(This is also the reason why we chose to treat some variables as continuous - when they have too many categories - as then we can use another tool to visualize and examine them, as we will see later.) Thus, below I introduce the simplest form of a contingency table, a 2×2 crosstab (i.e., 2 rows and 2 columns). While a contingency table can have any number of rows and columns, too large a number of either/or both can easily make the table unreadable as it would contain too much data to contemplate at once. The potential association between discrete variables can be examined both visually and numerically via a special table called cross-tabulation table (“cross-table” or “crosstab” for short) or contingency table. Again, this sounds far worse than it actually is, as you will see in the examples that follow. Examining a potential statistical association between two discrete variables amounts to comparing groups (as per the categories of one of the variables) on the number (and proportion) of their respective members that fall in the categories of the other variable.
0 Comments
Leave a Reply. |