Tools for Changing the World

You probably haven’t ever carried out a telephone survey – at least from the telephoning end. Most of us haven’t, and given how thankless a task it can be, most of us wouldn’t want to.

However, it’s also an invaluable way to gather information, and there’s always a possibility that your group, charity or organization will one day realize that it needs knowledge that it can’t gain any other way. Fortunately, a number of researchers have studied what works and doesn’t in this situation and I’m going to save you time by summarizing some of it for you.

Send a letter first. Researchers who analyzed the results of 29 studies learned that receiving a letter in advance made people more likely to agree to participate in a telephone survey. The most effective letters offered information about the survey in addition to asking the reader to participate.

Ask if this is a good time – and wait for an answer. A recent telephone survey study compared two approaches. Half the time, the caller began the conversation by identifying him- or herself and saying, “I hope I’m not disturbing you, am I?” then waited for an answer before explaining the survey. The rest of the time, the caller said exactly the same thing but didn’t wait for an answer before continuing to speak. People answering the phone were more likely to agree to the survey if they had had a chance to answer the question. (Interestingly, research I discussed in a previous post found that only questions with a “yes” answer had this effect.)

Say that you’re not selling anything (if you aren’t). One study compared a regular survey approach to one in which the callers added, “We are not selling anything” after describing the purpose of the survey. The added phrase increased the percentage of respondents who agreed to participate in the survey.

Go ahead and say “um” and “ah” – but not too often. In public speaking, saying “um” or “er” in a pause – filler words that linguists call disfluencies – is strongly discouraged. However, a study of more than 1,000 recorded phone calls requesting survey participation counted the number of fillers used by various interviewers. They found that when interviewers used too many fillers, fewer respondents agreed to answer the survey questions – but this also happened when interviewers used none at all. The authors theorized that this was because a lack of filler words made the caller sound too scripted.

Don’t interrupt – unless the respondent is refusing you. The same study found that the more the interviewer and respondent talked at the same time, the less likely the respondent was to agree to participate in the survey. However, they also discovered that when the interviewer talked over the respondent’s attempt to refuse, the respondent was more likely to ask the interviewer to call back, rather than continuing to refuse. I don’t actually advocate doing that (because it’s rude), but it’s interesting information.

So, there you go, and good luck to you. Some of the most fascinating information I run across comes from survey results, so don’t discount their usefulness. For example, a recent survey of my regular readers found that three-quarters of them made up 75% of all those I polled. I’m sure I’ll find a use for that figure someday.