Separate names with a comma.
Discussion in 'TiVo Coffee House - TiVo Discussion' started by Bierboy, Oct 28, 2013.
Announced this morning in their blog...
It only took them 15 years!
eta: i'm concerned this will open opportunities for abuse (fraudulently inflated ratings), but i'm sure they think they have "safeguards" built in...
Nielsen's ratings have been skewed forever. When I was in elementary school, we had one of those boxes for a while. However, very few people did, so the ratings for that geographical area were based on what that one family watched. If that one family watched NBC news and the 10,000 people around them all watched ABC news, guess which news show got the nod?
They have claimed accuracy over the years but unless you are monitoring EVERY SINGLE TV, how can you make a claim of accuracy? Yeah, I know. Plus or minus 3% or some other such nonsense. The fact is, Nielsen has been wrong for a long time now. And the pity is, many good TV shows were cancelled because of Nielsen and many horrible ones were allowed to linger.
Nonsense? It's called statistics. There is an entire field of mathematics devoted to it. You may not like it or understand it but it's been proven time and again to be accurate.
What is your proof that Nielsen has been wrong all this time? Let me guess, you know a lot of people who watch a low rated show so Nielsen must be wrong. Or you don't know anyone that watches a high rated show so Nielsen must be wrong.
Oh and they already monitor DVR playback. You can see the Live +7 numbers here. Right now these numbers are pretty much meaningless to advertisers. What Nielsen is doing is trying to make them valuable by showing where people are watching shows so advertisers can place their ads accordingly.
I think it's absolutely ridiculous to believe that Nielsen numbers reflect reality anymore. I know very few people that are still watching live TV. NCIS really pulls those big numbers??? REALLY? I can't say I have ever talked to a single person who has ever watched an episode of this show and somehow it has managed to be a very successful show for a long time. It could be that they cater to the few people who are still watching live TV. I don't know what it is but I'm certain it's never been the most watched show on TV.
When I first started in radio, back when we rode dinosaurs in to do our air shifts, I was appalled at the small sample size Arbitron used, and how different the results seemed compared to what I knew about who, among those I knew or knew about, listened to what, but later learned a bit about statistics and some other stuff about life and people in general that I didn't know when I was younger, including insights gained from where people had their old car radios tuned or the pre-sets set when I was doing car stereo installs. You probably wouldn't expect a middle aged black woman to mostly listen to the local country station, but sure enough...
If Nielsen numbers were to reflect people you know then they would probably be way off. They're looking at a much bigger group though. Why are they focused more on the numbers of people watching live TV? Because that's what advertisers are most interested in.
Does anyone wonder how someone like Nate Silver can predict presidential results as well as he is able without talking to every voter? It's called statistics. There are plenty of situations in which you use the results of a sample size to predict the results of an entire population. It's important to take a sample size reflective of the entire population. If you don't do a good job of that, you're predictions are going to be off, that's why Nielsen spends a lot of time and energy trying to get reflective sample groups.
I think they're finally including the numbers because they can finally get accurate enough analytics on what people did. Did they watch the show, skip the ads? If so, don't count the viewing. Did they watch it, with ads? OK, add it.
It's important for the Live+7 numbers because you want to know how valuable those +7's are - do 90% of people watch the ads? or is it only 70%? Or 50%? Since a lot of shows have terrible Live numbers, but good +7s, people want to know how much those +7's need to be discounted by to get accurate numbers.
Remember, only ad viewers count. Apparently most DVR users watch ads, so it's important to find out the percentage that skip them.
As for shows like NCIS - well, CBS skews old - how can it only get 2.6 with 18M viewers, when other shows can score 2.7 with half that? (FYI, I watch NCIS and NCIS LA).
There is it folks. The "It doesn't apply to my experience so it must not be true" argument. Ignorance is bliss.
Nielsen already includes DVR viewing. A few months ago I got picked to keep a Nielsen diary for a week (just a paper log of my viewing, not a box on my TV). It asks you to keep track of what you watched, when you watched it, and when it originally aired. I think it only shows up in the official ratings if you watch something within so many hours of it originally airing, but they do keep track of it (and probably still sell the stats on later viewings to the content companies).
Statistics is an amazing science, and it works because it's science. These people are way smarter than you and I, and the entertainment industry wouldn't be a multiple billion dollar industry if Nielsen were wrong or didn't know what they were doing.
There's 300+ million people in the country. According to Zap2It, NCIS pulled 18.9 million total viewers for the week (live + DVR). So, that's 6.02% of the US population. That's an incredible number, to be sure, but that still means 295 million people didn't watch it at all. You are way, way more likely to not know anyone at all who watched NCIS than to know someone who did.
Also, the population is way larger these days, so even though in the past a larger percentage of the population may have watched a given TV show, lower percentages can see a show these days and still be "most watched"
(source for ratings: http://tvbythenumbers.zap2it.com/20...49-ncis-number-one-with-total-viewers/211922/ )
That may be true, but they're adding this -- "... Nielsen confirmed to clients that it will make the software developer kit (SDK) that enables this measurement available for implementation in mid-November."
Here's the actual blog post.
How do they count the paper logs? I assume you are not sending the logs daily so at what point are your views counted?
Ah, I see. So, I suppose this means that they will be able to count participating viewers in overnights, and they will have better compatibility with DVRs and knowing exactly what they did. I have no idea how the boxes work, so maybe the data they get from DVR viewers with boxes is limited?
This also would let them have a uniform set of results, which will improve the statistical model
You send them in at the end of the week, and they ask you to drop in the mail pretty immediately after the last day. So, they have one method (boxes) that collect nightly data for things like overnight ratings, and then the diaries for longer term viewing habits and ratings. I suppose the diaries are probably more accurate, but the boxes are there because the networks probably want *something* immediately
I thought they did this already?
Given that more than half of the US has a DVR in their home, you would think advertisers would be much more interested in those numbers. The whole Nielsen thing makes little sense to me.
Well, the problem is they need to figure out the percentage of viewers per show of DVR users actually skip the commercials so DVR ratings can be adjusted likewise.
Then again, most advertisers pretty much assume all DVR users skip commercials (even though it appears a good chunk don't), so it's up to them to tell Neilsen what kind of numbers they want. Finding out the percentage of users who do watch ads helps networks and programs, but advertisers won't like it (again, see "all users skip commercials" which lets them discount prices).
Internet ratings is a new area for advertisers, and is unskippable, so those numbers are more interesting.
Because you don't understand how advertising and the ratings system works.
I know people that do not skip ads on their DVR but leave the room, pit break, drink etc, some just like to walk a little when the ad is on. Neilsen may take this into account, but I don't know how and I not paid to know.