发帖
雷达卡
Data on employees from onejob category (skilled, entry–level clerical) of a bank that was sued for sexdiscrimination. The data are on 32 male and 61 female employees, hired between1965 and 1975.
A data frame with93 observations on the following 7 variables.
Bsal: Annual salary at time of hire
Sal77: Salary as of March 1975
Sex: Sex of employee
Senior: Seniority (months since first hired)
Age: Age of employee (in months)
Educ: Education (in years)
Exper: Work experience prior to employmentwith the bank (months)
u Did the females receive lower startingsalaries than similarly qualified and similarly experienced males?
(1) Use two graphical methods and a two-samplet-test to show whether the distribution of starting salary, i.e. Bsal, is different for males than forfemales.
(2) Explainwhy it is difficult to prove sexdiscrimination using the above two-sample t-test. Consider a more appropriate model to exam whether the difference betweenthe average starting salaries given to males and to females is likely due tosex discrimination. [Is a transformationof the response variable is necessary? Provide the reasons for your chosenmodel].
u Did females receive smaller annual salaryincreases than males?
If an annual raiseof 100r% is received in each of N successive years of employment, the salary in1977 is Sal77 = Bsal * (1+r)^N. Seniority measures the number of months of employment, so the number ofyears of employment is N=seniority/12.
Now consider r asa response variable (the average annual raise).
(3) Use atwo-sample t-test to see whether the distribution of raise is different formales than for females.
(4) Whatevidence is there of a sex effect after the effect of age on average raise hasbeen account for?
(5) Whatevidence is there of a sex effect after the effects of age and beginning salaryhave been accounted for?
京公网安备 11010802022788号
