Contingency Tables
This module tests the hypothesis about independence of two categorical variables A, B based on experimentally observed occurence of combinations of particular levels of A, B. Number of levels of A is denoted r, number of levels of B is denoted c. The data are organized in the table of frequencies. Only the thick bordered box needs to be specified, the totals are calculated automatically.

Contingency Tables - Pdf manual

Example:
The the following table we have volumes of products of three types 1, 2 3 sold to 4 different markets, Europe, Asia, USA, and Africa. We want to test if there is a difference in structure between products or between markets. Practically, we test if the proportion between products is the same for any market and simultaneously if the proportion between markets is the same for any product.

Data:

Output:

 Analysis of contingency table Task name : Market Analysis Table of counts Product1 Product2 Product3 Total Europe 114 25 301 440 Theoretical: -117 -31 -292 Asia 68 20 225 313 Theoretical: -83 -22 -208 USA 131 30 240 401 Theoretical: -107 -28 -266 Africa 57 23 159 239 Theoretical: -63 -17 -159 Total 370 98 925 1393 Table of ratios and probabilities Product1 Product2 Product3 Total Europe 0.081838 0.017947 0.21608 0.315865 Theoretical: -0.083898 -0.022222 -0.209745 Asia 0.048816 0.014358 0.161522 0.224695 Theoretical: -0.059682 -0.015808 -0.149205 USA 0.094042 0.021536 0.17229 0.287868 Theoretical: -0.076462 -0.020252 -0.191154 Africa 0.040919 0.016511 0.114142 0.171572 Theoretical: -0.045572 -0.01207 -0.11393 Total 0.265614 0.070352 0.664034 1 Conclusion Independence of variables is rejected Significance level 0.05 Degrees of freedom 6 Chi2 statistic 17.11555756 Critical value 12.59158724 p-value 0.008867805