Message/Author 

Anonymous posted on Sunday, June 20, 2004  11:08 am



I have a model with two continuous latent variables F1 and F2, and one categorical variable X (a binary 0/1 variable). I am interested by the standardized indirect effect from F1 to X. The statements of model are: F2 ON F1; X ON F2; X VIA F2 F1; Is it correct to consider the STDYX coefficient of the indirect effect as standardized probit regression coefficient ? 


Yes because the end dependent variable is binary. 

Anonymous posted on Sunday, June 20, 2004  1:44 pm



Thank you Linda Muthén for your response. 

Anonymous posted on Tuesday, October 05, 2004  10:00 am



Hi Dr. Linda, When we present results of StdYX we must know S.E.or pvalue of StdYX. I'm not sure whether we can use Est./S.E. offered in the default display? If not how can I know the significance level of StdYX? Thanks 


The standard errors given in the output are for the raw coefficients. They are not the standard errors for the standardized coefficients. You would need to compute these standard errors using the Delta method. 

Anonymous posted on Tuesday, October 05, 2004  12:02 pm



Do you mean in Analysis command using Parameterization=delta; ? This is the default paramerization of Mplus. The results are the same with or without using Parameterization=delta. if Type=general. I do not know anything else about Delta method. Could you tell me the syntax needed for calculating S.E. or pvalue of StdYX results? Thanks 


No, it is not PARAMETERIZATION = DELTA. You will need to read about the Delta method for computing standard errors in a book like Bollen's SEM book. 

Anonymous posted on Tuesday, October 05, 2004  12:34 pm



Thanks 

Daniel posted on Wednesday, March 30, 2005  10:34 am



I used bootstrap standard errors to assess the significance of an indirect effect on an ordered categorical dependent variable. The indirecte effect was signficant. Is it possible to compute an odds ratio (exponentiating the log odds Beta) and confidence interval using the indirect effect, or does that not make sense? 

BMuthen posted on Saturday, April 02, 2005  8:27 pm



I think that makes sense if you are using maximum likelihood estimation which uses the logit model. The indirect effect still refers to a slope. 


I have a couple of questions. First, the output for 'model indirect' in Mplus lists the estimates, standard errors, and twotailed pvalues for each direct and indirect effect. My question is: What are the listed pvalues testing? Are these pvalues an indication of whether the indirect effects are significant? Second, my model includes multiple mediators regressed upon each other. For example, one of the indirect paths is SES>Social support>Negative affect>Selfefficacy>Smoking relapse (categorical/binary). Is there a test to determine if this complex mediational/indirect path is significant? Is that what is already reported in the indirect output? thanks so much in advance. 


The test is whether the indirect effect is different from zero. The pvalue is the value for the ztest given in column three, the ratio of the indirect effect to its standard error. If you define the indirect effect as described above, it will be tested against zero. This is what is reported in the output. 

Michael B posted on Tuesday, July 28, 2009  6:20 am



A reviewer requested that I describe how the indirect effects were tested. Can you refer me to a paper that describes what Mplus does to test this type of complex indirect effect? Is there a name for this type of test? 


The standard errors for the indirect effects are estimated using the Delta method. The ratio of the parameter estimate to its standard error is a ztest. 

Jo Brown posted on Wednesday, July 04, 2012  8:30 am



In the post above you mention that the SEs of the indirect effects are estimated using the Delta method. Is this robust to potential bias or should I still use bootstrapping to estimate biascorrected SEs? 


I would use biascorrected confidence intervals. 

Jo Brown posted on Friday, July 06, 2012  2:55 am



Hi Linda, thanks for your reply. Would you this using the bootstrap option or is there an alternative command. Thanks 


See the BCBOOTSTRAP setting of the CINTERVAL option. 

Jo Brown posted on Monday, July 09, 2012  3:55 am



Thanks Linda, I now have one more question re. CIs for mediation analyses as I now need to repeat my analyses on a number of imputed datasets; however, when I specified output: CINTERVAL(bcbootstrap) I receive the following message: CINTERVAL option is not available with multiple imputation. Request for CINTERVAL is ignored. Is there a way around this problem that would allow me to obtain CI for imputed datasets? Could I simply calculate the CIs by using: estimate +or (1.96*SE) in this case? Many thanks, Jo 

Jo Brown posted on Monday, July 09, 2012  9:00 am



one more thing is: are the confidence intervals calculated using the CINTERVAL option based on a normality assumption? 


You can use plus/minus 1.96 times the standard error. The is not BCBOOTSTRAP. See the description of the settings of the CINTERVAL option in the user's guide. 

Jo Brown posted on Monday, July 09, 2012  12:10 pm



thanks Linda. If I use the formula above, should I use bootstrap to obtain bootstrapped SEs or is this not necessary? Are the CIs so calculated not accounting for potential bias? Many thanks again 


That would need to be your decision. The confidence intervals are symmetric. See the user's guide. 

Jo Brown posted on Tuesday, July 10, 2012  12:00 pm



thank you 


Hi Linda, I am using method WLSMV, and I have several binary mediators and binary covariates. I specified STDY in the OUTPUT, and no standardized estimates were produced. When I specified STANDARDIZED, the output only showed two columns w/ STDYX and STD (which are identical to the unstandardized estimates). Do we have to calculate the oneway standardization (STDY) by hand? Or is there a way to get it in the MPlus output? Thanks for considering this question, Selahadin 


You need to convert STDYX to STDY yourself. 


Thanks Linda. 

Elina Dale posted on Wednesday, September 18, 2013  9:37 am



Dear Dr. Muthen, I have 2 groups randomized into trx & contr. As there was a high % of noncompliance, I used CACE to estimate the effect of trx on M, which was my outcome here (Ex 7.23 & 7.24 in MPlus 7 Guide). It worked fine. M is a latent variable measured through 3 f's. Now I need to specify a mediation model (trx>M>y). I modified input commands from paper 1 (I still need to use CACE b/c of high % NC w/ trx)[see below]. But I got error message [below]. I don't know what to change. Please, advise! CATEGORICAL = u i1i9 ; CLASSES = c(2) CLUSTER = clus; Analysis: TYPE = COMPLEX MIXTURE ; Model: %OVERALL% f1 BY i1 i2 i3 ; f2 BY i4 i5 i6 ; f3 BY i7 i8 i9 ; f1 ON trx ; f2 ON trx ; f3 ON trx ; c ON z1 z2 z3 ; y ON f1 ; y ON f2 ; y ON f3 ; %c#1% [u$1@15] ; f1 ON trx ; f2 ON trx ; f3 ON trx ; %c#2% [u$1@15] ; f1 ON trx @0; f2 ON trx @0; f3 ON trx @0; f4 ON trx @0; *** ERROR The following MODEL statements are ignored: * Statements in the OVERALL class: Y ON F1 Y ON F2 Y ON F3 

Elina Dale posted on Wednesday, September 18, 2013  9:40 am



Sorry, it's me again! I am lost because I am not sure how to specify a model when I have to use CACE and I have a mediating latent variable. I couldn't find any such examples in MPlus Guide or the Shrout and other papers. I would greatly appreciate it if you could help me & modify my commands from the previous posting. Thank you! 


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


Dear Linda Muthen, I used the MODEL CONSTRAINT command to calculate indirect effects, and I understand that the standard errors of these indirect effects are computed in Mplus using the multivariate delta method. According to Bollen (1987), this method assumes a normal distribution of the direct paths. However, in my model, the indirect effects are calculated for a combination of linear and loglinear direct paths. In what way would this affect the interpretation of the standard errors of the indirect effects? Thanks in advance, Evelien 


You need to read the paper on our website: Muthén, B. & Asparouhov T. (2014). Causal effects in mediation modeling: An introduction with applications to latent variables. Forthcoming in Structural Equation Modeling. 


Thank you, that is a very helpful paper indeed! 


Dear Bengt Muthén, I tried to ‘translate’ the inputfile in Table 54 from Muthen (2011) to my own model (count Y, continuous X and M, no XM interaction term, only estimating PIE) and I believe I need the following command: MODEL: [DQ1](beta0); DQ1 on rpeer Gend ethn parm SC age(beta1); DQ1 on US Gend ethn parm SC age(beta2); [rpeer](gamma0); rpeer on US (gamma1); rpeer(sig); MODEL CONSTRAINT: new(ey0 mum1 mum0 ay0 bym01 bym00 eym01 eym00 pie); ey0=exp(beta0); mum1=gamma0+gamma1; mum0=gamma0; ay0=2*sig*beta1; bym01=(ay0/mum1+2)/2; bym00=(ay0/mum0+2)/2; eym01=exp((bym01*bym011)*mum1*mum1/(2*sig)); eym00=exp((bym00*bym001)*mum0*mum0/(2*sig)); pie=ey0*eym01ey0*eym00; Is this correct? The estimates for the direct paths to Y have strange values and the estimated indirect effect is unlikely large. Is this because I don’t use Monte Carlo simulations? Evelien 


And as a second question: my model is actually multilevel. I can apply the proposed approach (Muthen, 2011; Muthen & Asparouhov, 2014) at the betweenlevel, but I don’t think I can apply it at the withinlevel, since I cannot specify a mean for Y at the withinlevel. What would be a smart way to calculate the indirect effect at the withinlevel? Could I just use beta1*sig+beta1*gamma1 (given the parameters as specified above)? Thanks in advance! The 2011 and 2014 papers are a great help by the way, Evelien 


Mplus Version 7.2 does the counterfactuallydefined (causal) indirect and direct effects automatically  see the Version 7.2 Mplus Language Addendum. This includes the case of a count outcome. Multilevel counterfactual effects are more complex  see references at the end of the 2011 paper. Perhaps you should instead use Type=Complex to take care of it. 


Hi, I have the following path model with a binary dependent variable and several continuous independent variables. I am using WLSMV to estimate the model: Categorical are Rang_CDU; (all other variables are continuous / dummys). Analysis: Type = general; Model: Rang_CDU on RE_W3_Au W1_UmfrW W1_Dummy W1_Dumm0 alter CDU_Dumm I_Wiss; W1_UmfrW on W1_InBTW I_Strate I_Heuris RE_W3_Au I_Wiss; RE_W3_Au on W1_InBTW I_Wiss; Model indirect: Rang_CDU ind I_Wiss; Rang_CDU ind W1_InBTW; Output: sampstat standardized stdyx modindices; I have some very basic questions: 1. The StdYX path coefficients between the continious dependents and continious independents are standardized regression coefficients, right? 2. The StdYX path coefficients between the binary dependent and continious independents are standardized probit coefficients, right? How can they be interpreted? 3. And can they be compared to other standardized probit coefficients within the same model in terms of strenght? 4. Can they be compared to standardized regression coefficients within the same model in terms of strenght? 5. How can the indirect effects be interpreted including paths with standardized regression coefficients and standardized probit coefficients (e.g. Rang_CDU ind I_Wiss)? Thank you very much for your help! 


1. Yes. 2. Yes, it is the change in the latent response variable. 3. Yes. 4. Only to other probits. 5. Using WLSMV, all dependent variables are continuous so this is not a problem. See the following paper on the website for further information: Muthén, B. & Asparouhov, T. (2014). Causal effects in mediation modeling: An introduction with applications to latent variables. Forthcoming in Structural Equation Modeling. 

Shiny posted on Friday, September 05, 2014  10:36 am



I am also testing a mediaton model with categorical data. I used Model constraint as WLSMV produces latent Response variables. Is the indirect effect coefficient under the new Parameter unstandardized? Can I get standardized indirect effect coefficient? Thanks! 


Q1. Yes. Q2. Yes, as usual, just divide by the estimated SD(Y*) and multiply by the sample SD(X). 

Back to top 