OPINION POLLS
SIR,—I am gratified that Mr. W. E. Styler has read both my article and my book, but I am not confused. In my article I wrote that polls had their subsidiary uses ; in my book I made subsidiary use of Dr. Roper's admirable polls.
It is a question of level of accuracy. A five or ten per cent, error does not invalidate the type of conclusion I draw ; it completely destroys the pollers' claim to reflect accurately the will of the electorate. Even from the election polls under discussion one can draw confirmation for a number of possible statements about the people of the United States: for example, the relative stability of the American electorate, and the very small percentage of changed votes necessary to determine election results ; the lack of appeal of "third parties," and so on. Of course, in the present case such conclusions are more properly founded on the election results ; but similar deductions from unproved and unprovable polls seem to me legitimate ; and I cannot see that such use of them invalidates my criticism of the pollers' excessive claims and influence.—
I am, Sir, yours, &c., GEOFFREY GORER. The Croft, Bradford-on-Tone, Taunton.
Previous page
