Confirmatory Factor Analysis and Measurement Models
http://openmx.psyc.virginia.edu/taxonomy/term/21/0
enUnstandardized ML factor scores
http://openmx.psyc.virginia.edu/thread/4005
<p>I'd like to obtain unstandardized factor scores for a single factor CFA, with the factor scores on the same metric (approximately the same mean, SD, range, distribution) as the indicators. There is missingness in the data, so ML factor scores are preferred. How can I do this in OpenMX? I read Appendix A from Estabrook and Neale's (2013) paper on estimating ML factor scores in OpenMx:<br />
<a href="http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3773873/" title="http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3773873/">http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3773873/</a></p>
<p>Does this approach calculate standardized or unstandardized factor scores? If standardized, how can I generate unstandardized factor scores with ML instead? Note that I don't want to transform standardized factor scores to unstandardized ones because the distributions of my indicators are non-normal (i.e., the normal distribution of standardized factor scores cannot be easily transformed to the raw metric of my indicators). In other words, I'd like to estimate unstandardized factor scores on the raw metric of the indicators without first estimating them on a standardized metric.</p>
http://openmx.psyc.virginia.edu/thread/4005#commentsConfirmatory Factor Analysis and Measurement ModelsTue, 16 Jun 2015 04:26:23 +0000dadrivr4005 at http://openmx.psyc.virginia.eduNeed help
http://openmx.psyc.virginia.edu/thread/4001
<p>Hello Everyone, Please HELP</p>
<p>I am trying to test CFA with AMOS and I have some error messages that I couldn't understand.</p>
<p>I parceled the items in some scales but I recieve messages such as (Probability level cannot be computed)<br />
OR</p>
<p>The model is probably unidentified. In order to achieve identifiability, it will probably be necessary to impose 1 additional constraint.</p>
<p>Anyone has any idea why the models didn't run? if you need more info please let me know</p>
<p>Thank you so much<br />
Tamara</p>
http://openmx.psyc.virginia.edu/thread/4001#commentsConfirmatory Factor Analysis and Measurement ModelsWed, 03 Jun 2015 03:43:10 +0000TQum4001 at http://openmx.psyc.virginia.eduKernel smoothing and factor analysis
http://openmx.psyc.virginia.edu/thread/3945
<p>Good day,</p>
<p>For my data set, I need to perform kernel smoothing with factor analysis to get plots for the parameters of the factor model. Using OpenMx to do factor analysis, how do I integrate kernel smoothing (especially kernel weights) with OpenMx? Could somebody provide an example? I have to show how a certain parameters evolves through time.</p>
<p>Thank you for your input.</p>
<p>Rgeards,</p>
<p>metavid</p>
http://openmx.psyc.virginia.edu/thread/3945#commentsConfirmatory Factor Analysis and Measurement ModelsTue, 27 Jan 2015 21:58:14 +0000metavid3945 at http://openmx.psyc.virginia.eduHigher-order factors, Matrix Specification
http://openmx.psyc.virginia.edu/thread/3872
<p>Hi,</p>
<p>Is this possible to model higher-order factors using matrix specification instead of path specification in OpenMx?</p>
<p>regards,<br />
Krzysiek</p>
http://openmx.psyc.virginia.edu/thread/3872#commentsConfirmatory Factor Analysis and Measurement ModelsThu, 14 Aug 2014 14:31:15 +0000krzysiek3872 at http://openmx.psyc.virginia.eduIntegration of sampling weights in a factor model
http://openmx.psyc.virginia.edu/thread/3661
<p>Good day,</p>
<p>I saw a thread discussing about integration of sample weights with likelihood but I still was not able to work out the code given in that thread. The code I used and worked out is stated below (from an earlier thread):</p>
<p>fullModel <- mxModel("ThisIsHowYouDoWeights",<br />
mxModel("firstModel",<br />
....<br />
mxFIMLObjective("myCov", "myMeans", dims, vector=TRUE)),<br />
mxData(myData, "raw"),<br />
mxAlgebra(-2 * sum(data.weight %x% log(firstModel.objective), name="obj"),<br />
mxAlgebraObjective("obj")<br />
)</p>
<p>with data.weight being a vector (in my case) and the firstModel.objective (also a vector) is calculated differently (not with vector=TRUE) because it was not working for me.</p>
<p>My analysis is about estimation of parameters from a one-factor model (with sampling weights). The model without the sampling weights (normal model) works alright but there were some issues with extracting the likelihood vector when vector=TRUE on mxFIMLObjective. I then decided to use another code from the website of OpenMx (using the formula of row-likelihood) and I was able to extract the likelihoods with it. Now, with available sample weights and the vector of likelihoods, I was wondering if there is another way to tackle this issue? Also, in my case, because data.weight is a vector as well the firstModel.objective, I multiplied them both (*) instead of using Kronecker multiplication (I thought it was correct?).</p>
<p>Thank you in advance for your suggestions and tips.</p>
<p>metavid</p>
http://openmx.psyc.virginia.edu/thread/3661#commentsConfirmatory Factor Analysis and Measurement ModelsTue, 08 Jul 2014 19:01:05 +0000metavid3661 at http://openmx.psyc.virginia.eduConfidence Intervals generated by OpenMx do not seem to look right to me.
http://openmx.psyc.virginia.edu/thread/3446
<p>Hi all,<br />
I ran a simple regression model in AMOS and repeated it in OpenMx in order to obtain the confidence interval for my regression coefficients and see that the a certain coefficient in my output came out significant (p <.001) in amos but not significant in OpenMx (i.e., confidence interval contains zero). I am not sure what to make of it. Here is the info from the AMOS output:<br />
Estimate S.E. C.R. (Critical Ratio) P.<br />
.276 .077 3.589 ***</p>
<p>The OpenMx output for that estimate and CI are:<br />
Estimate Std. Error Std. Estimate Std.SE<br />
.2757 .0756 .3895 .1069<br />
confidence intervals:<br />
Lbound estimate ubound<br />
.000 .276 .413<br />
And if I didn’t set the lower bound=.00, I would get the confidence interval of [ -.413, .413]</p>
<p>Just by calculating the CI by hand: .276 +/- (1.96 * .0756), I get a CI of (.127524, .423876), which converges with the significance of the AMOS output. This is very mysterious because the rest of the CIs seem to converge with Amos results and my calculations by hand so I am very eager to hearing what your thoughts on this are.<br />
Thank you for your time,<br />
Anne</p>
http://openmx.psyc.virginia.edu/thread/3446#commentsConfirmatory Factor Analysis and Measurement ModelsThu, 26 Jun 2014 04:24:39 +0000AnneN.3446 at http://openmx.psyc.virginia.edulatent scores
http://openmx.psyc.virginia.edu/thread/3245
<p>Hi,<br />
Does anyone know how to estimate latent scores in the following model?<br />
What package should I use?<br />
best regards,<br />
Krzysztof</p>
<table id="attachments" class="sticky-enabled">
<thead><tr><th>Attachment</th><th>Size</th> </tr></thead>
<tbody>
<tr class="odd"><td><a href="http://openmx.psyc.virginia.edu/sites/default/files/Rplot.jpeg">Rplot.jpeg</a></td><td>125.44 KB</td> </tr>
</tbody>
</table>
http://openmx.psyc.virginia.edu/thread/3245#commentsConfirmatory Factor Analysis and Measurement ModelsWed, 11 Jun 2014 20:25:09 +0000krzysiek3245 at http://openmx.psyc.virginia.eduBasic questions on CFA and OpenMx
http://openmx.psyc.virginia.edu/thread/2835
<p>Hi all,<br />
I am VERY new to SEM and also to OpenMx and I am trying to run a very simple CFA in OpenMx, which has 1 latent factor (F1) and 3 indicators of A, B, and C with C as my scaling variable and I have a couple of questions about the output that I got:<br />
(1) The output provides information on "Estimate", "Std. Error", "Std. Estimate", "Std. SE" and my question is: how do I derive the Std. Estimate based on the other info in the output that I got? For example, from indicator A to F1, Estimate is .6987, Std. Error is: .1245, Std. estimate is .6571, and Std. SE is .11715045. So let's say that the output didn't provide the Std. estimate. How do I obtain .6571 based on the other results?</p>
<p>(2) I compared the answers from OpenMx to the answers I got from Amos, and I got all identical results as far as standardized path loadings. However, in Amos, it gives me the standardized estimate for the path between C and F1, but in OpenMx, it only gives me standardized path coefficients for A to F1, and B to F1. Therefore, in my script, how do I get to specify that C is my scaling indicator and still allow OpenMx to estimate its standardized path loading just like in Amos? </p>
<p>Thank you so much in advance for your help on this, and please let me know if I need to clarify anything.<br />
Sincerely,<br />
Anne</p>
http://openmx.psyc.virginia.edu/thread/2835#commentsConfirmatory Factor Analysis and Measurement ModelsWed, 09 Apr 2014 19:26:08 +0000AnneN.2835 at http://openmx.psyc.virginia.eduNeed help in CFA
http://openmx.psyc.virginia.edu/thread/2825
<p>Guys,</p>
<p>I Have a basic question...though I am confused :( I wanted to know whether EFA and CFA is applied construct wise or it is applied on overall model by considering all the construct together. Please help me in this and also share if you have any reference material for this.</p>
http://openmx.psyc.virginia.edu/thread/2825#commentsConfirmatory Factor Analysis and Measurement ModelsWed, 09 Apr 2014 03:32:40 +0000Smriti2825 at http://openmx.psyc.virginia.eduNon Normal Data
http://openmx.psyc.virginia.edu/thread/2630
<p>Dear All,<br />
I need your help, using OpenMx for CFA (non normal data). Now, I am very enjoy with OpenMx than the other SEM-Software.<br />
Thank you very much</p>
http://openmx.psyc.virginia.edu/thread/2630#commentsConfirmatory Factor Analysis and Measurement ModelsFri, 28 Feb 2014 01:09:53 +0000toni.toharudin2630 at http://openmx.psyc.virginia.eduIs a total measurement model always needed in SEM??
http://openmx.psyc.virginia.edu/thread/2621
<p>Greetings!</p>
<p>Is a total measurement model always needed? I have a model with 14 latent variables, with 5-9 indicators for each variables. A total measurement model is such a huge total measurement model. Can I just simply use individual measurement models for each construct and then use path analysis (to use the mean of the indicators for the latent variables, which means, transform the latent variables to observed variables)?</p>
<p>Many thanks!</p>
<p>Best regards</p>
<p>Dan</p>
http://openmx.psyc.virginia.edu/thread/2621#commentsConfirmatory Factor Analysis and Measurement ModelsThu, 20 Feb 2014 14:51:05 +0000sunflowerdd2621 at http://openmx.psyc.virginia.eduEQS 6.2 - error: variable exceeds the number of variable read
http://openmx.psyc.virginia.edu/thread/2459
<p>Hi,<br />
I am non-expert SEM user. I am trying to run a CFA with the EQS 6.2 program. After creating the EQS input file, I tried to run EQS but in the output file I found the following sentence:<br />
*** ERROR *** VARIABLE NUMBER 1, A DEPENDENT VARIABLE, EXCEEDS THE NUMBER OF VARIABLES READ. and so on for all other variables. Could you help me to solve the problem?<br />
Thanks</p>
http://openmx.psyc.virginia.edu/thread/2459#commentsConfirmatory Factor Analysis and Measurement ModelsFri, 22 Nov 2013 15:12:15 +0000flow2459 at http://openmx.psyc.virginia.eduOpenMx runs a Two Factor Model but only returns Zero and NA Values
http://openmx.psyc.virginia.edu/thread/1899
<p>Hi,</p>
<p>I'm trying to reproduce an analysis of a congeneric two- factor- model from a textbook and I can't figure out where I am going wrong. The model has two latent factors (selfevaluation and foreignevaluation scales) with three indicators each. The output I get looks like this:</p>
<p>> summary(congeneric)<br />
free parameters:<br />
[1] lbound ubound<br />
<0 Zeilen> (oder row.names mit Länge 0)</p>
<p>observed statistics: 0<br />
estimated parameters: 0<br />
degrees of freedom: 0<br />
-2 log likelihood: NA<br />
saturated -2 log likelihood: NA<br />
number of observations: 0<br />
chi-square: NA<br />
p: NA<br />
Information Criteria:<br />
df Penalty Parameters Penalty Sample-Size Adjusted<br />
AIC: NA NA NA<br />
BIC: NA NA NA<br />
CFI: NA<br />
TLI: NA<br />
RMSEA: NA<br />
timestamp: NULL<br />
frontend time: NULL<br />
backend time: NULL<br />
independent submodels time: NULL<br />
wall clock time: NULL<br />
cpu time: NULL<br />
openmx version number: NULL </p>
<p>I have tried all sorts of modifications to my code, but nothing helped. It seems like I must be overlooking something terribly important. I would really appreciate some help! I attached my code and the data.</p>
<p>Jens</p>
<table id="attachments" class="sticky-enabled">
<thead><tr><th>Attachment</th><th>Size</th> </tr></thead>
<tbody>
<tr class="odd"><td><a href="http://openmx.psyc.virginia.edu/sites/default/files/TwoFactorModel.r">TwoFactorModel.r</a></td><td>1.42 KB</td> </tr>
<tr class="even"><td><a href="http://openmx.psyc.virginia.edu/sites/default/files/TwoFactor_Data.dat">TwoFactor_Data.dat</a></td><td>3.66 KB</td> </tr>
</tbody>
</table>
http://openmx.psyc.virginia.edu/thread/1899#commentsConfirmatory Factor Analysis and Measurement ModelsMon, 11 Feb 2013 19:30:01 +0000kouros7111899 at http://openmx.psyc.virginia.educhoice of using FIML for binary data
http://openmx.psyc.virginia.edu/thread/1722
<p>Hello,<br />
my question is quite a general newby one on whether to use FIML, and therefore OpenMX. I'd appreciate any comments on which direction to take.</p>
<p>I'm considering using it (and therefore using OpenMX) because I understood it is very well adapted to use with binary data, compared to conventional techniques (using tetrachoric correlation is often problematic, I believe, and WLS requires large numbers). However, the OpenMx Manual cites handling of missing data (which didn't worry me that much, as I was expecting to use listwise deletion):</p>
<p>'The intelligent handling of missing data is a primary reason to use FIML over other estimation techniques' (4.3.1)</p>
<p>In fact, in the data sets I will be analysing, I have between 3,000 and 30,000 cases (test candidates) and around 30 variables (test items), so my worry was mainly tetrachoric correlation, rather than WLS. My aim is to determine and verify dimensional models of two language tests and then compare them.</p>
<p>many thanks for any enlightenment,</p>
<p>Michael Corrigan</p>
http://openmx.psyc.virginia.edu/thread/1722#commentsConfirmatory Factor Analysis and Measurement ModelsFri, 09 Nov 2012 15:32:47 +0000MC1722 at http://openmx.psyc.virginia.eduiteration in one factor model not converge
http://openmx.psyc.virginia.edu/thread/1659
<p>i have four observed vaiables and fit one factor model on these four observed variable.when i am try to fit one factor model on my data which consist of ordered categorical variable with fiv ecategories. when r code run the whole programme warning message appear that "NPSOL is returned a non zero status code 1.The final iterate satisfies the optimality conditions to the accuracy requesed but the sequence of iterates has not yet converged. NPSOL was terminated because no further improvement could be made in the merit function (Mx status GREEN). "</p>
<p>what it mean please give help</p>
<p>iram</p>
<table id="attachments" class="sticky-enabled">
<thead><tr><th>Attachment</th><th>Size</th> </tr></thead>
<tbody>
<tr class="odd"><td><a href="http://openmx.psyc.virginia.edu/sites/default/files/F3.r">F3.r</a></td><td>3.32 KB</td> </tr>
</tbody>
</table>
http://openmx.psyc.virginia.edu/thread/1659#commentsConfirmatory Factor Analysis and Measurement ModelsTue, 16 Oct 2012 14:44:58 +0000iram1659 at http://openmx.psyc.virginia.edu