ICANN/DNSO
DNSO Mailling lists archives

[nc-whois]


<<< Chronological Index >>>    <<< Thread Index >>>

RE: [nc-whois] [fwd] WHOIS: Statistical note. (from: roessler@does-not-exist.org)

  • To: "Thomas Roessler" <roessler@does-not-exist.org>, <nc-whois@dnso.org>
  • Subject: RE: [nc-whois] [fwd] WHOIS: Statistical note. (from: roessler@does-not-exist.org)
  • From: "Cade,Marilyn S - LGA" <mcade@att.com>
  • Date: Fri, 9 Aug 2002 07:50:32 -0400
  • Sender: owner-nc-whois@dnso.org
  • Thread-Index: AcI/i7tNjqjLOzEhR1iJEJDzbCog+gADvvpQ
  • Thread-Topic: [nc-whois] [fwd] WHOIS: Statistical note. (from: roessler@does-not-exist.org)

Thanks, Thomas. All of us are aware that the data gathering was 
not intended to be statistically based, but that we would be doing 
our best to organize it and analyze it, using tools available to
the TF... 

At this point, I think we just should make a note to be sure that
we are very clear about this and our recognition of limitations in the
final report. I think we have tried to do that.. but it bears another
look to be sure we are clear in our explanations.

Regards, Marilyn

-----Original Message-----
From: Thomas Roessler [mailto:roessler@does-not-exist.org]
Sent: Friday, August 09, 2002 5:58 AM
To: nc-whois@dnso.org
Subject: [nc-whois] [fwd] WHOIS: Statistical note. (from:
roessler@does-not-exist.org)


----- Forwarded message from Thomas Roessler <roessler@does-not-exist.org> -----

From: Thomas Roessler <roessler@does-not-exist.org>
To: ga@dnso.org
Date: Fri, 9 Aug 2002 11:56:12 +0200
Subject: WHOIS: Statistical note.

   One reader of the [1]WHOIS Task Force's report sent me a private
   comment about the statitsical considerations section of that report,
   pointing out that there is no reason to suppose that the statistics of
   a self-selected sample can be modeled by a Gaussion or any other
   random model. This is, of course, true: The fact that the respondents
   were [2]self-selected can add any systematic bias, and we don't know
   what that bias looks like. However, we know that this bias makes the
   results worse. Thus, error margins derived from a random model
   underestimate the errors. Put differently: If you couldn't derive a
   conclusion assuming the sample is random, you certainly can't derive
   it when the results come from a self-selected sample.

References

   1. http://www.dnso.org/dnso/notes/whoisTF/
   2. http://www.capitalcentury.com/1935.html

-- 
Thomas Roessler                        http://log.does-not-exist.org/

----- End forwarded message -----


<<< Chronological Index >>>    <<< Thread Index >>>