[R] Results of CFA with Lavaan

John Fox jfox at mcmaster.ca
Thu Jun 9 00:45:00 CEST 2011


Dear Sam,

> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org]
> On Behalf Of R Help
> Sent: June-08-11 5:57 PM
> To: John Fox
> Cc: r-help
> Subject: Re: [R] Results of CFA with Lavaan
> 
> Yes, that is the difference.  For the last SEM I built I fixed the
> factor variances to 1, and I think that's what I want to do for the CFA
> I'm doing now.  Does that make sense for a CFA?

Sure -- then the factor covariances are correlations. The point is that you
have to do something to fix the metrics of the factors and identify the
model.

> 
> I'll try figuring out how to do that with lavaan later, but my model
> takes so long to fit that I can't try it right now.

Maybe that should tell you something about the conditioning of the problem.

Best,
 John

> 
> Thanks,
> Sam
> 
> On Wed, Jun 8, 2011 at 5:58 PM, John Fox <jfox at mcmaster.ca> wrote:
> > Dear Sam,
> >
> > In each case, the first observed variable is treated as a "reference
> > indicator" with its coefficient fixed to 1 to establish the metric of
> > the corresponding factor and therefore to identify the model. If you
> > didn't do the same thing (or something equivalent, such as fixing the
> > factor variances to 1) in specifying the model to sem::sem(), that
> > might account for the problems you encountered.
> >
> > Best,
> >  John
> >
> > --------------------------------
> > John Fox
> > Senator William McMaster
> >  Professor of Social Statistics
> > Department of Sociology
> > McMaster University
> > Hamilton, Ontario, Canada
> > http://socserv.mcmaster.ca/jfox
> >
> >
> >
> >> -----Original Message-----
> >> From: r-help-bounces at r-project.org
> >> [mailto:r-help-bounces at r-project.org]
> >> On Behalf Of R Help
> >> Sent: June-08-11 4:15 PM
> >> To: r-help
> >> Subject: [R] Results of CFA with Lavaan
> >>
> >> I've just found the lavaan package, and I really appreciate it, as it
> >> seems to succeed with models that were failing in sem::sem.  I need
> >> some clarification, however, in the output, and I was hoping the list
> >> could help me.
> >>
> >> I'll go with the standard example from the help documentation, as my
> >> problem is much larger but no more complicated than that.
> >>
> >> My question is, why is there one latent estimate that is set to 1
> >> with no SD for each factor?  Is that normal?  When I've managed to
> >> get sem::sem to fit a model this has not been the case.
> >>
> >> Thanks,
> >> Sam Stewart
> >>
> >> HS.model <- ' visual  =~ x1 + x2 + x3
> >>               textual =~ x4 + x5 + x6
> >>               speed   =~ x7 + x8 + x9 '
> >> fit <- sem(HS.model, data=HolzingerSwineford1939) summary(fit,
> >> fit.measures=TRUE) Lavaan (0.4-8) converged normally after 35
> >> iterations
> >>
> >>   Number of observations                           301
> >>
> >>   Estimator                                         ML
> >>   Minimum Function Chi-square                   85.306
> >>   Degrees of freedom                                24
> >>   P-value                                        0.000
> >>
> >> Chi-square test baseline model:
> >>
> >>   Minimum Function Chi-square                  918.852
> >>   Degrees of freedom                                36
> >>   P-value                                        0.000
> >>
> >> Full model versus baseline model:
> >>
> >>   Comparative Fit Index (CFI)                    0.931
> >>   Tucker-Lewis Index (TLI)                       0.896
> >>
> >> Loglikelihood and Information Criteria:
> >>
> >>   Loglikelihood user model (H0)              -3737.745
> >>   Loglikelihood unrestricted model (H1)      -3695.092
> >>
> >>   Number of free parameters                         21
> >>   Akaike (AIC)                                7517.490
> >>   Bayesian (BIC)                              7595.339
> >>   Sample-size adjusted Bayesian (BIC)         7528.739
> >>
> >> Root Mean Square Error of Approximation:
> >>
> >>   RMSEA                                          0.092
> >>   90 Percent Confidence Interval          0.071  0.114
> >>   P-value RMSEA <= 0.05                          0.001
> >>
> >> Standardized Root Mean Square Residual:
> >>
> >>   SRMR                                           0.065
> >>
> >> Parameter estimates:
> >>
> >>   Information                                 Expected
> >>   Standard Errors                             Standard
> >>
> >>
> >>                    Estimate  Std.err  Z-value  P(>|z|) Latent
> variables:
> >>   visual =~
> >>     x1                1.000
> >>     x2                0.554    0.100    5.554    0.000
> >>     x3                0.729    0.109    6.685    0.000
> >>   textual =~
> >>     x4                1.000
> >>     x5                1.113    0.065   17.014    0.000
> >>     x6                0.926    0.055   16.703    0.000
> >>   speed =~
> >>     x7                1.000
> >>     x8                1.180    0.165    7.152    0.000
> >>     x9                1.082    0.151    7.155    0.000
> >>
> >> Covariances:
> >>   visual ~~
> >>     textual           0.408    0.074    5.552    0.000
> >>     speed             0.262    0.056    4.660    0.000
> >>   textual ~~
> >>     speed             0.173    0.049    3.518    0.000
> >>
> >> Variances:
> >>     x1                0.549    0.114    4.833    0.000
> >>     x2                1.134    0.102   11.146    0.000
> >>     x3                0.844    0.091 9.317 0.000
> >>     x4                0.371    0.048    7.778    0.000
> >>     x5                0.446    0.058    7.642    0.000
> >>     x6                0.356    0.043    8.277    0.000
> >>     x7                0.799    0.081    9.823    0.000
> >>     x8                0.488    0.074    6.573    0.000
> >>     x9                0.566    0.071    8.003    0.000
> >>     visual            0.809    0.145    5.564    0.000
> >>     textual           0.979    0.112    8.737    0.000
> >>     speed             0.384    0.086    4.451    0.000
> >>
> >> ______________________________________________
> >> R-help at r-project.org mailing list
> >> https://stat.ethz.ch/mailman/listinfo/r-help
> >> PLEASE do read the posting guide http://www.R-project.org/posting-
> >> guide.html and provide commented, minimal, self-contained,
> >> reproducible code.
> >
> >
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-
> guide.html
> and provide commented, minimal, self-contained, reproducible code.



More information about the R-help mailing list