On 23 Apr 2002 07:58:08 -0700, [EMAIL PROTECTED] (F.
Goldhammer) wrote:

> Hi all,
> 
> I want to calculate the reliability of the following quotient:
> quotient=MEAN(item7, item8 ... item18 )/MEAN(item1, ... item6). The 18
> items form a time series.
> Analogous to the testhalf-reliability I have devided the 18 items in
> odd an even items. I have calculated the quotient separately with the
> odd and the even items. At last I have interpreted the correlation
[ ... ]

 - Does "internal consistency" as the measure of reliability 
make much sense for a time series?   How tightly linked are
the measures?   If there is a high lagged correlation, then 
any 'internal consistency'  is foolish to think about.  Before
you worry about that ratio, in fact, I think the nature of the
time-series needs to be worked out.

Is there an external criterion for this quotient, or is it
intended to predict itself-in-the-future, and nothing more?

Maybe a real-life example would impress me otherwise, 
but what comes to mind here, in the abstract, is that 
 a)  test-retest  reliability  would be more convincing; and 
 b)  even more than the usual cross-sectional report, 
your conclusions will be heavily tied to the character 
of the 'sample';  however well the quotient works, that
depends on various variances remaining similar.

-- 
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to