A New Way Of Monitoring Malware? >>
<< Does Putting Sundance Shorts Online Harm Movie...
 tdicon 


Surprises

by Mike Masnick

Mon, Jan 23rd 2006 7:21pm





How Many DVRs Is Nielsen Watching?

from the numbers-please? dept

We noted last month that after years of talking about it, TV watching data from Nielsen was finally going to include a recognition that people watch TV differently when they have a DVR by including data on when people watched shows, not just whether or not they were watched. However, David Card has noticed an article claiming that this data comes from a grand total of 60 households that have DVRs and are passing along that data to Nielsen. There's a good chance it's a typo (600 seems more likely, considering the article notes plans to add 100 more per month), but it does raise some questions about the sample size Nielsen uses and whether or not it's a relevant sample.

7 Comments | Leave a Comment..


If you liked this post, you may also be interested in...
 

Reader Comments (rss)

(Flattened / Threaded)

  1.  

    Nielson Dvr Watching

    identicon
    Ummm, Jan 23rd, 2006 @ 8:40pm

    They are coming to put that shit on my DVR on the tenth(shhhh, i'm not supposed to tell anyone).....And gonna pay me to do it too:)
    only 150 bones....but hey it'll pay for part of my cable.....

    [ reply to this | link to this | view in thread ]

  2.  

    Re: Nielson Dvr Watching

    identicon
    Dan, Jan 23rd, 2006 @ 8:55pm

    How do you sign up for that?

    [ reply to this | link to this | view in thread ]

  3.  

    No Subject Given

    identicon
    Adam W, Jan 23rd, 2006 @ 10:07pm

    My understanding is that Nielsen has always been irrelevant, because everytime they came up with a more acurate way to chart TV viewing it showed there were less viewers than the previous method indicated, and was rejected by their customers in favour of the previous method. But that could just be apocrypha.

    [ reply to this | link to this | view in thread ]

  4.  

    Statistical significance

    identicon
    Brian Bartlett, Jan 23rd, 2006 @ 11:47pm

    The statistical difference between a sample size of 60 and an infinite sample (i.e. the entire statistical universe or all watchers of TV) is very slight. Increasing the sample size to 600 isn't going to make that much difference unless your sample is non-random, which technically the Nielsen is, btw. In that case, a self-selected sample (all volunteers) no sample size is going to give you an accurate picture of your statistical universe which is very likely the entire problem with the Nielsen.

    [ reply to this | link to this | view in thread ]

  5.  

    Re: Nielson Dvr Watching

    identicon
    Jim, Jan 24th, 2006 @ 8:23am

    I doubt that. Nielsen doesn't pay for people to be included in its sampling.

    [ reply to this | link to this | view in thread ]

  6.  

    No Subject Given

    identicon
    haggie, Jan 24th, 2006 @ 10:48am

    If Nielsen paid my cable bill, I would gladly send them all the data from my home-built (and far superior to Tivo) Sage PVR.

    [ reply to this | link to this | view in thread ]

  7.  

    Re: Statistical significance

    identicon
    gc, Jan 24th, 2006 @ 11:21am

    go back to school.

    [ reply to this | link to this | view in thread ]


Add Your Comment

Have a Techdirt Account? Sign in now. Want one? Register here
Get Techdirt’s Daily Email
Save me a cookie
  • Note: A CRLF will be replaced by a break tag (<br>), all other allowable HTML will remain intact
  • Allowed HTML Tags: <b> <i> <a> <em> <br> <strong> <blockquote> <hr> <tt>


A New Way Of Monitoring Malware? >>
<< Does Putting Sundance Shorts Online Harm Movie...
 tdicon 

A word from our Sponsors...
Follow Techdirt
Flattr rss rss

Powered by Postrank

From the Techdirt Archive...
A word from our Sponsors...

Close

Email This