Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Reliability (Spearman/Brown)


From   nicola.baldini2@unibo.it
To   vmahlberg@netcologne.de
Subject   Re: st: Reliability (Spearman/Brown)
Date   Thu, 08 Mar 2007 12:27:55 +0100

Sorry for not having underscored it in my previous post. I hope from Conroys' mail that it is now clear that Spearman/Brown split-half reliability is NOT Spearman's correlation and is NOT calculated by Stata's command -spearman-. 
If Stata commands which I suggested in my reply to your first post are not satisfactory, you may think to use SPSS, or to calculate it by yourself (formula found in SPSS' help) as k*r/[1+(k-1)*r] where k is the number of items and r is the (Pearson's) correlation between the sums of the items in each of the two groups in which you splitted them.

Nicola

At 02.33 07/03/2007 -0500, you wrote:
>Date: Tue, 6 Mar 2007 17:27:17 +0000
>From: =?ISO-8859-1?Q?Ron=E1n_Conroy?= <rconroy@rcsi.ie>
>Subject: Re: st: Reliability (Spearman/Brown)
>
>On 4 Márta 2007, at 11:03, Vanessa Mahlberg wrote:
>
>>
>> I?ve got a (maybe very simple) question:
>>
>> I would like to calculate the split-half reliability (Spearman/ 
>> Brown) for a whole scala (including 14 items).
>
>The split half reliability is a very old method now. Essentially, it  
>depends on the split. You can think of Cronbach's alpha as the  
>average of all possible split half reliability coefficients.


*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index