Tue Apr 30 19:50:02 1996
From majordom Tue Apr 30 19:50:02 1996
Return-Path:
Received: by scholar.cc.emory.edu (5.0/SMI-SVR4)
id AA13857; Tue, 30 Apr 1996 19:50:02 +0500
Date: Tue, 30 Apr 96 16:46:28 PDT
From: broman@Np.nosc.mil (Vincent Broman)
Message-Id: <9604302346.AA29092@Np.nosc.mil>
To: tc-list@scholar.cc.emory.edu
In-Reply-To: (waltzmn@skypoint.com)
Subject: Sampling and Vulgate
Content-Length: 1324
Sender: owner-tc-list@scholar.cc.emory.edu
Precedence: bulk
Reply-To: tc-list@scholar.cc.emory.edu
waltzmn@skypoint.com said:
> I was doing what I could to improve on the heavily
> biased readings of UBS. But if a sample is large enough, it matters
> less.
>
> It's also important to remember the difference between RELATIVE and
> ABSOLUTE statistics. Changing our sample will change the actual
> RATE of agreement between, say, vg and E (the absolute statistic).
> It is far less likely to change the amount of difference between vg
> and E as opposed to vg and B (the relative statistic).
If your sampling method is biased, increasing the sample size won't help,
the larger sample will still be biased, it just has a smaller variance.
The RELATIVE statistics are =exactly= what gets botched up
by bias in your sampling. What you need is a sampling method
that you can convince your readers is totally independent of
what you're trying to measure.
And you would have a hard time convincing me that the selection
of variants in the UBS3 apparatus is =independent= of the attestation
by texttypes.
Vincent Broman Email: broman@nosc.mil = o
2224 33d St. Phone: +1 619 284 3775 = _ /- _
San Diego, CA 92104-5605 Starship: 32d42m22s N 117d14m13s W = (_)> (_)
___ PGP protected mail preferred. For public key finger broman@np.nosc.mil ___
Back