Thanks a lot ! I did not know that the fit measures could be computed from the covariance matrix. I'll definitely share that when I submit my paper for publication !
This video really helped me getting started with Lavaan, and has allowed me to do what I otherwise would have had to maybe do in Stata although I like to do as much as possible in R. One question though, but is there any way to in a nice (perhaps tidy) way to create tables from Lavaan output? An example would be with 3 latent variables each measured by 3 variables. To get each variable in it's own row with factors in the columns, and thus in the first column I'd get 3 factor loadings for the three variables composing the factor in row 1:3, then in the second column I'd get factor loadings for the 3 next variables in row 4:6, so on?
Hi - I assume the data is heavily skewed and problem with normality will occur hence the ordinal nature of the Likert scale. Do you have any suggestions to overcome this in the matrix ? I assume a poly matrix would be better off and more reliable. Regards !
hell of a good job, thank you Sacha!
Thanks a lot ! I did not know that the fit measures could be computed from the covariance matrix. I'll definitely share that when I submit my paper for publication !
This video really helped me getting started with Lavaan, and has allowed me to do what I otherwise would have had to maybe do in Stata although I like to do as much as possible in R. One question though, but is there any way to in a nice (perhaps tidy) way to create tables from Lavaan output? An example would be with 3 latent variables each measured by 3 variables. To get each variable in it's own row with factors in the columns, and thus in the first column I'd get 3 factor loadings for the three variables composing the factor in row 1:3, then in the second column I'd get factor loadings for the 3 next variables in row 4:6, so on?
really good tutorials. just as a note the code is rather hard to read due to your color theme in RStudio.
Hi - I assume the data is heavily skewed and problem with normality will occur hence the ordinal nature of the Likert scale. Do you have any suggestions to overcome this in the matrix ? I assume a poly matrix would be better off and more reliable. Regards !