I've sent you some links written in Persian

https://astat.ir/clustering-analysis-with-spss/

https://astat.ir/clustering-analysis-with-minitab/

https://astat.ir/discriminate-analysis-with-spss/

Many thanks Abolfazl!

Cluster analysis followed by PCA could be a workable option

Hi all !

I appréciate the responses given to Mayhar especialy the references that would optimize my procedure. I often used SPSS and a major problem is that it does NOT orvide -- as much I know, if I'm wrong please correct -- a sort of cut-off criteria to choose the best solution on the number of clusters in a data set according to parameters/variables.

There is only a rule of thumb artisanal procedure. Here is how I proceed :

1. observing the data with descriptive statistics - a scaterplot matrixs is of great help to get a hypothesis of an approximative range of minimum and maximum clusters - sometimes, when not too much parameters, a multidimensional scaling may be instructive.

2. run a hierarchical cluster analysis using a agglomerative algorithme (mostly a Ward but you have different choices of methods and the one you use depends on your research question / purpose. In this step you can choose the type of data you use; but this is only une type so mosly I read observed a multiple clustering with a clustering for all type of data and than a general clustering on the cluster solutions obtained with the diffrent type of data. I ask SPSS to provide the dendrogramme on which I apply the Best-cut criteria (first longest iterration distance) with the agglomeration schedule I choose the first longest distance also.

3a. Using another agglomerative method is sometimes useful but you may know that a perfect cluster fit is difficult to reach as you use different algorithm. To examine clustering fit, I use Chi2 (is there fit ? p