Easy way to deselect individual cases?

[Rick Oliver]

AV wrote:
>
> Good afternoon, people!
>
> I'm (still) using SPSS for Win 6.1. and my problem is this:
>
> I haven't found any easy way to temporarily deselect _several_
> individual cases (outliers). There must be some really easy way to do
> this...?
>
> I have previously used "Select if" procedure, but problems will emerge
> when there is several individual cases what I want to deselect.
>
> Thanks.

You can select cases using the FILTER command. Cases with a value of 0
for the specified fitler variables are not selected; cases with a value
of 1 (or any value > 0, I think) are selected.

compute filtvar=1.
if (idvar=20) filtvar=0.
filter by filtvar.

In this example, the case with an ID of 20 won't be used in subsequent
procedures. By changing the value of the specified filter variable for
individual cases and then running the FILTER command, you can deselect
those cases.

[AV]

Good afternoon, people!

I'm (still) using SPSS for Win 6.1. and my problem is this:

I haven't found any easy way to temporarily deselect _several_
individual cases (outliers). There must be some really easy way to do
this...?

I have previously used "Select if" procedure, but problems will emerge
when there is several individual cases what I want to deselect.

Thanks.

av

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
aki.vai...@joensuu.fi

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

[David Mark Marso]

Yes there is!!!

I am assuming you have some sort of numeric criteria which
indicates one is an outlier or other (say OutlVar > valuexxx).
(I am also assuming you are removing these cases all at once ???)

If so Try:

COMPUTE MyOLFilt=(OutlVar > valuexxx).
FILTER BY MyOLFilt.

*Do whatever...
..........
*RESTORE FILE TO ALL CASES * .

FILTER OFF.

Regards, David

[AV]

Solved (in three different ways)! Thank you all.

av

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
aki.vai...@joensuu.fi

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^