Science topics: Materials ScienceSEM Analysis

Science method

# SEM Analysis - Science method

Explore the latest questions and answers in SEM Analysis, and find SEM Analysis experts.

Questions related to SEM Analysis

My study is using the moderated mediation Model B and has three moderators. I want to get more clarification on index calculation and conditional indirect effect result interpretation. I would really appreciate your guidance and thank you n advance.

Hello,

I saw there is two different morphologies in SEM images of PCN-224, including cubic and spherical. I would appreciate if someone could explain this issue to me.

I have a SEM study where some variables were collected with a 7-points scale, while others with a 5-point. Is there any literature that I can look at? or any opinion? Thanks.

Hello,

I have survey data that I am attempting to use in IBM AMOS to create a SEM UTAUT model. However, during output, for "Result" I get:

Minimum was achieved

The model is probably unidentified. In order to achieve identifiability, it will probably be necessary to impose 7 additional constraints.

Chi-square = 2417.406

Degrees of freedom (corrected for nonidentifiability) = 82

Probability level = .000

I know close to nothing about statistics, and am a total newbie when it comes to AMOS, but from what I gather, that chi-square is bad; also for P CMIN/DF I get 29.481, which I also think is not great since it is greater than 3.

The SEM Model is supposed to be UTAUT model. Worst case scenario, can I still use the data as is? If so, how to correctly interpret the data? I have provided the model.

Any help is appreciated. Thank you.

I am trying to conduct an SEM analysis using Mplus7 with planned missing data. But I got this error: one or more variables in the data set have no non-missing values. Check your data and format statement. How can I fix this?

In evaluating second-order structure, I have faced two common approaches in applied studies:

1) Some fit the second-order structure in an explicitly hierarchical model: the first-order structure models the indicators and the second-order structure models the [implied] covariance between first-order latent factors.

2) Some fit the second-order structure on the first-order factors that are represented with sum scores (i.e., each subscale as a parcel).

To me, it seems that the first approach is more accurate (as it correctly models the measurement error), but I have some doubts about how to assess the fit of the second-order structure. Fit indices seem to put much more weight on the first-order structure. This goes to the extent that in a large model with a good first-order and a very poor second-order structure, the fit indices tend to show a good fit. Many authors consider the good fit of this hierarchical model as evidence of the good fit for second-order structure as well. (Some authors compare fit indices such as CFI and RMSEA from models with and without second-order factors and if there is little difference they conclude that the second-order structure is a good fit.)

Is this practice OK? And am I missing something here?

And is there any way to use the first approach and still calculate fit indices that exclusively evaluate the second-order structure?

(Something like calculating a chi-square for the discrepancy between the implied covariance matrix from the first-order structure and the implied covariance matrix from the second-order structure and using this for calculating other fit indices!)

Thank you in advance.

Hello!

In general, as a rule of thumb, what is the acceptable value for standardised factor loadings produced by a confirmatory factor analysis?

And, what could be done/interpretation if the obtained loadings are lower than the acceptable value?

How does everyone approach this?

Hi All,

I have run a default model and it can be calculated and has good fit both in two data sets (different topics) separately. Now I want to check for the moderating effect using multiple group analysis with AMOS. However, the unconstrained model, measurement weights model, structural weights model and structural covariances model are not identified.

My question is:

1. What is the reason for the unidentifiability of the unconstrained model? I am a bit confused since the default model without multiple group analysis definitely works.

2. What should I do to make unconstrained model, measurement weights model and structural weights model identified?

3. If unconstrained model, measurement weights model, structural weights model cannot be identified, is there some other way to test the moderating effect?

I attached some pictures. Hope someone can enlighten me, thanks so much for the help!

I am preparing sample for SEM analysis, but lacking the facility of liquid.co2 and HMDS.

So, is there any other method to dry the sample ?

Your suggestions will be highly appreciated .

We have finished writing an article about the psychological and emotional dimensions of the covid-19 disease, but it needs revision and final editing in English. In case of tendency, if someone has enough experience in this field and is completely fluent in English, I would appreciate it if you could send me a message.

Hi to everyone, I'm trying to obtain information about the morphology of small cylinders of GelMA. As they were too thick we cut them prior to standard treatment with Pt/pd coating required for SEM imaging. However, this process seems to alter the hydrogel structure leading to no significant data. Can GelMA films work better? We want to try heat-drying the samples instead of freeze-drying them but my concern is creating a temperature gradient that can alter the porosity. Has anyone ever tried this method?

Dear SEM experienced Users 😊, I’m thinking whether it is possible to make sthg like “variance decomposition” of latent variable in Structural Equation Modelling (SEM)?

Let’s imagine we’ve got some latent factor Y “determined” by two other latent variables X and Z. We have standardized parameter estimates between X->Y (0,5) and Z->Y (-0,4). Is it possible somehow to use theses two estimates to say which latent factor is more “important” for Y determination? Ideally, is it possible to say that X accounts for x% of variability in Y, and Z for z%? Thanks a lot in advance for any hints.

my ZnO nanoparticles sample was used for XRD and I do not have enough samples for SEM. Can the NPs samples be reused for SEM??

I synthesized gold nanoparticle thin films under four conditions, three of them show agglomerated islands, like any metallic particle observed before, highly agglomerated. However this fourth condition presents with a lot of circles and round objects ( as in attachement).

what it could be this structures in your opinion? is it gold nanoparticles? why are they so monodisperesed ?

i make (SEM analysis) in my research, then I modify the research model based on the correlation between independent variables. now i need to justify this modification in research model because the high value of the correlation. i am really confusing about how can I justify that change. I was add the research model before modification and after.

#SEM #Structural_equation_model #correlation

Simpson’s paradox is a statistical phenomenon where the relationship between two variables changes if the population is divided into subcategories. In the following animation, we can see how the linear relationship between two variables is inversed, if we take into account a third categorical variable. Simpson's paradox highlights the fact that analysts should be diligent to avoid mistakes.

How to identify this phenomenon in SmartPLS?

Dear all,

I am conducting a multiple-group path analysis (with observed variables) using AMOS. The unconstrained model reveals that several paths coefficient were different according to p values.

When I compare the fully constrained model to the unconstrained model, chi-square differences were insignificant. However, when I compare the models by constraining only one path at a time, I have various variant path coefficients (significant chi-square change).

Should I assume that the model is invariant based on the comparison between fully unconstrained and fully constrained models? Or, should I use the "one path at a time approach" to decide and freely estimate all variant paths? So, I would like to have your opinion on the best approach.

Ps: Although the chi-square did not change, the CFI decreased in the constrained model. Could it be a reason to examine specific path variances?

Thanks

Best

Pinar

I am building a model for the Five Factor Model of personality in AMOS (as measured by BFI-44 in a large >300,000 dataset). I am doing this to test for MI across groups so I can be sure of my conclusions. (using a small sample c. 10,000 for the CFA)

I based the model on preliminary EFA, using this to specify some cross loadings.

All fine... improved model slightly

I then added a method factor (to represent Halo Bias)... again good improvement

The part I am having trouble with is adding an "Acquiescence Bias" factor.

I am attempting to use either:

1. normalised sum of all scores (5point likert) or

2. sum of reverse scored items (normalised)

..... it won't run with either of these

CoVarience doesn't see to make a difference nor does adding error variables.

When I use a dummy variable for the observed variable, it runs (this is the sum of reversed divided by participant SD).

- I get a slight improvement in fit but I am expecting a lot more.

I get this error:

Minimization was unsuccessful

The results that follow are therefore incorrect.

The model is probably unidentified. In order to achieve identifiability, it will probably be necessary to impose 101 additional constraints.

Chi-square = 222005.662

Degrees of freedom (corrected for nonidentifiability) = 953

Probability level = .000

I am self teaching this over the past 3 days so please forgive my naivety.

Hi,

I am looking at the three-way interaction of a latent variable (F) with ordered-categorical indicators (mostly likert scales) and two observed variables (M1, M2), using Mplus. The two following issues arise:

1. How do I account for the ordered structure of the indicators? Do they need to be defined as categorical?

2. What does this mean for the standardization of the latent construct (F)? Standardizing the indicators does not seem appropriate!

Any advice or guidance is greatly appreciated. Thank you in advance!

Hello everyone, i've already did a characterization of SEM-EDX on iron electrodes before and after electrolysis. From this picture, I can only explain that this electrode:

- Before treatment: the iron surface is smooth and regular, the iron element is more dominant

- After treatment: the surface of the iron has cracks, holes and irregularities caused by the electrolysis process, there is also an oxide layer on the surface of the iron which means the surface of the iron has been oxidized, the element of oxygen is more dominant

Any additional suggestions to describe this SEM-EDX result?

Also, I have some questions about SEM-EDX :

1. Why is the oxygen level higher after the treatment?

2. How does Fe react with oxygen in water during electrolysis?

3. How to write down the chemical reaction that occurs between Fe and oxygen in water to form Fe-oxide?

4. Talking about the morphology of SEM on the image after treatment, how can a needle-like layer be formed?

5. If correlated with the pourbaix diagram, how to explain the formation process of iron oxidation, when there is water, the presence of dissolved oxygen in the water?

Thank you in advance

I synthesized ZIF-93 by the method of aqueous phase in the article, but unlike the article, I synthesized a tetrahedral structure instead of rhombic dodecahedron. zif-93 is composed of zinc acetate dihydrate and 4-methyl-5-imidazolecarboxaldehyde. Can you tell me what causes the formation of tetrahedra?

Hello everybody,

I developed a SEM model. Model fit was good and my other hypothesized pathways were in line with the theory and significant, but one pathway is showing totally meaningless relationship. my model was predicting social anxiety from social stress. the model fit analysis found the path significant but with a negative value. It says a decrease in social stress predicts an increase in social anxiety. What should I do know? Should I exclude the path by stating that it is meaningless or something else?

Thank you!

Hi Everybody,

I am a research scholar presently working on tourism, conflict and peace studies. I developed a conceptual model and want to use Structural Equation Modelling Approach. I have a total of 4 constructs and 23 indicators in my model. I want to know what should be the minimum sample size for my study to do SEM analysis.

Thanking you very much in advance

Wanie Mehraj

Hello everyone. Requesting your kind suggestion. I'm struggling with SEM analysis for my doctoral study by AMOS. My modified model result showed GFI & AGFI values under the recommended value of 0.90. May I ask .... which alternative indices will you suggest instead of GFI and AGFI? Which references should I use? Thank you.

Hi all,

I've done a CLF test and could you recommend what's the threshold value of a good CLF (if possible could you pls add some references)? I used '0.2' recommended by James Gaskin (on YouTube) and found 4 out of 17 exceeded 0.2 (ranged from 0.2-0.24), is it acceptable?

p.s. the test of common method variance is the last step of my data analysis, other tests all have good results. (preliminary analysis and SEM)

or do you recommend any other threshold value?

MANY THANKS!!!!!!!!!!!!!!

What is the best method for drying microfibrillated and nanofibrillated cellulose (MFC/NFC) for SEM analysis?

I would like to take a clear SEM images for MFC and NFC, from which I expect to see the morphological characteristics of fibrils. Please give your suggestions or comments. Thanks.

Dear all,

I have a question about a mediation hypothesis interpretation.

We have a model in which the direct effect of X on Y is significant, and its standardized estimate is greater than the indirect effect estimate (X -> M -> Y), which is significant too.

As far as I can understand, it should be a partial mediation, but should the indirect effect estimate be larger than the direct effect estimate to assess a partial mediation effect?

Or is the significance of the indirect effect sufficient to assess the mediation?

THanks in advance,

Marco

when conducting the SEM analysis, if RMSEA, GFI, CFI, and Chisq/df achieved the required levels except for CFI was 0.80, can I consider the model as a good fit?

Dear researchers:

Through my read some of the papers, I find that the welding efficiency may reach 100%, so the separation of the metal occurs away from the welding area (during the tensile test).

From your point of view: How do you evaluate the microstructure of welding zone?

With Regards

Hi, everyone, i want to know if someone ever done the SEM analysis for activated carbon using silicon wafers instead carbon adhesive tape. Thank you.

I have synthesized silver nanoparticles using PVA as surfactant and silver nitrate as precursor. I need to do SEM analysis and hence require powdered form of the silver sample. I tried freeze drying the solution but the lyophilised sample is not in powdered form but in cotton candy state . I tried to grind it in a mortar & pestle but the sample is getting contaminated and visible color change could be seen.

I'm anticipating that it's because of higher percentage of PVA that is being used that gives it the cotton candy state. Please correct me if I'm wrong..if so what could be done to get powdered form or to powder the lyophilised sample

Also, I tried to centrifuge it as a part of purification process. 10,000 rpm for 30mins at 4 degrees. I got small amount of brownish precipitate like literally the size of a sugar crystal. And the supernatant was yellowish in color which was the colour of colloidal silver sample.

Is there any other method for purification or to proceed with SEM analysis?

In confimatory factor analysis (CFA) in Stata, the first observed variable is constrained by default (beta coefficient =1, mean of latent variable =constant).

I don't know what is it! Because, other software packages report beta coefficients of all observed variables.

So, I have two questions.

1- Which variable should be constrained in confirmatory factor analysis in stata?

2- Is it possible to have a model without a constrained variable like other software packages?

Hi! I am trying to prepare hydroxyapatite scaffold samples for SEM imaging of cell growth. I have the Karnovsky's fixative kit but the procedure provided in the tech sheet (attached) is not sufficient for my applications. First, does anyone have a standard protocol for this SEM fixation using Karnovsky's fixative kit? Second, do I need to do the post-fix using OsO4 or is there an alternative method to the post-fix mentioned in the tech sheet? Can I do the fixation procedure without it, followed by the graded ethanol dehydration or will it have a negative impact on my sample preparation?

I would really appreciate any help answering this question. Thanks!

Hi, everyone!

I just received the comments of a reviewer who said:

you conducted EFA and based on EFA results, run the SEM modeling. You are supposed to conduct a CFA to confirm the EFA results and finalize the measurement model before proceeding SEM. You can fairly use half of the sample to test EFA and another half to test CFA.

Actually, in my study, i used the EFA to explore the possible dimensions of the high-order constructs, and then build a PLS-SEM with the results of EFA. However, i don´t think I should do also the CFA.

So, how can I answer the reviewer ? and is my method wrong??

Thanks!!!

I have a SEM model (with 9 psychological and/or physical activity latent variables) with cross-sectional data in which, guided by theory, different predictor and mediator variables are related to each other to explain a final outcome variable. After verifying the good fit of the model (and after being published), I would like to replicate such a model on the same sample, but with observations for those variables already taken after 2 and after 5 years. My interest is in the quasi-causal relationships between variables (also in directionality), rather than in the stability/change of the constructs. Would it be appropriate to test an identical model in which only the predictor exogenous variables are included at T1, the mediator variables at T2 and the outcome variable at T3? I have found few articles with this approach. Or, is it preferable to use another model, such as an autoregressive cross-lagged (ACL) model despite the high number of latent variables? The overall sample is 600 participants, but only 300 have complete data for each time point, so perhaps this ACL model is too complex for this sample size (especially if I include indicator-specific factors, second-order autoregressive effects, etc.).

Thank you very very much in advance!!

Hi,

Could you please tell me how can we calculate effect size measures for structural equation modelling? Could we do it via AMOS? Are there any practical resources?

Thanks in advance,

Tahani

Hello everyone,

I am surprised to see that PLS-SEM is an accepted tool in different areas: Management, Marketing, Tourism, ..., having become an "alternative" tool to the previously prevalent CB-SEM. I think I'm not wrong if I say that in recent years PLS-SEM is more widely used than CB-SEM in these areas. However, this does not seem to be the case in psychology, where PLS-SEM does not have a significant presence.

Are the objectives or premises really different in these areas of knowledge to justify that PLS-SEM is really valid in some and not in others?

The areas in which PLS-SEM is accepted, are they less rigorous?

Is it a matter of time before PLS-SEM succeeds in displacing CB-SEM in psychology?

I would appreciate if someone could help me understand this.

Thank you

Articles have only mentioned that cell-laden hydrogel scaffolds were lyophilized before SEM analyses for cell adhesion. However, no details were mentioned.

The particle size of samples for XRD and SEM analyses is crucial for obtaining relevant results.

In my SEM analysis, all the paths from constructs to the outcome construct were shown to be insignificant, although the model fit indices were all acceptable. My particular focus is on whether an A variable is directly related to the B variable or the A variable is fully mediated by C.

Considering this result is related to type 2 error by the multicollinearity among the latent constructs, I tried a regression analysis to prove if there is a significant direct effect between the A variable to the outcome variable B. In this regression analysis, measured variables for the A were used. My question is whether this process, which is to use regression analysis to see a signigicant direct effect that was not shown in the SEM analysis with latent variables, is statistically valid.

Dear fellow researchers,

Usually we use lavaan for continuous variable, so can we still use lavaan for categorical variable (e.g. high and low ethnic diversity composition)?

Thank you very much!

Best,

Edita

Hello everyone!

I currently have 2 measurement models. All are correlated factor models and the factors reflect subscales of anxiety related constructs. Estimator is robust maximum likelihood (to account for lack of multivariate normal distribution). This happens in the context of construct independence. First model implies independence between all subscales. Second model clusters factor 1 and factor 2 together, and factor 4 and factor 5 together.

model 1:

f1 =~ item 1 + item 2 + item 3

f2 =~ item 4 + item 5 + item 6

f3 =~ item 7 + item 8 + item 9

f4 =~ item 10 + item11 +item12

f5 =~ item 13 + item14 + item15

f6 =~ item 16 + item 17 + item18

f7 =~ item 19 + item 20 + item21

model 2:

f1,2 =~ item1 + item2 + item3 + item4 + item5 +item6

f3 =~ item 7 + item 8 + item9

f4,5=~ item 10 + item11 + item12 +item13 + item 14 +item15

f6 =~ item 16 + item 17 + item18

f7 =~ item 19 + item 20 + item21

Nested models are models wherein all parameters of a more restricted model are included within a less restrictive one. I am new to this, and I reviewed examples, but I cannot make a conclusion. I would appreciate an answer and the reasoning behind it a lot!

(note: based on the fit indices, I know that the second model does not work. Fit indices are well below the acceptable threshold and the differences are enormous. But in the case of non-nested models, comparison of absolute and relative fit indices is not the case and I want to do the comparison anyways to learn how to do it correctly)

Dear experts,

For my thesis, I want to clarify that the difference between “moderate variable “and “control variable’’ in SEM analysis techniques. If this is same, would it be possible to consider same variable to be both case?.

Please help me.

Kind regards,

Thathsarani

I'm currently doing SEM (Structural Equation Modeling) in R using the lavaan package and found that my data

**violated the normality and homoscedasticity**assumptions. However I get**good numbers on CFI and RMSEA**values. How is this possible? Does this mean the model is good? Do I still need to check the model assumptions? Thanks in advance.Hello everyone

I hope you are doing well

- AA6061-T6 or AA7075-T6 Al alloys fusion-welded plates contain the FZ (fusion zone) with a dendritic structure. I want the dendrites to be identified
**separately**and in the form of grains (whether they can be called grains or not is another issue). The figure shows the dendritic structure in the FZ, but it isn't easy to separate dendrites from each other. (Figure shows the FZ in fusion welded AA7075 (not AA6061) etched with Keller). - What do you suggest as the etchant solution for the SEM investigation of the PMZ and FZ
**grain boundaries**of the AA6061 fusion weld sample? - If you have experience in this field, I would appreciate writing it here.

Multigroup analysis in SEM is an excellent method to estimate the measurement invariance across different groups. JASP software has a user-friendly GUI for the application of R package lavaan with embedded multigroup analysis. My experience is that the new analytical method should be used not only through the theoretical framework, but also by the insight into the excellent yet simple examples. The following paper presents the simple approach that can be useful for novice researchers applying multigroup analysis in SEM.

1. What is your experience with multigroup analysis in SEM?

2. Which software do you use?

3. Can anyone share the syntax for the constraints in JASP?

Hello,

I need to synthesize a lithium-rich cathode using the sol-gel method. What is the best molar ratio between citric acid and transition metals in this method? Does citric acid have an effect on particle size?

thanks

I intend to pursue a Ph.D. and intend to apply Structural Equation Modelling on '

*Factors influencing student success*'. I have the following potential latent factors:1. Academic Success Factors

2. Student Success,

3. Student Retention

There are many studies building on Vincent Tinto's 1995 model of student departure, and many factors have been suggested to retain students. I also intend to build on Tinto's model, however applying SEM.

I am wondering if the 'success factors' can be modelled especially through SEM as exogenous variable/s to influence 'student success' and or 'student retention'.

Would appreciate any advice, suggestions, or comments on this potential research (I have attached the proposed model).

Hi there,
I definitely do need your help!!!
Looking through studies and books I got a little confused by the different approaches used to conduct factor analyses for reflective scales before running PLS-analysis.
Some recommend carrying out exploratory factor analysis (EFA) using SPSS first, followed by covariance-based confirmatory factor analysis (CB-CFA) using e.g. AMOS. The "stepwise" received results (items) are then applied to PLS for further analyses.
Others are pro EFA (in SPSS) but advice against using CB-CFA (e.g. AMOS) before PLS-analysis, criticizing they have different underlying assumptions. Instead they recommend doing the CFA directly in PLS (using the EFA's results).
But even within the field of EFA there seems to be some confusion about what extraction method (principal component vs. principal axis vs. ...) and which rotation procedure (oblique vs. Varimax) are most appropriate when using PLS afterwards.
So, my question: Are there any rules or is there a generally accepted way of how to conduct EFA and CFA when using PLS? Could you provide me with corresponding references (published articles etc.)?
Hope, someone can help!
Thanks in advance!

Greetings,

According to Hair et al. (2020) about the Confirmatory Composite Analysis (CCA) to assess quality of the measurement model, Nomological and Predictive validity steps are suggested. would someone explain how can i apply these in Smart-PLS software and what exactly the indices or measures that i should extract?

thank you in advance.

A common threshold for standardized coefficients in structural equation models is 0.1. But is this also valid for first difference models?

If I use SmartPLS to test the structural model then how I can measure the Goodness of Fit Index (GFI). What are the indices I need to observe for validating the research model?

I have to do SEM analysis of root samples colonized by a selected bacterium. After removal of roots from the pots, how long I can store the samples in 2.5% Glutaraldehyde solution. Is there any special treatment to increase the the storage period without affecting the colonized microbes?

I need to analyze plant tissue samples using SEM for which I am following the 2.5 % Glutaraldehyde fixation method. But, I cannot access SEM for a month and so need to store my plant samples. Should I fix my samples first and then refrigerate them? Or should I refrigerate them first for days and fix them before SEM analyses?

I run my model which consist of 5 factors but factor 4 and 5 are not significant in Pearson correlation. Is this ok? because I would like to run for SEM.

Any suggestions? Thanks/...

I am doing research on natural fiber composites. For composite analysis the tests are as following

1. XRD

2.SEM

3.FTIR

4.DMA

5.DSC

6.TGDTA

7.UTM

8.UV Visible

Is that possible a sample can be use in more than one test?

Please help me

Hi everyone,

in my SEM study, I have some problems with the Fornell-Larcker criterion assessing discriminant validity. The square root of the respective AVE is smaller than some of the correlations of this factor with other factors which is why I assume that discriminant validity is problematic.

I know that it is possible to merge factors and look at cross-loadings, however this really does not make much sense in my case. Is there any other way of dealing with a lack of discriminant validity or is it maybe possible to just mention the lack of discriminant validity as a limitation and continue with the interpretation of the model? It's for a Master's thesis, I am not planning to publish in a top journal ;)

Thanks in advance for your help!

Hi, I am running path analysis with latent variables. My model fit indices are good, however some of the factor loadings are negative. Also some of the standardized estimate are more than 1 like chemical on N2O is more than 1 (1.60) and topographical on N2O is -1.03.

Is it alright to have negative loadings in the attached path diagram. How can i correct this diagram?

Thanks

Hello.

I am doing path analysis with exogenous variables from my common garden experiments.

For this analysis, I the following steps;

1) structure hypothetical model with exogenous (A) and endogenous (B, C, D, E, and Z) variables

Z ~ p1*A + p2*B + p3*C + p4*D + p5*E

C ~ p6*B + p9*A

D ~ p7*B + p10*A

E ~ p8*B + p11*A

B ~ p12*A

2) data processing;

2-1) calculate mean values of observed individuals

2-2) standardizing the mean values (mean = 0, sd = 1)

2-3) Dataset consists of 90 < n < 100 values of the above variables.

So I ran cfa from the lavaan package in R using the dataset and path model.

However, I met severe troubles from the results

The model did not fit.

----------------------------------------------------------------------------------

lavaan 0.6-10 ended normally after 27 iterations

Estimator ML

Optimization method NLMINB

Number of model parameters 27

Used Total

Number of observations 89 94

Model Test User Model:

Test statistic 129.907

Degrees of freedom 6

P-value (Chi-square) 0.000

Model Test Baseline Model:

Test statistic 273.261

Degrees of freedom 21

P-value 0.000

User Model versus Baseline Model:

Comparative Fit Index (CFI) 0.509

Tucker-Lewis Index (TLI) -0.719

Loglikelihood and Information Criteria:

Loglikelihood user model (H0) -678.739

Loglikelihood unrestricted model (H1) -613.785

Akaike (AIC) 1411.478

Bayesian (BIC) 1478.671

Sample-size adjusted Bayesian (BIC) 1393.464

Root Mean Square Error of Approximation:

RMSEA 0.482

90 Percent confidence interval - lower 0.412

90 Percent confidence interval - upper 0.555

P-value RMSEA <= 0.05 0.000

Standardized Root Mean Square Residual:

SRMR 0.191

----------------------------------------------------------------------------------

1) Chi-square was significant.

2) CFI and RMSEA values also said that model is not fit.

To fit the model, I tried to modify the dataset, change the model, remove NA or add extra value into the dataset, but the un-fit was not changed.

Here're my questions,

What can I do to fit the model? Change or modify something? Try to use the other functions or packages?

And

Can I run separate multiple regression for the model? Then Do I use the coefficients from the regressions for my papers?

Please kind reply.

Thank you.

It is usual to say that causality is time-dependent. However, a reviewer sent this message for a study "The authors state that one limitation is the cross-sectional design, which does not enable establishing a cause-and-effect relationship. However, the paper intends to evaluate causal mechanisms. Please be coherent: if you believe causal effects are not plausible to be evaluated, then the whole conceptualization of your study is unsupported".

May you help me to understand the conception of causality in SEM?

Hi everyone,

I have longitudinal data for the same set of 300 subjects over seven years. Can I use '''year' as a control variable? Initially, I used one way ANOVA and found no significant different across seven years in each construct.

Which approach is more appropriate?. Pooling time series after ANOVA (if not significant) or using 'year' as a control variable?

I want to analyse cross-section samples by SEM microscopy whose constituent layers have been applied by dip-coating using SS and glass as substrates. However, I do not know if I would have to prepare the sample in a special way for its analysis and if it is the case, how should I previously prepare the sample without altering the applied layers? I have already tried to cut the glass samples using diamond tip cutter but the cut was neither clean nor precise.

Thank you for your help.