Different procedures have different options for handling missing values. Marta is correct that by default Frequencies excludes missing values variable by variable. The only other option in Frequencies is MISSING=INCLUDE, which treats user-missing values as valid.
As the following example demonstrates, Descriptives has three options for handling missing values: variable, listwise, and include. As with Frequencies, the default is to exclude user-missing values variable by variable.
data list free /var1 var2 var3.
begin data
1 1 1
1 2 6
2 1 6
6 1 2
2 6 1
1 6 1
2 6 2
end data.
missing values var1 var2 var3 (6).
frequencies variables=all /format notable /statistics mean.
frequences variables=all /format notable /statistic mean /missing=include.
descriptives variables=all /statistics mean /missing=variable.
descriptives variables=all /statistics mean /missing=listwise.
descriptives variables=all /statistics mean /missing=include.
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of Marta Garcia-Granero
Sent: Thursday, October 04, 2007 10:22 AM
To:
[hidden email]
Subject: Re: Fw: Exclude value from analysis
Hi Rhonda
Since I don't know Jonathon's reply, I can only answer partially.
Theoretically, the solution I gave to you should exclude rating 6 on a
variable by variable basis, not listwise. Please, send (to the list) a
small sample of your data and I'll try to see what's happening.
Marta
> Thanks for reply Jonathon and Marta, but I need to be more specific about
> the outcomes I want. Here's an example. In the following
> descriptives, I
> want SPSS to exclude rating 6 on a variable by variable basis. I've
> tried
> both options you suggested and ended up with the same cases being deleted
> from all analyses. And when I say I'm a novice, I mean no idea (but
> want to
> learn). In the example, where do I put the command to get the outcome I
> want?
>
> FREQUENCIES
> VARIABLES=Item9StudentLists Item10StudentLists Item11StudentLists
> /STATISTICS=STDDEV MEAN
> /ORDER= ANALYSIS .
Locked
