41 3 Tests Two Samples

Tests Concerning
Two Samples
41.3
Introduction
So far we have dealt with situations in which we either had a single sample drawn from a
population, or paired data whose dierences were considered essentially as a single sample.
In this Section we shall look at the situations occurring when we have two random samples each
drawn from independent populations. While the basic ideas involved will essentially repeat those
already met, you will nd that the calculations involved are more complex than those already
covered. However, you will nd as before that calculations do follow particular routines. Note
that in general the samples will be of dierent sizes. Cases involving samples of the same size,
while included, should be regarded as special cases.
#
Prerequisites
Before starting this Section you should . . .
"
Learning Outcomes
After completing this Section you should be
able to . . .
be familiar with the normal t-, F - and

chi-squared distributions
be familiar with the contents of Section
41.1 and 41.2
be able to apply the ideas of hypothesis testing to a range of problems underpinned by a substantial range of statistical distributions and involving two samples of dierent sizes
1. Tests concerning Two Samples

Two Independent Populations each with a Known Variance
We assume that the populations are normally distributed. This may not always be true and
you should note this basic assumption while studying this section of the booklet.
A standard notation often used to describe the populations and samples is:
Population
Sample
2
X1 N (1 , 1 ) x11 , x12 , x13 , , x1n1 with n1 members.
If you are not familiar with the double sux notation used to represent the samples, simply
remember that a random sample of size n1 is drawn from X1 N (1 , 12 ) and a random sample
of size n2 is drawn from X1 N (1 , 12 ).
In diagrammatic form the populations may be represented as follows:
X1 N (1 , 12 )
X2 N (2 , 22 )
When we look at hypothesis testing using two means, we will be considering the dierence
1 2 of the means and writing null hypotheses of the form
H0 : 1 2 = Value
As you might expect, Value will often be zero and we will be trying to detect whether there is
any statistically signicant evidence of a dierence between the means.
We know, from our previous work on continuous distributions (see Workbook 38) that:
2 ) = E(X
1 ) E(X
2 ) = 1 2
1 X
E(X
and that
2 ) = V (X
1 ) V (X
2) =
1 X
V (X
12 22
+
n1 n2
1 and X
2 are independent. Given the assumptions made we can assert that the quantity
since X
Z dened by
1 X
) (1 2 )
(X
2 2
Z=
1
2
+ n22
n1
follows the standard normal distribution N (0, 1).
We are now ready to apply this formula to practical problems in which random samples of
dierent sizes are drawn from normal populations. The conditions for the rejection of H0 at the
HELM (VERSION 1: March 18, 2004): Workbook Level 1
41.3: Tests Concerning Two Samples
5% and the 1% levels of signicance are exactly the same as those previously used for single
sample problems.
Example A motor manufacturer wishes to replace steel suspension components by aluminium components to save weight and thereby improve performance and fuel
consumption. Tensile strength tests are carried out on randomly chosen samples of two possible components before a nal choice is made. The results
are:
Component Sample
Mean Tensile
Standard Deviation
2
Number
Size
Strength (kr/mm )
(kr/mm2 )
1
15
90
2.3
2
10
88
2.2
Is there any dierence between the measured tensile strengths at the 5% level
of signicance?
Solution
The null and alternative hypotheses are:
H 0 : 1 2 = 0
H1 : 1 2 = 0
The null hypothesis represent the statement there is no dierence in the tensile strengths of
the two components. The test statistic Z is calculated as:
) (1 2 )
1 X
(90 88) (0)
2
(X
2 2
= 2.186
=
=
Z=
2
1
2
0.3527
+
0.484
2.32
2.22
+ 10
+ n2
15
n1
Since 2.186 > 1.96 we conclude that, on the basis of the (limited) evidence available, there is a
dierence in tensile strength between the components tested. The manufacturer should carry
out more comprehensive tests before making a nal decision as to which component to use.
The decision is a serious one with safety implications as well as economic implications. As well
as carrying out more tests the manufacturer should consider the level of rejection of the null
hypothesis, perhaps using 1% instead of 5%. Component 1 appears to be stronger but this may
not be the case after more tests are carried out.
A motor manufacturer is considering whether or not a new fuel formulation

will improve the maximum power output of a particular type of engine. Tests
are carried out on randomly chosen samples of the two fuels in order to inform
a decision. The results are:
Fuel
Type
1
2
Sample
Mean Maximum
Standard Deviation
Size
Power Output (bhp)
(bhp)
20
1350
10
16
131
8
Is there any dierence between the measured power outputs at the 5% level of
signicance?
3

Your solution
Since 1.33 < 1.96 we conclude that, on the basis of the (limited) evidence available, there is
insucient evidence to conclude that there is a dierence in the maximum power output of the
engines tested when run on the dierent types of fuel.
12
n1
22
n2
102
20
82
16
The null hypothesis represent the statement there is no dierence in the measured maximum
power outputs. The test statistic Z is calculated as:
) ( )
X
(135 131) (0)
4
1
2
2

= 1.33
=
=
5+4
Z=
(X1
H 0 : 1 2 = 0
H1 : 1 2 = 0
The null and alternative hypotheses are:
Two Independent Populations each with an Unknown Variance

Again we assume that the populations are normally distributed and use the same standard
notation used previously to describe the populations and samples, namely:
Population
Sample
2
There are two distinct cases to consider. Firstly, we will assume that although the variances are
unknown, they are in fact equal. Secondly, we will assume that the unknown variances are not
necessarily equal.
Case (i) - Unknown but Equal Variances
Again, when we look at hypothesis testing using two means, we will be considering the dierence
H0 : 1 2 = Value
and again Value will often be zero and we will be trying to detect whether there is any statistically
signicant dierence between the means.
We will take 12 = 22 = 2 so that in diagrammatic form the populations are:
X2 N (2 , 2 )
X1 N (1 , 2 )
The results from our work on continuous distributions (see Workbook *.*) tell us that:
1 X
2 ) = E(X
1 ) E(X
2 ) = 1 2
E(X
as before, and that
2
2
1 X
2 ) = V (X
1 ) V (X
2 ) = 1 + 2
V (X
n1 n2
Given that we do not know the value of , we must estimate it. This is done by combining (or
pooling) the sample variances say S12 and S22 for samples 1 and 2 respectively according to the
formula:
Sc2 =
(n1 1)S12 + (n2 1)S22

n1 + n2 2
Notice that
Sc2 =
(n1 1)S12 + (n2 1)S22

(n1 1)S12
(n2 1)S22
=
+
n1 + n2 2
n1 + n2 2 n1 + n2 2
so that you can see that Sc2 is a weighted average of S12 and S22 . In fact, each sample variance
is weighted according to the number of degrees of freedom available. Notice also that the rst
sample contributes degrees of freedom and the second sample contributes degrees of freedom so
that has degrees of freedom.
Since we are estimating unknown variances, the quantity T dened by
T =
1 X
2 ) (1 2 )
(X

Sc n11 + n12
will follow Students t-distribution with n1 + n2 2 degrees of freedom.

We are now ready to apply this formula to practical problems in which random samples of different sizes with unknown but equal variances are drawn from independent normal populations.
The conditions for the rejection of H0 at the 5% and the 1% levels of signicance are found from
tables of the t-distribution, a copy of which is attached to the end of this booklet.

Example A manufacturer of electronic equipment has developed a circuit to feed current

to a particular component in a computer display screen. While the new design
is cheaper to manufacture, it can only be adopted for mass production if it
passes the same average current to the component. In tests involving the two
circuits, the following results are obtained.
Test Number
1
2
3
4
5
6
7
8
9
10
11
12
Circuit 1 - Current (mA) Circuit 2 - Current (mA)

80.1
80.7
82.3
81.3
84.1
84.6
82.6
81.7
85.3
86.3
81.3
84.3
83.2
83.7
81.7
84.7
82.2
82.8
81.4
84.4
85.2
84.9
On the assumption that the populations from which the samples are drawn
have equal variances, should the manufacturer replace the old circuit design
by the new one? Use the 5% level of signicance.
Solution
If the average current ows are represented by 1 and 2 we form the hypotheses
H 0 : 1 2 = 0
H1 : 1 2 = 0
2 = 83.72.
1 = 82.42 and X
The sample means are X
The sample variances are S12 = 2.00 and S22 = 2.72.
The pooled estimate of the variance is
Sc2 =
(n1 1)S12 +(n2 1)S22

n1 +n2 2
92.00+112.72
20
= 2.396
The test statistic is

2 ) (1 2 )
1 X
82.42 83.72
(X

=
= 1.267
T =
1
1
1
1
Sc n1 + n2
2.396 10 + 12
From ttables, the critical values with 20 degrees of freedom and a two-tailed test are . Since
2.086 < 1.267 < 2.086 we conclude that we cannot reject the null hypothesis in favour of
the alternative. A 95% condence interval for the variance is given by where . The condence
interval is

A manufacturer of steel cables used in the construction of suspension bridges

has experimented with a new type of steel which it is hoped will result in
the cables produced being stronger in the sense that they will accept greater
tension loads before failure. In order to test the performance of the new cables
in comparison with the old cables, samples are tested for failure under tension.
The following results were obtained, the failure tensions being given in kg103 .
Test Number
1
2
3
4
5
6
7
8
9
10
New Cable
92.7
91.6
94.7
93.7
96.5
94.3
93.7
96.8
98.9
99.9
Original Cable
90.2
92.4
94.7
92.1
95.9
91.1
93.2
91.5
The cable manufacturer, on looking at health and safety legislation, decides

that a 1% level of signicance should be used in any statistical testing procedure adopted to distinguish between the cables. On the basis of the results
given, should the manufacturer replace the old cable by the new one? You
may assume that the populations from which the samples are drawn have
equal variances.
Your solution


In the case where we assume unequal variances, there is no exact statistic which we can use
to test the validity or otherwise of the null hypothesis H0 : 1 2 = Value. However, the
following approximation allows us to overcome this problem.
and again Value will often be zero and we will be trying to detect whether there is any statistically
signicant dierence between the means.
H0 : 1 2 = Value
Again, when we look at hypothesis testing using two means, we will be considering the dierence
1
2
X2 N (2 , 22 )
X1 N (1 , 12 )
In this case we will take 12 = 22 so that in diagrammatic form the populations may be represented as shown below.
Case (ii) - Unknown and Unequal Variances
If the average tensions are represented by 1 (new cable) and 2 (old cable) we form the
hypotheses
H 0 : 1 2 = 0
H 1 : 1 2 > 0
in order to test the hypothesis that the new cable is stronger on average than the old cable.
2 = 92.64.
1 = 95.28 and X
The pooled estimate of the variance is
Sc2 =
(n1 1)S12 +(n2 1)S22

n1 +n2 2
96.47+73.14
16
= 5.013

1 X
2 ) (1 2 )
(X
95.28 92.64
2.64

T =
=
= 2.486
1
2.239 10
+ 81 2.239 0.225
Sc
1
n1
1
n2
Using t-distribution tables with 16 degrees of freedom, we see that the critical value at the
1% level of signicance is 2.583. Since 2.486 < 2.583 we conclude that we cannot reject the
null hypothesis in favour of the alternative. However, the close result indicates that more tests
should be carried out before making a nal decision. At this stage the cable manufacturer
should not replace the old cable by the new one on the basis of the evidence available.
Key Point
Provided that the null hypothesis is true, the statistic
2 ) (1 2 )
1 X
(X

T =
Sc n11 + n12
will approximately follow Students distribution with the number of degrees of freedom given by
the expression:
2
2
S1
S22
+
n1
n2
= 2 2 2 2 2
S1
n1
n1 +1
S2
n2
n2 +1
Essentially, this means that the actual test procedure is similar to that used previously but with
T and the number of degrees of freedom calculated using the above formulae.
We are now ready to apply these formulae to practical problems in which random samples
of dierent sizes with unknown and unequal variances are drawn from independent normal
populations. We will illustrate the test procedure by reworking the examples done previously
but we will assume unequal rather than equal variances.

circuits, the results are obtained re those given in the Worked Example on
Page 6. On the assumption that the populations from which the samples are
drawn do not have equal variances, should the manufacturer replace the old
circuit design by the new one? Use the 5% level of signicance.

Solution
If the average current ows are represented by 1 and 2 we form the hypotheses
H 0 : 1 2 = 0
H1 : 1 2 = 0
2 = 83.72.
1 = 82.42 and X
The sample variances are 21 = 2.00 and S22 = 2.72.
) (1 2 )
1 X
82.42 83.72
1.3
(X
2 2
=
=
= 1.990
T =
S1
S22
2.00
2.72
0.427
+ 12
+ n2
10
n1
The number of degrees of freedom is given by
2
2
S1
S22
2.00 2.72 2
+
+ 12
n1
n2
0.182
10
2
=
2 21
= 2 2 2 2 2 = (2.00/10)
2
2
(2.72/12)
S1
S2
0.004 + 0.004
+
11
13
n1
n2
+ n2 +1
n1 +1
From t-tables, the critical values (two-tailed test, 5% level of signicance) are 2.080. Since
2.080 < 1.990 < 2.080 we conclude that there is insucient evidence to reject the null
hypothesis in favour of the alternative at the 5% level of signicance.

10

The results obtained are those given in the Worked Example on Page 23, the
failure tensions being given in kg103 .
Test Number
1
2
3
4
5
6
7
8
9
10
New Cable
92.7
91.6
94.7
93.7
96.5
94.3
93.7
96.8
98.9
99.9
Original Cable
90.2
92.4
94.7
92.1
95.9
91.1
93.2
91.5
The cable manufacturer, on looking at health and safety legislation, decides

that a 1% level of signicance should be used in any statistical testing procedure adopted to distinguish between the cables. On the basis of the results
given and assuming that the populations from which the samples are drawn
do not have equal variances, should the manufacturer replace the old cable by
the new one?
Your solution
11


12
against one of the alternatives

H0 : 12 = 22
Essentially, we will test the null hypothesis
1
2
X2 N (2 , 22 )
X1 N (1 , 12 )
In the tests above, we distinguished between the cases of equal and unequal variances of samples
chosen from independent normal populations. As you have seen, the analysis changes according
to the assumptions made, conclusions reached and recommendations made - accepting or rejecting a null hypothesis for example - may also change. In view of this, we may wish to test in
order to decide whether the assumption that the variances 12 and 22 of the independent normal
populations shown in the diagram below, may be regarded as equal.
The F -test
If the average tensions are represented by 1 (new cable) and 2 (old cable) we form the
hypotheses
H 0 : 1 2 = 0
H 1 : 1 2 > 0
in order to test the hypothesis that the new cable is stronger on average than the old cable.
2 = 92.64.
1 = 95.28 and X
1 X
2 ) (1 2 )
(X
95.28 92.64
2.64

= 2.589
=
=
1.017
T =
S12
n1
S12
n1
S22
n2
6.47
10
3.14
8
The number of degrees of freedom is given by

2

=
2
S1
n1
2
n1 +1
S22
n2

+ f rac
S22
n2
2=
2
n2 + 1
6.47
3.14 2
+
10
8
2
(6.47/10)2
+ (3.14/8)
11
9
2=
1.081
24
0.038 + 0.017
Using t-distribution tables with 18 degrees of freedom, we see that the critical value at the 1%
level of signicance is 2.552. Since 2.589 < 2.552 we conclude that we reject the null hypothesis
in favour of the alternative. Notice that the result could still be considered marginal. The cable
manufacturer should exercise caution if the old cable is replaced by the new one on the basis
of the evidence available.
H1 : 12 = 22
H1 : 12 > 22
H1 : 12 < 22
In order to do this, we use the F distribution. The hypothesis test for the equality of two
variances 12 and 22 is encapsulated in the following Key Point.
Key Point
Consider a random sample of size n1 taken from a normal population with mean 1 and variance
12 and a random sample of size n1 taken from a second normal population with mean 2
and variance 22 . Denote the respective sample variances by S12 and S22 and assume that the
populations are independent. The ratio
F =
S12 S22
/
12 22
follows an F distribution in which the numerator has n1 1 degrees of freedom and the denominator has n2 1 degrees of freedom. Note that if the null hypothesis H0 : 12 = 22 is
true, then the value of F reduces to the ratio of the sample variances and that in this case
S12
F = 2
S2
Note
Recall that if a random sample of size n1 is taken from a normal population with mean 1 and
variance 12 and if the sample variance is denoted by S12 , the random variable
X12 =
(n1 1)S12
12
has a 2 distribution with n1 1 degrees of freedom. Similarly, if a random sample of size n2

is taken from a normal population with mean 2 and variance 22 and if the sample variance is
denoted by S22 , the random variable
X22
(n2 1)S22
=
22
has a 2 distribution with n2 1 degrees of freedom. This means that the ratio
F =
S12 S22
/
12 22
is a ratio of 2 random variables with n1 1 degrees of freedom in the numerator and n2 1

degrees of freedom in the denominator. Under the null hypothesis
H0 : 12 = 22
we know that the expression for F reduces to
13

F =
S12
S22
and we say that F has an F -distribution with n1 1 degrees of freedom in the numerator and
n2 1 degrees of freedom in the denominator. This distribution is denoted by
Fn1 1,n2 1
and some tabulated values are given in tables at the end of this booklet. If you check the tables,
you will nd that only right-tail values are given. The left-tail values are calculated by using
the following formula:
Flefttail
value, n1 1, n2 1
1
Frighttail
value, n2 1, n1 1
Note the reversal in the order in which the expressions for the number of degrees of freedom
occur.
Example The following is an extract from the F distribution tables (5% tail) given at
the end of this booklet.
f
5%
f0.05,u,
Degrees of Freedom for the Numerator (u)
1
2
3
4
5
6
7
8
1
161.4
18.51
10.13
7.71
6.61
5.99
5.59
5.32
2
199.5
19.00
9.55
6.94
5.79
5.14
4.74
4.46
3
215.7
19.16
9.28
6.59
5.41
4.76
4.35
4.07
4
224.6
19.25
9.12
6.39
5.19
4.53
4.12
3.84
5
230.2
19.30
9.01
6.26
5.05
4.39
3.97
3.69
6
234.0
19.33
8.94
6.16
4.95
4.28
3.87
3.58
7
236.8
19.35
8.89
6.09
4.88
4.21
3.79
3.50
8
238.9
19.37
8.85
6.04
4.82
4.15
3.73
3.44
9
240.5
19.38
8.81
6.00
4.77
4.10
3.68
3.39
10
241.9
19.40
8.79
5.96
4.74
4.06
3.64
3.35
20
248.0
19.45
8.66
5.80
4.56
3.87
3.44
3.15
30
250.1
19.46
8.62
5.75
4.53
3.81
3.38
3.08
40
251.1
19.47
8.59
5.72
4.46
3.77
3.34
3.04
60
252.2
19.48
8.55
5.69
4.43
3.74
3.30
3.01
254.3
19.50
8.53
5.63
4.36
3.67
3.23
2.93
Write down or calculate as appropriate, the following values of F from the

table
Right-tail Values Left-tail Values
F5%,4,3
F5%,4,3
F5%,8,2
F5%,8,2
F5%,7,8
F5%,7,8

14
Solution
The right-tail values are read directly from the tables. The left-tail values are calculated using
the formula given above.
Right-tail
Values Left-tail Values
1
1
F5%,4,3 = 9.12 F5%,4,3 = F5%,3,4
= 6.59
= 0.152
F5%,8,2 = 19.37
F5%,8,2
F5%,7,8 = 3.50
F5%,7,8
Write down or calculate as appropriate, the following values of F from the

tables given at the end of this booklet.
Right-tail Values Left-tail Values
F5%,10,20
F5%,10,20
F5%,5,30
F5%,5,30
F5%,20,7
F5%,20,7
F2.5%,10,10
F2.5%,10,10
F2.5%,8,30
F2.5%,8,30
F2.5%,20,30
F2.5%,20,30
Your solution
15

Right-tail
Values Left-tail Values
1
F5%,10,20 = 2.35
F5%,10,20 =
= 1 = 0.361
F
2.77
5%,20,10
1
F5%,5,30 = 2.53
F5%,5,30 = F 1 = 4.53
= 0.221
5%,30,5
1
F5%,20,7 = 3.44
F5%,20,7 = 1 = 2.51
= 0.398
F
5%,7,20
1
F
= 3.72 F
=
= 1 = 0.269
2.5%,10,10
2.5%,10,10
F5%,10,10
3.72
1
F2.5%,8,30 = 2.65
F2.5%,8,30 = F 1 = 3.89
= 0.257
5%,30,8
1
1
F2.5%,20,30 = 2.20 F2.5%,20,30 = F5%,30,20
= 2.35
= 0.426
Flefttail
value, n1 1,n2 1
Frighttail
value, n2 1,n1 1
The right-tail values are read directly from the tables. The left-tail values are calculated using
the formula
We are now in a position to use the F -test to solve engineering problems. The application of
the F -test will be illustrated by using the data given in a previous worked example in order to
determine whether the assumption of equal variability in the samples used is realistic.

circuits, the results obtained are those given on Page 6.
Last time, we worked on the assumption that the populations from which the
samples are drawn do not have equal variances. Is this assumption valid at
the 5% level of signicance?
Note that the manufacturer may also be interested in knowing whether the
variances are equal as well as the means. We shall not address the problem
here but it can be argued that equality of variances will facilitate consistent
performance from the components.

16
Solution
We form the hypotheses
H1 : 12 = 22
H0 : 12 = 22
and perform a two-tailed test.

2.00
S2
= 0.735
F = 12 =
S2
2.72
which has an F -distribution with 9 degrees of freedom in the numerator and 11 degrees of
freedom in the denominator.
We require two 2.5% tails, that is we require right-tail F2.5%,9,11 = 3.59 and left-tail F2.5%,9,11 .
The latter may be roughly estimated as
F2.5%,9,11 =
1
F2.5%,11,9
1
F2.5%,10,9
F2.5%,10,9 F2.5%,20,9
10
1
3.96
3.963.67
10
0.254
Since 0.254 < 0.735 < 3.59 we conclude that we cannot reject the null hypothesis in favour of
the alternative at the 5% level of signicance. The evidence supports the conclusion that the
samples have equal variability.
Note that we can adopt the rule (many statisticians do this) of always dividing the larger S 2
value by the smaller S 2 value so that you only need to look up right tail values.

The results obtained are those given on Page 8, the failure tensions being given
in kg103 . Last time we assumed that the populations from which the samples
are drawn did not have equal variances. Is this assumption valid at the 5%
level of signicance?
17


18
We form the hypotheses
H0 : 12 = 22
H1 : 12 = 22
and perform a two-tailed test.

6.47
S12
=
= 2.061
S22
3.14
F =
which has an F -distribution with 9 degrees of freedom in the numerator and 7 degrees of freedom
in the denominator. We require two 2.5% tails, that is we require right-tail F2.5%,9,7 = 4.42 and
left-tail F2.5%,9,7 which may be calculated as
F2.5%,9,7 =
1
1
=
= 0.238
F2.5%,7,9
4.20
Since 0.238 < 2.061 < 4.82 we conclude that we cannot reject the null hypothesis in favour of
the alternative at the 5% level of signicance. The evidence does not support the conclusion
that the populations have unequal variances.
Your solution
The Normal Probability Integral

The area is denoted by A and is measured from the mean z = 0 to any ordinate z = z1 .
A
0
Z = X
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2.0
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
2.9
3.0
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
0.00
.0000
.0398
.0793
.1179
.1554
.1915
.2257
.2530
.2881
.3159
.3413
.3643
.3849
.4032
.4192
.4332
.4452
.4554
.4641
.4713
.4772
.4621
.4861
.4893
.4918
.4938
.4953
.4965
.4974
.4981
.4986
.4990
.4993
.4995
.4997
.4998
.4998
.4999
.4999
.4999
0.01
0040
0438
0832
1217
1591
1950
2291
2611
2910
3186
3438
3665
3869
4049
4207
4345
4463
4564
4649
4719
4778
4826
4865
4896
4920
4940
4955
4966
4975
4982
4987
4991
4994
4995
4997
4998
4998
4999
4999
4999
0.02
0080
0478
0871
1255
1628
1985
2324
2642
2939
3212
3461
3686
3888
4066
4222
4357
4474
4573
4656
4726
4783
4830
4868
4898
4922
4941
4956
4967
4976
4983
4987
4991
4994
4995
4997
4998
4999
4999
4999
4999
0.03
0120
0517
0910
1293
1664
2019
2357
2673
2967
3238
3485
3708
3907
4082
4236
4370
4485
4582
4664
4732
4788
4835
4871
4901
4925
4943
4957
4968
4977
4983
4988
4991
4994
4996
4997
4998
4999
4999
4999
4999
z1
0.04
0159
0557
0948
1331
1700
2054
2389
2704
2995
3264
3508
3729
3925
4099
4251
4382
4495
4591
4671
4738
4793
4838
4875
4904
4927
4945
4959
4969
4977
4984
4988
4992
4994
4996
4997
4998
4999
4999
4999
4999
0.05
0199
0596
0987
1368
1736
2088
2422
2734
3023
3289
3531
3749
3944
4115
4265
4394
4505
4599
4678
4744
4798
4842
4878
4906
4929
4946
4960
4970
4978
4984
4989
4992
4994
4996
4997
4998
4999
4999
4999
4999
0.06
0239
0636
1026
1406
1772
2123
2454
2764
3051
3315
3554
3770
3962
4131
4279
4406
4515
4608
4686
4750
4803
4846
4881
4909
4931
4948
4961
4971
4979
4985
4989
4992
4994
4996
4997
4998
4999
4999
4999
4999
0.07
0279
0657
1064
1443
1808
2157
2486
2794
3078
3340
3577
3790
3980
4147
4292
4418
4525
4616
4693
4756
4808
4850
4884
4911
4932
4949
4962
4972
4980
4985
4989
4992
4995
4996
4997
4998
4999
4999
4999
4999
0.08
0319
0714
1103
1480
1844
2190
2518
2823
3106
3365
3599
3810
3997
4162
4306
4430
4535
4625
4699
4762
4812
4854
4887
4913
4934
4951
4963
4973
4980
4986
4990
4993
4995
4996
4997
4998
4999
4999
4999
4999
0.09
0359
0753
1141
1517
1879
2224
2549
2852
3133
3389
3621
3830
4015
4177
4319
4441
4545
4633
4706
4767
4817
4857
4890
4916
4936
4952
4964
4974
4981
4986
4990
4993
4995
4997
4998
4998
4999
4999
4999
4999
Note that some text books give the nal line entries as 0.5 rather than 0.4999.
In these workbooks we shall use 0.4999.
19

Percentage Points of the Students t -distribution
t,
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
40
60
120
.40
.25
.10
.05
.025
.01
.005
.0025
.001
.0005
.325
.289
.277
.271
.267
.265
.263
.262
.261
.260
.260
.259
.259
.258
.258
.258
.257
.257
.257
.257
.257
.256
.256
.256
.256
.256
.256
.256
.256
.256
.255
.254
.254
.253
1.000
.816
.765
.741
.727
.718
.711
.706
.703
.700
.697
.695
.694
.692
.691
.690
.689
.688
.688
.687
.686
.686
.685
.685
.684
.684
.684
.683
.683
.683
.681
.679
.677
.674
3.078
1.886
1.638
1.533
1.476
1.440
1.415
1.397
1.383
1.372
1.363
1.356
1.350
1.345
1.341
1.337
1.333
1.330
1.328
1.325
1.323
1.321
1.319
1.318
1.316
1.315
1.314
1.313
1.311
1.310
1.303
1.296
1.289
1.282
6.314
2.902
2.353
2.132
2.015
1.943
1.895
1.860
1.833
1.812
1.796
1.782
1.771
1.761
1.753
1.746
1.740
1.734
1.729
1.725
1.721
1.717
1.714
1.711
1.708
1.706
1.703
1.701
1.699
1.697
1.684
1.671
1.658
1.645
12.706
4.303
3.182
2.776
2.571
2.447
2.365
2.306
2.262
2.228
2.201
2.179
2.160
2.145
2.131
2.120
2.110
2.101
2.093
2.086
2.080
2.074
2.069
2.064
2.060
2.056
2.052
2.048
2.045
2.042
2.021
2.000
1.980
1.960
31.825
6.965
4.514
3.747
3.365
3.143
2.998
2.896
2.821
2.764
2.718
2.681
2.650
2.624
2.602
2.583
2.567
2.552
2.539
2.528
2.518
2.508
2.500
2.492
2.485
2.479
2.473
2.467
2.462
2.457
2.423
2.390
2.358
2.326
63.657
9.925
5.841
4.604
4.032
3.707
3.499
3.355
3.250
3.169
3.106
3.055
3.012
2.977
2.947
2.921
2.898
2.878
2.861
2.845
2.831
2.819
2.807
2.797
2.787
2.779
2.771
2.763
2.756
2.750
2.704
2.660
2.617
2.576
127.32
14.089
7.453
5.598
4.773
4.317
4.029
3.833
3.690
3.581
3.497
3.428
3.372
3.326
3.286
3.252
3.222
3.197
3.174
3.153
3.135
3.119
3.104
3.091
3.078
3.067
3.057
3.047
3.038
3.030
2.971
2.915
2.860
2.807
318.31
23.326
10.213
7.173
5.893
5.208
4.785
4.501
4.297
4.144
4.025
3.930
3.852
3.787
3.733
3.686
3.646
3.610
3.579
3.552
3.527
3.505
3.485
3.467
3.450
3.435
3.421
3.408
3.396
3.385
3.307
3.232
3.160
3.090
636.62
31.598
12.924
8.610
6.869
5.959
5.408
5.041
4.781
4.487
4.437
4.318
4.221
4.140
4.073
4.015
3.965
3.922
3.883
3.850
3.819
3.792
3.767
3.745
3.725
3.707
3.690
3.674
3.659
3.646
3.551
3.460
3.373
3.291

20
Percentage Points of the F Distribution (5% tail)
5%
f0.05,u ,
Degrees of Freedom for the Numerator (u)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
40
60
21
1
161.4
18.51
10.13
7.71
6.61
5.99
5.59
5.32
5.12
4.96
4.84
4.75
4.67
4.60
4.54
4.49
4.45
4.41
4.38
4.35
4.32
4.30
4.28
4.26
4.24
4.23
4.21
4.20
4.18
4.17
4.08
4.00
3.84
2
199.5
19.00
9.55
6.94
5.79
5.14
4.74
4.46
4.26
4.10
3.98
3.89
3.81
3.74
3.68
3.63
3.59
3.55
3.52
3.49
3.47
3.44
3.42
3.40
3.39
3.37
3.35
3.34
3.33
3.32
3.23
3.15
3.00
3
215.7
19.16
9.28
6.59
5.41
4.76
4.35
4.07
3.86
3.71
3.59
3.49
3.41
3.34
3.29
3.24
3.20
3.16
3.13
3.10
3.07
3.05
3.03
3.01
2.99
2.98
2.96
2.95
2.93
2.92
2.84
2.76
2.60
4
224.6
19.25
9.12
6.39
5.19
4.53
4.12
3.84
3.63
3.48
3.36
3.26
3.18
3.11
3.06
3.01
2.96
2.93
2.90
2.87
2.84
2.82
2.80
2.78
2.76
2.74
2.73
2.71
2.70
2.69
2.61
2.53
2.37
5
230.2
19.30
9.01
6.26
5.05
4.39
3.97
3.69
3.48
3.33
3.20
3.11
3.03
2.96
2.90
2.85
2.81
2.77
2.74
2.71
2.68
2.66
2.64
2.62
2.60
2.59
2.57
2.56
2.55
2.53
2.45
2.37
2.21
6
234.0
19.33
8.94
6.16
4.95
4.28
3.87
3.58
3.37
3.22
3.09
3.00
2.92
2.85
2.79
2.74
2.70
2.66
2.63
2.60
2.57
2.55
2.53
2.51
2.49
2.47
2.46
2.45
2.43
2.42
2.34
2.25
2.10
7
236.8
19.35
8.89
6.09
4.88
4.21
3.79
3.50
3.29
3.14
3.01
2.91
2.83
2.76
2.71
2.66
2.61
2.58
2.54
2.51
2.49
2.46
2.44
2.42
2.40
2.39
2.37
2.36
2.35
2.33
2.25
2.17
2.01
8
238.9
19.37
8.85
6.04
4.82
4.15
3.73
3.44
3.23
3.07
2.95
2.85
2.77
2.70
2.64
2.59
2.55
2.51
2.48
2.45
2.42
2.40
2.37
2.36
2.34
2.32
2.31
2.29
2.28
2.27
2.18
2.10
1.94
9
240.5
19.38
8.81
6.00
4.77
4.10
3.68
3.39
3.18
3.02
2.90
2.80
2.71
2.65
2.59
2.54
2.49
2.46
2.42
2.39
2.37
2.34
2.32
2.30
2.28
2.27
2.25
2.24
2.22
2.21
2.12
2.04
1.88
10
241.9
19.40
8.79
5.96
4.74
4.06
3.64
3.35
3.14
2.98
2.85
2.75
2.67
2.60
2.54
2.49
2.45
2.41
2.38
2.35
2.32
2.30
2.27
2.25
2.24
2.22
2.20
2.19
2.18
2.16
2.08
1.99
1.83
20
248.0
19.45
8.66
5.80
4.56
3.87
3.44
3.15
2.94
2.77
2.65
2.54
2.46
2.39
2.33
2.28
2.23
2.19
2.16
2.12
2.10
2.07
2.05
2.03
2.01
1.99
1.97
1.96
1.94
1.93
1.84
1.75
1.57
30
250.1
19.46
8.62
5.75
4.53
3.81
3.38
3.08
2.86
2.70
2.57
2.47
2.38
2.31
2.25
2.19
2.15
2.11
2.07
2.04
2.01
1.98
1.96
1.94
1.92
1.90
1.88
1.87
1.85
1.84
1.74
1.65
1.46
40
251.1
19.47
8.59
5.72
4.46
3.77
3.34
3.04
2.83
2.66
2.53
2.43
2.34
2.27
2.20
2.15
2.10
2.06
2.03
1.99
1.96
1.94
1.91
1.89
1.87
1.85
1.84
1.82
1.81
1.79
1.69
1.59
1.39
60
252.2
19.48
8.55
5.69
4.43
3.74
3.30
3.01
2.79
2.62
2.49
2.38
2.30
2.22
2.16
2.11
2.06
2.02
1.93
1.95
1.92
1.89
1.86
1.84
1.82
1.80
1.79
1.77
1.75
1.74
1.64
1.53
3.32
254.3
19.50
8.53
5.63
4.36
3.67
3.23
2.93
2.71
2.54
2.40
2.30
2.21
2.13
2.07
2.01
1.96
1.92
1.88
1.84
1.81
1.78
1.76
1.73
1.71
1.69
1.67
1.65
1.64
1.62
1.51
1.39
1.00

Percentage Points of the F Distribution (2.5% tail)
2.5%
f0.025,u,
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
40
60
1
647.8
38.51
17.44
12.22
10.01
8.81
8.07
7.57
7.21
6.94
6.72
6.55
6.41
6.30
6.20
6.12
6.04
5.98
5.92
5.87
5.83
5.79
5.75
5.72
5.69
5.66
5.63
5.61
5.59
5.57
5.42
5.29
5.02
2
799.5
39.00
16.04
10.65
8.43
7.26
6.54
6.06
5.71
5.46
5.26
5.10
4.97
4.86
4.77
4.69
4.62
4.56
4.51
4.46
4.42
4.38
4.35
4.32
4.29
4.27
4.24
4.22
4.20
4.18
4.05
3.93
3.69
3
864.2
39.17
15.44
9.98
7.76
6.60
5.89
5.42
5.08
4.83
4.63
4.47
4.35
4.24
4.15
4.08
4.01
3.95
3.90
3.86
3.82
3.78
3.75
3.72
3.69
3.67
3.65
3.63
3.61
3.59
3.46
3.34
3.12
Degrees of
4
5
899.6 921.8
39.25 39.30
15.10 14.88
9.60 9.36
7.39 7.15
6.23 5.99
5.52 5.29
5.05 4.82
4.72 4.48
4.47 4.24
4.28 4.04
4.12 3.89
4.00 3.77
3.89 3.66
3.80 3.58
3.73 3.50
3.66 3.44
3.61 3.38
3.56 3.33
3.51 3.29
3.48 3.25
3.44 3.22
3.41 3.18
3.38 3.15
3.35 3.13
3.33 3.10
3.31 3.08
3.29 3.06
3.27 3.04
3.25 3.03
3.13 2.90
3.01 2.79
2.79 2.57
Freedom for the Numerator (u)

6
7
8
9
10
937.1 948.2 956.7 963.3 968.6
39.33 39.36 39.37 39.39 39.40
14.73 14.62 14.54 14.47 14.42
9.20 9.07 8.98 8.90 8.84
6.98 6.85 6.76 6.68 6.62
5.82 5.70 5.60 5.52 5.46
5.12 4.99 4.90 4.82 4.75
4.65 4.53 4.43 4.36 4.30
4.32 4.20 4.10 4.03 3.96
4.07 3.95 3.85 3.78 3.72
3.88 3.76 3.66 3.59 3.53
3.73 3.61 3.51 3.44 3.37
3.60 3.48 3.39 3.31 3.25
3.50 3.38 3.29 3.21 3.15
3.41 3.29 3.20 3.12 3.06
3.34 3.32 3.12 3.05 2.99
3.28 3.16 3.06 2.98 2.92
3.22 3.10 3.01 2.93 2.87
3.17 3.05 2.96 2.88 2.82
3.13 3.01 2.91 2.84 2.77
3.09 2.97 2.87 2.80 2.73
3.05 2.93 2.84 2.76 2.70
3.02 2.90 2.81 2.73 2.67
2.99 2.87 2.78 2.70 2.64
2.97 2.85 2.75 2.68 2.61
2.94 2.82 2.73 2.65 2.59
2.92 2.80 2.71 2.63 2.57
2.90 2.78 2.69 2.61 2.55
2.88 2.76 2.67 2.59 2.53
2.87 2.75 2.65 2.57 2.51
2.74 2.62 2.53 2.45 2.39
2.63 2.51 2.41 2.33 2.27
2.41 2.29 2.19 2.11 2.05

20
993.1
39.45
14.17
8.56
6.33
5.17
4.47
4.00
3.67
3.42
3.23
3.07
2.95
2.84
2.76
2.68
2.62
2.56
2.51
2.46
2.42
2.39
2.36
2.33
2.30
2.28
2.25
2.23
2.21
2.20
2.07
1.94
1.71
30
1001
39.46
14.08
8.46
6.23
5.07
4.36
3.89
3.56
3.31
3.12
2.96
2.84
2.73
2.64
2.57
2.50
2.44
2.39
2.35
2.31
2.27
2.24
2.21
2.18
2.16
2.13
2.11
2.09
2.07
1.94
1.82
1.57
40
1006
39.47
14.04
8.41
6.18
5.01
4.31
3.84
3.51
3.26
3.06
2.91
2.78
2.67
2.59
2.51
2.44
2.38
2.33
2.29
2.25
2.21
2.18
2.15
2.12
2.09
2.07
2.05
2.03
2.01
1.88
1.74
1.48
60
1010
39.48
13.99
8.36
6.12
4.96
4.25
3.78
3.45
3.20
3.00
2.85
2.72
2.61
2.52
2.45
2.38
2.32
2.27
2.22
2.18
2.14
2.11
2.08
2.05
2.03
2.00
1.91
1.96
1.94
1.80
1.67
1.39
22
1018
39.50
13.90
8.26
6.02
4.85
4.14
3.67
3.33
3.08
2.88
2.72
2.60
2.49
2.40
2.32
2.25
2.19
2.13
2.09
2.04
2.00
1.97
1.94
1.91
1.88
1.85
1.83
1.81
1.79
1.64
1.48
1.00

41 3 Tests Two Samples

Uploaded by

Copyright:

Available Formats

41 3 Tests Two Samples

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

41 3 Tests Two Samples

Uploaded by

Copyright:

Available Formats

Tests Concerning

be familiar with the normal t-, F - and

1. Tests concerning Two Samples

A motor manufacturer is considering whether or not a new fuel formulation

HELM (VERSION 1: March 18, 2004): Workbook Level 1

The null and alternative hypotheses are:

Two Independent Populations each with an Unknown Variance

We will take 12 = 22 = 2 so that in diagrammatic form the populations are:

(n1 1)S12 + (n2 1)S22

(n1 1)S12 + (n2 1)S22

will follow Students t-distribution with n1 + n2 2 degrees of freedom.

HELM (VERSION 1: March 18, 2004): Workbook Level 1

Example A manufacturer of electronic equipment has developed a circuit to feed current

Circuit 1 - Current (mA) Circuit 2 - Current (mA)

(n1 1)S12 +(n2 1)S22

The test statistic is

HELM (VERSION 1: March 18, 2004): Workbook Level 1

A manufacturer of steel cables used in the construction of suspension bridges

The cable manufacturer, on looking at health and safety legislation, decides

HELM (VERSION 1: March 18, 2004): Workbook Level 1

HELM (VERSION 1: March 18, 2004): Workbook Level 1

(n1 1)S12 +(n2 1)S22

The test statistic is

Example A manufacturer of electronic equipment has developed a circuit to feed current

HELM (VERSION 1: March 18, 2004): Workbook Level 1

HELM (VERSION 1: March 18, 2004): Workbook Level 1

A manufacturer of steel cables used in the construction of suspension bridges

The cable manufacturer, on looking at health and safety legislation, decides

HELM (VERSION 1: March 18, 2004): Workbook Level 1

HELM (VERSION 1: March 18, 2004): Workbook Level 1

against one of the alternatives

The number of degrees of freedom is given by

has a 2 distribution with n1 1 degrees of freedom. Similarly, if a random sample of size n2

is a ratio of 2 random variables with n1 1 degrees of freedom in the numerator and n2 1

HELM (VERSION 1: March 18, 2004): Workbook Level 1

Write down or calculate as appropriate, the following values of F from the

HELM (VERSION 1: March 18, 2004): Workbook Level 1

Write down or calculate as appropriate, the following values of F from the

HELM (VERSION 1: March 18, 2004): Workbook Level 1

Example A manufacturer of electronic equipment has developed a circuit to feed current

HELM (VERSION 1: March 18, 2004): Workbook Level 1

and perform a two-tailed test.

A manufacturer of steel cables used in the construction of suspension bridges

HELM (VERSION 1: March 18, 2004): Workbook Level 1

HELM (VERSION 1: March 18, 2004): Workbook Level 1

and perform a two-tailed test.

The Normal Probability Integral

HELM (VERSION 1: March 18, 2004): Workbook Level 1

Percentage Points of the Students t -distribution

HELM (VERSION 1: March 18, 2004): Workbook Level 1

Percentage Points of the F Distribution (5% tail)

HELM (VERSION 1: March 18, 2004): Workbook Level 1

Percentage Points of the F Distribution (2.5% tail)

Freedom for the Numerator (u)

HELM (VERSION 1: March 18, 2004): Workbook Level 1

You might also like