Want to Know What I Think? ASK ME!

I’m not really sure why, but I seem to get a lot of calls from survey companies. And by “a lot”, I mean once or twice a month now and, during the run-up to an election, it’s several times a week. That may not seem like a lot, but survey results are only accurate if the original sample of respondents is a good cross-section of the population. If the same people are called again and again, the survey results can legitimately be questioned.

The sample size is an issue, too. The smaller the number of people surveyed, the wider the margin of error. Back when my agency handled political clients, we always tried to get at lease 1,000 valid responses to any survey, particularly a “horserace” poll. Even then, the margin-of-error is about 3 percent.
There are several very good research firms here in Hawaii, but a lot of the political candidates—almost always Republicans, for whatever reason—insist on using mainland firms. How can we tell? Because the interviewer invariably cannot pronounce the names of politicians with Japanese or Hawaiian surnames.
A lot of the surveys are automated now. A recorded voice tells you to “Press one if your answer is Yes; press two if your answer is No.” And even on these surveys, they mispronounce some of the names. Sheesh! How hard can it be?
The surveys always conclude with a few questions that help to ensure a representative sample: age, gender, ethnicity, household income … things like that.
Still, assuming the sample of respondents is large enough and is representative of the universe in question (registered voters, condominium owners, frequent fliers), you can learn a great deal about what those folks are thinking … how they’re likely to vote … what they want.
Which is the long way ‘round to noting that during a recent World Symposium sponsored by IATA (the International Air Transport Association), there were conclusions drawn about what passengers want and what they like and don’t like. But, as passenger advocate Charlie Leocha points out in this interesting column from Consumer Traveler, all the conclusions were drawn from a computer analysis of data collected by the airlines themselves. There were no surveys of actual airline passengers. The airlines decided what we like and what we don’t like without ever actually asking us! No wonder they keep getting it wrong.