Question: Schei and Bakketeig (1989) took an SRS of 150 women between 20 and 49 years of age from the city of Trondheim, Norway. Their goal was to investigate the relationship between sexual and physical abuse by a spouse and certain gynecological symptoms in the women. Of the 150 women selected to be in the sample, 15 had moved, I had died, 3 were excluded because they were not eligible for the study, and 13 refused to participate. Of the 118 women who participated in the study, 20 reported some type of sexual or physical abuse from their spouse: Eight reported being hit, 2 being kicked or bitten, 7 being beaten up, and 3 being threatened or cut with a knife. Seventeen of the women in the study reported a gynecological symptom of irregular bleeding or pelvic pain. The numbers of women falling into the four categories of gynecological symptom and abuse by spouse are given in the following contingency table:
Gynecological Symptom Present?
Abuse
No Yes
No 89 12 101
Yes 9 8 17
98 20 118
a If abuse and presence of gynecological symptoms are not associated, what are the expected probabilities in each of the four cells?
b Perform a chi-square test of association for the variables abuse and presence of gynecological symptoms.
c What is the response rate for this study? Which definition of response rate did you use? Do you think that the nonresponse might affect the conclusions of the study? Explain.