Back in the old days (around the time I was born) a man named George Gallup (photo below) started conducting polls. The pollsters went from home to home and interviewed respondents in person. The results of these sample surveys proved surprisingly accurate. Then in the mid-1980s, the company started using telephone interviews instead. The reason: by then just about every household in America had a telephone, and phone calls are a lot cheaper than personal interviews.
Gallup’s polls still proved amazingly reliable, considering the company uses a few hundred respondents to represent the views of millions of people. But surveys are not so simple in today’s America.
Most American families now use cell phones as well as – or instead of – landlines. Reacting to this trend, Gallup has started to include cell phones in its surveys.
Here’s a statement from the company: As of Jan. 1, 2008, Gallup has made the decision to include cell phone interviewing as part of the sample used for its general population studies. This is a complex and costly modification in methodology. Our statisticians and methodologists have spent a great deal of time reviewing the procedures and implications of the change. Essentially, in addition to sampling from the traditional database of all landline telephone exchanges, Gallup now also adds in sampling from a new database of all cell phone telephone exchanges in the country. We screen for those individuals using cell phones who report not having a landline, and then interview a random sample thereof. We then weigh into the sample a proportionate percentage of these interviews conducted via cell phone.
I don’t know where the company gets these cell phone numbers. I sincerely hope my cell phone company wouldn’t give my number to some stranger who could call me and run up my bill! In any event, how do you think I would react if someone called my cell phone and wanted to use up my minutes with a survey? I would hit the “off” button so hard my thumb would hurt. So the cell phone users included in Gallup’s survey must be a special breed who don’t mind strangers running up their phone bills. I don’t know who they represent but it sure isn’t me!
That’s why I wondered about the “analyst” who came to the conclusion in Yahoo News today that there will be no landslide in November’s U.S. presidential election. I don’t know whether there will be a landslide or not. But that’s not the point. The point is that this writer uses recent Gallup polls as the basis of his argument.
My gut feeling is that Gallup polls are no longer reliable. Here’s one reason why:
Let’s say the Gallup pollster calls a cell phone number listed for a Mr. John Jones and it rings in his daughter Millie’s handbag. Millie just turned 14 and loves to talk on the phone, but how does the pollster know she is being truthful? She could pretend to be Mrs. Jones, and who would know any different?
Believe me, I know the dangers of phone interviews from personal experience. I was working on the news desk late one night, trying to run down a story about an entrepreneur who was believed to have sold a local attraction. I dialed a number I’d dug up and had a long conversation confirming the sale. Just as I was about to hang up and write my story for the next day’s front page, the guy at the other end burst out laughing.
The phone that rang was in a bar somewhere, and the guy had picked it up on a whim. If he hadn’t relented and admitted the hoax I would have been in a lot of trouble.
So you’ll understand why I’m skeptical of telephone polls, and why I have a jaundiced view of analysts who base their conclusions on them.