By Ron Sellers
I’ve seen a lot of discussion on various boards and blogs about making surveys more fascinating for respondents, in order to encourage higher response rates. Ideas have ranged from short-term solutions (have in-person interviewers use iPads, to create interest in the project) to the slightly absurd (give each respondent a free trip). There have even been lengthy arguments about replacing the word survey with something “sexier,” such as market intelligence or research science.
I think these arguments tend to have things a little backwards. In my own humble opinion, response rates are falling because we’ve been abusing respondents for years, and they’re tired of it.
I participate in surveys whenever I can, and each year I get more and more depressed for our industry. Telephone interviewers with accents so strong I can’t understand them, or who read each question in a disinterested monotone. Leading questions that make it obvious how I’m expected to answer. Lengthy, in-depth surveys on topics on which I have no knowledge and no interest. Repetitive ratings of 23 brands using the same scale (in a brand category in which I’ve never heard of most of the brands). Absurdly lengthy questionnaires (like 75 minutes with the incentive of a sweepstakes entry – and no, I’m not kidding). Questions that are impossible for me to answer, like telling the interviewer exactly how many nights I spent at a Hilton or a Marriott over the past year (when I’ve probably spent 50 nights in a wide variety of hotels). CATI programs that make me wait on the phone while the interviewer figures out what he’s supposed to do next or two-finger types my response to the open-end. “Surveys” that are badly disguised sales or fundraising efforts. And the list goes on and on and on.
When we repeatedly subject people to garbage like this, is it any wonder that so few of them want to continue helping us out?
Trying to come up with a better word for survey to solve the research industry’s problems is a little like the worst team in the league hiring cheerleaders to bring back its fans, rather than getting some decent players. It’s like American carmakers in the seventies saying, “Hey, the Japanese are making cars that are better designed, better built, longer lasting, and more efficient – we’ll bring back customers by putting whitewall tires on ours!” It’s like a restaurant with broken glass in the food trying to attract customers by buying new napkins. It’s like Congress turning the thermostat down to 68 degrees to solve the federal budget deficit. It’s like putting mascara on a pig so it’ll make a better-looking prom date.
I think you’re probably picking up on my subtle analogies by now. Until bad research companies stop getting business, until clients stop insisting that we need to shove another 12 questions into the 30-minute questionnaire, until companies stop gathering “research” by having the intern use SurveyMonkey, until online panels stop sending six invitations to the same panel members each day, until questionnaire designers use intelligible language, until phone rooms stop hiring people who can’t read Dick and Jane out loud, we’re going to have poor response rates, whether we call it a survey or a scientific investigation into your cognitive process.
So while the team drops another game by 30 points, why are we out designing pom-poms for the cheerleading crew?