Understanding a dendrogram (no cursework)

[hexa]

Hello,

I tried to find a fitting forum but did not quiet succeed. The following diagram is an example of something I've been given at work. I try to understand it but fail with that.
http://img98.imageshack.us/img98/4770/testwk3.png

So how do I understand such diagram? The upper black dot comprises all the data in all 4 groups or colours, the second highest black dot assumes two groups? But where's the boundary then: between the light blue and green? What are the boundaries of the third black dot? Basically: how do I understand such diagram?

Thanks a lot,
H.

 

[chemisttree]

The Dendogram is meaningless without some context.

(Whew! I'm glad I got through that without cursing...)

 

[Blueshift5]

Hello H.,

Thank you for reaching out and sharing your confusion about the dendrogram. A dendrogram is a type of hierarchical clustering diagram that is commonly used to visualize relationships between groups or data points. In this diagram, the vertical axis represents the measure of similarity or distance between the different groups or data points, while the horizontal axis represents the different groups or data points themselves.

In this specific diagram, it looks like the data points are being clustered into four distinct groups or categories, represented by the different colors. The black dots at the top of the diagram represent the overall groupings of the data, with the first black dot encompassing all four groups and the second black dot representing a split between two groups. The boundaries between the groups are shown by the horizontal lines connecting the different colored data points.

The third black dot represents a further split within one of the groups, with the boundary between the light blue and green groups being shown by the horizontal line connecting those two data points. The boundaries of the third black dot are determined by the measure of distance or similarity between the data points within that group.

Overall, the dendrogram is showing the relationships and groupings between the data points, with the distance or similarity between them being used to determine the boundaries between the different clusters. I hope this helps clarify the diagram for you. If you have any further questions, please don't hesitate to ask.