Re: Filtering a specific sample of cases (not random)

[Alan Mead]

Or possibly:

Temporary.
Select if( Cluster = '2A' ).
freq / Code.

The temporary makes the select if statement active only for the next analysis. 
So you'd have to repeat it if you had a second analysis. You could also omit it 
but then your dataset will be altered... I hope you're not analyzing your only 
copy of the data...

-Alan

--

Alan D. Mead, Ph.D.
President, Talent Algorithms, Inc.
+815.588.3846 o
+267.334.4143 m 
science + technology = better workers

On Apr 16, 2015 8:11 AM, Alan Mead <address@hidden> wrote:
>
> Robert,
>
> Something like this?
>
> sort cases by Cluster.
> execute.
> split file by Cluster.
> execute.
> freq  / Code.
> split file off.
> execute.
>
> All that was from memory and untested; my apologies if I remembered wrong but 
> hopefully it suggests the solution.
>
> -Alan
>
> On 4/16/2015 5:14 AM, Robert Orzanna wrote:
>>
>> Dear all,
>>
>> I recently started using PSPP to analyse sav data. I must say that I have no 
>> previous experience with SPSS. But from what I have seen so far, I very much 
>> like PSPP. I have one (maybe simple) question that I couldn't find an answer 
>> for in the documentation. 
>>
>> In an Excel file I have a list that looks as follows: 
>>
>> Cluster | Code
>> 2A         | 33edfdkeew
>> 2A         | 29dnewdnd
>> 2B         | sd2034wss
>> 2B         | 29jsnn2221
>>
>> Then I have a sav file that I analyse with PSPP. This file contains ~ 400 
>> cases, all with unique entries for the "Code" variable. I want to perform 
>> the analysis of the data for the different clusters. Thus, I want to filter 
>> the file for cases whose codes belong to a specific cluster. 
>>
>> What is the best way of doing this in PSPP?
>>
>> Select cases only allows me to select a random case sample, based on a range 
>> (but only allowing me to select case numbers) or a filter variable (which I 
>> don't have in my default sav file). 
>>
>> My current idea was to create 2 new variables for the two different clusters 
>> and then assign a 1 to the variable for each case that belongs to the 
>> respective cluster. However, this seems to be very inconvenient and I am 
>> wondering if there is a better way to do this.
>>
>> Thanks ahead!
>>
>> Robert
>>
>> -- 
>>
>> mindfully present. just be and let go. 
>> passionate. simple life. user-centred design. 
>> impulsive thinking. human hearted. 
>> Co-Creator of @MindfulMeerkats.
>>
>> Website: robert.orzanna.de 
>> Twitter: @orschiro
>>
>>
>>
>> _______________________________________________
>>
>> Pspp-users mailing list
>>
>> address@hidden
>>
>> https://lists.gnu.org/mailman/listinfo/pspp-users
>>
>
> -- 
>
>
>
> Alan D. Mead, Ph.D.
>
> President, Talent Algorithms Inc.
>
>
>
> science + technology = better workers
>
>
>
> +815.588.3846 (Office)
>
> +267.334.4143 (Mobile)
>
>
>
> http://www.alanmead.org
>
>
>
> Announcing the Journal of Computerized Adaptive Testing (JCAT), a
>
> peer-reviewed electronic journal designed to advance the science and
>
> practice of computerized adaptive testing: http://www.iacat.org/jcat