k-means and hierarchical clustering

I’m about to manage my SPSS research myself by to do it i need to know the difference between a k-means clustering and hierarchical clustering. I will appreciate every your advice!

1 comment to k-means and hierarchical clustering

  • Cluster analysis in SPSS research is an analysis which is used for data segmentation with verity of goals. This analysis can include a collection of objects related with a group of clusters. The aim of this analysis is to assess the degree of similarity/dissimilarity between the clustered object. There are two important methods of clustering; one is hierarchical clustering which is used to identify relatively homogeneous groups of cases and another is k-means clustering which can assign each point to the cluster whose center is nearest among all points. Moreover, Hierarchical clustering can create a hierarchy which is present in a tree structure, that termed as Dendrogram. The root of this tree comprises of a solo cluster which has all the observations and leaves represent the individual observations. Again, in k-means clustering researcher needs to choose the number of clusters K then randomly generate the K cluster is required. Then the cluster center should be determined. After that, the nearest point should be estimated by using a distance measure.

    For doing this analysis in SPSS research first you need to click on Analyze option on your data set. After that you have to choose Classify and then Hierarchical Cluster option. Then put the variable names that you want to in cluster cases and display both statistics and plots. Further, click Statistics and specify if you would like to see an Agglomeration schedule and followed by click Continue. Again, click plots and specify if you want a Dendogram. If you want to have the Between-groups linkage method, Euclidian distances and z scores then choose your options. Again, click Continue and then OK.

    For k-means clustering similarly need to click Analyze option then choose Classify and then choose k-Means Cluster Analysis option. Then put the variables and then click Iterate and put the value of maximum Iterates. After that, go Options sub menu and click the initial cluster centers. If you want the ANOVA table in your SPSS research then click on it and followed by Continue and Ok option.

Leave a Reply

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>