Nielsen Measuring TV/Online Usage. You Gotta Problem With That?

If you want a good insight into how screwed up the specialty of media measurement is, you could do worse than to read this story outlining Nielsen's attempts to measure Internet usage, and, most importantly, online viewing habits, in some of the households that make up its TV research panel. Sounds like a no-brainer, right?

Well, no. Some (unnamed) researchers believe that it will "impact the stability of Nielsen's core TV research" in part because the 375 households who will participate in the TV/online research panel might not accurately depict where they go on the Web. What a pathetically weak argument. Like other, previous, challenges to Nielsen's attempts to move forward, this one relies on a fallacy: That Nielsen's traditional TV panel is somehow sacrosanct, even though it represents only a tiny fraction of the nation's homes and also relies on humans, and not just technology, to gather data. Its core People Meter service employs a box connected to the TV which keeps record of what's being watched, but each member of the family, when the TV is on, is also supposed to press "a personal viewing button" which syncs up the viewing data with the age and sex of the viewer. Does anyone really believe that that happens 100 percent of the time? In fact, for the all-important sweeps measurements, two million households send Nielsen paper diaries, which, of course, rely entirely on humans to record data.

The truth is that media measurement will always be imperfect, but with media consumption habits changing so fast, years can't be spent in debate over the best way to measure things. Will Nielsen's initial research be flawed? Probably. But, particularly as more TV is viewed online, Nielsen has to start somewhere. It can refine its methodology over time.