Python & PASW

[Anthony Bressy]

Hi,

I would like to know if python is a good way to work with PASW statistics in my job.

What is the added value of python to generate tables or transform & export data (in xls or sav) compared with the syntax or macros.

Thanks in advance.

Anthony

====================To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Jon K Peck]

To get a good idea of how to use Python with Statistics, look at the Programming and Data Management book pdf available in the Books and Articles section of the SPSS Community website at www.ibm.com/developerworks/spssdevcentral.  There are also a number of Ppt presentations in that area that show the benefits of Python and R programmability.


Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: [phone]




From:        Anthony Bressy <[hidden email]>
To:        [hidden email],
Date:        09/06/2013 02:25 AM
Subject:        [SPSSX-L] Python & PASW
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

Hi,

I would like to know if python is a good way to work with PASW statistics in my job.

What is the added value of python to generate tables or transform & export data (in xls or sav) compared with the syntax or macros.

Thanks in advance.

Anthony

====================To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

[Ruben Geert van den Berg]

Dear Anthony,

Take a look at www.pythonforspss.org or just Google "spss python". Having used both Python and Macros, I didn't feel like dedicating more than a few lines to this. The difference speaks for itself.

Around March this year I wrote a module for generating tables. It runs through the data and depending on variable name, label, value labels and measurement level figures out what kind of variable(s) or MR set(s) it's dealing with. It generates and runs syntax (mostly SET TLOOK, then CTABLES including captions with variable labels after cleaning those up temporarily and adding some info). After tidying up the viewer (hiding logs and stuff) with a script, the end result was a very professionally looking report, automatically generated in about a minute (for large data, that is).

My client used to do that with SaxBasic which took about 15 times(!) more code, 10 time more processing time and would crash with more than some 500 variables.