Professional Documents
Culture Documents
MEASUREMENTS TECHNIQUES
12 22 12 22
( y1 y2 ) Z 2 1 2 ( y1 y2 ) Z 2
n1 n2 n1 n2
Example:
A standard chemistry test was given to 50
girls and 75 boys. The girls made an
average of 82 while the boys made an
average of 76.
Construct a 95% confidence interval
of the difference between the population
means assuming that both populations
are normally distributed with variances
64 and 36 for boys and girls respectively.
From sample data:
36 64
y y 1.25
1 2
50 75
Using Tables,
Z/2= 1.96
Thus;
3.54 < 1-2 < 8.46
Conclusions
At the 95% confidence level, the true
1 1 1 1
( y1 y2 ) t 2 S p 1 2 ( y1 y2 ) t 2 S p
n1 n2 n1 n2
where ,
2 2
( n1 1) S1 ( n2 1) S 2
Sp
n1 n2 2
= The Pooled Standard deviation
will be accepted
P-value
Accept
c
IN TERMS OF Z-STATISTIC
Upper Tailed:
p-value = P(Z>Zc)
Lower Tailed:
p-value = P(Z<Zc)
Two Tailed:
p-value = P(Z>|Zc|)
WHY P-VALUE?
Using the p-value allow to reporting test results and
leave the selection of () to the decision maker.
The decision criterion is to reject Ho if > p-value.
100
90
80
70
60
50
40
30
20
10
0
16.2 16.4 16.6 16.8 17 17.2 17.4
Estimation of the difference
between two means (1-2):
Sd Sd
d t 2 d d t 2
n n
Example
Independent
The Two Variances are known
The Two Variances are Unknown but Equal
Dependent
INDEPENDENT
THE TWO VARIANCES ARE KNOWN
OR N>30
Test Statistic:
_ _ _ _
( y1 y 2 ) D0 ( y1 y 2 ) D0
z
_ _ s12 s22
( y1 y 2 )
n1 n2
INDEPENDENT
THE TWO VARIANCES ARE UNKNOWN
BUT EQUAL
Test Statistic
_ _ Degrees of Freedom =
( y1 y 2 ) D0 n1+n2-2
t
1 1
Sp(2
)
n1 n2
( n1 1) S12 ( n2 1) S 22
Sp
n1 n2 2
INDEPENDENT
THE TWO VARIANCES ARE UNKNOWN AND NOT
EQUAL
_ _
Test Statistic ( y1 y 2 ) D0
t
S12 S 22
( )
n1 n2
S12 S 22 2
( )
n1 n2
S12 2 S 22 2
( ) ( )
n1 n2
n1 1 n2 1
EXAMPLE
A quality control inspector compares the ultimate tensile
strength (UTS) measurements for class I and class II pre-
stressing wire by taking a sample of 5 specimens from a
roll of each class for laboratory testing.
Sample data (in 1000 psi) are as follows:
Test Statistic
_
d d0
tc
sd
n
P ( F1 2 ( 1 , 2 ) F F 2 ( 1 , 2 ) ) 1
F/2 and F1-/2 are the F-values corresponding
to an area (/2) to the right and left of the F-
distribution with:
1 = (n1-1) = Numerator d.f.
And,
2= (n2-1) = Denominator d.f.
From the F-distribution Tables.
The statistic F is given by :
S1 2
2 2
F ( 2 )( 2 )
S2 1
Substituting :
S12
1 1
2 2
S1
( 2) 2 ( 2 ) F 2( 2 ,1 )
S 2 F 2(1 , 2 ) 2 S2
Note :
1
F1 2 ( 1 , 2 )
F 2 ( 2 , 1 )
Example 1
Using the F distribution Tables, find:
a) F 0.025 (6,8)
b) F 0.975 (6,8)
f(F)
a) F 0.025 (6,8)
F(6,8)
1
b) F0.975( 6,8)
F0.025(8,6)
1
0.1786
5 .6
Example 2
A random sample of n1=16 measurements on
the breaking strength of a certain type of
material has s12=3.68 (psi)2. Repeated
measurements on a second machine with
n2=10 shows s22=2.3 (psi)2.
Assuming that the measurements are
normally distributed, would you conclude
that the two variances are significantly
different at the 90% confidence level?
At 1-= 0.90 , /2 = 0.05
Using the F distribution Table,
F0.05(15,9) = 3.01
F0.05(9,15) = 2.59
Thus,
1
2
0.532 4.144
2
2
Conclusions
At the 90% confidence level, the ratio
between the two variances is expected to
lie within 0.532 and 4.144.
Since the interval includes a ratio of
1.0, it can be further concluded that the
difference is not significant.
TEST OF HYPOTHESIS FOR THE RATIO
BETWEEN TWO POPULATIONS
VARIANCES
Hypothesis Test Statistic Criteria for Rejection
H0: 12 = 22
S12 F0 F / 2,n1 1,n2 1
H1: 12 ≠ 22 F0 2
S2 or
F0 F1 / 2,n1 1,n2 1
S 22
H0: 12 = 22 F0 2 F0 F ,n2 1,n1 1
S1
H1: 12 < 22
H0: 12 = 22 S12
F0 2 F0 F ,n1 1,n2 1
H1: 1 > 2
2 2 S2
Similar results would have been obtained
by testing the hypothesis:
12 12
H 0 : 2 1.0 vs. H 1 : 2 1.0
2 2
And using the test statistic:
f(F)
3.68
F 1 .6
2.30
F(6,8)