To the list of courses || RASA2018 || To the theme || Estonian

Exercise 4041. Points 1, theme: Cluster Analysis

Open exercise
Which 3 tree species set up common stands (grow most often together in the same wood) according to cluster analysis using Euclidean distance and Single linkage?
The coverage points ranging from 0 to 10 are in columns from Kuusk to muu_puuliik in the attached file. The tree names in Latin are in the worksheet Puud.
Data: SAAREMAA.xls


The Euclidean distance means distance function to use, Single Linkage is the grouping rule.

Solution using the SDC. See also attached figure.
  • Open Non-spatial clusters in the SDC, select the above mentioned distance function and grouping method.
  • Check Group variables and uncheck Case names are in the first column.
  • Copy the columns D-Q (Kuusk … muu_puuliik) to the SDC input window. You can select columns by clicking on the header as the SDC automatically removes empty rows from the end.
  • As the first row contains feature names, the checkbox Variable names are in the first row must be checked.
  • Press Calculate,
  • Cut notionally the cluster tree at different levels beginning from the stem until the last 3-trees cluster remains. Write the names of trees in this cluster to the answer.
  • By default, the tree is growing upward. You can also try other directions.

Solution using the IBM SPSS Statistics 22 (written by Najmeh Mozaffaree Pour)
  • Open the attached file and copy the rows and columns and paste them into SPSS software.
  • Change the variables name (as is mentioned in excel file to tree’s name) and change types of trees columns by double click on type (string to numeric).
  • Delete the first row which is the name of variables after changing the name of variables.
  • Select Analysis tab, classify and then hierarchical cluster.
  • Select columns 4 to 17 (trees) and cluster variables.
  • On the right of the window, select statistics and check proximity matrix and press continue, click to plots and check the Dendrogram and press continue, and click to Methods (select nearest neighbor (single Linkage), Euclidean distance) and continue, and then press OK.
  • The output gives use the whole useful information to solve the question.
  • You can change the tree orientation as vertical or horizontal.
For more information about SPSS (Statistical Package for the Social Sciences), press SPSS -- Wikipedia and, of course, for doing statistics online, use Free Trial -- MarketSight, Data and Visualization Software or install this useful statistical package.
SPSS is a powerful software for statistics which is used for market researchers, health researchers, survey companies, government entities, education researchers, marketing organizations, data miners, and many more for the processing and analyzing of survey data.
Log in to send your results and to see the expected answer and responses from other students.