Why is there no chi-square test for model fit when using multiple imputation (Mplus)?

More Marcel Grieger's questions See All

Anybody to help me with Code source for DOUBLE CONCAVE FRICTION PENDULUM BEARINGS Please?

Hello, would you please share the codes for Base Isolation of structures using Double Concave Friction Pendulum Bearings? I am doing a related research but as I am just a beginner in MATLAB, I...

18 January 2021 6,683 1 View

Recommendations for Fluorescent Western Blotting?

Hi everyone, we're planning to do some protein quantification using western blot. Since a lot of our proteins of interest are of similar size, I would either have to strip and reblot a lot or...

04 November 2020 5,033 2 View

What outcomes can be expected when a leader creates a positive informal environment?

I am looking for outcomes (individual or organizational) related to the creation of a positive informal work environment. Can anybody recommend me articles on the topic? Thank you in advance....

01 August 2020 7,142 3 View

How to correct typos in RG refs?

how can one modify / correct an (annoying) typo in Research Gate mention of some publication? sometimes the name, sometimes the first name, sometimes the reference, ... is incorrect; what can one do ?

07 January 2020 9,089 5 View

How to simulate an imbalance of a rotating shaft with pointmass in abaqus?

Hi everyone, I am working on a model of a rotor to examine the deviation and critical speeds due to high speed rotation. Therefore I want to integrate an eccentricity of the shaft by creating a...

23 July 2019 4,726 17 View

How can you do statistics with normalized data?

Just briefly: I have WB quantifications from 4 different blots (4 biological repilcates) and I normalized my signal intensity values to the untreated ctrl sample. Therefore, for all 4 replicates...

14 July 2019 3,258 3 View

Internal normalization in Western Blot quantification and statistics. How to?

This might seem like a rather simple problem, but I am actually unable to find a helpful solution that specifically refers to this problem. I did several Western Blots of biological replicates...

10 July 2019 6,476 5 View

What is the significance of the graph in anova two way?

We finalized some results using ANOVA Two Way and now facing difficulties to report the statistical significance in the figures. We do not know if we use the factor significance (Factor 1, Factor...

28 October 2018 3,750 2 View

Is there a list of chemicals which lead to a higher food consumption in animals?

I am looking for some sort of list with information about how (and which) chemicals increase the amount of consumed food of animals.

12 April 2017 5,646 3 View

Boundary Conditions for Porous Contact among Two Surfaces?

I'm currently doing simulations related with contact against porous materials (Tissues) in ABAQUS. As long as I use Pore Pressure=0 as Boundary Condition everything works fine and I'm able to...

30 May 2016 3,561 2 View

How to compare two groups with only two measurements?

Hiiiii everyone! I have an enquiry on statistical analysis. I was looking for many forum and it's still cannot solve my problem. I want to compare means of two groups of data but only with two...

03 March 2021 8,796 3 View

Can we classify HV and LV lines based on line length?

I have dataset which shows the length of power lines. I need to classify the lines based on the line length. Is there a rule to classify the High voltage (HV) and low voltage (LV) lines based on...

03 March 2021 4,116 4 View

What SPSS analysis should I run with one IV (with two levels) and two DVs?

Hi, I am trying to figure out which test would be most appropriate to run for the following question: Is there a difference in task blood pressure between smokers and non-smokers? There are two...

01 March 2021 8,235 2 View

What test should I use to compare two or three means, if data is not normally distributed but sample size is > 30 in each group?

I personally feel this as a most confusing question. As per my understanding, if sample size is > 30, and sample is skewed, we can still use paramteric approach, as it will follow Z...

01 March 2021 1,587 6 View

Distribution of particles in metal matrix?

How would define the "good" and "poor" distribution of particles within the matrix. Is there a quantitative solution to assess the distribution of the particles, which then having denotation of...

28 February 2021 7,397 8 View

Statistics: include groups of respondents made by one person for analysis?

Good morning, I could not find literature about this so I am asking. I am organizing data for a questionnaire (N=100) that foresees some personal questions like level of education. For the aim...

25 February 2021 2,090 4 View

Sample size determination for non-parametric tests?

How do I run a a priori sample size determination for non-parametric tests, like the Mann-Whitney U-test and the Kruskall-Wallis or other non-parametric tests that accord with the unpaired t-test...

24 February 2021 166 6 View

How to examine multicolineairty in backwards regression with binary variables?

I did a backwards regression with various binary variables, but some of the binary variables are negatively correlated with the outcome, which would make no sense cause the correlation univariate...

21 February 2021 4,392 3 View

Statistical Tests for Quantitative Study?

Hello Everyone, I have one population, sample size 200. and needed to find the correlation between Variable A and B. Also, which T-test should I choose for hypothesis testing Please advise...

21 February 2021 7,591 4 View

How to model rupture in geogrid in abaqus?

Hi researchers I wanna model tensile test on geogrid in abaqus, so I added displacement to one side of geogrid and the other side of geogrid is constrained completely, everything seems normal but...

19 February 2021 2,195 2 View

Christian Geiser

Marcel Grieger This is odd given that a paper by the Mplus folks themselves (http://statmodel.com/download/MI7.pdf) states that the chi-square will be computed, at least for ML estimation with continuous variables. Also, they have example output on their website that shows the chi-square results for an imputed analysis of a growth model (http://www.statmodel.com/usersguide/chapter11.shtml). I would email Bengt Muthen directly--this seems strange. I'm sure he can explain this. Perhaps you can share the response here if you get one--I'd be interested in the explanation for this also.

Marcel Grieger

Thank you Christian Geiser . Contacting Bengt Muthen is the next step. Perhaps I should have been more precise: What is missing is the p-value of the chi-square test as well as the confidence interval for RMSEA. I used the WLSMV estimator because of the ordinal indicator variables. Please find attached parts of the output showing the fit indices.

Best

Marcel

I believe there isn't a valid chi-square test available yet for multiple imputation when WLSMV estimation is used. You can look at the average chi-square across imputations, but there is no valid p value for a test of significance available yet. Something that might be developed in the future...

I remember reading about this drawback in an older statmodel post from 2017 (https://www.statmodel.com/discussion/messages/22/24740.html?1510106241), but had hoped for some developments in the last five years. Unfortunately, no modification indices are available either. "One of the bells and whistles lost with MI compared to 'FIML'." (https://www.statmodel.com/discussion/messages/14/27094.html?1600179943).

Thanks again!