Reliability of phone polling

Factual Questions
1

To determine public opinion, particularly for issues of a political nature, companies will poll people over the telephone. These numbers come from those already listed in the area’s white pages as I don’t believe they random dial for these types of issues.

What this means is that certain people are not being called. We’ve known about those with unlisted home numbers for awhile. They can’t participate in these though they choose not to do so by the very nature of their unlisted status.

There’s a new and growing subset of people who also do not get called. They have no home phone line, instead using a cell phone as their main number. I am a member of this group. Pollsters do not call me.

Now, I’ve done no research on this area, but my WAG would be that people who choose this option tend to be young, childless and spouseless. I couldn’t hazzard a guess as to this group’s average political leanings or outlook on various social and economic policies but there must be leanings of some kind and these aren’t being represented in polling numbers.

How does this affect the reliability of public opinion polls when we know a certain segment of the population is being cut off from having their opinions recorded. How does this change future polls as we get into the sociological aspects of opinion poll results influencing opinion poll results? They feed off each other and we’re left out of the loop.

2

I work for an opinion research firm. Maybe I can shed some light on this.

Our phone lists are normally provided by the client. The lists are randomly generated, or can be taken from public records. We get unlisted numbers fairly regularly, based on the complaints I hear about “How did you get this number? It’s unlisted!”

We get cell phone numbers too, but we normally don’t do surveys with those people. I’ve heard, though, that the percentage of people with no home phone (such as myself) is smaller than might be expected.

3

The people in market research are way ahead of you. They are aware of each of your caveats (and more that you didn’t mention), and they design and implement corrective procedures for each of them.

4

There’s a third group that I’m included in, and with the general hatred for telemarketers I wonder if it’s growing (but doing a survey to test that would be rather tricky :slight_smile: ). It’s people that refuse to answer their phones if there isn’t someone they know showing up on their call display.