Monday, November 16, 2015

Jill Lepore on political polls

Jill Lepore is a Harvard history professor who writes sensibly and knowledgeably about everything she writes about. (She’s been a New Yorker staff writer for 10 years).

Recently Lepore took a stab at describing the squalid reality of public opinion polling and political polling.

Take the time to read the whole piece here:

A blog-size summary won’t do her piece justice. It suffices to repeat this excerpt:
“As these and other critics have demonstrated again and again, a sizable number of people polled either know nothing about the matters those polls purport to measure or hold no opinion about them.”
And this one:
“ ‘The first question a pollster should ask,’ the sociologist Leo Bogart advised in 1972, is “Have you thought about this at all? Do you have an opinion?” ’ ”

There’s also the technical problem, which Lepore mentions: response rates to modern polling are down below 10 per cent. Thus, no poll results represent a legitimate sample of any group of people, and, thus, no poll results can be statistically validated within an error range, and, thus, no one can say with any realistic confidence what the poll results actually mean. Period.

I managed market research and opinion surveys for more than 35 years. I understand too well that pollsters today are furiously cooking the results to pretend that they have surveyed a representative sample.

The survey results that are reported now for public consumption are garbage. Period.

Don’t bother participating the next time you get a call asking for your opinion. Your answers are going to be french-fried by the pollster before the report is published. Whatever you might say really doesn’t matter a whole lot.

Copyright © Richard Carl Subber 2015 All rights reserved.

No comments:

Post a Comment