4A
Thursday, October 31, 1974
University Daily Kansan
Student poll
Bill Roy, Bob Bennett have big leads
By TOM BILLAM Wire Editor
More than 70 per cent of the University of Kansas students polled favored Robert Bennett over Vern Miller for Kansas governor, and more than 60 per cent favored Bill Roy over Robert Dole for U.S. Senator from Kansas, according to a poll conducted Oct. 17 through Oct. 21.
A random sample of 385 students, 82.6 per cent of which are registered in Kansas, showed 71.7 per cent would have voted for Robert Bennett for governor if the election would have been during that period. Fifteen per cent were undecided and 13.3 per cent would have voted for Vern Miller.
In the race for the U.S. Senate, Bill Roy garnered 61.8 per cent of the students' favor while Robert Dole got 26 per cent, and 12.2 per cent were undecided.
Of those registered, 23.7 per cent said they were Republicans, 25.9 per cent said they were Democrats and 50.4 per cent were independent of any party affiliation.
With a sample size of 385, the tolerated error could be as high as 10.1 per cent either
way of the given percentages. Most elections are decided by a smaller margin than 10 per cent, so these results can't be considered conclusive, yet they do show a reasonable perception of what the student body thinks of Kansas candidates.
These results differed greatly from statewise polls sponsored by WIBW in Topeka and KAKE in Wichita.
Roy has gained popularity during the year, according to the WIBW polls, conducted by Central Research Corporation of Topeka.
In January, Roy had 32 per cent, Dole had 56 per cent and 12 per cent were undecided. In June, Roy had 36, Dole 49 and 15 per cent were undecided.
At the end of August, Roy led Dole 47 to 45, with only 8 per cent undecided. By Sept. 24 Roy had strengthened his lead to 50 per cent to Dole's 42 per cent, with 8 per cent undecided.
However, the WIBW poll of Oct. 13 showed Roy and Dole were tied with 46 per cent each, and 8 per cent undecided.
The KAKE poll of Wichita, released the
In the gubernatorial race Miller has held a substantial lead, according to the WIBW polls. In June, Miller had 55 per cent, Bennett had 28 and 17 per cent undecided,
same day, gave Dole 48 per cent and Roy
47.5 per cent, a .5 per cent Dole lead.
In August, Miller gained one point to have 56 per cent, Bennett got 32 per cent and 12 per cent were undecided. Miller lost points in September and got only 50 per cent, Bennett got 36 per cent and 13 per cent were undecided. The Prohibition candidate received 1 per cent.
The latest poll, Oct. 13, showed Miller with 47 per cent, Bennett with 35 per cent and 17 per cent undecided.
The wide divergence in the results of Senate race, in the campus poll and in the state-wide polls is understandable considering the magnitude of student comments in the poll concerning Vern Miller's qualifications for office. Few were complimentary.
Commentators have said Dole's trouble was being the national Republican chairman during the Watergate burglary.
Poll Results
Bennett . . . . . . . . 71.7%
Miller . . . . . . . . 13.3%
Roy . . . . . . . . 61.8%
Dole . . . . . . . . 26.0%
Polling accurate but controversial
By TOM BILLAM Wire Editor
Public opinion polls emerge in election years with increasing regularity, particularly because everybody wants to think his hand is on the pulse of America.
But the paramedics of the body politic have come under fire in recent years regarding not only the accuracy of their results but also their research methods.
One result of these charges is a question of validity—can we believe what the polls say? To determine whether a poll is credible, one must know how it was taken.
In a public opinion poll the number of respondents isn't as important as how they
are chosen. It would be ideal if everyone could be asked his opinion on any given subject, but such an undertaking is impossible.
If 1,500 persons were polled each week, it would take more than 1,400 years to include everyone, assuming a stable population.
The next best method of determining opinion is a random sample of the entire population you wish to poll. The random sample method is based on the statistical law of probability, which means that each individual has an equal chance of being selected for his opinion.
A true random sample would involve listing each voter, assigning him a number,
Campaign ads ...
From Preceding Page
commercial was beatific footage of Hawaii shown while the announcer said that if all the roads Rockefeller had built or improved were put end to end, they would stretch from New York to Hawaii. Rockefeller's opponent charged that Rockefeller was attempting to take personal credit for projects accomplished with public funds. Rockefeller later agreed to drop some of the commercials.
The most controversial of all campaign advertising appeared in the 1964 presidential race. Early in the campaign, Republican candidate Sen. Barry Goldwater stated on ABC's "Issues and Answers" that atomic weapons might be used to defoliate the Vietnam jungles. Democrats jumped on the opportunity to portray Goldwater as a trigger-happy militarist who would plunge the world into nuclear war.
The "Daisy" commercial first ran during NBC's "Monday Night at the Movies" on Sept. 7, 1964. This epic showed a little girl picking petals off a daisy while she was standing in a meadow. She counts "One, two, three..." and then she is overcome by a male voice counting "three, two, one, zero." The screen goes black, and then there is the mushroom cloud of an atomic explosion. The voice of Lyndon Johnson says "These are the stakes. To make a world in which all of God's children can live, or go into the dark." The announcer returns to say "The stakes are too high to for you to remain at home." Republican National Chairman Dean Burch called the ad "a new low in American politics."
Another commercial with a similar purpose showed a little girl licking an ice cream cone. A woman's voice explains that the Strontium-90 released into the air by atomic bombs will kill children. She says President Johnson wants a nuclear test ban treaty, but another man who wants to be president voted against it. "His name is Barry Goldwater." A ticking Geiger counter ends the commercial.
The uproar over the "ice cream" commercial aborted plans for showing another such commercial. This one showed a pregnant woman strolling through a park while an announcer told of the dangers of radiation to those children still unborn. Democrats dropped all three spots but said they couldn't be blamed for the image Goldwater had created for himself.
Republicans responded with a film entitled "Choice," which was to be shown on daytime TV. Goldwater ultimately vetoed the use of the half-hour commercial, but copies of the film fell into Democratic hands. Sponsored by a group hastily put together to front for the film, the "Mothers for a Moral America," it was to depict the low moral state of America under Johnson. It offered scenes of racial violence, strip teasers, teenagers doing erotic dances, drug use and a pornographic bookstore. The film ends with a man resembling Johnson racing along in a Lincoln Continental while drinking a can of beer. Democrats called it "an appallingly tasteless film."
A 1968 Nixon-Agnew spot showed Vietnam battle scenes, rioting in the streets and starving children interspersed with a photograph of a smiling Hubert Humphrey. The Democratic National Committee call it "the epitome of gutter campaign practices."
Four years later, both presidential candidates removed commercials that some people said were offensive.
A Johnson-Humphrey spot showed a man laughing as he watched a commercial with the message "Spiro Agnew for Vice President." The announcer says "If it weren't so serious, it would be funny." Another spot showed an electrocardiogram with the sound of a beating heart in the background. It reminded viewers that Agnew would be just a heartbeat away from the presidency.
The Washington Post said both sides should draw a finer line between what is fair comment and what is sick.
The Dole-Roy controversy pales by comparison.
randomly selecting a sample of these numbers, converting the numbers back into names and interviewing those people.
With a large population, this is a costly and complicated process. With a small population, such as this campus, where every student's name and telephone number is listed in the student directory, it is much easier.
The practical way to determine public opinion with a large population is a modified random sample, in which geographical areas, such as precincts, are selected randomly. The residents to be interviewed within that area are also randomly selected.
Most polls today are of the modified random sample type, which reveals essentially the same results as a pure random sample type but is easier and cheaper.
One discredited technique is quotas sampling, in which certain characteristics such as income or education are used to attain a sample distribution.
The problem with this type of sample is that the distribution of people with these characteristics is unknown, so that quotas of the population based on this distribution may not give everyone an equal chance of being selected.
This was the case with the Literary Digest poll of 1936. That magazine interviewed more than 2,250,000 people, but all were
selected from either telephone directories or lists of automobile owners.
The middle of the Depression, automobile and telephone users were far from the majority. Only the affluent were polled, and as a result the Literary Digest predicted Alf Landon would receive 57 per cent of the vote. He got 37.5 per cent.
Modern pollsters are undergoing criticism for making adjustments in their results. The pollster is faced with the problem of making judgments about voter behavior from voter opinion. The two are different things.
Suppose voting behavior in the South showed voter turnout was continually lighter than expected. A pollster might decide to make the South's opinions count only two-thirds as much for each million people as the East's or the West's.
Is this science or pseudo-science? Accounting for the problems associated with making predictions solely from voter opinion, many people say adjustments are allowable, so long as they are pointed out and explained.
Another major criticism of polls concerns the effects they have on the opinion they are supposed to measure.
Lattice F. Coor, a Washington University political scientist charged in 1964 that polls may kill off potential candidates and See POLLING Page 6.
See POLLING Page 6
Clark, Javits ...
From Preceding Page
better receptions from Jewish audiences than Javits has. And although Javits attacks Clark's stand on Israel and the Middle East, their stands are essentially the same. The only difference is that Clark, like McGovern, wants to cut defense spending by $25 billion, which Javits says will eliminate military aid to Israel. Javitis proposes that the defense budget be cut by$ 10 billion.
Many political observers think Javits has two things going against him—he is a Republican and an incumbent. Whether the problems that many Republicans are facing in the election will affect the support that Javits has been building for 18 years as New York's senator remains to be seen. He didn't do the politically popular thing and denounce Nixon every time he was in front of a television camera, and some think that will hurt him in the election.
But speculation will remain speculation, even after November. What is certain is that Javits has a strongly liberal record, and many New Yorkers see Javits as a respectable elderly statesman. His recent trip to Havana also will help a great deal, and Clark can only applaud Javits' chat with Fidel Castro.
Clark might, as he asserts, represent the "politics of virtue" as opposed to Javits and the "old politics," but a Texan with virtue
who wears clothes that never match isn't necessarily what New Yorkers are looking for in a senator. He can hope on election day either that Republicans in rural New York stay home or that they vote for the conservative candidate, Keating.
If conservatives do become distraught over making a choice between two liberals, they may well opt for Keating. That will hurt Javits and help Clark. But Javits is soliciting support from conservative Republicans and many see him as an acceptable liberal as opposed to a radical one. Most likely they will play it safe with Javits rather than split the vote three ways and give to Clark what they gave to Buckley in 1970.
Also important is that Sen. Buckley can't actively support or campaign for Keating. He is a Republican now, and good Republicans don't campaign against other Republicans, no matter how much they oppose one another.
Clark has a tough road ahead until the election if he plans to win. Javits, too, has a lot of campaigning to do if he wants to prevent Clark from attracting any more of his liberal support. Javits' base, which has been considered impregnable in the past, is getting its first major threat in many years, from, of all things, a "funny talking radical" Texan.