Use the data in Table 12.11 to test the null hypothesis that gender is independent of department. Show that this hypothesis is rejected at the. = 0.01 level. We conclude that department is associated with both gender and admissions. Therefore, department is a confounder for the relationship between gender and admissions. This means that it is possible that the differences in admissions rates between men and women may be due to differences in the departments they apply to, rather than to gender. To determine whether this is the case, we must look at the admissions rates for men and women in each department separately. The following six tables present these data.

Department B Gender Accept Reject Total 313 825 108 Department A pepej Gender Accept Reject Total 353 207560 25 Male Female Total 512 Male Female Total 17 19 332 933 370215 585 601 Department D Department C Gender Accept Reject Total 120 205 325 391 593 322 596 918 Male Female Total Gender Accept Reject Total 138 279 417 AA192 Male Female Total 13 244 375 269 523 792 Department E Department F Gender Accept Reject Total Gender Accept Reject Total 53 138 191 299 393 Male Female Total Male Female 437 584Total 351 373 317 341 668 714 94 147 24 46