PCAnet function

PCAnet

Convenience function for converting a qgraph object to a layout determined by principal components analysis

PCAnet(
  qgraph_net,
  cormat,
  varTxt = F,
  repulse = F,
  repulsion = 1,
  principalArgs = list(),
  ...
)

Arguments

• qgraph_net: an object of type qgraph
• cormat: the correlation matrix of the relevant data. If this argument is missing, the function will assume that the adjacency matrix from qgraph_net is a correlation matrix
• varTxt: logical. Print the variance accounted for by the PCA in the lower left corner of the plot
• repulse: logical. Add a small repulsion force with wordcloud package to avoid node overlap?
• repulsion: scalar for the repulsion force (if repulse=T). Larger values add more repulsion
• principalArgs: additional arguments in list format passed to psych::principal
• ...: additional arguments passed to qgraph

Details

A network plotted with PCA can be interpreted based on coordinate placement of each node. A node in the top right corner scored high on both the first and second principal components

References

Jones, P. J., Mair, P., & McNally, R. J. (2018). Visualizing psychological networks: A tutorial in R. Frontiers in Psychology, 9, 1742. https://doi.org/10.3389/fpsyg.2018.01742