Interpreting STDY vs. STDYX PreviousNext
Mplus Discussion > Structural Equation Modeling >
Message/Author
 Linda Lathroum posted on Monday, April 04, 2011 - 8:06 am
Dr. Muthen,
What is the difference between STDY Standardization and STDYX Standardization in the standardized model results? Which of the two should I interpret when I have direct and indirect effects unto one Dependent latent variable?
Thank you,
Linda
 Linda K. Muthen posted on Monday, April 04, 2011 - 9:09 am
You should use StdYX when covariates are continuous and StdY when covariates are binary.
 Linda Lathroum posted on Tuesday, April 05, 2011 - 8:22 am
Dear Dr. Muthen,
Thank you very much for this clarification.

I have one other question. At what point is skewness problematic (as an indicator of a latent)?

Thank you!
Linda
 Linda K. Muthen posted on Tuesday, April 05, 2011 - 9:33 am
Unless you have a preponderance of zeroes, estimators like MLR that are robust to non-normality should handle skewness for continuous indicators. Floor or ceiling effects for categorical indicators are handled by categorical data methodology.
 Linda Lathroum posted on Wednesday, April 06, 2011 - 12:31 pm
Thank you Dr. Muthen,
Linda
 Ahmed Shafik posted on Thursday, June 14, 2012 - 3:59 am
Dear Dr. Muthen,

I am estimating a logistic regression (MLR) with some covariates that are continuous and others that are binary. I am using exclusively observed variables and no indirect effects. It is a simple logistic regression as you use to estimate in Stata or SPSS.

Should I use StdYX for continous and StdY for the binary covariates even when the two types of covariates are in one and the same model? I have a strange feeling by using for some covariates the coefficients from one part of the output and for the other covariates the coefficients from an other part of the output. Is this really advisable?


Best Regards
 Linda K. Muthen posted on Thursday, June 14, 2012 - 11:17 am
With logistic regression, I would standardize the coefficients with continuous covariates with respect to x only. I would not standardize the coefficients with binary covariates.
 Ahmed Shafik posted on Thursday, June 14, 2012 - 3:26 pm
If I understand right, I should use the coefficients under the output block titled StdYX for the continuous covariates while for the binary covariates I should use the non-standardized coefficients at the top of the output?

Is it really advisable to use for some covariates one specific part of the output while for other covariates a different part of the output is used?


Best Regards
 Linda K. Muthen posted on Thursday, June 14, 2012 - 3:49 pm
No, for continuous covariates you need StdX which you will have to create yourself. See the STANDARDIZED option in the user's guide where you will find formulas to help you do that.

You should report all of the raw results and report standardized only for variables with continuous covariates.
 Michael Lorber posted on Thursday, January 17, 2013 - 11:53 am
I am trying to make sense of differences in the p-values and meanings of the unstandardized vs. standardized (STDYX) covariances in the following output. All the variables in the model are continuous and both have missing data, so I am using FIML. Estimator = MLR. Typically, I like to report the STDYX results because they’re in an easily understood metric, but I am unnerved by such differences. I have several similar examples in regression models.

If the only difference is that STDYX is calculating the covariance having rescaled each variable to have M=0 and SD=1, I would expect the p-values to agree perfectly. Since that’s not the case, I’m unsure what Mplus is doing…how these inconsistent p-values occurred. Also, which results are “more correct.” And is the answer any different if I’m doing regression models with continuous DVs and predictors?

Thanks for your input!


MODEL RESULTS

Y WITH X 0.031 0.019 1.645 0.100


STDYX Standardization
Y WITH X 0.322 0.156 2.062 0.039
 Linda K. Muthen posted on Thursday, January 17, 2013 - 5:35 pm
The p-values differ for unstandardized and standardized parameters because the parameters have different sampling distributions.
 Bengt O. Muthen posted on Thursday, January 17, 2013 - 5:39 pm
You can try Bayes and see what this says about the 95% CIs in this case.
 Michael Lorber posted on Friday, January 18, 2013 - 11:46 am
Thanks for the feedback!

Leaving Bayes est aside for the moment, which results would you put more stock in, the unstandardized (n.s.) or the standardized (sig)? Is it arbitrary because they're both "correct"? Sigh, I'm sure this is where the people who advocate getting rid of null hypothesis testing would pipe up.
 Bengt O. Muthen posted on Friday, January 18, 2013 - 12:23 pm
See if Bayes shows disagreement for the 2. Bayes also tells you if the distribution is non-normal. You wouldn't trust an ML z-score if the distribution is non-normal.
 Michael Lorber posted on Sunday, January 27, 2013 - 11:51 am
So you're suggesting that Bayes is "better" than either the regular unstandardized or standardized solutions? I should trust the result that is more consistent with the Bayes result? If so, does that imply that I should be using Bayes all the time?

By the way, I've been using MLR because the IV and DV distributions are nonnormal.

Thanks again.
 Linda K. Muthen posted on Monday, January 28, 2013 - 10:33 am
Bayes does not assume that the sampling distributions of the parameters are normal as frequentist methods do. It is not clear how robust Bayes is to non-normality of the data. If your data are non-normal and the sampling distributions of the parameters are non-normal, you may be best off with ML and bootstrapping.
 Sarah  posted on Thursday, May 15, 2014 - 5:12 am
Dear Dr Muthen

I am running a multiple mediation model with two dummy independent variables, 5 continuous mediators and 4 latent outcome variables. When reporting standardized output am I right in thinking I should report STDY for the path estimates between the independent variables and the latent factors and STDYX for the path estimates between each of my mediators and the latent factors?
Also when reporting standardized indirect effects should I use STDY as the independents are binary variables?

Many thanks for your help.
 Linda K. Muthen posted on Thursday, May 15, 2014 - 10:09 am
Use StdYX unless the covariate is binary. Then use StdY.
 Stat posted on Wednesday, September 03, 2014 - 9:21 pm
Hello there,

For factor loadings, should we use STDY for binary variables and STDYX for continuous/ordinal/latent variables, or if we can use both, without consideration for the scale of the variables ?

Thank you
 Linda K. Muthen posted on Thursday, September 04, 2014 - 9:01 am
The factor model is

factor indicators ON factors

You should standardize by the factor (continuous covariate) and the factor indicator so StdY. This is for both binary and continuous factor indicators.
 Stat posted on Friday, September 05, 2014 - 12:32 pm
Thank you for the answer.

I would have one more question for factor loading coefficients.

When I am using a WLSMV estimator, there is no StdY coefficients in the output. Does that mean that it is not possible to obtain StdY coefficents when using WLSMV estimator ?
 Bengt O. Muthen posted on Friday, September 05, 2014 - 2:11 pm
If you are talking about standardizing loadings, just use STDYX. In this case, it is the same as what you would get with STDY.
 Athanasios Mouratidis posted on Monday, February 01, 2016 - 1:49 am
Dear Dr. Muthen,
I have a practical question regarding the numerical values I am getting in a two-level model.
Although the observed scores of the variables in my model are between 1 and 5, and although I have already centered the scores at the higher level, the values that I am getting at the between level are excessively high under the STDYX standardization section (the same applies under the STDY standardization section). Specifically, although the paths from the between level to the outcomes seem reasonable (i.e., .71 and -.74) the intercepts are as high as 11.46 and 10.56.
In the non-standardized solution the respective intercepts fall within logical boundaries (i.e., 3.56 and 3.04). I am getting exactly these values under the STD standardization section. In those sections however the paths from the between level to the outcomes are excessively high (.995 and -.965).
So, my question is what am I missing (or doing wrong)?
I thank you for your consideration
 Linda K. Muthen posted on Monday, February 01, 2016 - 11:05 am
Please send the output and your license number to support@statmodel.com.
 yvette xie posted on Monday, January 16, 2017 - 12:50 am
Dear Professor,

In my model, I have continuous latent variables as DV, and both continuous and bivariate variables as IV. Should I report the results in STDYX for continuous covariates and STDY for bivariate covariates?

Many thanks!

Lili
 Bengt O. Muthen posted on Monday, January 16, 2017 - 9:42 am
Yes.
 yvette  posted on Monday, January 16, 2017 - 5:09 pm
Thank you for your prompt reply.

I have a follow-up question.
Some studies argued that we can no longer use standardised coefficient to compare the effect sizes of different IVs within a study. Then, why should we report the standardised solution?
 Bengt O. Muthen posted on Monday, January 16, 2017 - 6:01 pm
You may want to ask this general question on SEMNET.
 yvette  posted on Monday, January 16, 2017 - 6:18 pm
Thanks for your advice.

Best,
Yvette
 Jessica Tripp posted on Sunday, April 02, 2017 - 10:52 am
Is it unusual or problematic to have all standardized values using LVMM to be positive? I understand that it is typical for at least some values to be negative and want to be sure that having all positive values for each indicator wasn't indicative of an error or problem with the data? This is using STDyx. Thank you.
 Bengt O. Muthen posted on Sunday, April 02, 2017 - 11:53 am
If these are standardized factor loadings, no, that is not unusual.
 Danique van de Laar posted on Thursday, April 13, 2017 - 6:04 am
Dear professors,

I'm running several models without an interaction and one model with a latent variable interaction. Before entering the interaction I was reporting the stdyx results of my models (in my opinion easier to interpret). However, when I add the interaction effect to my model (XWITH) MPlus does not show me the stdyx results, only the model results. Is there a way to get the stdyx-results for a model with XWITH. If not, would you advise me to report the model results solely for the model with the interaction effect or do I need to be consistent and report the model results for all my models instead of the stdyx-results?

I look forward hearing from you soon.

Best,
Danique
 Bengt O. Muthen posted on Thursday, April 13, 2017 - 4:09 pm
Version 7.4 will give standardized results. If not, send to Support along with your license number.
 Danique van de Laar posted on Friday, April 14, 2017 - 8:21 am
Dear Bengt Muthen,

Thank you for your quick reply. I’ll try out running my model in the MPlus 7.4 version soon. I have one follow-up question: how does it come that the p-value for the model results and the stdyx results can be different (I use z-scores)? Sometimes there are only slight differences, however, two parameters in my model have a non-significant standardized coefficient (e.g., p = .069) whereas the non-standardized coefficient is significant (e.g., p = .025). Which one should I interpreted and report?

Best,
Danique
 Bengt O. Muthen posted on Friday, April 14, 2017 - 3:51 pm
See our FAQ:

Standardized coefficient can have different significance than unstandardized

This is also discussed in our new book.
 Hyunsik Kim posted on Thursday, February 01, 2018 - 7:44 pm
Dr. Muthen,

I'm running two types of survival analyses (semi-parametric model):
(1) A model where continuous survival time is regressed on a latent factor (continuous indicators; factor variance is fixed at 1 and mean at 0), controlling categorical (gender) and continuous (age) CVs.
(2) A model where continuous survival time is regressed on factor SCORES saved (continuous indicators; factor variance is fixed at 1 and mean at 0 when saving the factor scores), controlling categorical and continuous CVs.

In each model, which standardized output should I look at?

Thank you,
 Tihomir Asparouhov posted on Friday, February 02, 2018 - 11:28 pm
I would recommend both stdyx and stdy. For the continuous predictor use stdyx (so the predictor is standardized) while for the binary predictor use stdy so the predictor is not standardized. If that is too confusing you can standardize the continuous predictor with the define command and then use stdy.
 ywang posted on Monday, February 19, 2018 - 1:06 pm
Dear Dr. Muthen,

If there is a dummy mediator variable and the Y and X are continuous, should I report the stdyx or std? Thank you very much!
 Bengt O. Muthen posted on Monday, February 19, 2018 - 4:39 pm
If you have a binary mediator, the indirect and direct effects are best handled via counterfactually-defined effect formulas as discussed in our mediation book. But if you are not ready to go there, you can use WLSMV where instead the mediator is considered as the continuous latent response variable M* underlying the binary mediator M. In this case, you should use stdyx (which is the same as stdy for Y ON M*).
 Hillary Gorin posted on Tuesday, June 05, 2018 - 1:17 pm
Hello,

I am running a few parallel process mediation model and am assessing the slopes of the individual growth curves in the model.

I have two growth curve model with count distributions, one growth curve model with categorical data, and two other growth curve models with continuous data.

Should I interpret the unstandardized results in all models?

Or should I interpret the unstandardized results for the count data, the STDY results for the categorical, and the STDYX results for the continuous models?

Thanks!
Hillary
 Bengt O. Muthen posted on Tuesday, June 05, 2018 - 3:18 pm
If you regress i and s on x, you could use standardization of those coefficients using STDYX=STDY. But growth curves aren't directly comparable across those 3 types of outcomes.
 Hillary Gorin posted on Tuesday, June 12, 2018 - 9:08 am
Hello,

Thank you for your response! So to determine which growth curves to include in the parallel process mediation model, I should use the unstandardized results for all growth curve models?

Hillary
 Bengt O. Muthen posted on Tuesday, June 12, 2018 - 5:45 pm
I don't know what it means to determine which growth curves to include in a parallel process model - substantive motivation would seem to be prime, not whether you use raw or standardized coefficients. This question may be more suitable for SEMNET.
 Hillary Gorin posted on Tuesday, June 12, 2018 - 6:05 pm
Thank you for your response. Sorry for my lack of clarity. In my parallel process mediation model, I have three growth curves. I am trying to determine if change in one variable mediates the relationship between change in two other variables. I have one growth model with count data, one growth model with categorical data, and one growth model with continuous data.

Before running a parallel process model, it must be determined if all growth curve models have significant slopes. If the slope is significant, it is included in the model. If the slope is not significant, it is not included in the model. I am trying to determine if I should use the standardized or unstandardized p-values to decide if slopes are/ are not significant.

I can reach out to the SEMNET if this is not clearer. Thanks again for your help!
 Bengt O. Muthen posted on Wednesday, June 13, 2018 - 11:47 am
Usually, one uses the p-values or CIs of the unstandardized coefficients. Usually, they are the same as for the stand'd. If they are not, I would use bootstrapping so I could look at non-symmetric CIs because the 2 versions may differ in how close they approximate a symmetric distribution. This is discussed in our RMA book.
 mboer posted on Thursday, January 03, 2019 - 6:00 am
Dear Prof. Muthen,

I have a model with binary as well as continuous observed independent variables. The dependent variable is observed continuous. When I report the results in a table, should I report the STDY for the binary covariates and STDYX for the continous covariates? I am not sure whether it is appropriate to report results based on different standardizations within one table. Or should I just report the unstandardized results here?

Thank you in advance.
 Bengt O. Muthen posted on Thursday, January 03, 2019 - 4:10 pm
Q1: Yes

Q2: Report both
 Aurelie Lange posted on Friday, June 21, 2019 - 3:26 am
Dear Dr Muthen,

I need to calculate stdX for several paths as the outcome/y-variable is dichotomous and the x-variable is continuous. I have used the following formula:
stdx = b/(square root of variance of y)
To my surprise, the stdx value is larger than 1.
- Is this possible, or have I made a mistake in the formula?
- How can I calculate standardized S.E. and p-value for stdX? Or should I report all the raw results and only the standardized B?
- What standardization should I use for ordinal y-variables?

Thank you so much for your advice!

Sincerely,
Aurelie
 Bengt O. Muthen posted on Saturday, June 22, 2019 - 6:21 am
If you want to standardize with respect to x you should compute

stdx = b* SD(X);

where SD(X) is the standard deviation of x. You can do that in Model Constraint.
 Aurelie Lange posted on Monday, June 24, 2019 - 12:56 am
Dear Dr Muthen,

Thank you for your help!

1. What do you mean with 'you can do that in model constraint'?

2. Is it possible to also compute a standardized S.E. and p-value? An s.e. and p-value associated with the sdtx

3. I was told I could use the odds ratio, by using MLR instead of WLSMV (y is dichoutomous). Would that also by an option? What would be preferable?

Thank you for your help and advice!

Sincerely,
Aurelie
 Bengt O. Muthen posted on Monday, June 24, 2019 - 3:57 pm
1. See UG ex 5.21 for an example.

2. Model Constraint will give you that.

3. Yes, that is an option. Probably preferable. See also the paper:


Nguyen, T.Q., Webb-Vargas, Y., Koning, I.K. & Stuart, E.A. (2016). Causal mediation analysis with a binary outcome and multiple continuous or ordinal mediators: Simulations and application to an alcohol intervention. Structural Equation Modeling: A Multidisciplinary Journal, 23:3, 368-383 DOI: 10.1080/10705511.2015.1062730
 Jamie Griffith posted on Thursday, April 02, 2020 - 6:15 pm
Dear Mplus team

I have an RDSEM model for which we are interested in gender ("male", coded 1/0 below). I understand that SDTYX standardises on both X and Y and is used for continuous covariates, and STDY is for binary covariates, standardising only on Y.

When I run the model below, I get exactly the same values for both STDYX and STDY. Do you have any insight as to why these are the same?

MODEL:
%WITHIN%
spp | p^ ON p^1;
suu | u^ ON u^1;
spu | p^ ON u^1;
sup | u^ ON p^1;

sp | p ON time;
su | u ON time;

%BETWEEN%
sp su spp suu spu sup p u WITH
sp su spp suu spu sup p u;

spu sup ON male;
 Bengt O. Muthen posted on Saturday, April 04, 2020 - 11:14 am
STDYX and STDY are the same when x's are brought into the model, turning them into y's. But I don't see that in your model. We have to see your full output - send to Support along with your license number.
Back to top
Add Your Message Here
Post:
Username: Posting Information:
This is a private posting area. Only registered users and moderators may post messages here.
Password:
Options: Enable HTML code in message
Automatically activate URLs in message
Action: