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]

st: RE: Moderation effect by splitting the sample

From   "David Radwin" <>
To   <>
Subject   st: RE: Moderation effect by splitting the sample
Date   Thu, 20 Dec 2012 11:50:45 -0800 (PST)


I can't say for sure without knowing more about what you are trying to do,
but regarding splitting continuous variables, see this post and the
references therein:

If you do split the variable anyway, consider comparing upper and lower
terciles or quartiles (and not the upper and lower halves as a previous
email intimated) to minimize loss of efficiency:

Gelman, A., & Park, D. K. (2009). Splitting a predictor at the upper
quarter or third and the lower quarter or third. The American
Statistician, 63(1), 1-8.

David Radwin
Senior Research Associate
MPR Associates, Inc.
2150 Shattuck Ave., Suite 800
Berkeley, CA 94704
Phone: 510-849-4942
Fax: 510-849-0794

> -----Original Message-----
> From: [mailto:owner-
>] On Behalf Of Ebru Ozturk
> Sent: Thursday, December 20, 2012 11:42 AM
> To:
> Subject: st: Moderation effect by splitting the sample
> Dear All,
> For non-linear models, I want to test the moderation effect of X
> Can I test this moderation effect by spliting the sample according to X
> variable (moderator)?
> Kind regards
> Ebru

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index