Re: Warppls

Posted by Rich Ulrich on Sep 10, 2014; 5:50am
URL: http://spssx-discussion.165.s1.nabble.com/Warppls-tp5727195p5727198.html

1.  No.  Kind of by definition:  a constant is not a "variable."

2.  Write a model that makes sense for the data you have. 
Those constants must belong in the description of the data set,
not in the model.

> Date: Tue, 9 Sep 2014 12:28:53 -0700

> From: [hidden email]
> Subject: Warppls
> To: [hidden email]
>
> I have a structural equation model with multiple latent variables. Some have
> four indicators, some have more than ten indicators and one has has three
> indicators but the latent variable with the three indicators has three
> constant values for example (0.55, 0.42, 0.56). /I'm using Warppls software/
> !!!
> Here are my questions:
> 1. Should I consider a variable with the three indicators (or constant
> values) as a latent variable.
> 2. The software automatically delete all identical columns so how can I
> overcome this issue.
> Many thanks,
>

--
Rich Ulrich

===================== 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