Dear friends,
I have been working with Parscale for some time, and I was
always anxious when the Newton Cycles (after the E-M Cycles
converge) did not converge (actually usually diverge). This
time, its the same story. The Newton Cycles diverge. When I
fiddle with the settings, Parscale manages 2-3 cycles and
then diverges again. Anyone knowing any tricks to make this
creature converge?
Jason
----- Original Message Follows -----
From: Paul Barrett <[log in to unmask]>
To: [log in to unmask]
Subject: Very important paper on SEM modeling
Date: Fri, 5 Jan 2007 13:50:22 +1300
> Hello again
>
> Almost forgot - but I think this is a very important and
> readable paper for anyone contemplating using hierarchical
> factor models in SEM ... It's clearly written, and that
> nested (bifactor) model is a very nice way of modeling a
> general factor. I've used this myself recently ...
>
> Gignac, G. (2007) Multi-factor modeling in individual
> differences research: Some recommendations and
> suggestions. Personality and Individual Differences, 42, 1
> , 37-48.
>
> Abstract
> This paper offers some commentary and recommendations
> relevant the multi-factor modeling in individual
> differences research. Several similarities and
> distinctions between oblique factor modeling, higherorder
> modeling, Schmid-Leiman transformations, and nested
> factors modeling are discussed. An empirical illustration
> of the various multi-factor models is presented, based on
> 18 items derived from three Neuroticism facets within the
> NEO PI-R. Researchers are encouraged to always perform a
> Schmid-Leiman transformation to a higher-order model
> solution, as well as to consider the possibility that a
> nested factors model will yield superior model fit, in
> comparison to a higher-order model, as well as less
> ambiguous factor solutions.
>
>
> Another recent paper on the same topic - but focused more
> in the Quality of Life literature is:
>
> Chen, F.F., West, S.G., and Sousa, K.H. (2006) A
> comparison of bifactor and second order models of quality
> of life. Multivariate Behavioral Research, 41, 2, 189-225.
>
> Abstract
> Bifactor and second-order factor models are two
> alternative approaches for representing general constructs
> comprised of several highly related domains. Bifactor and
> second-order models were compared using a quality of life
> data set (N = 403). The bifactor model identified three,
> rather than the hypothesized four, domain specific factors
> beyond the general factor. The bifactor model fit the data
> significantly better than the second-order model. The
> bifactor model allowed for easier interpretation of the
> relationship between the domain specific factors and
> external variables, over and above the general factor.
> Contrary to the literature, sufficient power existed to
> distinguish between bifactor and corresponding
> second-order models in one actual and one simulated
> example, given reasonable sample sizes. Advantages of
> bifactor models over second-order models are discussed.
>
>
> Regards .. Paul
> ___________________________________________________
>
> Paul Barrett Mob:
> +64-021-415625 www.pbmetrix.com <http://www.pbarrett.net/>
> Skype: pbar088
> [log in to unmask]
>
>
|