Oh, joy. Another thread of let’s bash Nielsen even though we don’t understand what ratings are.
Ratings don’t measure the viewership of programs. The ones you see are a meaningless sop to public curiosity, just as weekend box office grosses are for movies. Ratings measure the viewership of commercials. That changes everything. And yes, that means that PBS doesn’t have official Nielsen ratings. They don’t need them, so they don’t need to spend the enormous sums of money they cost. (They may contract for special surveys, although many firms will do that.)
Measuring commercial viewership (or listenership for radio) means that the real ratings aren’t what you get to see. If you’re old enough you may remember that in tv programs like Mary Tyler Moore or WKRP in Cincinnati, the executives were always struggling with giant, thick piles of printouts called rating books. Whenever the book came out Andy Travis would announce to the DJs that he hadn’t had time to go through it all yet but that they looked good. Or horrible.
Why was the book so thick and hard to read? Because it was designed to measure commercialship not programship. In fact, Nielsen ratings are given in blocks of six minutes. Ten every hour. Two hundred forty every day (except daylight time switch days). Yes, they do bundle those into larger groups, but the real numbers are the six-minute averages. Why? Because it makes it much easier to measure the commercials they care about, not the programs they don’t
And those really don’t count all that much because advertisers discovered demographics many decades ago. Each of the 240 blocks are broken down by age, location, sex, race, and anything else they can check off. And also in combinations thereof. Are you a black female in Chicago between 18 and 24? There’s a box for you.
Advertisers pay not merely for gross numbers but for numbers in their targeted demographics. They don’t care whether the program is new or old or good or bad or a rerun or a special. They pay by expectations, and their contracts call for minimum viewership in the right demographics or else the programmers have to give them money back or run extra commercials for free. The programmers hate this. As soon as they have to start making good, as it is known, they switch to Plan B. But they have an additional problem. They know they can only cater to a specific demographic set so far or else it starts alienating viewers not in that set. No network will program a night of MTV style programs, even if that would give them a chunk of a desired demographic. It would turn everybody else off.
Wait, it gets worse. Because the point is getting commercials watched, the ratings only care about watched commercials. They know that people leave the room during commercials. They always have, so it’s part of the deal. But how do they count viewership for time-shifted programming? More to the point, do they even care about viewership of time-shifted programming? Commercials are timely. Sure, many of them are part of on-going national campaigns, but even those often have a tag about a special sale or promotion. Watching how sales fluctuate with these specials helps advertisers measure the effectiveness of their advertising. It doesn’t help them at all if you record a Christmas program and watch it in January. The compromise is - and has been for a long time - that time-shifted viewership only counts if it is watched within a week or ten days.
Wait, again. How do they know it was watched? What if the tv was on and the person left the room or muted or fast-forwarded through it? (You think that because you do it, everybody fast-forward through commercials. They don’t. A surprisingly high percentage, something like one-third to one-half, actually watch commercials on DVRs.) Anyway, that’s where the diaries come in.
The regular Nielsen ratings come from the meters attached to sets. Another and larger group is given what are called diaries to fill out. They are supposed to record not only what is being watched but who is watching at any given moment. Everybody knows how flawed a system this is, but everybody knows it’s better than nothing. Many systems have been tried that try to count automatically, by some sensor arrangement, but I don’t think any is widespread and practical. And the diaries are so expensive, since they require so many more respondents to get an adequate sample size in each breakdown box, that they are limited. Except for sweeps months. Which feature so many specials that the point obviously can’t be to measure programming. They don’t. They measure demographics.
Added to all these flaws is the problem of trying to measure group viewing. Dorms, bars, prisons, hotels. Can these be properly measured? As far as I know, they can’t. After-the-fact memory surveys are the best alternative, although obviously inadequate. So some demographics do get short-changed. That means that network programming is always going to be disproportionally aimed at stay-at-home families. Family-friendly programming is therefore an extremely sensitive issue and any deviation makes headlines.
What happens when the tv is taken out of the equation? You can now watch tv on your computer. You can’t watch tv with the original set of commercials on your computer, though. So why would those advertisers care who watches? A different set of advertisers might, true. But not the originals. How do you program for two entirely different sets of advertisers with different demographic targets and different time needs? Nobody has come close to an answer for that.
The short answer is what I said upfront. The system is set up to make the most money by selling commercials. Programming is geared almost entirely toward this goal. Viewers are good, but the wrong set of viewers is not as valuable. The right set is valuable, but they can’t be seen to control. IIRC, there are a few stations in Europe that broadcast nothing but commercials. That’s an ideal that every station owner aspires to.
Commericals, commercials, commercials. [del]Programs[/del], [del]programs[/del], [del]programs[/del].