To investigate a possible relationship between breast cancer and late child birth a number of women with breast cancer and controls were sampled. Early first birth (<20) was recorded as well as late first birth (≥35).

(a) Test if there is a relationship. In the original investigation there was surely 2368 rows of data representing the women, but if you want to do a test in SPSS you need only 4 rows of data representing the four cells. You need one variable for disease status and one for early/late first child birth. In addition you need a third variable which you may call e.g. COUNT. In this variable you put the numbers of women in each cell. To tell SPSS that you want this interpretation you choose Data followed by Weight Cases and finally you set COUNT as the Frequency Variable. Now you are ready to do the test.

(b) What is the OR of getting cancer fora late child birth woman compared to an early first birth woman?