SAS learning Edition Marketing Research Help

The instructions given here and in all the data analysis chapters (14 to 22) will work with the SAS Learning Edition as well as with the SAS Enterprise Guide. For a point-and-click approach for performing cluster analysis, use the Analyze task within the SAS Learning Edition. The Multivariate>Cluster Analysis task creates hierarchical clusters from data that contains either coordinate or distance data. If the data set contains coordinate data, the task computes Euclidean distances before applying the clustering methods. Alternatively, the task can create” nonhierarchical clusters of coordinate data by using the k-means method. The task also produces dendrograms.

To select this procedure using SAS Learning Edition, click:
Analyze>Multivariate>Cluster Analysis

The following are the detailed steps for running hierarchical cluster analysis on attitudinal data (VI to V6) of Table 20.1.

  1. Select ANALYZE from the SAS Learning Edition menu bar.
  2. Select MULTIVARIATE>CLUSTER ANALYSIS.
  3. Move VI-V6 to the ANALYSIS VARIABLES task role.
  4. Click CLUSTER and select WARD’S MINIMUM VARIANCE METHOD under CLUSTER METHOD.
  5. Click RESULTS and select SIMPLE SUMMARY STATISTICS.
  6. Click RUN.

The following are the detailed steps for running nonhierarchical (k-Means) cluster analysis on the attitudinal data of Table 20.1.

  1. Select ANALYZE from the SAS Learning Edition menu bar .
  2. Select MULTIVARIATE>CLUSTER ANALYSIS.
  3. Move VI-V6 to the ANALYSIS VARIABLES task role.
  4. Click CLUSTER and select K-MEANS ALGORlTIIM as the CLUSTER METHOD and 3 for the MAXIMUM NUMBER OF CLUSTERS.
  5. Click RUN.

SAS does not provide TwoStep cluster analysis.

Cluster Analysis

In the department store project, respondents were clustered on the basis of self-reported importance attached to each factor of the choice criteria utilized in selecting a department store. ‘The results indicated that respondents could be clustered into four segments. Differences among the segments were statistically tested.Thus, each segment contained respondents who were relatively homogeneous with respect to their choice criteria. The store choice model was then estimated separately for each segment. This procedure resulted in choice models that better represented the underlying choice process of respondents in specific segments.

Project Activities 

Download the SPSS or SAS data file Sears Data 17 from the Web site for this book. for a description of this fIle.

  1. Can the respondents be segmented based on the factor scores (that you generated) for the 21 lifestyle statements? Use Ward’s procedure to deterniine the number of clusters fIlen cenduct cluster analysis (use the k-means procedure) by selecting all the factor scores.
  2. Can the respondents be segmented based on the importance attached to the eight factors of the choice criteria? Use Ward’s procedure to determine the number of clusters. Then conduct cluster analysis (use the k-means procedure) by selecting all the factors. Interpret the resulting benefit segments.
A Concept Map for Cluster Analysis

A Concept Map for Cluster Analysis

Posted on November 30, 2015 in Cluster Analysis

Share the Story

Back to Top