Nodes sliding off path diagram in R

Question

This is a sample dput since the dataset is huge:

> dput(head(dat, n=20))
structure(list(q01 = c(2, 1, 2, 3, 2, 2, 2, 2, 3, 2, 2, 2, 3, 
2, 2, 3, 1, 2, 2, 2), q02 = c(1, 1, 3, 1, 1, 1, 3, 2, 3, 4, 1, 
1, 1, 2, 2, 1, 2, 2, 3, 1), q03 = c(4, 4, 2, 1, 3, 3, 3, 3, 1, 
4, 5, 3, 3, 1, 3, 2, 5, 3, 4, 1), q04 = c(2, 3, 2, 4, 2, 2, 2, 
2, 4, 3, 2, 3, 4, 2, 4, 2, 2, 3, 2, 2), q05 = c(2, 2, 4, 3, 2, 
4, 2, 2, 5, 2, 2, 4, 3, 2, 2, 2, 1, 3, 3, 3), q06 = c(2, 2, 1, 
3, 3, 4, 2, 2, 3, 1, 1, 3, 2, 2, 2, 2, 1, 4, 1, 4), q07 = c(3, 
2, 2, 4, 3, 4, 2, 2, 5, 2, 2, 3, 3, 3, 3, 2, 1, 3, 1, 4), q08 = c(1, 
2, 2, 2, 2, 2, 2, 2, 5, 2, 2, 1, 3, 2, 2, 2, 1, 2, 1, 1), q09 = c(1, 
5, 2, 2, 4, 4, 3, 4, 3, 3, 5, 3, 2, 2, 2, 2, 4, 5, 5, 5), q10 = c(2, 
2, 2, 4, 2, 3, 2, 2, 3, 2, 2, 2, 3, 3, 3, 3, 1, 2, 2, 1), q11 = c(1, 
2, 3, 2, 2, 2, 2, 2, 5, 2, 1, 2, 3, 2, 2, 2, 1, 3, 1, 2), q12 = c(2, 
3, 3, 2, 3, 4, 2, 3, 5, 3, 3, 3, 4, 4, 3, 3, 2, 3, 3, 5), q13 = c(2, 
1, 2, 2, 3, 3, 2, 2, 5, 2, 1, 2, 4, 2, 2, 2, 1, 3, 1, 2), q14 = c(2, 
3, 4, 3, 2, 3, 2, 2, 5, 1, 2, 2, 4, 4, 3, 3, 1, 3, 2, 5), q15 = c(2, 
4, 2, 3, 2, 5, 2, 3, 5, 2, 1, 3, 4, 4, 3, 2, 1, 4, 2, 5), q16 = c(3, 
3, 3, 3, 2, 2, 2, 2, 5, 3, 2, 3, 4, 4, 4, 3, 2, 3, 3, 5), q17 = c(1, 
2, 2, 2, 2, 3, 2, 2, 5, 2, 2, 2, 3, 2, 2, 2, 2, 2, 1, 2), q18 = c(2, 
2, 3, 4, 3, 5, 2, 2, 5, 2, 2, 2, 3, 4, 3, 3, 1, 2, 1, 5), q19 = c(3, 
3, 1, 2, 3, 1, 3, 4, 2, 3, 5, 3, 2, 1, 3, 2, 4, 2, 4, 1), q20 = c(2, 
4, 4, 4, 4, 5, 2, 3, 5, 3, 3, 4, 4, 5, 4, 3, 2, 3, 2, 5), q21 = c(2, 
4, 3, 4, 2, 3, 2, 2, 5, 2, 2, 3, 4, 5, 4, 2, 1, 3, 2, 5), q22 = c(2, 
4, 2, 4, 4, 1, 4, 4, 3, 4, 5, 4, 3, 3, 4, 3, 4, 3, 4, 5), q23 = c(5, 
2, 2, 3, 4, 4, 4, 4, 3, 4, 5, 4, 4, 1, 4, 4, 4, 4, 4, 5)), variable.labels = c(q01 = "Statistics makes me cry", 
q02 = "My friends will think I'm stupid for not being able to cope with SPSS", 
q03 = "Standard deviations excite me", q04 = "I dream that Pearson is attacking me with correlation coefficients", 
q05 = "I don't understand statistics", q06 = "I have little experience of computers", 
q07 = "All computers hate me", q08 = "I have never been good at mathematics", 
q09 = "My friends are better at statistics than me", q10 = "Computers are useful only for playing games ", 
q11 = "I did badly at mathematics at school", q12 = "People try to tell you that SPSS makes statistics easier to understand but it doesn't", 
q13 = "I worry that I will cause irreparable damage because of my incompetenece with computers", 
q14 = "Computers have minds of their own and deliberately go wrong whenever I use them", 
q15 = "Computers are out to get me", q16 = "I weep openly at the mention of central tendency", 
q17 = "I slip into a coma whenever I see an equation", q18 = "SPSS always crashes when I try to use it", 
q19 = "Everybody looks at me when I use SPSS", q20 = "I can't sleep for thoughts of eigen vectors", 
q21 = "I wake up under my duvet thinking that I am trapped under a normal distribtion", 
q22 = "My friends are better at SPSS than I am", q23 = "If I'm good at statistics my friends will think I'm a nerd"
), codepage = 65001L, row.names = c(NA, 20L), class = "data.frame")

I mostly copied another semPath model but edited it to fit the dataset I was using. First the nodes:

nodeNames <- c(
  "Statistics makes me cry.",
  "My friends think I'm stupid for not being able to cope with SPSS.",
  "Standard deviations excite me.",
  "I dream that Pearson is attacking me with correlation coefficients.",
  "I don't understand statistics.",
  "I have little experience with computers.",
  "All computers hate me.",
  "I've never been good at mathematics.",
  "SPSS Anxiety"
)

Then the actual semPath:

semPaths(onefac8items_a,
         what = "std", # this argument controls what the color of edges represent. In this case, standardized parameters
         whatLabels = "est", 
         style = "lisrel", 
         residScale = 8, 
         theme = "colorblind",
         manifests = paste0("q",1:8),
         nCharNodes = 0, 
         reorder = FALSE, 
         nodeNames = nodeNames, 
         legend.cex = 0.5, 
         rotation = 2, 
         layout = "tree2", 
         cardinal = "lat cov",  
         curvePivot = TRUE, 
         sizeMan = 4,
         sizeLat = 10, 
         mar = c(2,5,2,5.5), 
         filetype = "pdf", width = 8, height = 6, filename = "SPSS Anxiety" 
)

So I really only have one question here. When I try to run my path diagram, the nodes look like they are sliding off to the right of the page. How do I fix this? Below is a picture of what I'm referring to:

Answer 1

I ended up just shortening my questions down in the nodes and it fixed the problem. I guess there's a limit to how much text you can put into your legend:

nodeNames <- c(
  "Statistics makes me cry.",
  "Friends think I'm stupid because I cant do SPSS.",
  "Standard deviations excite me.",
  "I dream that Pearson is attacking me with correlations.",
  "I don't understand statistics.",
  "I have little experience with computers.",
  "All computers hate me.",
  "I've never been good at mathematics.",
  "SPSS Anxiety"
)

Answer 2

Your page isn't big enough.

There are two graphics systems in R, base and grid. The one semPaths uses is the base package which sort of mimics how you draw on a paper: first you set up the size, then you draw things; you can't go back. The other, grid, is used in lattice and ggplot2 which saves the plotting until you call for it. grid plots typically do not run off the page as base graphics can, the plots are usually scaled to fit with the plotting region.

Here is basically your problem using an example from lavaan::cfa

library('lavaan')
library('semPlot')

nodeNames <- c(
  "Statistics makes me cry.",
  "My friends think I'm stupid for not being able to cope with SPSS.",
  "Standard deviations excite me.",
  "I dream that Pearson is attacking me with correlation coefficients.",
  "I don't understand statistics.",
  "I have little experience with computers.",
  "All computers hate me.",
  "I've never been good at mathematics.",
  "SPSS Anxiety"
)

?semPlot::semPaths
example(cfa)
semPaths(
  fit,
  what = "std", # this argument controls what the color of edges represent. In this case, standardized parameters
  whatLabels = "est", 
  style = "lisrel", 
  residScale = 8, 
  theme = "colorblind",
  # manifests = paste0("q",1:8),
  nCharNodes = 0, 
  reorder = FALSE, 
  nodeNames = nodeNames, 
  legend.cex = 0.5, 
  rotation = 2, 
  layout = "tree2", 
  cardinal = "lat cov",  
  curvePivot = TRUE, 
  sizeMan = 4,
  sizeLat = 10,
  mar = c(2,5,2,5.5),
  filetype = "pdf", width = 8, height = 6, filename = "SPSS-Anxiety"
)

I'm not sure what semPaths is doing here with the size because it is definitely not coming out 8x6

$ identify -verbose SPSS-Anxiety.pdf | grep "Print size"
8:  Print size: 11.1944x6

I'm guessing it compensates for the extra features to fit everything on, but it is not doing a very good job.

The typical way to save base plots to file is

pdf() ## or png() or jpg() etc
plotting code
dev.off() ## or graphics.off() to close everything not just the current device

And to do this you need to remove the filetype part from your code

pdf('SPSS-Anxiety-2.pdf', width = 8, height = 6)
par(oma = c(0, 2, 0, 25), xpd = NA)
semPaths(
  fit,
  what = "std", # this argument controls what the color of edges represent. In this case, standardized parameters
  whatLabels = "est", 
  style = "lisrel", 
  residScale = 8, 
  theme = "colorblind",
  # manifests = paste0("q",1:8),
  nCharNodes = 0,
  reorder = FALSE, 
  nodeNames = nodeNames, 
  legend.cex = 0.5, 
  rotation = 2, 
  layout = "tree2", 
  cardinal = "lat cov",  
  curvePivot = TRUE, 
  sizeMan = 4,
  sizeLat = 10,
  mar = c(2,5,2,5.5)
)
dev.off()

Now I am getting something 8x6

$ identify -verbose SPSS-Anxiety-2.pdf | grep "Print size"
8:  Print size: 8x6

I increased the size of the outer margins, oma see ?par, which gives me 2 extra lines of space on the left and 25 on the right. Also, note xpd = NA which turns off clipping, ie, anything printed outside of the plotting area will be shown--this also comes up a lot in base plots.

But this is a lot of wasted space for some text. I would either scale it down or split the text into multiple lines. You can use strwrap to split each label at white space into <= some maximum width:

par(oma = c(0, 0, 3, 0))
semPaths(
  fit,
  what = "std", # this argument controls what the color of edges represent. In this case, standardized parameters
  whatLabels = "est", 
  style = "lisrel", 
  residScale = 8, 
  theme = "colorblind",
  # manifests = paste0("q",1:8),
  nCharNodes = 0,
  reorder = FALSE, 
  nodeNames = sapply(nodeNames, function(x)
    paste(strwrap(x, 30), collapse = '\n      ')),
  legend.cex = 0.5, 
  rotation = 2, 
  layout = "tree2", 
  cardinal = "lat cov",  
  curvePivot = TRUE, 
  sizeMan = 4,
  sizeLat = 10,
  mar = c(2,5,2,5.5)
)
title('Anxiety and Depression SEM Path Diagram', outer = TRUE)

Answer 3

Since you didn't share your model, I reproduced a dummy model. It seems semPaths doesn't allow us to adjust nodeNames, maybe you could save this graph as an object and try to reproduce with the "plot()" function in order to rescaling since semPaths has a lot of attributes.

semPaths(fit,
     what = "std",
     style = "lisrel", 
     residScale = 8, 
     theme = "colorblind",
     nCharNodes = 4, 
     reorder = FALSE, 
     nodeNames = nodeNames, 
     legend.cex = 0.35, 
     rotation = 2, 
     layout = "tree2", 
     cardinal = "lat cov",  
     curvePivot = TRUE)

Or we could change the GLRatio in the plotOptions:

a<-semPaths(onefac8items_a,
         what = "std",
         whatLabels = "est", 
         style = "lisrel", 
         residScale = 8, 
         theme = "colorblind",
         nCharNodes = 0, 
         reorder = FALSE, 
         nodeNames = nodeNames, 
         legend.cex = 0.5, 
         rotation = 2, 
         layout = "tree2", 
         cardinal = "lat cov",  
         curvePivot = TRUE, 
         sizeMan = 4,
         sizeLat = 10, 
         mar = c(2,5,2,5.5)
)

a$plotOptions$GLratio<-1  # you may need to play with this number
plot(a)