Display the p-values in the weights matrix of network analysis
Dear all,
I was invited to post my request about the network analysis from the JASP facebook page.
I have a database made by ordered + continuous variables, thus I used the "correlation" as estimator and the "cor_auto" as option. By reading JASP guide, in this case the weights matrix is computed according to the polychoric or tetrachoryc correlation rather than Pearson correlation.
Therefore, I would know if it is possible to display the p-values in the weights matrix arising from the network analysis, since JASP does not enabel to compute polychoric and tetrachoryc correlations with another module.
Thank you very much in advance for your help and for this great software.
Giovanni
Comments
I'm afraid that JASP does not return the p-values right now. However, the cor_auto function is just a wrapper for the lavaan method, which is also implemented through the SEM module. You can get polychoric/tetracoric correlation estimates with p-values there. Just enter a saturated model in the SEM module (e.g. the model 'x ~~ y' if you have columns x and y) and make sure your variables are encoded as ordinal.