Kendall's W

[drfg2008]

(1) How can I comput Kendall's W (not Kendall's tau) with SPSS?

(concordance coefficient Kendall & Babington-Smith 1939)

(2) What is the minimum number of raters: two? Or more than two (my table starts at three raters)

Thank you.

[Bruce Weaver]

It is an option in both the old (legacy) and new nonparametric test procedures for k related  samples.  Something like:

* Old school method .
NPAR TESTS
  /KENDALL= a b c
  /MISSING LISTWISE.

* New-fangled method.
NPTESTS
  /RELATED TEST(a b c) KENDALL(COMPARE=PAIRWISE)
  /MISSING SCOPE=ANALYSIS USERMISSING=EXCLUDE
  /CRITERIA ALPHA=0.05  CILEVEL=95.

HTH.

drfg2008 wrote

(1) How can I comput Kendall's W (not Kendall's tau) with SPSS?

(concordance coefficient Kendall & Babington-Smith 1939)

(2) What is the minimum number of raters: two? Or more than two (my table starts at three raters)

Thank you.

[Mike]

Look at the NPAR procedure for the subcommand Kendall --
it provides W.  The minimum is number of raters is two.
If you have the PDF of the SPSS v18 syntax guide, look on
pages 1367-1368 (look for NPAR in other version's syntax
guides).

-Mike Palij
New York University
[hidden email]



----- Original Message -----
From: "drfg2008" <[hidden email]>
To: <[hidden email]>
Sent: Friday, April 29, 2011 11:43 AM
Subject: Kendall's W

> (1) How can I comput Kendall's W (not Kendall's tau) with SPSS?
>
> (concordance coefficient Kendall & Babington-Smith 1939)
>
> (2) What is the minimum number of raters: two? Or more than two (my table
> starts at three raters)
>
> Thank you.
> -----Dr. Frank Gaeth
> FU Berlin
> --
> View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Kendall-s-W-tp4359053p4359053.html
> Sent from the SPSSX Discussion mailing list archive at Nabble.com.
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[drfg2008]

Thanks,

I seem to have missed it.

NPAR TESTS
  /KENDALL=v1 v2 v3
  /MISSING LISTWISE.