« Friday night life | Main | National Survey On Drug Use and Health »

Saturday, August 04, 2007

Surveys should be voluntary

It's late Saturday afternoon, the sky is dark with black clouds. It's raining and thundering. There's a knock at the door. As I'm expecting no one, I consider not answering it. But, there's a middle-aged woman standing on the porch, so I open the door a crack.

She's here from the National Survey on Drug Use and Health and she wants to ask me a few questions. I have a vague recollection of getting a letter from someone in Bloomington a week or so ago saying this house had been selected for something or other. She said she put a brochure in my door a few days ago. That I do not recall getting.

She flashed some kind of hand-sized computer device which she seemed quite anxious to use.

I ask her if she's from the government; she says, yes, and shows me her I.D. But, frankly, it looked like something anyone could have whipped up on the computer and laminated.

I tell her I don't give out personal information to people I don't know. That's my standard reply to any survey, as a matter of fact. (Except for surveys I go seeking out.)

She says, it's not personal. We just need your age, how many people live in the house, and some information on your health and drug use.

I tell her that is absolutely the most personal information there is and to go away.

She says fine, but they'll just send someone else out.

I tell her fine, keep sending them.

She did give me a brochure which I will scan and upload later. Also, if I haven't thrown away the letter, I'll scan and upload that too. I have too much to do right now, which hopefully which shake off my paranoia.

Posted by Marie at August 4, 2007 4:14 PM


You were absolutely right to say "No."

Posted by: rich miller at August 4, 2007 6:01 PM

I used to work in survey research--on the phones, door-to-door (for the Census), and behind the scenes, running projects, writing questions, and doing reports. Just setting the context, so you know I'm as biased as you can get when I say: I think it's pretty crucial to respond to surveys that are studying important topics.

Some surveys are pretty severely intrusive, but they're that way because that's the only way to get good information, to help shape policy or monitor what's going on. Reputable surveys have to go to great leaps to protect the identities of the people involved, and anything touching on sensitive issues increases those leaps.

You should NOT take my word for this. I would recommend verifying that it's this survey, since it looks like you got information on the survey center, not the people ultimately doing the survey (US Health & Human Services) or the survey itself. If you're concerned about privacy, ask what precautions there are to protect your identity and keep your answers confidential.

If you get a survey taker who is pushy or rude, I would try to speak with a manager, because you're absolutely right that the survey is intruding on you. With good reason, I would say, but they should definitely not expect to demand that you respond.

The goal of the little hand-held device is that you don't even need to tell the survey taker your answers. They get uploaded, and immediately divorced from your identifying information. Of course, that won't assure you at all if you don't trust the survey or the survey taker in the first place. But it's intended as a way to relieve the social pressure in reporting on potentially illegal or embarrassing activity.

Sorry for the long response. Like I said, I feel pretty strongly about this, and I hate to see crappy survey takers (which it sounds like she was, since she made hash of your "personal information" concern--I imagine what she meant was that they wouldn't be collecting personally identifying information or stuff like bank info) mess up important studies.

Posted by: greg claxton at August 5, 2007 7:25 PM

Greg, based on your comments, I'm taking it seriously and checking out this thing more.

Posted by: Marie at August 5, 2007 11:45 PM

There are three URLs on the flyer. I linked to the shortest one. You found a page from another one. And, here's the third: National Survey On Drug Use and Health (NSDUH)—Homepage.

Posted by: Marie at August 5, 2007 11:50 PM