There is a whole family of transformations for proportions usually refered 
as the Lambda-Tukey. Arcsine, logit and probit are members of this family. 
Logarithms can also work well but only under some conditions. You may check 
the chapter of Emerson in the book about Exploratory analysis of Variance 
by Hoagle, Mosteller and Tukey. ViSta has a graphical demonstration for 
these transformations.

Pedro

At 15:50 17/10/2001 +0200, Tobias Richter wrote:


>We have collected variables that represent proportions (i. e., the
>proportion of sentences in a number of texts that belong to a certain
>category). The distributions of these variables are highly skewed (the
>proportions  for most of the texts are zero or rather low). So my
>question is: Is there a function that transforms the proportions into
>symmetrically distributed variables? And is there a reliable statistics
>text that discusses such transformations?
>
>Thank you very much,
>
>Tobias Richter
>
>
>
>--
>-------------------------------------------------------------
>Tobias Richter, Psychological Dpt. (General Psychology)
>University of Cologne (Germany)
>Phone +49-(0)221-470 38 48 E-Mail [EMAIL PROTECTED]
>http://www.allg-psych.uni-koeln.de/richter/trichter.htm
>
>
>=================================================================
>Instructions for joining and leaving this list and remarks about
>the problem of INAPPROPRIATE MESSAGES are available at
>                   http://jse.stat.ncsu.edu/
>=================================================================

==========================================================
Dr. Pedro M. Valero-Mora.                       WWW: http://www.uv.es/~valerop
Director M�todos de Investigaci�n               Phone: +34 96 3393880
                                                                       Fax: 
+34 96 3393881
Universitat de Val�ncia
INTRAS-Instituto de Tr�fico y Seg. Vial.
INTRAS-Institut of Traffic and Road Safety.
C/ Hugo de Moncada 4. Entresuelo.
Valencia, 46010, Espa�a (Spain).
==========================================================



=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
                  http://jse.stat.ncsu.edu/
=================================================================

Reply via email to