I’m finally recovering from jet lag after presenting last month at the Groningen Symposium on Language and Social Interaction in the Netherlands!

The crux of my presentation is this: Telephone interviewers who sound and act like people rather than robots will do better at convincing people to take their surveys. (If you’re really interested in the details, you can read my articles in Field Methods and the Journal of Official Statistics!)

I looked at how scripted interviewers sounded during the first few seconds of a phone call, and then how responsive they were over the course of the survey invitation. Do they answer questions from the person on the other end of the phone or engage in real conversation, like this interviewer?

Answerer: Just so you know, the next time Nebraska plays Michigan {laugh} we’re going to root for Nebraska even though you’re giving us fifty bucks.

Interviewer: That’s all right. I’ll root for Nebraska if you do the interview. How’s that?

Or do they just repeat their lines, like this one:

Answerer: There’s only two of us and my husband’s in the shower and I’m in the middle of making dinner.

Interviewer: Well, this is a very important study.

The results are pretty strong: When interviews sound scripted and non-responsive, only 3% agree to take the survey, but when they sound less scripted and more conversational, response rates rise to 16%.
Through my work on the respondent experience I’ve heard a lot from respondents about how the selection experience and the online survey process are tedious and treat them as just “cogs in wheel” — but my GSLI presentation reminded me that the importance of humanizing respondents applies to phone surveys as well.