In: Statistics and Probability
To test the belief that sons are taller than their fathers, a student randomly selects 6 fathers who have adult male children. She records the height of both the father and son in inches and obtains the accompanying data. Are sons taller than their fathers? Use the alpha equals 0.1 level of significance. Note that a normal probability plot and boxplot of the data indicate that the differences are approximately normally distributed with no outliers.
Observation |
1 |
2 |
3 |
4 |
5 |
6 |
|
---|---|---|---|---|---|---|---|
Height of father (in inches),
Xi |
69.8 |
66.3 |
71.8 |
67.1 |
71.9 |
70.5 |
|
Height of son (in inches),
Yi |
73.2 |
68.5 |
68.0 |
67.4 |
68.0 |
76.4 |
Here we want to test,
Ho: The average height of father and son is same
Vs
H1: The average height of father is less than his son's height (ie. son are taller than their father)
x: father's height , y: son's height
Here we will use the R software to test which is free software,
the R-codes as follows also i will provide the procedure of test by
theory.
> father_Ht=c(69.8,66.3,71.8,67.1,71.9,70.5)
> son_Ht=c(73.2,68.5,68.0,67.4,68.0,76.4)
> xbar=mean(father_Ht)
> ybar=mean(son_Ht)
> xbar
[1] 69.56667
> ybar
[1] 70.25
> sx=sqrt(var(father_Ht))
> sx
[1] 2.371216
> sy=sqrt(var(son_Ht))
> sy
[1] 3.683341
> sp=(5*sx+5*sy)/10
> sp
[1] 3.027279
> k=sqrt((1/6)+(1/6))
> k
[1] 0.5773503
> t_stat=(xbar-ybar)/(sp*k)
> t_stat
[1] -0.3909677
> ttab=qt(0.99,10)
> ttab
[1] 2.763769
xbar= 69.56667 ybar= 70.25
The test statistic under Ho
t-stat=(xbar-ybar)/(sp*k)
= -0.3909677
and table value is ttab=qt(0.99,10)=2.763, here t-stat < -ttab, therefor we may accept ho at the 1% level of significance. and conclude that the average of father's height and son's height is same, there is no statistical evidance that the son are taller than the father.
OR
> tcal=t.test(father_Ht,son_Ht,var.equal=T,level=0.99)
> tcal
Two Sample t-test
data: father_Ht and son_Ht
t = -0.3821, df = 10, p-value = 0.7104
alternative hypothesis: true difference in means is not equal to
0
95 percent confidence interval:
-4.668077 3.301410
sample estimates:
mean of x mean of y
69.56667 70.25000
table value is ttab=qt(0.99,10)=2.763, here t-stat < -ttab, also P-value > 0.01, therefore we may accept Ho at the 1% level of significance. and conclude that the average of father's height and son's height is same, there is no statistical evidance that the son are taller than the father.