problem-7.27

problem-7.27  This data is paired off, as it is reasonable to assume that there is a correlation between the IQ scores of twins. We assume that differences in the scores can be viewed as a random sample from a normally distributed population. Then a confidence interval is found as follows:
> foster = c(80, 88,75, 113, 95, 82, 97, 94, 132, 108)
> biological = c(90, 91, 79, 97, 97, 82, 87, 94, 131, 115)
> plot(foster, biological)
> boxplot(foster - biological)
> t.test(foster, biological, conf.level=0.90, paired=TRUE)

        Paired t-test

data:  foster and biological
t = 0.041, df = 9, p-value = 0.9682
alternative hypothesis: true difference in means is not equal to 0
90 percent confidence interval:
 -4.369  4.569
    
The exploratory plots show the correlation between the two sets of scores and the distribution of the differences, indicating that a paired t-test is appropriate.