UNIT IV

NON – PARAMETRIC TESTS

 TABLE VALUE : Table value is taken from chi-square
table with ϑ = 𝑛 − 1 degrees of freedom

If calculated value 0f 𝜒 2 < Table value , Then

accept 𝐻0
Calculated value (10) > Table value(4.57)

We reject 𝐻0 .
 A random sample of size 25 from a population
gives the sample standard deviation 8.5. Test the
hypothesis that the population s.d. is 10.

Given : n=25, s=8.5, 𝜎 = 10 ⇒ 𝜎 2 = 100

𝐻0 : 𝜎 2 = 100
𝐻1 : 𝜎 2 ≠ 100
(𝑛−1)𝑠 2 24×8.52
𝑇𝑒𝑠𝑡 𝑠𝑡𝑎𝑡𝑖𝑠𝑡𝑖𝑐: 𝜒 2 = = = 17.34
𝜎2 100
𝑇𝑎𝑏𝑙𝑒 𝑣𝑎𝑙𝑢𝑒 𝑤𝑖𝑡𝑕 24 𝑑. 𝑜. 𝑓 = 36.415
Calculated value < Table value. 𝐻0 is accepted.
Nominal data are used to label variables without any
quantitative value. Common examples
include male/female, hair color, nationalities, names of
people, and so on.

Ordinal data, on the other hand, is a type of data that

has a natural ordering or ranking. It is categorical data
that can be ranked or ordered in accordance with a
specific attribute or characteristic. Examples of ordinal
data are the level of education, the range of income, or
the grades.
Note: IF observed frequencies are <10, then regroup them.
Also adjust n according to regrouping
200 digits were chosen at random from a set of tables, the frequency of the digits were
Digit 0 1 2 3 4 5 6 7 8 9
Freq. 18 19 23 21 16 25 22 20 21 15

Using chi-square test, find the digits were distributed in equal number in the tables.
Table value at 9
d.o.f at 5% =
16.919

Calculated
value < Table
value

𝐻0 𝑖𝑠 𝑎𝑐𝑐𝑒𝑝𝑡𝑒𝑑.
Note:
• We convert all frequencies to whole number so that 𝑂 = 𝐸
• All the frequency should be ≥ 10, 𝑜𝑡𝑕𝑒𝑟𝑤𝑖𝑠𝑒 𝑟𝑒𝑔𝑟𝑜𝑢𝑝.
Fit a Poisson distribution for the following data and also test the
goodness of fit.

X 0 1 2 3 4 5
f 142 156 69 27 5 1

Solution:
𝐻0 : 𝑃𝑜𝑖𝑠𝑠𝑜𝑛 𝑓𝑖𝑡 𝑖𝑠 𝑔𝑜𝑜𝑑

𝑒 −𝜆 𝜆𝑥
𝐹𝑜𝑟 𝑎 𝑃𝑜𝑖𝑠𝑠𝑜𝑛 𝑑𝑖𝑠𝑡𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛, 𝑃 𝑋=𝑥 = , 𝑤𝑕𝑒𝑟𝑒 𝑥 = 0,1,2, …
𝑥!
𝑓𝑖 𝑥𝑖 400
𝑤𝑕𝑒𝑟𝑒 𝜆 = 𝑚𝑒𝑎𝑛 = = =1
𝑓𝑖 400
 𝑥𝑖 O=𝑓𝑖 𝑥𝑖 𝑓𝑖 𝑁 𝑒 −𝜆 𝜆𝑥 (𝑂 − 𝐸)2
𝐸=
𝑥! 𝐸
0 142 0 400 × 𝑒 −1 10 0.17
= 147
0!
1 156 156 400 × 𝑒 −1 11 0.55
= 147
1!
2 69 138 400 × 𝑒 −1 12 0.34
= 74
2!
3 27 81 400 × 𝑒 −1 13 0.03
= 25
3!
4 5 33 20 400 × 𝑒 −1 14
=6 32
4!
5 1 5 400 × 𝑒 −1 15
=1
5!

N= 𝑂 = 𝑓𝑖 𝑥𝑖 =400
𝐸 = 400
400

1.09
 Table value of 𝜒 2 𝑎𝑡 𝑛 − 2 = 4 − 2 = 2 𝑑. 𝑜. 𝑓 =
5.99 (𝑎𝑡 5% 𝑙𝑒𝑣𝑒𝑙)

Calculated value < Table value

H0 is accepted. The given distribution is nearly Poisson

A.U 2022

Fit a Poisson distribution and test the goodness of

fit

X 0 1 2 3 4
f 419 352 154 56 19
A survey of 320 families with 5 children each revealed the following
distribution:

No. of 5 4 3 2 1 0
Boys
No. of girls 0 1 2 3 4 5
No. of 14 56 110 88 40 12
families

Is this result consistent with the hypothesis that male and female births are equally
probable?

H0: Male and female births are equally probable.

O E=𝑵 × 𝒏𝑪𝒙 𝒑𝒙 𝒒𝒏−𝒙 O-E (𝑂 − 𝐸)2
𝐸
14 1 0 4 1.6
320 × 5𝐶0 × ×
2
1 5
= 10
2
56 1 1 6 0.72
320 × 5𝐶1 × ×
2
1 5−1
=50
2
110 100 10 1
88 100 -12 1.44
40 50 -10 2
12 10 2 0.4
𝜒 2 =7.16
Table value of 𝜒 2 at 5% with n-1=6-1=5 d.o.f = 11.07

Calculated 𝜒 2 < Table 𝜒 2

Accept H0.
 (𝑎 + 𝑏)(𝑎 + 𝑐) (𝑎 + 𝑏)(𝑏 + 𝑑)
𝐸 𝑎 = 𝐸 𝑏 =
𝑁 𝑁

(𝑐 + 𝑑)(𝑎 + 𝑐) (𝑐 + 𝑑)(𝑏 + 𝑑)
𝐸 𝑐 = 𝐸 𝑑 =
𝑁 𝑁

𝐻0 : 𝐴𝑡𝑡𝑟𝑖𝑏𝑢𝑡𝑒𝑠 𝑎𝑟𝑒 𝑖𝑛𝑑𝑒𝑝𝑒𝑛𝑑𝑒𝑛𝑡

𝐷𝑒𝑔𝑟𝑒𝑒𝑠 𝑜𝑓 𝑓𝑟𝑒𝑒𝑑𝑜𝑚 = 𝑟 − 1 𝑐 − 1 ,
𝑟 = 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑟𝑜𝑤𝑠, 𝑐 = 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑐𝑜𝑙𝑢𝑚𝑛𝑠
On the basis of information given below about the treatment of 200
people suffering from a disease, state whether the new treatment is
comparatively superior to the conventional or customary treatment.

Treatments Favourable Non-Favourable

New 60 30
Conventional 40 70

Solution: 𝐻0 : 𝐴𝑡𝑡𝑟𝑖𝑏𝑢𝑡𝑒𝑠 𝑎𝑟𝑒 𝑖𝑛𝑑𝑒𝑝𝑒𝑛𝑑𝑒𝑛𝑡

Treatments Favourable Non- Total

Favourable
New 60 30 90
Conventional 40 70 110
Total 100 100 N=200
O E O-E (𝑂 − 𝐸)2
𝐸
60 90 × 100 15 5
=
200
= 45
30 90 × 100 -15 5
=
200
= 45

40 110 × 100 -15 4.09

=
200
= 55

70 110 × 100 -15 4.09

=
200
= 55

𝜒 2 =18.18
TABLE VALUE OF 𝜒 2 AT (r-1)(c-1)=1 d.o.f = 3.841

CALCULATED 𝜒 2 > TABLE VALUE OF 𝜒 2

Reject H0. There is some difference between the

new and conventional treatments
 From the following information, state whether the two attributes viz, condition of house and
condition of child are independent

Condition of Condition of house Total

child
clean dirty
Clean 69 51 120
Fairly clean 81 20 101
Dirty 35 44 79
Total 185 115 300

𝐻0 : 𝐴𝑡𝑡𝑟𝑖𝑏𝑢𝑡𝑒𝑠 𝑎𝑟𝑒 𝑖𝑛𝑑𝑒𝑝𝑒𝑛𝑑𝑒𝑛𝑡

O E O-E (𝑂 − 𝐸)2 Table value at (3-
1)× 2 − 1 =
𝐸
2 𝑑. 𝑜. 𝑓 = 5.991
69 185 × 120 -5 0.34
=
300
= 74
CALCULATED 𝜒 2 > TABLE VALUE OF 𝜒 2
51 120 × 115 5 0.54
=
300 Reject H0. There is an association between the condition
= 46 Of child and condition of house.
81 101 × 185 18.72 5.62
=
300
= 62
20 39 -18.72 9.05
35 49 -13.72 3.86
44 30 13.72 6.21
𝜒 2 =25.62
𝑂 = 300 𝐸 = 300
A brand manager is concerned that her brand's share may be
unevenly distributed throughout the country.
In a survey in which the country was divided into 4 geographic
regions, a random sampling of 100 consumers in each region was
surveyed with the following results.
In the North East region, 40 purchased the brand and the rest did not
purchase. In the North West region, 55 purchased the brand and the
rest did not purchase.
In the South East region, 45 purchased the brand and the rest did not
purchase.
In the South West region, 50 purchased the brand and the rest did
not purchase.
At 𝛼 = 0.05, Use Chi Square test to check whether the brand share is
the same across the four regions.
𝐻0 : 𝑇𝑕𝑒 𝑎𝑡𝑡𝑟𝑖𝑏𝑢𝑡𝑒𝑠 𝑎𝑟𝑒 𝑖𝑛𝑑𝑒𝑝𝑒𝑛𝑑𝑒𝑛𝑡
O E (𝑂 − 𝐸)2
𝐸
40 100×190 1.18
= = 47.5
400
60 52.5 1.07
55 47.5 1.18
45 52.5 1.07
45 47.5 0.13
55 52.5 0.11
50 47.5 0.13
50 52.5 0.119
𝜒 2 = 4.989
Table value at (4-1)(2-1)=3 d.o.f at 5%
level =7.815
Calculated value < Table value
𝐻0 𝑖𝑠 𝑎𝑐𝑐𝑒𝑝𝑡𝑒𝑑
450 150 600
550 850 1400
1000 1000 N=2000

O E (𝑂 − 𝐸)2
𝐸
450 300 75
550 700 32.14
150 300 75
850 700 32.14
𝜒 2 = 214.28
𝑇𝑎𝑏𝑙𝑒 𝑣𝑎𝑙𝑢𝑒 𝑎𝑡 1 𝑑. 𝑜. 𝑓 𝑎𝑡 5% = 3.841
Calculated value > Table value
𝐻0 𝑖𝑠 𝑟𝑒𝑗𝑒𝑐𝑡𝑒𝑑