I have a theoretical framework of 5 latent factors, each with 10 questions (big 5, personality questionaire). I want to create a semPlot of all 10*5 = 50 questions and 5 latent factors, and also visualize relationships between those. However, since there is a large amount of variables, the plot is unreadable and somehow very low quality, no matter what I do.
Here is my code (I created dummy DF):
df2 <- data.frame(O1 = seq(1,16,by=2),
O2 = seq(1,16,by=2),
O3 = seq(1,16,by=2),
O4 = seq(1,16,by=2),
O5 = seq(1,16,by=2),
O6 = seq(1,16,by=2),
O7 = seq(1,16,by=2),
O8 = seq(1,16,by=2),
O9 = seq(1,16,by=2),
O10 = seq(1,16,by=2),
C1 = seq(1,16,by=2),
C2 = seq(1,16,by=2),
C3 = seq(1,16,by=2),
C4 = seq(1,16,by=2),
C5 = seq(1,16,by=2),
C6 = seq(1,16,by=2),
C7 = seq(1,16,by=2),
C8 = seq(1,16,by=2),
C9 = seq(1,16,by=2),
C10 = seq(1,16,by=2),
E1 = seq(1,16,by=2),
E2 = seq(1,16,by=2),
E3 = seq(1,16,by=2),
E4 = seq(1,16,by=2),
E5 = seq(1,16,by=2),
E6 = seq(1,16,by=2),
E7 = seq(1,16,by=2),
E8 = seq(1,16,by=2),
E9 = seq(1,16,by=2),
E10 = seq(1,16,by=2),
A1 = seq(1,16,by=2),
A2 = seq(1,16,by=2),
A3 = seq(1,16,by=2),
A4 = seq(1,16,by=2),
A5 = seq(1,16,by=2),
A6 = seq(1,16,by=2),
A7 = seq(1,16,by=2),
A8 = seq(1,16,by=2),
A9 = seq(1,16,by=2),
A10 = seq(1,16,by=2),
N1 = seq(1,16,by=2),
N2 = seq(1,16,by=2),
N3 = seq(1,16,by=2),
N4 = seq(1,16,by=2),
N5 = seq(1,16,by=2),
N6 = seq(1,16,by=2),
N7 = seq(1,16,by=2),
N8 = seq(1,16,by=2),
N9 = seq(1,16,by=2),
N10 = seq(1,16,by=2))
model_fa <- ' Openness =~ O1 + O2 +O3 + O4 + O5 + O6 + O7 + O8 + O9 + O10
Contientiousness =~ C1 + C2 + C3 + C4 + C5 + C6 + C7 + C8 + C9 + C10
Extraversion =~ E1 + E2 + E3 + E4 + E5 + E6 + E7 + E8 + E9 + E10
Agreeableness =~ A1 + A2 + A3 + A4 + A5 + A6 + A7 + A8 + A9 + A10
Neuroticism =~ N1 + N2 + N3 + N4 + N5 + N6 + N7 + N8 + N9 + N10'
#fit model - select just items
fit_cfa <- cfa(model_fa, data = df2,
std.lv=TRUE)
semPaths(fit_cfa, what = 'par', weighted = FALSE, layout = 'tree', nCharNodes = 1,
residuals = F, thresholds = F, curvePivot=TRUE, rotation = 2, sizeMan = 3)
No matter what I change, I end up with an unreadable plot. Can I do some "Stacking" or "dodging", or maybe a hierarchical, tree-like structure?
Also, why is it so low quality? Can I change resolution?
