Analyzes Typical Reports of polls—
by Roper, Gallup, and TIME—and asks
the Reader to determine for himself
what Meaning the Reports have.
SAVE IN THE CASE of their election predictions, the pollsters do not measure, but are pseudo-measurers. What they think they are examining— public opinion—does not exist save in terms of the instruments they use. I am surprised that they have not realized this when they fashioned some of their questions and looked at the replies they got. It is even more surprising that when the pollsters prepare their reports of what "public opinion" is, say on Russia or the United Nations, they do not then see a great light. I think that they would if they paused and pondered their prose with care. To illustrate, I select three reports: one from Roper, one from Dr. Gallup, and one from Time.
Roper weekly column in the New York Herald Tribune has the title:
"What People are Thinking."Perhaps a more accurate title would be:
"What People Say They are Thinking."But the point is a minor one. The column dated August 5, 1948, had the subheading:
"Poll Favors 'Get-Tougher' Policy with Russia."I have put the article under discussion in quotation marks. What I have to say about it will be without quotation marks and between brackets: