Iasonas Lamprianou
2010-Sep-10 11:21 UTC
[R] Standardized logistic regression coefficients
Dear all, I am looking for ways to compute standardized logistic regression coefficients. I found papers describing at least 6 different ways to standardize logistic regression coefficients. I also found a very old (Thu May 12 21:50:36 CEST 2005) suggestion by Frank E Harrell (one of the colleagues who frequently contribute on this list) saying... Design doesn't implement those because they have terrible properties. Instead consider interquartile-range odds ratios (done by summary.Design by typing summary(. . .)). 1. Is this still the case, or is there any package today in R which computes some sort of standardized logistic regression coefficients widely accepted by the community? 2. Also, if anyone knows, how can I implement this interquartile-range odds ratios of the Design package? I checked on the manual and found no reference to interquartile-range odds ratios Thanks Dr. Iasonas Lamprianou Assistant Professor (Educational Research and Evaluation) Department of Education Sciences European University-Cyprus P.O. Box 22006 1516 Nicosia Cyprus Tel.: +357-22-713178 Fax: +357-22-590539 Honorary Research Fellow Department of Education The University of Manchester Oxford Road, Manchester M13 9PL, UK Tel. 0044 161 275 3485 iasonas.lamprianou at manchester.ac.uk
Iasonas Lamprianou
2010-Sep-10 11:23 UTC
[R] Standardized Logistic Regression coefficients
Dear all, I am looking for ways to compute standardized logistic regression coefficients. Before asking for your time I did some homework. I found papers describing at least 6 different ways to standardize logistic regression coefficients. I also found a very old (Thu May 12 21:50:36 CEST 2005) suggestion by Frank E Harrell (one of the colleagues who frequently contribute on this list) saying... Design doesn't implement those because they have terrible properties. Instead consider interquartile-range odds ratios (done by summary.Design by typing summary(. . .)). 1. Is this still the case, or is there any package today in R which computes some sort of standardized logistic regression coefficients widely accepted by the community? 2. Also, if anyone knows, how can I implement this interquartile-range odds ratios of the Design package? I checked on the manual and found no reference to interquartile-range odds ratios Thanks Dr. Iasonas Lamprianou Assistant Professor (Educational Research and Evaluation) Department of Education Sciences European University-Cyprus P.O. Box 22006 1516 Nicosia Cyprus Tel.: +357-22-713178 Fax: +357-22-590539 Honorary Research Fellow Department of Education The University of Manchester Oxford Road, Manchester M13 9PL, UK Tel. 0044 161 275 3485 iasonas.lamprianou at manchester.ac.uk