Bill Tancer is the general manager of research at the Internet-data firm Hitwise, and has appeared on this blog before. He is the author of a new book, Click: What Millions of People Are Doing Online and Why it Matters, and has filed a new guest post here.
Say what you will about the addition of Sarah Palin to the Republican ticket (here I said that most searches sought “hot” pictures of the Alaska governor); regardless of personal or political viewpoint, it is clear that McCain’s selection altered the political blogosphere.
Looking at U.S. Internet users’ interactions with political websites for the week ending with August 30, 2008 (the week of the Palin announcement), visits to over 800 campaign and blog sites grew 33.8 percent to reach the highest level of online political activity in the last three years.
The massive surge in online political interest was also coincident with a change in the demographic composition of visitors to the category — but maybe not in the way you might expect.
Checking the demographics of visitors to political sites, I expected that gender composition would show the greatest difference pre- and post-Palin. In fact, the percentage of female versus male visitors was the least affected by the Palin announcement; age was the most affected.
If we look at the age demographics of online politics over time, the 55-plus crowd continues to be the predominant age group (39 percent at the end of 2007). In fact, if we look at online political interest by age group, there is a strong positive correlation: as age increases, so does percentage composition in political site visits.
The age group with the largest increase during the Palin surge during the week ending with August 30, 2008, was the 25- to 34-year-olds increasing to 14.4 percent of all visits from 13.4 percent the previous week (the 18 to 24 and the 55-plus age groups both decreased about 1 percent in the same time period).
During the last week in August, McCain visitors in the 25 to 34 age group increased from 15.5 percent of all site visitors to 22.2 percent compared to the week ending with September 6, 2008, while Obama visitors dropped from 24.7 percent to 20.7 percent during the same week.
Before we get too far ahead of ourselves, however, it’s important to remember that political site visits do not necessarily correlate very well to popular votes. In fact, one phenomenon that we’ve noticed in the past is that the closer you get to election time, the more crossover traffic happens between political viewpoints (liberals visiting conservative websites and vice versa).
Regardless of end-vote, online behavioral data indicate that recent events have ignited our interest in November’s election, especially for the 25- to 34-year-old voter.