Rasmussen has Obama up five points over McCain in its latest poll in Virginia. Really. Go see for yourself:
Obama Takes Five-Point Lead in Virginia
Filed under Polling data, Swing States
Tagged as 2008 election, Barack Obama, John McCain, Obama leads McCain in Virginia, Presidential polling, Rasmussen poll, Virginia
Currently my mind has been on the methodology of the presidential polls. I know that there are Internet polls and phone polls, two separate methods of collecting data. So, I was wondering, now as it is reported that Barack Obama is pulling ahead, as to whether his lead may actually be even greater. Here are my thoughts:
Fact: This election is predicted, by the number of new registrants, to be the biggest presidential election ever.
Fact: Many new registrants, thought not all, are young people energied by a palpably exciting presidential campaign.
Fact: Many young people have only cellular phones and do not pay the additional costs to buy a landline.
Educated guess: Some of the “newly minted” voters may not be as inclined as people who are already “seasoned” voters to use the Internet for presidential polling purposes.
(This guess is made because many new registrants are fairly new to the political scene and that this group may vary statistically in their attitudes toward how they wish to be involved polically. I am inclined to believe these people may just be getting registered or have just gotten registered within the last couple of months. They fully intend to vote, but are frankly too busy with their lives to spend too much time on the Internet.)
While I know of no empirical research to support this, it seems reasonalble to believe that there is bias in the typical presidential polling sampling methodology. Since cell phone numbers are not listed publically, relatively few of the people in this subgroup are being surveyed. Also, the new registrants who may have a landline–but who lead busy lives working many hours to make ends meet–may not be represented fairly in these samples. They simply may not be at home when the polling calls are made. This is very important because if your sample is distorted, you can’t generalize your results to predict the election. It is a problem of validity.
My conclusion is that the polls may be off. I predict given the demongraphic of the people described, that overall, this may provide Obama with a boost. In a conventional election where there is not such excitement, I would suggest that such subgroups are not large enough to matter. But in a fairly close election, and when these groups have dramatically increased in size, it is something to pay serious attention to.
Fill in your details below or click an icon to log in:
You are commenting using your WordPress.com account. ( Log Out / Change )
You are commenting using your Twitter account. ( Log Out / Change )
You are commenting using your Facebook account. ( Log Out / Change )
You are commenting using your Google+ account. ( Log Out / Change )
Connecting to %s
Notify me of new comments via email.