, Iraq, gay marriage, abortion, the economy, and health care. But 83% of blacks say they are personally affected a great deal by taxes v. 67% of whites. Further, 72% of blacks say they are affected by crime compared to 52% for whites. And 86% of blacks are highly affected by the educational issues compared to 65% of whites. The gap in race relations is also marked, with 69% of blacks and only 24% of whites saying they are affected a great deal.

Taxes, crime, education, economic wellbeing, foster care, and race relations, those are the fault lines in the state between the races, and they have persisted across time. But on health care and the economy, where blacks and whites rate closer together, neither race seems too optimistic.

The small sample of those selecting other races limits further examination.

To whom are you willing to speak your mind?

(Contact Ken Blake, 615-210-6187)

Imagine your opinion on whether sending troops to Iraq was a mistake differs from the trend in the opinions held by most people around you. Or at least you think it does. Now imagine that, during a casual conversation, somebody offers an opinion opposite of the one you hold. What would you do? Would you say what your opinion is, or would you keep quiet? And would it make a difference whether you were among strangers, friends, or family members?

Answers to those questions say important things about where public opinion concerning one of the most critical issues of the day comes from, and how it develops.

Most Tennesseans, it turns out, say they would speak their minds no matter what. Over three fourths (76%) say they would be either very likely or somewhat likely to say what they think while among strangers. Slightly more would be willing to speak out among friends (84%) and among family members (86%).

Anxiety about expressing a minority view does make a difference – but in a somewhat paradoxical way. While among strangers, people who consider themselves in the minority are just as likely to speak out as are people who consider themselves in the majority. Interestingly, though, people who hold a minority view grow more willing to dissent among friends or family. But in the same setting, people aligned with the majority view grow more willing to confront dissent by expressing the majority view.

Why would one become more willing to dissent when in the company of friends or relatives who are more likely to confront such dissent? One explanation is that, despite whatever levels of tension and dysfunction typically exist among friends and family members, such relationships offer relatively safe spaces for discussing political issues, even ones as touchy as the Iraq war. Private debate, in short, may be more common than public debate, and although there may be more at stake when we disagree with friends and family over issues, there’s also may be more goodwill with which to smooth things over.

President’s approval rating still low

(Contact Ken Blake, 615-210-6187)

Just over a third (35%) of Tennesseans say they approve of how George W. Bush is handling his job as president. The figure remains essentially unchanged from the 34% approval rating he had in last spring’s MTSU Poll, and substantially lower than his 42% approval rating a year before that.

Bush still enjoys a 66% approval rating among the state’s Republicans, but the figure stands at 25% among Independents and 10% among Democrats. Behind party identification, his highest approval figures lie among college-educated evangelical Christians (66% approval).

Appendix A: Measuring attitudes in polls

Attitudes toward many issues – such as taxes, military actions, or immigrants – are complex. The same person may hold several contradictory notions and balance them off against each other to determine an overall attitude.

For example, in forming an attitude toward abortion, the same person may believe that abortion should not be used as a method of birth control. When asked bluntly whether he or she is in favor of abortion, that person might reply either “No” or “Yes.” This is because the same person could also believe that abortion is acceptable in cases of incest, rape, or serious defects in the fetus. To learn the person’s attitude, a survey researcher must therefore ask more than one question, then report the results in all their complexity.

Appendix B: Evangelicalism in Tennessee

The label “Evangelical” is claimed by subgroups within a wide array of Christian types including Southern Baptists, Pentecostals, Catholics, Lutherans, and more. To complicate matters, many self-described Evangelicals attend non-denominational churches, and still others hold Evangelical beliefs and exhibit Evangelical behaviors without identifying with or even recognizing the term “Evangelical.” In short, Evangelicalism is an abstraction, and there is no perfect way to measure it. But whatever it is, Evangelicalism is a strong force in Tennessee politics, and an interpretation of Tennessee attitudes would be incomplete without some attempt to account for it.

The MTSU Poll assesses Evangelicalism by asking individuals whether they consider themselves an “Evangelical or born-again” Christian and also by measuring three themes often found in Evangelical belief and practice: Belief that the Bible is the “actual word of God” and should be “taken literally, word for word,” belief that “Jesus will return to earth and take all true Christians to heaven, leaving non-Christians here to face tribulation and the Antichrist,” and a personal history of having “tried to encourage someone to believe in Jesus Christ or to accept Jesus Christ as his or her savior.” In Tennessee, all three measures correlate positively and strongly – although not perfectly – with self-identification as an Evangelical, and when one or more of these measures emerges as a significant predictor, the predictor is assumed to accurately characterize the attitudes of Evangelicals.

Appendix C: Sample and method

The poll was conducted by telephone Oct. 17-27 by students in the College of Mass Communication at Middle Tennessee State University. Students interviewed 593 people age 18 or older chosen at random from the state population. The poll has an estimated error margin of ± 4 percentage points at the 95% level of confidence. Theoretically, this means that a sample of this size should produce a statistical portrait of the population within 4 percentage points 95 out of 100 times. Other factors, such as question wording, also affect the outcome of a survey. Error margins are greater for sample subgroups.

The sample varied somewhat from the U.S. Census Bureau’s latest available projections for age, race and gender proportions within the state. Such variation commonly occurs because certain demographic groups are more difficult to contact. The data were thus weighted to more closely match Census projections for these demographics. Here are the Census data, the sample data, and the weights:

Small variations in reported percentages (1% or less) sometimes result for rounding variations in different statistical procedures or the way different programs handle population weights.