A contingency table also known as a two-way frequency table is a tabular mechanism with at least two rows and two columns used in statistics to present categorical data in terms of frequency counts.
A contingency table shows the observed frequency of two variables arranged into rows and columns. The intersection of a row and a column of a contingency table is called a cell.
For example, a cell count nij represents a joint occurrence of row i and column j where i is a value between 1 to r (total number of rows) and j is a value between 2 to c (total number of columns).
- The First column represents the first category, gender, and has two labels, female and male which are represented by two rows.
- The second category, habits has two labels, smokers and non-smokers which are represented by the second and third columns.
The second category can have at most 2046 unique labels. The function ignores NULL values in the table.
Maximum label length is 64000 for category_1, 128 for all other columns.
For a valid test output, the value of each observed frequency in the CONTINGENCY table must be at least 5.
CONTINGENCY Table Schema
| Column | Data Type | Description |
|---|---|---|
| Name of categorical column 1 | Any | Columns can have one or multiple labels. Can either be an integer, LATIN, or UTF8 code. |
| category_2_label_1 | INTEGER, SMALLINT, BYTEINT, or BIGINT | Joint frequency of category 1 label i and category 2 label 1, where i has a value between 1 to r. |
| category_2_label_2 | INTEGER, SMALLINT, BYTEINT, or BIGINT | Joint frequency of category 1 label i and category 2 label 2, where i has a value between 1 to r. |
|
. . . . |
||
| category_2_label_c | INTEGER, SMALLINT, BYTEINT, or BIGINT | [Column appears zero or more times.] Joint frequency of category 1 label i and category 2 label c, where i has a value between 1 to r. |