Running the Analysis

  1. The setup of the data requires that the cases be weighted by the variable freq. To do this, from the menus choose:

    Data > Weight Cases...

    Figure 1. Weight Cases dialog box
    Weight Cases dialog box with freq selected as the Frequency Variable
  2. In the Weight Cases dialog, choose to weight cases by freq.
  3. Click OK.
  4. To obtain an initial solution in five dimensions with principal normalization, from the menus choose:

    Analyze > Dimension Reduction > Correspondence Analysis...

    Figure 2. Correspondence Analysis dialog box
    Correspondence Analysis main dialog box with image selected as the Row variable
  5. Select image as the row variable.
  6. Click Define Range.
    Figure 3. Define Row Range dialog box
    The Correspondence Analysis Define Row Range dialog box.
  7. In the Define Row Range dialog, type 1 as the minimum value.
  8. Type 23 as the maximum value.
  9. Click Update.
  10. Click Continue.
  11. In the Correspondence Analysis dialog, select brand as the column variable.
  12. Click Define Range.
  13. In the Define Column Range dialog, type 1 as the minimum value.
  14. Type 6 as the maximum value.
  15. Click Update.
  16. Click Continue.
  17. Click Model in the Correspondence Analysis dialog box.
    Figure 4. Model dialog box
    The Correspondence Analysis Model dialog box.
  18. In the Model dialog, select Principal as the normalization method.
  19. Click Continue.
  20. Click Plots in the Correspondence Analysis dialog box.
    Figure 5. Plots dialog box
    The Correspondence Analysis Plots dialog box.
  21. In the Plots dialog, select Row points and Column points in the Scatterplots group.
  22. Click Continue.
  23. Click OK in the Correspondence Analysis dialog box.

Next