I wanted to know if there were any issues with performing a cluster analysis using subscales as the clustering variables? The three subscales are part of a multidimensional measure and they have good internal consistency. They're correlated with one another from .43-.52. I haven't seen anything that says this is inappropriate, but I just want to make sure I'm not missing something. Any guidance or resources would be greatly appreciated. Thanks!
http://personality-project.org/revelle/publications/iclust.pdf
http://www.uoguelph.ca/~manand/Papers/DaleEt07.pdf
Article A Cluster Analysis Investigation of Workaholism as a Syndrome
Article Complexities in Complex Posttraumatic Stress Disorder in Inp...
Have a look at Article Pattern Clustering by Multivariate Mixture Analysis, Multiva...
https://sci2s.ugr.es/keel/pdf/specific/articulo/schafer_yucel02.pdf
Article "Multivariate Statistical Analyses: Cluster Analysis, Factor...
Article Exploring the Correlation Between Multiple Latent Variables ...
Scale values have a specific effect on the modeling object and can be clustered. And scales and subscales contain gradations that have different strength and direction of influence. Therefore, they are not variables that can be used in clustering. But if you calculate for example the average values for the gradations of each scale, you can replace the scale with this average. This will already be a variable and can be used for clustering. In principle I can do this in the Eidos system
- Badges
- Science topic
- Similar topics
- Philosophy
- Philosophy Of Science
- Reasoning