The Truth about the ‘Lieberpoll’

September 18, 2007


Daily Kos gets egg on its face with a new poll

As should be obvious by now, I do not particularly sympathise with either the antiwar left or the anti-immigration right. However, it is necessary for the serious gambler to regularly check sites like Daily Kos and Michelle Malkin’s website, because of their (some would say undue) influence on the Republican and Democratic parties and on the fringes of the media. Five days ago Daily Kos released an opinion poll. What made it especially interesting was that this was on a race that had already taken place last November, namely the Connecticut Senate race. This poll claims that if the race was re-run Ned Lamont would have beaten Joe Lieberman by 48% to 40%. Coming on the heels of a poll that claims John McCain could be beaten in 2010 by Janet Napolitano, Kos uses this to support his claim that the tide is turning against supporters of the war. This poll has received a large amount of publicity, even prompting an article in the British press.

However, not only is this an extreme case of a sore loser it also has several methodological flaws:

1. Any poll about a past event is not going to get many respondents. People do not like responding to polls at the best of times and even fewer will be willing to respond to a race that they were already bored with. It goes without saying that those who will repsond are more likely to have stronger views – making them less representative as a whole. If Daily Kos want to be taken at all seriously they should publish their response rate.

2. The weighting of their poll is greatly different from the weighting that the CNN recorded. If you re-weight the poll by age to make it consistent with the 2006 Exit Poll you get Lieberman winning by 51% to 40% (50.96 to 39.96%). In fact Lieberman’s vote is slightly higher overall than it was in November 2006.

3. Instead of using a respected pollster like Gallup or Rasmussen they use Research 2000, a rather minor and poorly respected pollster.

In any case this poll deserves to be taken with a big pinch of salt.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: