In: Statistics and Probability
| 
 Assume that both populations are normally distributed. a) Test whether 2μ1≠μ2 at the α=0.05 level of significance for the given sample data. b) Construct a 95% confidence interval about 2μ1−μ2.  | 
 Sample 1  | 
 Sample 2  | 
|||
|---|---|---|---|---|---|
| 
 n  | 
 19  | 
 19  | 
|||
| 
 x overbarx  | 
 11.7  | 
 14.4  | 
|||
| 
 s  | 
 3.4  | 
 3.9  | 
|||
SOLUTION:
From given data,
Assume that both populations are normally distributed
| Sample 1 | Sample 2 | |
![]()  | 
  = 19 | 
  = 19 | 
![]()  | 
 
=11.7 | 
  =
14.4 | 
![]()  | 
  = 3.4 | 
  = 3.9 | 
a) Test whether 2μ1≠μ2 at the α=0.05 level of significance for the given sample data
H0 : 2μ1 = μ2 ( Null hypothesis )
H1 : 2μ1 ≠ μ2 ( Alternative hypothesis )
 -
 =
11.7 - 14.4 = -2.7
SE( 
 -
 ) =
= 
= 1.18699
Test statistic : t =  ( 
 -
 ) /
SE( 
 -
)
= -2.7 / 1.18699
= - 2.2746
t = - 2.2746
degree of freedom =df = 
 + 
 - 2
= 19 + 19 -2 = 36
α=0.05 = > critical value = tcritical = 
 2.028
b) Construct a 95% confidence interval about 2μ1−μ2
95% confidence interval is
( 
 -
 )
tcritical * SE( 
 -
)
- 2.7   
 2.028 * 1.18699
- 2.7   
 2.407215
( - 2.7 - 2.407215 ) , ( - 2.7+ 2.407215 )
-5.1072 , -0.2927
confidence interval about 2μ1−μ2 is from -5.10 to -0.29