Those who continue to believe that the Internet has more to offer the public than pornography will probably be interested in a story that K. C. Jones filed yesterday for InformationWeek about the online metrics firm Hitwise:
Hitwise announced Tuesday that it has created a new tool for measuring the most popular political search terms and the most searched, visited and sought candidates online. Unlike most figures from Hitwise, this data sampling is open to the public.
"As the Internet continues to evolve as a major communications channel for candidates and a platform for the public to research issues, we are delighted to share insights on how American Internet users react to political campaigns and policies online as the election unfolds," Tessa Court, chief marketing officer of Hitwise, said in a prepared statement.
The data will be updated every Tuesday and available through RSS feeds. It covers the top candidates' Web sites, as well as breakdowns according to party affiliation. It also shows which candidate search terms are most popular and which political Web sites draw the most visitors. Based on the results, Hitwise said it will name "Fast Moving National Political Web sites" and the "Candidate of the Month."
In other words, having already imposed major changes on the normative practices of journalism, the Internet is now moving in on the turf of political polling and its legacy of discipline developed by George Gallup and those of his ilk. Here is the first round of Hitwise results:
As Hitwise announced the new features, [Ron] Paul a Republican Congressman from Texas, held the lead for the largest market share of visits to Republican Web sites. Paul's site, Ronpaul2008.com accounted for 44.16% of the market share for Republican visits during the week that ended August 4.
Mitt Romney's site, www.mitromney.com, ranks second among Republicans, but well behind, with 16.13% of the market share for Republican Web site visits during that week, according to Hitwise. Rudy Giuliani ranked third, holding 11.78% percent of the market share among Republicans for Joinrudy2008.com.
Barack Obama (Barackobama.com) held a lead over Hillary Clinton (Hillaryclinton.com) among Democrats, with 40.64% of the market share among Democrats, compared to Clinton's 24.20%.
John Edwards held third place among Democrats with his site, www.johnedwards.com, accounting for 18.36% of the market share of visits among Democratic candidates.
Film maker Michael Moore ranked top among political search terms, followed by Daily Kos.
The data is based on Hitwise' examination of Internet use and search behavior of more than 10 million users and 1 million Web sites that fit within more than 160 industry categories.
Back when I was in high school, I had an opportunity to see George Gallup interviewed. Back in those days, humility was often taken as a positive sign of professionalism; and, if he had not begun his work with a spirit of humility, he had no trouble talking about how he learned it when the best polling techniques of the time totally bungled the forecasting of the 1948 Presidential Election. As I recall, the lesson Gallup learned from this experience was that there was still a lot to be learned about selecting the right sample as a basis for representative results.
Both Hitwise and its subscribers should bear this in mind in light of the fact that the results of last spring's National Technology Scan are unlikely to have changed very much. Let me reproduce the "punch line" from my post about this survey:
A little under one-third of U.S. households have no Internet access and do not plan to get it, with most of the holdouts seeing little use for it in their lives, according to a survey released on Friday.
The issue is not one of how many data points Hitwise can summon up with the click of a mouse but of whether or not they are reporting on that great bugaboo of polling, a representative sample.
Now, in all fairness, I should cleave to my own Wittgensteinian principle that the value of information is only revealed in how it is used. If nothing else, Hitwise has identified a gap between the amount of attention Ron Paul is receiving from the mainstream media and the amount of attention he is receiving from Web surfers. This should constitute grounds for someone in the mainstream media poking around to see if there is really a story here or if it is either a statistical fluke or the result of a few hackers gaming the system (both alternatives being entirely feasible). Personally, I am very skeptical of the proposition that vox populi speaks through the Internet (at least about anything other that the latest site for hot pornography); but I am equally skeptical of the judgments currently imposed by mainstream media over what constitutes news. To invoke a phrase I previously used in a post about technology fads, I think we are dealing with another clash of over-inflated midgets. Unfortunately, in this case both midgets are more interested in selling soap (in one sense or another) than in promoting political discourse, meaning that, one again, our electoral process will get stuck with the short end of the stick.