Overview for Cluster Observations

Use Cluster Observations to join observations that share common characteristics into groups. This analysis is appropriate when you do not have any initial information about how to form the groups.

For example, a designer for a sporting goods company uses cluster observations to define different groups for testing a new soccer goalie glove. The designer wants to group the athletes who wear the glove by their similarities, such as height, weight, handedness, and so on.

Cluster observations uses a hierarchical procedure to form the groups. At each step, two groups (clusters) are joined, until only one group contains all the observations at the final step. At each step of the clustering process, Minitab calculates similarity and distance values for the groups to help you select the final grouping of observations. You can also display a dendrogram to visualize the grouping results at each step.

Where to find this analysis

To cluster observations, choose Stat > Multivariate > Cluster Observations.

When to use an alternate analysis

  • If you have enough information to provide initial starting points for the groups, use Cluster K-Means.
  • If you know the correct final groups for the observations and want to classify new observations into those groups, use Discriminant Analysis.
  • If you want to group variables, instead of observations, use Cluster Variables.
By using this site you agree to the use of cookies for analytics and personalized content.  Read our policy