Especially in an election year, the way news media polls are conducted and reported takes on added importance.
Polls on controversial national, state or local issues can pretty obviously influence the way political officeholders decide those issues. And when the issue is what will happen on Election Day, the way the polls are conducted and reported takes on added significance, of course.
Too often, as I see it, the polls have the potential effect of influencing election results rather than predicting them. For example:
A poll which, on the basis of a very limited sampling of potential voters, projects a comfortable margin for a candidate may influence some of that candidate’s supporters to relax and stay home on election day—an obvious advantage to the challenger if he has the drive—and the money—to run hard to the finish line.
Conversely, a poll can work to the disadvantage of the trailing candidate if—keeping in mind that polls have a margin of error ranging up to at least eight percentage points—he and his supporters become discouraged in the final crucial days, when he may be up to eight percentage points closer to the frontrunner than the poll indicates.
A couple of recent Omaha World-Herald polls indicate the problem, as I see it. (Let me stress that questionable interpretation and prominence given poll stories is a national problem. The New York Times and USA Today are involved, too. I concentrate today on World-Herald poll handling because it’s a close-to-home issue.)
Polls Can Mislead Because Of Margin Of Error
World-Herald poll results published last Monday were given top play on the front page, reporting that while Second District Republican Representative Lee Terry led by 44% to 39% over Democratic candidate Tom White among the 607 registered voters who were polled, Randy Adkins, a political scientist at the University of Nebraska at Omaha, commented:
“Anytime it’s within the margin of error [in this case eight percentage points], the candidate who is down can win.”
The UNO political scientist was not quoted as pointing out the obvious flipside of that equation; i.e., the margin of error could add up to eight points in favor of the candidate leading in the poll, in this case, Republican Lee Terry.
(Incidentally in a candid comment on his campaign strategy, White made it clear to a reporter that he was hoping for a big turnout from the North Omaha area, which has a heavy Democratic constituency. “If we can get North Omaha to vote, we win. It’s that simple,” White said.
(North Omaha, of course, is the home of most of the city’s black population. White’s hope for a major turnout there could certainly reasonably be interpreted as realistic but not necessarily citywide-popular political strategy, making clear that he is hoping for the voters of an ethnic minority to be the decisive factor if he is to win election.)
* * *
Recurring Problem: Misrepresenting The Results
A major, recurring problem that good journalistic practice would not allow is the recurring mistake—again, The New York Times and USA Today and other national publications are among the guilty parties—is reporting poll results as reflecting accurately the feelings of all members of the group from which a very small percentage of members are polled.
Obviously (obvious, that is, except to the reporters and editors handling the story) poll results are based on the responses of only a very, very small percentage of the group to which the poll respondents belong as, in a case currently at hand, registered voters in Nebraska’s Second Congressional District.
The practice is especially misleading when the poll has a significant margin of error, which may or may not be mentioned prominently in the story, and customarily without any mention the total members in the group whose sentiments are purportedly being accurately reflected by a small sample.
(In the case at hand, the poll produced responses from 607 registered voters in the Second Congressional District. No mention of the total of registered voters in the Second District, which would have given the reader a chance to judge whether the sample is large enough to justify drawing sweeping conclusions about the feelings of all registered voters in the Second District.)
On Tuesday, The World-Herald again gave top play to the poll of those 607 registered voters. (You could find the number, and the fact that the poll results were subject to a plus or minus margin of error of eight percentage points, if you read the agate type under some of the poll results.)
Small Sample Represented As The Whole
In the story itself, interpretation of the poll results was phrased in a way suggesting that the sampling of 607 registered voters represented the feelings of all registered voters within the Second Congressional District. For example:
“Forty-seven percent of the district’s registered voters still approve of Obama’s job performance, a level of support similar to Obama’s margin of victory in the district two years ago—when he won an electoral vote with 50% of the vote.”
The fact: 47% of the 607 voters who were polled—about 285 in all—indicated approval of Obama’s job performance. And there was no mention of the possible 8% margin of error.
One hopes that such polls, which have increased in popularity with all the news media, come at least reasonably close most of the time to reflecting the feelings of the much larger number of people in the group targeted by the pollsters. But in any case, the results of those polls should be reported accurately as the results of a small sampling of opinion, with the margin for error always mentioned high up in the story, whether it deals with politics or abortion or any other issue of significant interest to the public.
There is also the question of the “play” given to poll stories and their timing, particularly close to an election. The better policy, it seems to me, is not to give any poll story top play, especially not close to an election when it can influence results rather than simply reflect a small sample of opinion.
In political polling in a hotly-contested race, the risk of misleading is especially high when the polling is spread over several days, a time period in which the tabulated total of responses may not accurately reflect any change in sentiment within the group.
Then, too, there is the obvious (but often overlooked) question of how many of the polled “registered voters” will actually go to the polls. Being registered does not guarantee that the poll respondent will actually vote.
* * *
Don’t Give The ‘Whistle Blower’ Label
To Criminal Threats To National Security
A smorgasbord of comment, offered with the hope that you will find something to your taste:
–Let’s stop calling them “whistle-blowers”—the people who leak classified government information to liberal sources which broadcast them as evidence of what they consider to be proof of American complicity in brutality, torture—you name it—usually involving the United States military or its allies.
How about “traitors,” or at the very least, criminals who have placed the security of the United States and the lives of its military personnel at risk.
–The village of Hooper, Nebraska could hardly have chosen a better way to damage its community image than by, in effect, telling its part-time, award-winning city librarian, Carla Shafer, to drop her plans to spend her day off teaching English to immigrants in Nickerson, a village about nine miles from Hooper.
Shafer says that Hooper City Council President Gene Meyer told her that she shouldn’t teach English to immigrants in Nickerson because it would appear that the village of Hooper approved of the classes. Shafer said she told Meyer: “You can’t stop me. It’s my own vehicle, on my own day off with my own energy.” She quit her part-time job as director of the Hooper Public Library.
I doubt that any reader would have an ounce of sympathy for Hooper city officials, including Police Chief Matt Schott who, Shafer said, blocked her from retrieving her personal items until it was decided she had done nothing improper with the two national grants that she had been awarded to help finance teaching English to immigrants.
–A number of Democratic candidates appear to have reached a new low in dirty politics—if that’s possible—in their frantic effort to protect their majority in the House of Representatives.
In a number of hotly contested House elections, there is a third Tea Party-type candidate in addition to the Republican and Democratic candidates. Several Democratic candidates have acknowledged that they have arranged financing for telephone campaigns urging Republican-registered voters to support the Tea Party candidates, hoping, of course, to draw votes away from their Republican opponent.
Interestingly but not surprisingly, when asked whether these sleazy efforts were being coordinated on a national level, the Democratic National Committee ducked the question, saying only: “Republicans have no one to blame but their own ideological intolerance for the bloody civil war on their side.”
* * *
Four Former Governors Urge
‘No’ Vote On Amendment 2
I wholeheartedly concur with the judgment of four friends—all of whom have served as Nebraskas governor—in their opposition to Constitutional Amendment 2, which appears on the November 2 ballot and would eliminate the elected office of State Treasurer.
Governor Dave Heineman, a former state treasurer, Kay Orr, former governor and former state treasurer, Senator Mike Johanns, former governor, and Charles Thone, former First District Congressman and governor, joined in a statement that having an elected State Treasurer “provides an important and necessary check and balance on the handling of the billions of dollars that flow through the treasury each year. In our opinion, the elimination of our elected State Treasurer position would jeopardize this important safeguard of the taxpayers’ money…
“Please join us in voting ‘No’ on Amendment 2.”
I value the opinion of the four former governors. My feelings on the matter are buttressed by the fact that I was a Statehouse reporter for eight years and had occasion to observe the important role which elected State Treasurers play in Nebraska state government.
* * *
When It Comes To Huskers Vs. Tigers
I Hope The Gamblers Are Right
No column this week would be complete without some mention of this year’s “Game of the Century” for the Nebraska Cornhuskers.
I hardly need tell any reader that the winner of the Nebraska/Missouri game in Memorial Stadium in Lincoln on Saturday becomes the odds-on favorite to represent the North Division in the Big 12 championship playoff in Dallas November 4.
Some Husker fans have already provided for the trip to Dallas. A good friend told me that he ordered six tickets from one of those ticket-selling websites.
A foursome of close friends’ and their wives, including Marian and me, have made provision for lodging in Dallas, including a reservation of rooms in something called the Mansion at Turtle Creek. (It isn’t as expensive as it sounds.)
We read that in various rankings, Missouri is considered no lower than sixth best among the nation’s elite, while Nebraska’s ranking is from 12th to 14th.
But the much-quoted Las Vegas gambling line has Nebraska favored by seven points.
Let’s hope that the gamblers are right.
# # #