pspp-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [bug #48040] GLM produces wrong output


From: Alan Mead
Subject: Re: [bug #48040] GLM produces wrong output
Date: Sat, 28 May 2016 21:31:25 -0500
User-agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:45.0) Gecko/20100101 Thunderbird/45.1.0

John,

Sorry for the delay.

Attached is the SPSS output.

I also ran PSPP and I notice that PSPP calculates negative SS for this analysis, which produces negative MS, Negative F's, etc. 

If I add "select if( not sysmis(agree_score))." before GLM, I reproduce the SPSS output.  Seems like not treating missing values is part (or the entirety?) of the problem? I wondered if perhaps SPSS 24 might mark missing values in a new way that is confusing to PSPP? But that's pure hypothesizing...

-Alan


On 5/28/2016 3:43 AM, John Darrington wrote:
On Fri, May 27, 2016 at 12:11:20PM -0500, Alan Mead wrote:
     
     
     From the output, it's possible this is all (or much) of the issue is due
     to the df being calculated wrong.  The error df somehow came out
     negative, which caused the error MS to be negative, which caused all the
     F's to be negative.

I had a brief look, and it seems that the first problem is the Mean Square Error
is negative.  If that is wrong, then almost everything else will be.
This comes from Jason's sweep operator.  Unfortunately Jason is no longer involved
in PSPP.
     

Let's start with a simpler example.  Can you tell me what SPSS does with this
syntax (using your original personality.sav):


GET FILE="personality.sav".

recode caution_score (lo thru 35=1) (36 thru hi=2) (missing=2) (else=copy) into x1.
recode extra_score (lo thru 31=1) (32 thru hi=2) (missing=2) (else=copy) into x2.
frequencies x1 x2.

GLM agree_score BY x1 x2.


     


-- 

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

I've... seen things you people wouldn't believe...
functions on fire in a copy of Orion.
I watched C-Sharp glitter in the dark near a programmable gate.
All those moments will be lost in time, like Ruby... on... Rails... Time for Pi.

          --"The Register" user Alister, applying the famous 
            "Blade Runner" speech to software development

Attachment: glm_output3.pdf
Description: Adobe PDF document


reply via email to

[Prev in Thread] Current Thread [Next in Thread]