Re: Entering Variables in Blocks Using PSPP

[Alan Mead]

What version are you using?

My initial thought is that PSPP doesn't support this, which may well be true. The PSPP manual, does not include the METHOD subcommand in the reference, although the examples show an example like your second one with a single METHOD subcommand:

https://www.gnu.org/software/pspp/manual/html_node/Syntax.html#Syntax
https://www.gnu.org/software/pspp/manual/html_node/Examples.html#Examples

But I opened the src/language/stats/regression.c file in the most recent version of PSPP on my machine (which is 1.2), it looks like it will parse variables after a METHOD unless there was already a VARIABLES command (see below; variables_seen is a true value if /VARIABLES has already been parsed). The code below seems to parse METHOD, then require an equals, then require that the method is ENTER (no other methods are allowed), then IFF variables_seen is false, parse dependent variables. I don't know enough about PSPP internals to know if parse_variables_const() imposes additional conditions.

So, removing the /VARIABLES may allow it to run. But , I don't think PSPP has the logic to perform hierarchical regression. My guess is that if you get it to run, it will ignore blocks and simply perform one regression on all the variables that you enter.

-Alan

On 2/24/2019 4:06 PM, mag wrote:

First, thanks very much to Alan for answering my previous question.  I am now trying to replicate the following SPSS regression language (simplified here for clarity).

* SPSS, Entering Two Blocks, No Errors

REGRESSION
   /STATISTICS COEFF OUTS R ANOVA CHANGE
   /DEPENDENT y
   /METHOD=ENTER var1
   /METHOD=ENTER var2.

Now, I try to do the same thing in PSPP, but am not successful.

* PSPP, Example 1, Entering Two Blocks, ERROR

REGRESSION

   /STATISTICS COEFF CI R ANOVA 

   /VARIABLES= var1 var2

   /DEPENDENT=y

   /METHOD=ENTER var1

   /METHOD=ENTER var2.

Output= syntax error at var1. 

Please note that the following, with one block only and using "/method=enter", works fine.

* PSPP Example 2, Entering 1 Block, No Errors

REGRESSION

   /STATISTICS COEFF CI R ANOVA 

   /VARIABLES= var1 var2

   /DEPENDENT=y

   /METHOD=ENTER.

Please reply if you know how to enter variables in blocks, as shown in the SPSS example (and, if you used a book, website, or other resource,where you found the information). Thank you!

   

_______________________________________________
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

http://www.alanmead.org

"You're an interesting species. An interesting mix. 
You're capable of such beautiful dreams, and such 
horrible nightmares. You feel so lost, so cut off, 
so alone, only you're not. See, in all our 
searching, the only thing we've found that makes 
the emptiness bearable, is each other."

-- Carl Sagan, Contact