Libel in Fact: Pol vs. Poll
A member of a polling dynasty enlightens the court
Posted September 20, 2009
One of the foremost experts on polling in the 1960s was Elmo Roper (1900-1971). Mr. Roper, who had been born in Nebraska on July 31, 1900, had directed the Fortune Survey from 1935 until 1940. Soon after, he founded the Roper Survey. In 1946, Elmo Roper went on to found the Roper Center, now located at the University of Connecticut. To this day, the center serves as a clearinghouse for polling centers to send their data.
Elmo Roper's voice of was one of responsible public-opinion polling when many Americans were still getting used to the idea that public opinion could be measured. Not only had Elmo Roper's polls been among the most accurate available at the time, but he also was known for accepting responsibility when a poll was wrong and trying to improve his techniques.
Mr. Roper's son, Burns, had worked with his father, on national polling projects, and had since become a New York marketing research consultant.
On May 15th, 1968, Burns W. Roper testified at the Goldwater v. Ginsburg libel trial. His appearance was sure to have brought a touch of polling royalty into the New York courtroom - a courtroom already graced, it should be noted, by titans of law, publishing and politics.
Mr. Roper brought with him a number of criticisms of the Fact poll.
A key set of criticisms concerned the phrasing of the question to the psychiatrist. First, Mr. Roper pointed out that the poll asked only whether Senator Goldwater was fit for office, but included no question about the fitness of President Lyndon Johnson. Because no comparable question was asked concerning Mr. Goldwater's opponent, Mr. Roper suggested, the key question of the poll conjured up the idea that Mr. Goldwater's character was more suspect or questionable than was then-President Johnson's.
Overall, 1960's psychiatry (and psychology) was not known for seeing positive health in its clients (with some notable exceptions such as Carl Rogers). It therefore would have been interesting to see how this same group of psychiatrists might have depicted President Johnson at that time.
In addition, Mr. Roper pointed out, the question asked, "Do you believe Barry Goldwater is psychologically fit to serve as President of the United States?" and was followed by two boxes for respondents to select. Usually, these would be ordered "Yes....No," but in the poll, they were ordered "No....Yes," perhaps drawing for a "No" (unfit) answer.
Perhaps the most damaging issue about the questionnaire, from the juror's standpoint (and from those who read Fact magazine) was that the polling question had been accompanied by a letter from Fact to each psychiatrist informing them that Mr. Goldwater had suffered nervous breakdowns, and asking the psychiatrists such leading questions to consider as, "Does he seem prone to aggressive behavior and destructiveness?"
The magazine article never had mentioned this letter - a letter that could have gone some considerable distance toward biasing responses.
According to the Times' reporter, Mr. Burks, many of the trial's most scathing and sarcastic exchanges occurred as Mr. Steinberg, the attorney for Mr. Ginzburg, attempted to defend his client.
First, Mr. Steinberg was overruled by the judge for claiming that Mr. Roper's views were not relevant. Next, Mr. Steinberg tried to dismiss the idea that methods and procedures were important to polling. The key thing in polling, Mr. Steinberg argued, was just getting the outcome right, and Mr. Steinberg cited a second poll by the Medical Tribune, which indicated psychiatrists preferred Johnson to Goldwater 2-to-1.
Mr. Steinberg's equating of the two polls failed to catch the vituperative specifics of the Fact version. The defense attorney appeared to be edging out on a limb. He seemed to lose the sympathy of the Times reporter, and perhaps the courtroom. Mr. Burks, who never seemed entirely sympathetic to Mr. Steinberg, described him that day in court as favoring: "Ben Franklin spectacles and a poker face while making barbed remarks...".
Whether individual jury members were scientifically-minded or not, it was hard to miss the fact that Mr. Ginzburg had included a plainly prejudicial letter to the psychiatrists that he and Mr. Boroson had polled - and then had failed to disclose the existence of the letter in the article. In the face of such poor methods and incomplete reporting, Mr. Steinberg's arguments on behalf of his clients certainly must have seemed difficult to accept.
Mr. Roper, representing one of the most promiment families of polling in the United States, had presented a carefully reasoned critique of the Fact poll's methodology and its flaws. Mr. Roper raised another issue as well: Out that of the 12,356 psychiatrists contacted, only about 20 percent responded, suggesting that the opinion of the "typical" psychiatrist - the 80% who did not respond - remained unknown.
And so it was that a little science entered the courtroom that was not so easily dismissed.
More on the trial next week...
Burks, E. C. (1968, May 16). Expert condemns Goldwater poll. New York Times, p. 44.
Copyright (c) 2009 John D. Mayer