Politics, elections and piffle plinking

Nielsen Online Panel

Here’s something belatedly interesting – a Nielsen Online poll using their Your Voice panel from October.

For those of you that may be wondering what I’m on about, Nielsen is a large market research company that not only does the usual telephone polls on politics – the results of which we see every month in the Fairfax press on vote estimates – but they also run online surveys where the respondents are from the Your Voice panel that is over 100,000 people large (click the links if you want to find out more about it). Last year Nielsen started publishing some of the political polling results from this online panel in the lead up to the last election.

They ran one of these online panel polls in October that I missed as it was published in the Smage the day before the US Election.

What’s interesting about it is the way preferences were distributed.

Normally, Nielsen asks respondents how they will distribute preferences which is how they get their two party preferred results, but with this online poll they distributed preferences according to the way they were actually distributed at the 2007 election.

If we compare the Nielsen telephone poll of October 16-18 with the online poll of October 30-31, but distribute preferences from the phone poll on the basis of the 2007 election (making these polls comparable) we get something interesting:

ALP LNP Greens Others ALP TPP LNP TPP
Online Poll 43 39 9 7 55 45
Phone Poll 46 39 9 7 55 45

Although the numbers of each poll don’t sum exactly to 100 on the primaries because of rounding errors, they none the less distribute to 55/45 before the decimal points drop off.

This bodes well having similar phone poll and online panel results from the same pollster, as online panels in the US this year didn’t do as well as their phone poll peers.

And for those interested, here’s the Nielsen breakdown by age and State for their latest November telephone poll. Just remember the small sample sizes involved in those cohorts.

Post a Comment

You must be logged in to post a comment.