UNIT-6: NON-PARAMETRIC TEST

(Professor Asfar H Siddiqui)

1.The Sign Test (Wilcoxon-Test)

Ex-1: A salesman paid 12 visits to his area sales manager and noted that he had to wait for
10,15,20,17,11,25,30,27,36,40,5 and 26minutes respectively, before being called in his office.
The area sales manager claims that the salesman seeking to meet him does not have to wait for
more than 20 minutes before being called in. Using the sign test, verify ay 0.05 level of
significance the claim made by the area sales manager.

10 15 20 17 11 25 30 27 36 40 05 26

Solution: Let μ = 20

S.n Time(T) 𝑇−μ sign S.n Time(min) 𝑇−μ sign

1 10 -10 − 07 30 10 +
2 15 -5 − 08 27 07 +
3 20 0 0 09 36 16 +
4 17 -3 − 10 40 20 +
5 11 -9 − 11 05 -15 −
6 25 5 + 12 26 06 +
𝑁 = 𝑁+ + 𝑁− = 11 𝑁+ = 6 and 𝑁− = 5

1) Null Hypothesis 𝐻0: μ = 20

2) Alternate Hypothesis 𝐻𝐴: μ > 20

3) LOS = 0.05

4) Critical Region :𝑍𝑐𝑎𝑙 < 𝑍∝ or 𝑍𝑐𝑎𝑙 >− 𝑍∝ ( one tail test)

𝑍𝑐𝑎𝑙 < 1. 645 (accept)

5) Calculation Statistics:

Let 𝑃 = 0. 5 (always)

𝑋−μ 𝑁+−𝑁𝑃 6−11×0.5

𝑍𝑐𝑎𝑙 = σ
= = = 0. 302
𝑁𝑃(1−𝑃) 11×0.5×0.5

𝑍𝑡𝑎𝑏 = 1. 645

6) Decision:

𝑍𝑐𝑎𝑙 < 𝑍∝ accept NH

Ex-2: The PQR company claims that the lifetime of a type of battery that it manufactures is more
than 250 hrs. A consumer advocate wishing to determine whether the claim is justified measures
the lifetime of 24 of the company’s batteries; the result are as follows. Assuming the sample to
be random .Detemine whether the company’s claim is justified at 0.05 significance level
 271 230 198 275 282 225 284 219
253 216 262 288 236 291 253 224
264 295 211 252 294 243 272 268

Solution: Let μ = 250

S.N LT 𝐿𝑇 − μ sign S.n LT 𝑇−μ sign

1 271 21 + 13 282 32 +
2 253 3 + 14 236 -14 −
3 264 14 + 15 294 44 +
4 230 -20 − 16 225 -25 −
5 216 -34 − 17 291 41 +
6 295 45 + 18 243 -7 −
7 198 -52 − 19 284 34 +
8 262 12 + 20 253 3 +
9 211 -39 − 21 272 22 +
10 275 25 + 22 219 -31 −
11 288 38 + 23 224 -26 −
12 252 2 + 24 268 18 +
𝑁 = 𝑁+ + 𝑁− = 24 𝑁+ = 15, 𝑁− = 9

1) Null Hypothesis 𝐻0: μ = 250

2) Alternate Hypothesis 𝐻𝐴: μ > 250

3) LOS = 0.05

4) Critical Region :𝑍𝑐𝑎𝑙 < 𝑍∝ or 𝑍𝑐𝑎𝑙 >− 𝑍∝ ( one tail test)

𝑍𝑐𝑎𝑙 < 1. 645 (accept)

5) Calculation Statistics:

Let 𝑃 = 0. 5 (always)

𝑋−μ 𝑁+−𝑁𝑃 15−24×0.5

𝑍𝑐𝑎𝑙 = σ
= = = 1. 225 and 𝑍 = 1. 645
𝑁𝑃(1−𝑃) 24×0.5×0.5 𝑡𝑎𝑏

6) Decision:

𝑍𝑐𝑎𝑙 < 𝑍∝ accept NH

Ex-3: The production department of a company Typewriters has developed a new model. It is
claimed that new model is more efficient than the current model because on an average the new
one is capable of typing 20 words per minute. To verify the claim the test performance on a
random sample of 20 typists gave the following results. LOS = 0.05

Word typed on Word typed on

Typists No. New model Old model Typists no. New model Old Model
01 40 45 11 70 68
02 45 50 12 71 50
03 60 55 13 65 58
04 50 52 14 45 47
05 55 52 15 48 54
06 60 60 16 42 50
07 78 55 17 48 43
08 80 53 18 68 60
09 65 70 19 50 48
 10 62 65 20 55 50

Solution:

MODEL CAL MODEL CAL

T.N NEW OLD DIFF SIGN T.N NEW OLD DIFF SIGN
01 40 45 -05 − 11 70 68 02 +
02 45 50 -05 − 12 71 50 21 +
03 60 55 05 + 13 65 58 07 +
04 50 52 -02 − 14 45 47 -02 −
05 55 52 03 + 15 48 54 -06 −
06 60 60 00 0 16 42 50 -08 −
07 78 55 23 + 17 48 43 05 +
08 80 53 27 + 18 68 60 08 +
09 65 70 -05 − 19 50 48 02 +
10 62 65 -03 − 20 55 50 05 +
𝑁 = 𝑁+ + 𝑁− = 19 𝑁+ = 11 𝑁− = 8

1) Null Hypothesis 𝐻0: µ𝑁 = µ𝑂

2) Alternate Hypothesis 𝐻𝐴: µ𝑁 > µ𝑂

3) LOS = 0.05

4) Critical Region :𝑍𝑐𝑎𝑙 < 𝑍∝ or 𝑍𝑐𝑎𝑙 >− 𝑍∝ ( one tail test)

𝑍𝑐𝑎𝑙 < 1. 645 (accept)

5) Calculation Statistics:

Let 𝑃 = 0. 5 (always)

𝑋−μ 𝑁+−𝑁𝑃 11−19×0.5

𝑍𝑐𝑎𝑙 = σ
= = = 0. 688 and 𝑍 = 1. 645
𝑁𝑃(1−𝑃) 19×0.5×0.5 𝑡𝑎𝑏

6) Decision:

𝑍𝑐𝑎𝑙 < 𝑍∝ accept NH

Ex-4:The table below shows the hours of relief provided by two analgesic drugs in 12
patients suffering from arthritis. Is there any evidence that one drug provides longer relief
than the other?

Drug Drug
Case A B Case A B
01 2.0 3.5 7 14.9 16.7
02 3.6 5.7 8 6.6 6.0
03 2.6 2.9 9 2.3 3.8
04 2.6 2.4 10 2.0 4.0
05 7.3 9.9 11 6.8 9.1
06 3.4 3.3 12 8.5 20.9
Solution:

DRUG CAL DRUG CAL

C.N A B DIFF SIGN C.N A B DIFF SIGN
01 2.0 3.5 -1.5 − 7 14.9 16.7 -1.8 −
02 3.6 5.7 -2.1 − 8 6.6 6.0 0.6 +
03 2.6 2.9 -0.3 − 9 2.3 3.8 -1.5 −
04 2.6 2.4 0.2 + 10 2.0 4.0 -2 −
05 7.3 9.9 -2.6 − 11 6.8 9.1 -2.3 −
06 3.4 3.3 0.1 + 12 8.5 20.9 -12.4 −
𝑁 = 𝑁+ + 𝑁− = 12 𝑁+ = 03 𝑁− = 9

1) Null Hypothesis 𝐻0: µ𝐴 = µ𝐵

2) Alternate Hypothesis 𝐻𝐴: µ𝐴 > µ𝐵

3) LOS = 0.05

4) Critical Region :𝑍𝑐𝑎𝑙 < 𝑍∝ or 𝑍𝑐𝑎𝑙 >− 𝑍∝ ( one tail test)

𝑍𝑐𝑎𝑙 >− 1. 645 Accept NH

5) Calculation Statistics:

Let 𝑃 = 0. 5 (always)

𝑋−μ 𝑁+−𝑁𝑃 3−12×0.5

𝑍𝑐𝑎𝑙 = σ
= = =− 1. 73 and 𝑍 =− 1. 645
𝑁𝑃(1−𝑃) 12×0.5×0.5 𝑡𝑎𝑏

6) Decision:

𝑍𝑐𝑎𝑙 > 𝑍∝ Reject NH

Ex-5: The breaking strengths of a random sample of 25 ropes made by a manufacturer are given
in Table 10-32. On the basis of this sample, test at the 0.05 significance level the manufacturer’s
claim that the breaking strength of a rope is 25

41 28 35 38 23 37 32 24 46 30 25 36 22 41 37
43 27 34 27 36 42 33 28 31 24

Solution: Let μ = 25

S.n Time(T) 𝑇−μ sign S.n Time(min) 𝑇−μ sign

1 41 16 + 13 22 -3 −
2 28 3 + 14 41 16 +
3 35 10 + 15 37 12 +
4 38 13 + 16 43 18 +
5 23 -2 − 17 27 2 +
6 37 12 + 18 34 9 +
07 32 7 + 19 27 2 +
08 24 -1 − 20 36 11 +
09 46 21 + 21 42 17 +
10 30 5 + 22 33 8 +
11 25 0 0 23 28 3 +
12 36 11 + 24 31 6 +
 25 24 -1 −
𝑁 = 𝑁+ + 𝑁− = 24 𝑁+ = 20 and
𝑁− = 4

1) Null Hypothesis 𝐻0: μ = 25

2) Alternate Hypothesis 𝐻𝐴: μ≠20

3) LOS = 0.05
4) Critical Region :

− 𝑍𝑐𝑎𝑙 < 𝑍∝ < 𝑍𝑐𝑎𝑙

( Two tail test)

− 1. 96 < 𝑍∝ < 1. 96

5) Calculation Statistics:

Let 𝑃 = 0. 5 (always)

𝑋−μ 𝑁+−𝑁𝑃 20−24×0.5

𝑍𝑐𝑎𝑙 = σ
= = = 3. 266
𝑁𝑃(1−𝑃) 24×0.5×0.5

𝑍𝑡𝑎𝑏 = 1. 96

6) Decision:

𝑍𝑐𝑎𝑙 > 𝑍∝ Reject NH

Ex-6: A company claims that if its product is added to an automobile’s gasoline tank, the mileage
per gallon will improve. To test the claim, 15 different automobiles are chosen and the mileage
per gallon with and without the additive is measured; the results are shown in Table Assuming
that the driving conditions are the same, determine whether there is a difference due to the
additive at significance levels of (a) 0.05, (b) 0.01.

With additive 34.7 28.3 19.6 25.1 15.7 24.5 28.7 23.5 27.7 32.1 29.6 22.4 25.7 28.1 24.3
Without additive 31.4 27.2 20.4 24.6 14.9 22.3 26.8 24.1 26.2 31.4 28.8 23.1 24.0 27.3 22.9

Solution:

Additive CAL Additive CAL

S.N With Without DIFF SIGN S.N With Without DIFF SIGN
01 34.7 31.4 3.3 + 9 27.7 26.2 1.5 +
02 28.3 27.2 1.1 + 10 32.1 31.4 0.7 +
03 19.6 20.4 -0.8 − 11 29.6 28.3 1.3 +
04 25.1 24.6 0.5 + 12 22.4 23.1 -0.7 −
05 15.7 14.9 0.8 + 13 25.7 24.0 1.7 +
06 24.5 22.3 2.2 + 14 28.1 27.3 0.8 +
07 28.7 26.8 1.9 + 15 24.3 22.9 1.4 +
08 23.5 24.1 -0.6 −
𝑁 = 𝑁+ + 𝑁− = 15 𝑁+ = 12 𝑁− = 3

1) Null Hypothesis 𝐻0: µ𝐴 = µ𝑊𝐴

2) Alternate Hypothesis 𝐻𝐴: µ𝐴 ≠ µ𝑊𝐴

3) LOS = (a) 0.05, (b) 0.01.

4) Critical Region :

− 1. 96 < 𝑍∝ < 1. 96

− 2. 58 < 𝑍∝ < 2. 58

5) Calculation Statistics:

Let 𝑃 = 0. 5 (always)

𝑋−μ 𝑁+−𝑁𝑃 12−15×0.5

𝑍𝑐𝑎𝑙 = σ
= = = 2. 3237
𝑁𝑃(1−𝑃) 15×0.5×0.5

and 𝑍𝑡𝑎𝑏 = ∓1. 96

𝑍𝑡𝑎𝑏 = ∓2. 58

6) Decision:

For LOS = 0.05

𝑍𝑐𝑎𝑙 > 𝑍∝ Reject NH

For LOS = 0.01.

− 𝑍∝ < 𝑍 < 𝑍∝ Accept NH

𝑐𝑎𝑙

Ex-7: A training manager claims that by giving a special course to company sales personnel, the
company’s annual sales will increase. To test this claim, the course is given to 24 people. Of
these 24, the sales of 16 increase, those of 6 decrease, and those of 2 remain unchanged. Test at
the 0.05 significance level the hypothesis that the course increased the company’s sales.

Solution:

𝑁+ = 16 and 𝑁− = 6

𝑁 = 𝑁+ + 𝑁− = 22

1) Null Hypothesis 𝐻0: No difference

2) Alternate Hypothesis 𝐻𝐴: There is difference

3) LOS = 0.05
4) Critical Region :

− 𝑍𝑐𝑎𝑙 < 𝑍∝ < 𝑍𝑐𝑎𝑙

( Two tail test)

− 1. 96 < 𝑍∝ < 1. 96

5) Calculation Statistics:

Let 𝑃 = 0. 5 (always)
 𝑋−μ 𝑁+−𝑁𝑃 16−22×0.5
𝑍𝑐𝑎𝑙 = σ
= = = 2. 13
𝑁𝑃(1−𝑃) 22×0.5×0.5

𝑍𝑡𝑎𝑏 = 1. 96

6) Decision:

For LOS = 0.05

𝑍𝑐𝑎𝑙 > 𝑍∝ Reject NH

EXERCISE
1. A company claims that if its product is added to an automobile’s gasoline tank, the mileage
per gallon will improve. To test the claim, 15 different automobiles are chosen and the mileage
per gallon with and without the additive is measured; the results are shown in Table 10-31.
Assuming that the driving conditions are the same, determine whether there is a difference due to
the additive at significance levels of (a) 0.05, (b) 0.01.

With additive 34.7 28.3 19.6 25.1 15.7 24.5 28.7 23.5 27.7 32.1 29.6 22.4 25.7 28.1 24.3
Without additive 31.4 27.2 20.4 24.6 14.9 22.3 26.8 24.1 26.2 31.4 28.8 23.1 24.0 27.3 22.9

2. A weight-loss club advertises that a special program that it has designed will produce a weight
loss of at least 6% in 1 month if followed precisely. To test the club’s claim, 36 adults undertake
the program. Of these, 25 realize the desired loss, 6 gain weight, and the rest remain essentially
unchanged. Determine at the 0.05 significance level whether the program is effective.

3. A training manager claims that by giving a special course to company sales personnel, the
company’s annual sales will increase. To test this claim, the course is given to 24 people. Of
these 24, the sales of 16 increase, those of 6 decrease, and those of 2 remain unchanged. Test at
the 0.05 significance level the hypothesis that the course increased the company’s sales.

4. The MW Soda Company sets up “taste tests” in 27 locations around the country in order to
determine the public’s relative preference for two brands of cola, A and B. In eight locations
brand A is preferred over brand B, in 17 locations brand B is preferred over brand A, and in the
remaining locations there is indifference. Can one conclude at the 0.05 that brand B is preferred
over brand A?

5. The breaking strengths of a random sample of 25 ropes made by a manufacturer are given in
Table 10-32. On the basis of this sample, test at the 0.05 significance level the manufacturer’s
claim that the breaking strength of a rope is (a) 25, (b) 30, (c) 35, (d) 40.

41 28 35 38 23 37 32 24 46 30 25 36 22 41 37
43 27 34 27 36 42 33 28 31 24

2. The Mann-Whitney U Test

𝑁1(𝑁1+1)
i) 𝑈1 = 𝑁1𝑁2 + 2
− 𝑅1

(
𝑁2 𝑁2+1 )
ii) 𝑈2 = 𝑁1𝑁2 + 2
− 𝑅2

𝑁1𝑁2
iv) µ𝑈 = 2
1
 (
𝑁1𝑁2 𝑁1+𝑁2+1 )
v) σ𝑈 = 12
1

𝑈1−µ𝑈
vi) 𝑍𝑐𝑎𝑙 = σ𝑈
1

Ex: A professor has two classes in psychology a morning class of 9 students, and an afternoon
class of 12 students. On a final examination scheduled at the same time for all students, the
classes received the grades shown in Table Can one concludes at the 0.05 significance level that
the morning class performed worse than the afternoon class?

MC 73 87 79 75 82 66 97 75 70
AC 86 81 84 88 90 85 84 92 83 91 53 84
Sol :

CLASSES RANKS
MC AC R-1 R-2
73 86 04 15
87 81 16 8
79 84 07 12
75 88 5.5 17
82 90 09 18
66 85 02 14
97 84 21 12
75 92 5.5 20
70 83 03 10
91 19
53 1
84 12
𝑁1 = 9 𝑁2 = 12
∑ 𝑅1 = 73 ∑ 𝑅2 = 158

1) Null Hypothesis 𝐻0: No difference in Performances

2) Alternate Hypothesis 𝐻𝐴: 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 𝑖𝑛 𝑃𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒𝑠

3) LOS = 0.05

4) Critical Region:𝑍𝑐𝑎𝑙 < 𝑍∝ = 𝐴𝑐𝑐𝑒𝑝𝑡

𝑍𝑐𝑎𝑙 > 𝑍∝ = 𝑅𝑒𝑗𝑒𝑐𝑡 [ Using Z-distribution Table]

5) Calculation Statistics:
𝑁1(𝑁1+1)
𝑈1 = 𝑁1𝑁2 + 2
− 𝑅1 = 80

𝑁2(𝑁2+1)
𝑈2 = 𝑁1𝑁2 + 2
− 𝑅2 = 28

𝑁1𝑁2 9×12 𝑁1𝑁2(𝑁1+𝑁2+1)

µ𝑈 = 2
= 2
= 54 and σ𝑈 = 12
= 14. 07
1 1

𝑈1−µ𝑈
𝑍𝑐𝑎𝑙 = σ𝑈
1
= 1. 85
1

6) Decision:
𝑍𝑐𝑎𝑙 > 𝑍∝ = 1. 645 = 𝑅𝑒𝑗𝑒𝑐𝑡

Ex-2: Instructors A and B both teach a first course in chemistry at XYZ University. On a
common final examination, their students received the grades shown in Table Test at the 0.05
significance level the hypothesis that there is no difference between the two instructors’ grades.

A 88 75 92 71 63 84 55 64 82 96
B 72 65 84 53 76 80 51 60 57 85 94 87 73 61

Sol :

INSTRUCTORS RANKS
A B R-1 R-2
55 51 3 1
63 53 7 2
64 57 8 4
71 60 10 5
75 61 13 6
82 65 16 9
84 72 17.5 11
88 73 21 12
92 76 22 14
96 80 24 15
84 17.5
85 19
87 20
94 23
𝑁1 = 10 𝑁2 = 14
∑ 𝑅1 = 141.5 ∑ 𝑅2 = 158.5

1) Null Hypothesis 𝐻0: No difference in Instructors

2) Alternate Hypothesis 𝐻𝐴: 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 𝑖𝑛 Instructors

3) LOS = 0.05
4) Critical Region:

− 1. 96 < 𝑍𝑐𝑎𝑙 < 1. 96

5) Calculation Statistics:

𝑁1(𝑁1+1)
𝑈1 = 𝑁1𝑁2 + 2
− 𝑅1 = 53. 5
 𝑁2(𝑁2+1)
𝑈2 = 𝑁1𝑁2 + 2
− 𝑅2 = 86. 5

𝑁1𝑁2 9×12 𝑁1𝑁2(𝑁1+𝑁2+1)

µ𝑈 = 2
= 2
= 70 and σ𝑈 = 12
= 17. 08
1 1

𝑈1−µ𝑈
𝑍𝑐𝑎𝑙 = σ𝑈
1
=− 0. 966
1

6) Decision:

Calculated value lies in acceptance region

Accept Null hypothesis

EXERCISE

1.The following tables shows strength of cable made from 4 different alloys 1 and 2 .There are
two samples 10 cables of alloy 1 and 10 samples of alloy 2 using mann- whilney- U test whether
there is significant difference in them. LOS 0.05

Alloy 1 18.3 16.4 22.7 17.8 18.9 25.3 16.1 24.2

Alloy 2 12.6 14.1 20.5 10.7 15.9 19.6 12.9 15.2 11.8 14.7

2. Instructors A and B both teach a first course in chemistry at XYZ University. On a common
final examination, their students received the grades shown in Table Test at the 0.05 significance
level the hypothesis that there is no difference between the two instructors’ grades.

A 88 75 92 71 63 84 55 64 82 96
B 72 65 84 53 76 80 51 60 57 85 94 87 73 61

3. A farmer wishes to determine whether there is a difference in yields between two different
varieties of wheat, I and II. Table 10-34 shows the production of wheat per unit area using the
two varieties. Can the farmer conclude at significance levels of (a) 0.05, (b) 0.01 that a difference
exists?
Wheat 15.9 15.3 16.4 14.9 15.3 16.0 14.6 15.3 14.5 16.6 16.0
I
Wheat 16.4 16.8 17.1 16.9 18.0 15.6 18.1 17.2 15.4
II

4. A company wishes to determine whether there is a difference between two brands of gasoline,
A and B. table shows the distances traveled per gallon for each brand. Can we conclude at the
0.05 los (a) that there is a difference between the brands (b) that brand B is better than brand A?

A 30.4 28.7 29.2 32.5 31.7 29.5 30.8 31.1 30.7 31.8
B 33.5 29.8 30.1 31.4 33.8 30.9 31.3 29.6 32.8 33.0

3. The Kruskal Wallish H Test

Ex-1: A personal trainer is interested in comparing the anaerobic thresholds of elite athletes.
Anaerobic threshold is defined as the point at which the muscles cannot get more oxygen to
sustain activity or the upper limit of aerobic exercise. It is a measure also related to maximum
heart rate. The following data are anaerobic thresholds for distance runners, distance cyclists,
distance swimmers and cross-country skiers.  ∝ = 0. 05

Distance
Runners Cyclists Swimmers Skiers
185 190 166 201
179 209 159 195
192 182 170 180
165 178 183 187
174 181 160 215
Is a difference in anaerobic thresholds among the different groups of elite athletes?

Solution:

Distance Ranks
Runners Cyclists Swimmers Skiers 𝑅1 𝑅2 𝑅3 𝑅3
185 190 13166 201 15 4 18
179 209 8159 195 19 1 17
192 182 16170 180 11 5 9
165 178 3183 187 7 12 14
174 181 6160 215 10 2 20
R1=46 R2=62 R3=24 R4=78
1) Null Hypothesis 𝐻0: No difference in Performances

2) Alternate Hypothesis 𝐻𝐴: 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 𝑖𝑛 𝑃𝑒𝑟𝑓𝑜𝑟𝑚𝑎𝑛𝑐𝑒𝑠

3) LOS = 0.05
2 2
4) Critical Region :𝐻𝑐𝑎𝑙 < χ ∝
= 𝐴𝑐𝑐𝑒𝑝𝑡 and 𝐻𝑐𝑎𝑙 > χ ∝= Reject

5) Calculation Statistics:

⎰𝐽=𝐾
2
12 𝑅𝑗 ⎱
𝐻= ∑ − 3(𝑁 + 1)
𝑁(𝑁+1) ⎱𝐽=1 𝑛𝐽 ⎰

( ) − 3(20 + 1)⎤⎥⎦ = 9. 11
2 2 2 2
12 46 62 24 78
𝐻 = ⎡⎢ 20(20+1) 5
+ 5
+ 5
+ 5
⎣
2
And χ at ∝ = 0. 05 and 𝑑𝑜𝑓 = ϑ = 𝑘 − 1 = 4 − 1 = 3
2 2
χ [∝,ϑ]
χ [0.05,3]
= 7. 815

6) Decision
2
𝐻𝑐𝑎𝑙 < χ [∝,ϑ]
= 𝐴𝑐𝑐𝑒𝑝𝑡
Ex-2: During one semester a student received in various subjects the grades shown in Table. Test
at the (a) 0.05, (b) 0.01 significance levels whether there is a difference between the grades in
these subjects.
Mathematics 72 80 83 75

Science 81 74 77

English 88 82 90 87 80

Economics 74 71 77 70

Solution:

Subjects Ranks
Maths Science English Economics 𝑅1 𝑅2 𝑅3 𝑅3
72 74 80 70 3 4.5 9.5 1
75 77 82 71 6 7.5 12 2
80 81 87 74 9.5 11 14 4.5
83 - 88 77 13 - 15 7.5
- - 90 - - - 16 -
R1=31.5 R2=23 R3=66.5 R4=15
1) Null Hypothesis 𝐻0: No difference in Grades

2) Alternate Hypothesis 𝐻𝐴: 𝐷𝑖𝑓𝑓𝑒𝑟𝑒𝑛𝑐𝑒 𝑖𝑛 𝐺𝑟𝑎𝑑𝑒𝑠

3) LOS = (a) 0.05, (b) 0.01

2 2
4) Critical Region :𝐻𝑐𝑎𝑙 < χ ∝
= 𝐴𝑐𝑐𝑒𝑝𝑡 and 𝐻𝑐𝑎𝑙 > χ ∝= Reject

5) Calculation Statistics:

⎰𝐽=𝐾
2
12 𝑅𝑗 ⎱
𝐻= ∑ − 3(𝑁 + 1)
𝑁(𝑁+1) ⎱𝐽=1 𝑛𝐽 ⎰

⎰𝐽=4
2
12 𝑅𝑗 ⎱
𝐻= ∑ − 3(16 + 1)
16(16+1) ⎱𝐽=1 𝑛𝐽 ⎰

2 2 2 2
12 ⎰ 𝑅1 𝑅2 𝑅3 𝑅5 ⎱
𝐻= + + + − 3(16 + 1)
16(16+1) ⎱ 𝑛1 𝑛2 𝑛3 𝑛4 ⎰

( ) − 3(16 + 1)⎤⎥⎦ = 9. 2248

2 2 2 2
12 (31.5) (23) (66.5) (15)
𝐻 = ⎡⎢ 16(16+1) 4
+ 3
+ 5
+ 4
⎣
𝑑𝑜𝑓 = ϑ = 𝑘 − 1 = 4 − 1 = 3
2 2
χ [∝,ϑ]
= χ [0.05,3]
= 7. 815

2 2
χ [∝,ϑ]
= χ [0.01,3]
= 11. 34

6) Decision
2
𝐻𝑐𝑎𝑙 < χ [0.05,3]
= 𝑅𝑒𝑗𝑒𝑐𝑡
 2
𝐻𝑐𝑎𝑙 < χ [0.01,3]
= 𝐴𝑐𝑐𝑒𝑝𝑡

EXERCISE

1. During one semester a student received in various subjects the grades shown in Table. Test at
the (a) 0.05, (b) 0.01 significance levels whether there is a difference between the grades in these
subjects.
Mathematics 72 80 83 75

Science 81 74 77

English 88 82 90 87 80

Economics 74 71 77 70

2. A teacher wishes to test three different teaching methods: A, B, and C. To do this, the teacher
chooses at random three groups of five students each and teaches each group by a different
method. The same examination is then given to all the students, and the grades in following table
are obtained. Determine at the (a) 0.05 and (b) 0.01 significance levels whether there is a
difference between the teaching methods. [KWT]

Method A 78 62 71 58 73
Method B 76 85 77 90 87
Method C 74 79 60 75 80

3. A company wishes to purchase one of five different machines A, B, C, D, or E. In an

experiment designed to determine whether there is a performance difference between the
machines, five experienced operators each work on machines for equal times. Table shows
number of units produced by each machine. Test the hypothesis that there is no difference
between the machines at the (a) 0.05, (b) 0.01 significance levels.

A 68 72 77 42 53
B 72 53 63 53 48
C 60 82 64 75 72
D 48 61 57 64 50
E 64 65 70 68 53

4. A teacher wishes to test three different teaching methods: A, B, and C.To do this, the teacher
chooses at random three groups of five students each and teaches each group by a different
method. The same examination is then given to all the students, and the grades in following table
are obtained. Determine at the (a) 0.05 and (b) 0.01 significance levels whether there is a
difference between the teaching methods.

Method A 78 62 71 58 73
Method B 76 85 77 90 87
Method C 74 79 60 75 80

5. An experiment is performed to determine the yields of five different varieties of wheat: A, B,

C, D, and E. Four plots of land are assigned to each variety. The yields (in bushels per acre) are
shown in Table 10-38. Assuming that the plots have similar fertility and that the varieties are
assigned to the plots at random, determine whether there is a significant difference between the
yields at the (a) 0.05, (b) 0.01 levels.

A 20 12 15 19
B 17 14 12 15
C 23 16 18 14
D 15 17 20 12
E 21 14 17 18

6. A company wishes to test four different types of tires: A, B, C, and D. The lifetimes of the
tires, as determined from their treads, are given (in thousands of miles) in Table . Each type has
been tried on six similar automobiles assigned to the tires at random. Determine whether there is
a significant difference between the tires at the (a) 0.05, (b) 0.01 levels.

A 33 38 36 40 31 35
B 32 40 42 38 30 34
C 31 37 35 33 34 30
D 27 33 32 29 31 28
7. A teacher wishes to test three different teaching methods: I, II, and III. To do this, the teacher
chooses at random three groups of five students each and teaches each group by a different
method. The same examination is then given to all the students, and the grades in Table 10-40
are obtained. Determine at the (a) 0.05, (b) 0.01 significance levels whether there is a difference
between the teaching methods.

Method I 78 62 71 58 73
Method II 76 85 77 90 87
Method III 74 79 60 75 80

8. During one semester a student received in various subjects the grades shown in Table 10-41. Test at
the (a) 0.05, (b) 0.01 LOS whether there is a difference between the grades in these subjects.

Mathematics 72 80 83 75 -
Science 81 74 77 - -
English 88 82 90 87 80
Economics 74 71 77 70 -

4. The Run Test for Randomness

Ex-1: A sample of 48 tools produced by a machine shows the following sequence of good (G)
and defective (D) tools:
G G G G G G D D G G G G G G G G G G D D D D G G
G G G G D G G G G G G G G G D D G G G G G D G G
Test the randomness of the sequence at the 0.05 significance level.

Sol: Let 𝑁1 = 10 𝐷 and 𝑁2 = 38𝐷, and the number of runs is V = 11.

1) Null Hypothesis 𝐻0: Randomness

2) Alternate Hypothesis 𝐻𝐴: No Randomness

3) LOS = 0.05
4) Critical Region :

− 1. 96 < 𝑍 < 1. 96

5) Calculation Statistics:

Thus the mean and variance are given by

2𝑁1𝑁2 2×10×38
µ𝑣 = 𝑁1+𝑁2
+ 1= 10+38
+ 1 = 16. 8333

2𝑁1𝑁2(2𝑁1𝑁2−𝑁1−𝑁2) 2×10×38(2×10×38−10−38)
σ𝑣 = 2 = 2 = 2. 234
(𝑁1+𝑁2) (𝑁1+𝑁2−1) (10+38) (10+38−1)

𝑉−µ𝑣 11−16.8333
𝑍𝑐𝑎𝑙 = σ𝑣
= 2.234
=− 2. 6095

6) Decision

Here 𝑍𝑐𝑎𝑙 Lies outside of Critical Region therefore reject Hypothesis

Example:  In 30 tosses of a coin, the following sequence of heads (H) and tails (T) is obtained:
 H T T H T H H H T H H T T H T H T H H T H T T H T H H T H T
 (a) Determine the number of runs, V.
 (b) Test at the 0.05 significance level whether the sequence is random.

Sol: Let 𝑁1 = 16 𝐻 and 𝑁2 = = 14𝑇, and the number of runs is V = 22.

1) Null Hypothesis 𝐻0: Randomness

2) Alternate Hypothesis 𝐻𝐴: No Randomness

3) LOS = 0.05
4) Critical Region :

− 1. 96 < 𝑍 < 1. 96

5) Calculation Statistics:

Thus the mean and variance are given by

2𝑁1𝑁2 2×16×14
µ𝑣 = 𝑁1+𝑁2
+ 1= 16+14
+ 1 = 15. 93

2𝑁1𝑁2(2𝑁1𝑁2−𝑁1−𝑁2) 2×16×14(2×16×14−16−14)
σ𝑣 = 2 = 2 = 2. 679
(𝑁1+𝑁2) (𝑁1+𝑁2−1) (16+14) (16+14−1)

𝑉−µ𝑣 22−15.93
𝑍𝑐𝑎𝑙 = σ𝑣
= 2.679
= 2. 27

6) Decision

Here 𝑍𝑐𝑎𝑙 Lies outside of Critical Region therefore reject Hypothesis

 EXERCISE

1. Determine the number of runs, V, for each of these sequences:

(a) A B A B B A A A B B A B (b) H H T H H H T T T T H H T H H T H T

Test at the 0.05 significance level whether the sequence is random.

2. Twenty-five individuals were sampled as to whether they liked or did not like a product
(indicated by Y and N, respectively). The resulting sample is shown by the following sequence:
Y Y N N N N Y Y Y N Y N N Y N N N N N Y Y Y Y N N

Test at the 0.05 significance level whether the responses are random.