Friday, 12 July 2013


I've had a couple of surveys recently.  One was web-based (the request to participate being by email) as a follow up to the car purchase, and other was a phone call.  The phone call was automated and basically started off seeking voting intentions but moved on to the asylum seekers and the "pink batts" issues.

I don't normally respond to surveys of any sort, but in each case I was prepared to make an exception.   In the case of the car purchase survey, I felt that some sort of feedback (positive and negative) was appropriate.  In the case of the voting intentions survey, I was slightly re-assured by the fact that it was automated, and, as we read so much about polls regarding voting intentions, I thought it might be interesting to participate (even though there was no indication whether this particular one was being conducted on behalf of the media, a political party or someone else).

But in each case, after a few preliminary questions, the survey descended into questions that, to my mind, were quite intrusive.   For example, the car survey sought information about income, and the voting intentions survey sought quite detailed information along the lines about who in the household makes decisions about matters such as the purchase of insulation.

However, neither gave an option of, "decline to comment, move on to the next question"!   Needless to say, I drew the line at this type of question.   I think that the voting intention survey may nevertheless have recorded my responses up to the point where I hung up. On the other hand, the car purchase survey never reached a point where I could push the "submit" button, so they missed out on the responses to even the basic questions (as was witnessed by the fact that we received a "follow up" email a week or so later, reminding us that we had not responded).

I guess the people who design these surveys think they know what they're doing - but my own view is that the results they obtain are likely to be representative only of people who are prepared to give out a lot of personal information - and if that's a cross-section of society, then I for one would be very surprised.

No comments:

Post a Comment