Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: -metan- doesn't seem to do the Welch F-test

From   Roger Newson <>
To   "" <>
Subject   Re: st: -metan- doesn't seem to do the Welch F-test
Date   Mon, 11 Apr 2011 12:46:51 +0100

Thanks for the correction of my spelling mistake, which also appears in my on-line help for the SSC package -parmhet-. I'll fix this typo in the next update of -parmhet-. I was aware that William G. Cochran was not the same person as Archie Cochrane, but obviously forgot the difference in their surnames.

Best wishes


Roger B Newson BSc MSc DPhil
Lecturer in Medical Statistics
Respiratory Epidemiology and Public Health Group
National Heart and Lung Institute
Imperial College London
Royal Brompton Campus
Room 33, Emmanuel Kaye Building
1B Manresa Road
London SW3 6LR
Tel: +44 (0)20 7352 8121 ext 3381
Fax: +44 (0)20 7351 8322
Web page:
Departmental Web page:

Opinions expressed are those of the author, not of the institution.

On 11/04/2011 12:29, Nick Cox wrote:
I don't know, but I have a minor emendation.

Roger is in meta mode and probably thinking "Cochrane collaboration".

This author was W.G. Cochran.


On Mon, Apr 11, 2011 at 12:24 PM, Roger Newson<>  wrote:
I have a query re the -metan- command. When meta-analysing statistics from
continuous data (eg arithmetic mean differences and geometric mean ratios),
it seems to be using a chi-squared heterogeneity test, instead of the Welch
heterogeneity F-test, which was recommended in the seminal paper by Cochrane
(1954) for the case of continuous data. Is there a reason for this?

Best wishes (and thanks in advance)



Cochrane, W. G.  1954.  The combination of estimates from different
experiments.  Biometrics 10(1): 101-129.

*   For searches and help try:
*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index