Thank you very much! A quick question. Aren't we supposed to check model assumptions like we do after fitting a regression model? Or is it enough to look at only fit indices CFI, TLI , chi square, RMSEA and SRMR?
In SEM, we are primarily interested in the overall fit of the model to the data. This is where fit indices like the Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), Chi-square, Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR) come into play. These indices provide a measure of how well the model reproduces the data (or covariance matrix) and are crucial for evaluating the adequacy of the model. However, this doesn’t mean we ignore the assumptions (Linearity, Multivariate normality, independence of errors) entirely.
Thank you for the great video! I`m new with R and I do not understand how you could interpret the significance levels or the correlations between the variables from the output. Could you please explain this?
THANKS FOR NICE VEDIO. How did you handle variables such as priming, water and rep in your analysis , I mean did you consider those as well. Please help me as my problem is something similar . Advance thanks
Hi, thank you for this lecture. I would like to ask how to go about a case where I have two IV and two moderators. I think the model specification will be modelspec
When i calculate vaeiability through R Studio, the result shows negative values in fcal. Please suggest me your opinion and let me know if there is any corrections
in model estimation step the output is like this mod.est = sem(model = mod.id, + data = data) Error in if ((!is.matrix(model)) | ncol(model) != 3) stop("model argument must be a 3-column matrix") : argument is of length zero how to rectify this sir
The error message you’re seeing typically occurs when the model argument in the sem() function doesn’t recognize the input as a valid model specification. In your case, it seems like mod.id might not be correctly specified. mod.id = ' latent1 =~ observed1 + observed2 + observed3 latent2 =~ observed4 + observed5 + observed6 observed7 ~ latent1 + latent2 '
You are right fitmeasure() and fitMeasures() are functions of lavaan package however I have used fit.measures as an argument which is set to TRUE within summary function.
The dada importing process does not work. I cannot access the data I. literally tried everything but I cannot load a data file data_path.xls which I have not downloaded beforehand.
If you have downloaded data from the link in description and then you was unable to import, it means there is some error in path. Share the path you have used.
Thanks for your reply! I could not find a link in the description where I can download the data. I only find a link to your blog but there I also cannot find a file for download. Perhaps you could share the link here again?
@@Satoshi467 At the end of the blog post you will find a link "Download data file - Click here" When you will click the file will automatically download.
Thanks Sir, Very Informative!!
Thank you
You are the best! Thanks for this ❤️
Thank you
Thank you very much! A quick question. Aren't we supposed to check model assumptions like we do after fitting a regression model? Or is it enough to look at only fit indices CFI, TLI , chi square, RMSEA and SRMR?
In SEM, we are primarily interested in the overall fit of the model to the data. This is where fit indices like the Comparative Fit Index (CFI), Tucker-Lewis Index (TLI), Chi-square, Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR) come into play. These indices provide a measure of how well the model reproduces the data (or covariance matrix) and are crucial for evaluating the adequacy of the model.
However, this doesn’t mean we ignore the assumptions (Linearity, Multivariate normality, independence of errors) entirely.
Thanks dear for a good video about path analysis
Thanks for your support.
Thank you for the great video!
I`m new with R and I do not understand how you could interpret the significance levels or the correlations between the variables from the output. Could you please explain this?
Thank you very much, Sir!
You're welcome!
So interesting and knowledgeable for me.
Sir in another also includes fixed effect and random effect sem model. Thanks
Thanks for your support. I shall keep it in loop.
THANKS FOR NICE VEDIO. How did you handle variables such as priming, water and rep in your analysis , I mean did you consider those as well. Please help me as my problem is something similar . Advance thanks
There is a small mistake "data$rep = as.factor(data_corr$priming)” is incorrect. it should be priming on the left side too
Very good. Thanks.
Thanks for your support.
Hi, thank you for this lecture. I would like to ask how to go about a case where I have two IV and two moderators. I think the model specification will be modelspec
When i calculate vaeiability through R Studio, the result shows negative values in fcal. Please suggest me your opinion and let me know if there is any corrections
Please share your code and output at agron.infotech@gmail.com
in model estimation step the output is like this
mod.est = sem(model = mod.id,
+ data = data)
Error in if ((!is.matrix(model)) | ncol(model) != 3) stop("model argument must be a 3-column matrix") :
argument is of length zero
how to rectify this sir
The error message you’re seeing typically occurs when the model argument in the sem() function doesn’t recognize the input as a valid model specification. In your case, it seems like mod.id might not be correctly specified.
mod.id = '
latent1 =~ observed1 + observed2 + observed3
latent2 =~ observed4 + observed5 + observed6
observed7 ~ latent1 + latent2
'
@@AGRONInfoTechgot it rectified thank you so much sir
I'm delighted that it was helpful for you.
Very informative. Would you please like to share the code and dataset.
Please visit below link: www.agroninfo.com/how-to-perform-structural-equation-modeling-sem-in-r/
Thank you can u please make a tutorial video on redundancy analysis
Thanks for your support. I shall try to consider this topic in upcoming videos.
Please make video on DDRs, distance decay relationship plot. Will be very appreciated.
Thanks for your suggestion. I shall keep it in loop.
I can never find the function fit.measures, only fitmeasures or fitMeasures, do you know why this is?
You are right fitmeasure() and fitMeasures() are functions of lavaan package however I have used fit.measures as an argument which is set to TRUE within summary function.
Please Prepared SEM with Likert scale based data
Okay I shall create a tutorial on Likert base data. Thank you for your suggestion.
The dada importing process does not work. I cannot access the data I. literally tried everything but I cannot load a data file data_path.xls which I have not downloaded beforehand.
If you have downloaded data from the link in description and then you was unable to import, it means there is some error in path. Share the path you have used.
Thanks for your reply! I could not find a link in the description where I can download the data. I only find a link to your blog but there I also cannot find a file for download. Perhaps you could share the link here again?
@@Satoshi467 At the end of the blog post you will find a link "Download data file - Click here"
When you will click the file will automatically download.
I had the same problem. I tried with my data but it didn't work.
nakakaiyak. balikan ko to pag naka defense na kami. gajgass