Ch. 19: Two samples t-test
The following examples are based on the data, soc510hw2.csv in which female, married, and wage variables are found. Note that females are coded 1 and males are coded 0. How to read a data file into R is found here.
Two-tail t-test
Two-tail t-test of wage difference between male and female.
Ho: Mean wage for male is equal to mean wage for female
Ha: Mean wage for male is not equal to mean wage for femaleor> t.test(wage[female==0], wage[female==1]) Welch Two Sample t-test data: wage[female == 0] and wage[female == 1] t = 37.2086, df = 76223.97, p-value < 2.2e-16 alternative hypothesis: true difference in means is not equal to 0 95 percent confidence interval: 2.208620 2.454241 sample estimates: mean of x mean of y 15.34689 13.01546
> t.test(wage~female)
Look at this link first to learn how to get statistics by group and how to create a subset data.
Two-tail t-test of wage difference between male and female among married.
Ho: Mean wage for married male is equal to mean wage for married female
Ha: Mean wage for married male is not equal to mean wage for married female> t.test(wage[female==0 & married==1], wage[female==1 & married==1])
Right side 1 tail t-test
Right side 1 tail t-test of wage difference between male and female.
Ho: Mean wage for male is not higher than mean wage for female
Ha: Mean wage for male is higher than mean wage forfemale> t.test(wage[female==0], wage[female==1], c("greater"))
Left side 1 tail t-test
Left side 1 tail t-test of wage difference between male and female.
Ho: Mean wage for male is not lower than mean wage for female
Ha: Mean wage for male is lower than mean wage forfemale> t.test(wage[female==0], wage[female==1], c("greater"))
[ Going back to Using R: Index ]