After reading Dave's posting about the model reviewer, I was curious. While it's true that I've seen it (and didn't know what it was) for Npar, I was particularly curious about Genlinmixed. I generated some data and ran genlinmixed and also through mplus. The genlinmixed syntax is really complex compared to Mixed (and it may have to be to give the desired flexibility). I think I have the syntax written correctly but I'd like somebody with experience to comment. The generated data have a two level structure: 45 clusters of 50 records each. The dependent variable is y and is dichotomous. There are no level 1 predictors and one level 2 predictor, x. The cluster variable is ij. Thus the level 1 model is just the random intercept and the level 2 model is x regressed on the level 1 intercept. Given that is the following syntax correct?
GENLINMIXED /DATA_STRUCTURE SUBJECTS=ij/FIELDS TARGET=y/
TARGET_OPTIONS DISTRIBUTION=biNOMIAL LINK=LOGIT/fixed effects=x/
RANDOM USE_INTERCEPT=TRUE SUBJECTS=ij COVARIANCE_TYPE=VARIANCE_COMPONENTS.
Thanks, Gene Maguin
-----Original Message-----
From: SPSSX(r) Discussion [mailto:
[hidden email]] On Behalf Of David Marso
Sent: Sunday, March 10, 2013 4:20 PM
To:
[hidden email]
Subject: Re: Model Viewer: Wave of the future or Hideous abomination?
I know how to access the TTSNBN (thing that shall not be named).
My rant simply posits a screech in the dark. A dart aimed at the heart of the matter which will undoubtedly go unheeded as successive versions of TPFKASPSS devolve into glitzy, useless ghosts of their former glory. I suspect part of the IBM final solution is to port all output to the model viewer *CRAP* .
Another chapter in the ongoing "Real Stats Real Easy" undermining of the analytical professions!
Who ever decided statistics should be easy?> Carpentry isn't easy! Being a Chef isn't easy! Being an auto mechanic isn't EASY, Being a lawyer isn't easy.
In fact, training in just about any profession isn't EASY!!!!!!.
Why the f### should statistics be EASY??????
People go to school for years to master the tools and techniques! Why should mathematically illiterate users be coddled by software? WOW: Case studies! That way 'consultants' can be one full help screen ahead of their clients!!!!!
--
Alex Reutter wrote
> There are screenshots of the output in the online help. The top-level
> topic for Linear Models (ALM), for example, is at:
>
http://pic.dhe.ibm.com/infocenter/spssstat/v21r0m0/topic/com.ibm.spss.
> statistics.help/idh_alm.htm ; from there, you should see links to
> topics like "Model Summary", "Predictor Importance", etc.
>
> What these screenshots don't give a good sense of is how you access
> the Model Viewer. When you run ALM, it produces a "Model Viewer"
> object in the Viewer; double-clicking on this object opens the Model
> Viewer in a new window. This is the same view you would see when
> opening a model nugget produced by the Linear node in Modeler
> (formerly Clementine). ALM, NPTESTS, and GENLINMIXED all produce
> Model Viewer output, though as of v21 GENLINMIXED can now produce pivot table output instead.
>
> Alex
-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
--
View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/Model-Viewer-Wave-of-the-future-or-Hideous-abomination-tp5718520p5718525.htmlSent from the SPSSX Discussion mailing list archive at Nabble.com.
=====================
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
=====================
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
Locked
