How do I plot standardized loadings (Std.all) for a CFA model in R?

CFA
1

Using lavaan in R, I fit a CFA model. How can I plot the standardized loadings (Std.all) using a different plotting tool?

library(lavaan)
model_def <- 'F1 =~ varA + varB + varC'
data_set <- data.frame(varA = rnorm(20), varB = rnorm(20), varC = rnorm(20))
fit_mod <- cfa(model_def, data = data_set)
library(CustomCFAPlot)
plotCFA(fit_mod, label = 'std')
2

Based on my own experience working with CFA models in R using lavaan and alternative plotting packages, I found that the semPlot package can be an effective tool for visualizing standardized loadings. After fitting your model with lavaan, using semPaths allowed me to clearly see the standardized coefficients with options to fine-tune the plot appearance. This extra flexibility in adjusting aspects like node layout and edge thickness proved useful when interpreting the model structure, particularly on more complex models. The intuitive interface and robust documentation make it a strong alternative to device-specific plotting functions.

3

Hey everyone! I’ve been exploring another cool way to visualize standardized loadings using the qgraph package, and I thought I’d share my little experiment. After fitting your model with lavaan, you can extract the standardized coefficients (using something like standardizedSolution() or parameterEstimates()) and then plot these values as a network. The neat part about qgraph is that it allows you to create interactive graphs, which can help in understanding how variables connect in your model. I’ve been tweaking node sizes and edge colors to really bring out the structure, and it sparks some interesting debate on whether this interactive style helps in model interpretation compared to traditional static plots.

Has anyone else tried qgraph for CFA visualization? What tweaks did you find useful for emphasizing particular loadings or paths? Would love to hear your thoughts and maybe some sample tweaks you’ve used! :blush: