New Pew Survey: The Internet in Campaign 2008

April 16, 2009

The crazy-productive folks at Pew’s Internet and American Life project have a new survey published looking at The Internet’s Role in Campaign 2008.

There’s a lot of fun results to mine for anybody interested in political news consumption, participation and engagement via the Internet. I still need to read it more closely, but some of my favorite sound-bites so far:

  • A handy chart comparing where self-identified democrats and republicans get their online news. Statistically significant differences are marked with a “^” (Hint: look at CNN, Fox, Radio, and the Internet). Caveat: see my methodological comments below before interpreting this too deeply.
  • This staggering time-series graph illustrating the decline of newspapers as a primary source of political news over the past 10 years or so (respondents were only allowed to mention their top two sources of news)

Americans Top sources of Political News During the 2008 campaign

Americans' top sources of election news during the 2008 campaign

On a methodological note, it’s interesting that the surveyors chose to conduct the survey via land-line telephones only.

Some of you might recall that Pew also published some really interesting data in the middle of the campaign season suggesting that cell-only voters are disproportionately young, democratic, and Internet users.

Despite the fact that the surveyors weighted their results to try to reflect the demographics of telephone users in the U.S. as a whole, I take that to imply that the numbers in this latest survey should provide a conservative estimate the total Internet use in the population as a whole. At the same time, I think it undermines some of the comparisons between democratic and republican voters based on the land-line only data.


3 Responses to “New Pew Survey: The Internet in Campaign 2008”

  1. Pablo Says:

    I always wondered how they measured the “cell phone only” voting crowd…by cell phone plan or by cell phone possession. In East Chicago, Gary (IN) where I worked for Obama, a lot of people used the pay-per phone cards on their cell phone and a lot of cell phones were allegedly purchased stolen. When the Pew Research came out, even amended, I couldn’t help but think that a vast underclass that was overwhelmingly for Obama was still going uncounted.

  2. aaron Says:

    Hey Pablo! Thanks for another great comment 🙂

    Your observation of informal and transient cell-phone use practices in Gary illustrates a classic data collection and analysis paradox that I would call “the Hidden Population Catch 22.” I think the paradox goes something like this:

    (1) Survey researchers are unlikely to change their sampling techniques or population estimates unless someone can demonstrate conclusively that their data is biased in a systematic way.

    (2) Hidden populations are hard/impossible to get reliable data about using research techniques that meet the standards of survey researchers.

    As you might guess, it follows that it’s difficult to find evidence about “hidden populations” that meet survey researchers’ standards of rigor. Consequently, it’s tough to figure out whether the data about hidden populations is biased or not.

    The trick would be to figure out whether or not your gut suspicion about informal and transient cell-phone use is widespread enough to result in a systematic bias that Pew hadn’t already caught somehow in their weighing of the data. I’m sure they did a ton of internal analysis to make sure they thought their sample population was reasonable and that landline surveys of political opinion yield unbiased results. The trouble, as I mentioned above, is that the only basis for comparison are other studies and population estimates generated through methods that wouldn’t necessarily capture accurate information about the sub-populations we’re talking about. You could go in circles for hours…

  3. […] New Pew Survey: The Internet in Campaign 2008 | Aaron Shaw Detailed stats on the influence of the internet on election politics (tags: internet politics usa election) […]

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your 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: