Help! How do I exclude a particular value from all analyses?

[Simone Krell]

Hello there
I am currently looking at the data of a questionnaire, which had a
response scale from 1-6, and an option to use "not applicable".

I want to compute variables (the averages of my scales) in order to
further work with the data. However I am struggling to find a way to
exclude the values for the "n/a", without 'punishing' the respondents.

For example when you calculate an average it would be helpful of SPSS
could automatically adjust the N, rather than counting the n/a as a
zero and thereby reducing the average of that person.

I hope that makes sense, please feel free to ask any clarifying
questions you may need to help me.

Many thanks,
Simone

[Reeza]

Hi Simone,

How are you calculating the average of the person?

By default most calculations in SPSS do ignore the n/a, in addition,
most variables are either number or char and a numeric value would not
allow a value of n/a but replace it with a missing value.

Reeza

[Rich Ulrich]

Compute fact1= mean.7( item01 to item10) .

- says: Create Fact1 as the average of the non-missing items
of the ten items numbered from 1 to 10, so long as at least 7 of
them are not coded as missing. "7" can be whatever
number you want, when you want to indicate that the
result requires at least that minimum number of valid responses.
(I've usually used 2/3 or 3/4.)

If you do not have ".7" part, then the result is set to missing
if any of the items are missing.

When someone insists on a total score instead of a mean,
use (for example)
Compute fact1= 10*mean.7( item01 to item10) .

--
Rich Ulrich

[fred....@lsc.gov.uk]

Have you specified that the not applicable answere is to be treated as
tho it's a missing value? I'm presuming that there is a code for not
applicable that has been entered, and that it's zero. If you haven't
specified it as missing, then it will be included in all calculations,
if you do sepcify it then it will be excluded, which should give you
the means you want, and be able to use the above solution from Rich.

The syntax would be something like
missing values Question1 Question2 Question3 (0).

On May 11, 8:18 pm, Rich Ulrich <rich.ulr...@comcast.net> wrote:
> On Tue, 11 May 2010 08:10:18 -0700 (PDT), Simone Krell
>