Everything I need to know about polls I learned in Junior High

It’s funny how some episodes from your past stick in your memory, while more significant ones don’t.

In ninth grade, all students were asked to complete a written survey on the topic of alcohol and drugs. We were told that the survey was being conducted by a grad student at the local university.

Among my friends, this seemed like the perfect spoof. I don’t remember if I’d even tasted beer at the time, but according to my survey answers I was a frequent drinker who blacked out regularly. Drug use? Sure, why not: LSD, cocaine, pills. Heroin too, but no more frequently than once a month.

The induced paranoia of another crowd convinced them that “narcs” were really behind the survey. Answer honestly? Riiiight.

I’m sure the survey made a beautiful grad school paper, complete with line charts, bar graphs and R-squared factors.

But the data it was based on wasn’t worth ca-ca.

LESSON: People lie, especially to pollsters, for a host of reasons.

Political polls have been notoriously bad of late. Out of every 100 people who answer a poll, maybe a handful consciously lie out of fear of judgment, paranoia, or distrust of the polling organization. Enough to make the results unreliable.

Most polls are reported with a precise-sounding “margin of error”. The margin of error is a statistical device to account for the small size of the sample relative to the population whose attitudes it is supposed to represent.

So if a poll has a margin of error of 3.8%, but one in twenty people lie, what do you have? A poll that is very likely wrong.

Distrustful people may also avoid taking the poll altogether. The poll naturally skews to the opinions of people who trust the pollster.

People who self-report as “likely voters” may not all have the same motivation to vote come election day.

Some may argue that all these error factors should cancel each other out, and they’re right, the should. But these days, journalists are not unbiased observers. CNN, to name just one, has a stake in the outcome.

What is the value of a poll that has lost its predictive power? Lorem ipsum dolor sit amet.  News filler. Merely something to talk about until real news comes along.

Polls aren’t news. Polls aren’t news. Polls aren’t news.





This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s