[ snipping original question, from the end ]

On Tue, 01 Jul 2003 14:19:25 GMT, "Arthur J. Kendall"
<[EMAIL PROTECTED]> wrote:

> COMPUTE GRAT = SUM.6 (g101, g102, g103r, g104, g105, g106r).
> 
> many SPSS within-case statistical functions allow you to specify how 
> many of the argument-variables must be present.
> 
> COMPUTE GRAT2 = mean.4 (g101, g102, g103r, g104, g105, g106r).
> would give you a score that was on the original response metric of the 
> items as long as at least 4 of the items were present.
> 

And for a pro-rated sum of 6, insisting that 4 are present,

Compute GRAT2= 6* mean.4( G101 to g106r) .

If you don't put the ".4"  after the function, then GRAT2 
will become Missing  when any of the items are Missing.

For that pro-rated sum of 6,  rounded to one decimal point,
Compute GRAT2=   Rnd(60* mean.4( G101 to g106r) )/10 ) .


-- 
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
"Taxes are the price we pay for civilization."  Justice Holmes.
.
.
=================================================================
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