Message/Author 

Anonymous posted on Saturday, May 14, 2005  6:18 pm



Hi I am new to multilevel modeling/growth curve modeling and am in need of a VERY basic answer to a simple (and probably silly) question. Could you please help me understand the basic differences between HLM and MPlus? I have a sample of parentchild interactions over 4 time points and want to look at linear change in positive interaction patterns over time. I have some specific child factors (i.e. temperament, medical history) that I want to use as predictors of slope and intercept. Can I do this in either HLM or MPlus or is one better suited than the other? Thank you. 

bmuthen posted on Saturday, May 14, 2005  6:33 pm



In terms of standard growth modeling, Mplus and HLM largely overlap. Mplus has advantages of allowing more general modeling. For example, if you want your child factors as predictors of slope and intercept, in HLM you would have to enter a score such as the sum of the items as a proxy for the factor, whereas with Mplus you could have the factor model with its multiple indicators as part of the growth model, avoiding the bias of unreliable scores. Growth mixture modeling is another advantage of Mplus (see for instance Muthén, B. & Muthén, L. (2000). Integrating personcentered and variablecentered analyses: Growth mixture modeling with latent trajectory classes. Alcoholism: Clinical and Experimental Research, 24, 882891. Contact the first author; mention paper #85). 

Anonymous posted on Sunday, May 15, 2005  2:29 pm



another basic question: In standard LGM (no covariates, only one growth factor) the correlation of the intercept and the slope is identical to a direct effect of s ON i (regarding model fit, covariances etc.). Although this effect is simply a direct function of the choice of the position of i, I am interested in this effect. In GMM, with varying means, var and cov across 2 classes, this does not seem to hold true anymore, why? More precisely: %c#1% i s  t1@0 t2@1 t3* t4* t5*; i* s*; i WITH s*; %c#2% i s  t1@0 t2@1 t3* t4* t5*; i* s*; i WITH s*; converges without problems and yields reasonable results, whereas %c#1% i s  t1@0 t2@1 t3* t4* t5*; i* s*; s ON i; i WITH s@0; %c#2% i s  t1@0 t2@1 t3* t4* t5*; i* s*; s ON i; i WITH s@0; does not converge and gives me a couple of error messages such as: THE MODEL ESTIMATION DID NOT TERMINATE NORMALLY DUE TO AN ILLCONDITIONED FISHER INFORMATION MATRIX. Clearly, I am missing something here...what is it? Thank you very much, Manuel 


It's difficult to say what is happening from the information provided. First of all, if you are not using Version 3.12, you should download it from Product Support. If you are, please send your input/output, data, and license number to support@statmodel.com. 

Anonymous posted on Monday, May 16, 2005  3:30 pm



thank you very much for your quick reply! I was using version 3.11 and just updated to 3.12. Now the same syntax runs without any problems  problem resolved:)) Manuel 

Anonymous posted on Tuesday, June 21, 2005  8:29 pm



Here's real dumb one  Do all growth models have latent variables? Is that the nature of growth models? If so, what would be my latent variable if I was modeling the growth over time of kids reading levels? (Putting aside any classes they are taking, etc.) 

rich jones posted on Wednesday, June 22, 2005  1:02 am



Here's my perspective: There are several general approaches to modeling change over time. Many involve modeling the idea that individual persons differ from one another in their starting level and in the pace of change over time. Sometimes these individual differences are called 'random effects'. Sometimes they are called (and modeled as) latent variables. Unless you're doing something fancy, it dosen't really matter what you call them or how you model them. Convince yourself of the equivalence of various approaches by viewing worked examples on Patrick Curran's web page (http://www.unc.edu/%7Ecurran/example.html). So the situation you describe could be approached from a latent growth modeling persective, and the latent variables might represent individual differences in the starting value and rate of change over time. 

Anonymous posted on Wednesday, June 22, 2005  5:59 pm



Thanks, Rich. I see, I was just letting the terminology throw me! I appreciate your help! 

Ray posted on Friday, April 21, 2006  6:32 pm



I am running a CFA with 40 items and 6 factors using Mplus 4. I got the following warning and was trying to locate TECH4 Output for more information. Thanks so much for your help. THE MODEL ESTIMATION TERMINATED NORMALLY WARNING: THE LATENT VARIABLE COVARIANCE MATRIX (PSI) IS NOT POSITIVE DEFINITE. THIS COULD INDICATE A NEGATIVE VARIANCE/RESIDUAL VARIANCE FOR A LATENT VARIABLE, A CORRELATION GREATER OR EQUAL TO ONE BETWEEN TWO LATENT VARIABLES, OR A LINEAR DEPENDENCY AMONG MORE THAN TWO LATENT VARIABLES. CHECK THE TECH4 OUTPUT FOR MORE INFORMATION. PROBLEM INVOLVING VARIABLE F6. 


Ask for TECH4 in the OUTPUT command to obtain TECH4. 


Is there any problem having latent growth curve parameters predict two correlated continous manisfest variables using an MLR estimator. I don't think there are any problems, but could be wrong. The model is specified below: i s  cu1@0 cu2@.5 cu3@1 cu4@1.5 cu5@2 cu6@3 cu7@4; y1 y2 on i s; y1 with y2; 


I can't see any problems with this. 


Hi, I am trying to run growth model with longitudinal data. I have a file for all children. I want to use the same file for all the analyses. So I was thinking to use Useobservations and separate kids according to grade. I was writing (to take only 5th and 6th graders) : Useobservations = grade eq 5 or grade eq 6; But it doesn’t want to run. Am I doing something wrong? 


I can't answer that based on the information you have provided. Please send your input, data, output, and license number to support@statmodel.com. 


I am a PhD student working with LCA for a continuous variable. I have your book, and took the course at University of Maryland in May 2006, but I am struggling to work with model fit, and deciding the number of classes. For example, what represents an acceptable log likelihood, what targets for BIC, AIC are desirable. Are there a few seminal papers that use LCA that really walk the beginning student through that process of understanding these decision processes. LCA is a difficult technique for the new student to gain entry. Thanking you in advance. Sandra 


You want to look at several things to decide on the number of classes including whether the classes make sense. Bengt has a paper in the book edited by Kaplan. You can find this paper on the website under Recent Papers. I believe it looks at a growth model but the steps are the same. You are not looking for absolute loglikelihood and BIC values. You are looking for the highest loglikelihood and the lowest BIC when comparing 2, 3, and 4 classes for example. 

Wei Chun posted on Thursday, November 13, 2008  3:07 am



Dear Muthen, What's wrong with my data? Mplus gives the following message: *** ERROR Unexpected end of file reached in data file. Thank you for your help. 


Please send your input, data, output, and license number to support@statmodel.com. 

Anne Chan posted on Saturday, January 16, 2010  9:57 pm



Hello! I need to argue why I select using MPLUS but no HLM in my presentation in doing my LGC analysis. I am looking for what kind of analysis results are only provided in Mplus but not HLM. Is it the correlation between slope and intercept only provided in Mplus, but not HLM? What results are only able to achieve by Mplus? Thanks! 


You may want to take a look at: Muthén, B. & Asparouhov, T. (2009). Beyond multilevel regression modeling: Multilevel analysis in a general latent variable framework. To appear in The Handbook of Advanced Multilevel Analysis. J. Hox & J.K Roberts (eds). Taylor and Francis. which is on our web site under Papers, Multilevel SEM. The slopeintercept correlation is available in all multilevel programs. 

Anne Chan posted on Sunday, January 17, 2010  5:24 pm



Thanks! I have another basic question. I was following the mplus online video course, in which I learnt that the "intercept" of LGM can be interpreted as the "initial level" of the growth curve. But why does the value of the intercept not equal to the value of the mean of time one? Is that "initial level" not equal to "time one mean"? 


The intercept growth factor is defined by the time point where the time score is zero. If this is the first time point, the mean of the intercept growth factor is equal to the model estimated mean of the first measurement not the observed mean. They are the same only if the model fits perfectly. 


I also ask the very basic question. I ran a couple of growth modeling (chapter 6 in the guidebook). Is this comparable to GEE in SAS? If so, what is the default covariance structure in mplus analysis? Thank you in advance to answer my basic question. 


No Mplus uses maximum likelihood estimation, not GEE. The default covariance structure when doing growth modeling in the wide, singlelevel format of UG chapter 6 is that each time point has its own residual variance, where the residuals are uncorrelated across time. Many other residual correlation forms are possible in Mplus using WITH statements for residual covariances. For specification of autocorrelated residuals, see UG ex 6.17. 

Emil Coman posted on Thursday, May 17, 2012  3:37 pm



I prepared a response to reviewers on GEE vs GLVM in Mplus, summarized here [I may be wrong on some things, like 2!]: The advantages of generalized latent variable modeling (GLVM, Skrondal & RabeHesketh 2004) over GEE come down to: 1. GEE is primarily a regression modeling approach that cannot be used for simultaneous regressions estimation, and therefore GEE does not model links between DVs or IV variables or regression parameters for that matter, across different regressions (including indirect effects; see e.g. xtgee Stata command, a panel data 1DVatatime modeling). 2. It is also clear that GLVM is more flexible in modeling relatively few time points (‘short panel’), where crosslag causal links are known to vary dramatically by lag, especially when an intervention is conducted, while models like xtgee may be more appropriate for ‘long panels’ (Cameron & Trivedi, 2009, Ch. 8, p. 265). 3. The robust estimators used in Mplus, especially for categorical and count variables, WLSMV = robust weighted least squares estimator with diagonal matrix, perform as well or better than GEE (Muthén, du Toit, & Spisic, 1997) for oneregression models. Furthermore, GEE “treats the betweensubjects correlations as nuisance parameters” Cameron McIntosh, SEMNET (for clustered data, for repeated measures the withinsubject correlations are treated as nuisance). 

Emil Coman posted on Thursday, May 17, 2012  3:37 pm



part 2 4. The intraindividual clustering (the only clustering here due to repeated measures component of the model) is modeled specifically rather than ‘corrected for’ as in GEE. The repeated measurement nonindependence in the data is flexibly modeled in GLVM framework however through various means, e.g. correlated errors of repeated measures (of lag one, or possibly more). 5. Model fit/adequacy with GEE is not easy to evaluate, see Skrondal, & RabeHesketh, 2004 (p. 199). 


Hello, I am looking at change in x over time on an outcome variable. I am doing latent growth curve modelling. I understand that I can look at how the intercept (initial level of x) and slope (change in x over time) may relate to the outcome. But I am wondering whether I could also regress the last repeated measurement point of x onto the outcome to also look at whether the latter level of x uniquely impacts the outcome variable? That way, I get information about initial level, change, and later level of x on the outcome. Would this work? 


Yes, you can do this. 


I'm new to Mplus and trying to understand why my fairly simple model is not converging. I eventually want to build on this model, but starting simple. Variables names mean fall scores, winter scores, spring scores Here is my model: Model: i3 s3  f11score@0 w12score@1 s12score@2; Message: NO CONVERGENCE. NUMBER OF ITERATIONS EXCEEDED. I've read the manual and did TECH1 output, that gave me a new start point i3 s3  f11score@0 w12score@1 s12score@2; s3 WITH i3*0.65800; I've tried this, but still didn't work. I also read in the manual to up the iterations, but I'm not sure how to do that..Anyway suggestions? 


Please send the output and your license number to support@statmodel.com. 


I am trying to run a growth curve model with data from 3 time points using the following syntax: MODEL: intercept linear Y3_INT@0 Y5_INT@2 Y9_INT@6; [Y3_INTY9_INT@0]; [intercept linear]; LINEAR BY Y5_INT ANALYSIS: ESTIMATOR IS MLR; PLOT: Type= plot3; Series = Y3_INT Y5_INT Y9_INT (*); When I add LINEAR BY Y3_INT as a modification I get the error message: WARNING: THE LATENT VARIABLE COVARIANCE MATRIX (PSI) IS NOT POSITIVE DEFINITE. THIS COULD INDICATE A NEGATIVE VARIANCE/RESIDUAL VARIANCE FOR A LATENT VARIABLE, A CORRELATION GREATER OR EQUAL TO ONE BETWEEN TWO LATENT VARIABLES, OR A LINEAR DEPENDENCY AMONG MORE THAN TWO LATENT VARIABLES. CHECK THE TECH4 OUTPUT FOR MORE INFORMATION. PROBLEM INVOLVING VARIABLE LINEAR. In Tech4 I see negative and greater than 1 correlations and ESTIMATED COVARIANCE MATRIX FOR ESTIMATED COVARIANCE MATRIX FOR THE LATENT VARIABLES LINEAR INTERCEP ________ ________ LINEAR 0.001 INTERCEP 0.132 1.672 However, I am not sure exactly what this means or how it can be addressed. I would be grateful for some advice on this. Many thanks 


I don't understand what you want to accomplish by saying LINEAR BY Y5_INT The Y5_INT variable is already influenced by the LINEAR growth factor because of your growth model statement. 


I started with the following model syntax: intercept linear Y3_INT@0 Y5_INT@2 Y9_INT@6; [Y3_INTY9_INT@0]; [intercept linear]; This is the model fit: MSEA (Root Mean Square Error Of Approximation) Estimate 0.135 90 Percent C.I. 0.094 0.180 Probability RMSEA <=.05 0.000 CFI/TLI CFI 0.751 TLI 0.254 The output gives these MIs which all involve the linear or intercept growth factors, although as you say this would not make sense to add to the model: Minimum M.I. value for printing the modification index 3.830 M.I. E.P.C. Std E.P.C. StdYX E.P.C. BY Statements INTERCEP BY Y3_INT 27.420 0.372 0.439 0.253 INTERCEP BY Y5_INT 27.421 0.248 0.293 0.161 INTERCEP BY Y9_INT 27.422 0.745 0.879 0.549 LINEAR BY Y3_INT 27.423 8.076 1.840 1.060 LINEAR BY Y5_INT 27.423 5.384 1.227 0.674 LINEAR BY Y9_INT 27.417 16.151 3.680 2.298 So my question is about if and how I can improve the model fit? Any advice would be appreciated. Many thanks again. 


This is covered by the video and handout for Topic 3 of our short courses, slides 5152. Check the linearity assumption by looking at the means and the individual curves. This leads to ideas about freeing time scores as we discuss. Instead of freeing LINEAR BY Y5_INT, free the time score for Y5_INT. 

Back to top 