This article describes how to go from a default table with columns that represent different categories:
To a table which also shows the significant differences between the columns within each row:
- A document containing a table which is cross-tabbed by a variable so that each column represents a different category.
1. Select your table. For this example, we're using a table showing cola brand preference by age group.
2. From the object inspector on the right, in the Inputs sections select Statistics > Cells > Column Comparisons. Note that when you do this, Statistics > Below > Column Names will be selected automatically which will display the individual column letters below the table.
The letters are interpreted as follows:
- The letters at the base of the table indicate the names of the columns (i.e., in this case: A, B, C and D).
- Where letters appear in a cell, this indicates that the percentage in that cell is significantly higher than the percentage in the same row for the column represented by the letter. For example, the letter "C" that appears for Coca-Cola in the Under 30 column indicates that the 53% preference for Coca-Cola among people aged less than 30 is significantly higher than the value of 34% for people aged 40 or more in column C. Similarly, we can see that the preference for Coca-Cola among the 30 to 39 year old group is also significantly higher than for people aged 40 or more.
- The absence of column letters is also informative. Since no letters are shown on the Coke Zero row, this indicates that there are no significant differences in preference for Coke Zero among the different age groups.
- By default, letters that are significant at the 99.9% confidence level (i.e., p <= 0.001) are represented by capital letters. Lower-case letters are used to represent confidence levels between the 95% and 99.9% level. In the table above, all the results are significant at between the 95% and 99.9% levels of significance.