example of a normality test question

1. The influence of rice field area on rice production results

Rice Field Area (ha) Production Results (tons)

X Y
1 4
2 10
5 20
7 35
10 50
12 60

a. Normality test for rice field area (x) with α = 5%. Does the sample come from a
normally distributed population?
Solution
1) Ho: The sample comes from a normally distributed population
H 1 : the sample does not come from a population with a norm distribution

2) α = 0.05
3) test statistics used
L obs = Max | F (zi) – S (zi) |
4) computing
̅
Zi =

Diket:
̅ = 29.83
s = 22.36

5) Table looking for Lmax

Xi Zi Ztable F(zi) S(zi) |F(zi)- S(zi)|
1 -1.18 0.3810 0.1190 0.1667 0.0477
2 -0.95 0.3289 0.1711 0.3333 0.1622
 5 -0.26 0.1026 0.3174 0.5000 0.1826*
7 0.19 0.0753 0.5753 0.6667 0.0916
10 0.88 0.3106 0.8106 0.8333 0.0227
12 1.34 0.4099 0.9099 1,0000 0.0901

Lobs = 0.1826
6) Critical area
Ltable : Lα ; L 0.05; 6
= 03190
Accept Ho if Lobs < Ltable
0.1826 < 0.3190
7) Test Decision
Thank Ho
8) Conclusion
The sample comes from a normally distributed population

b. Normality test of production results (y) with α = 5%. Does the sample come from
a normally distributed population?
Solution:
1) Ho: the sample comes from a normally distributed population
H 1 : the sample does not come from a normally distributed population
2) α = 0.05
3) test statistics
Lobs = Max | F(zi) – S(zi) |
4) Computing
̅
Zi =

Diket:
̅ = 29.83
s = 22.36
5) Table looking for Lmax
 Xi Zi Ztable F(zi) S(zi) |F(zi)-S(zi)|
4 -1.15 0.3749 0.1251 0.1667 0.0416
10 -0.88 0.3106 0.1894 0.3333 0.1439
20 -0.44 0.1700 0.3300 0.5000 0.1700*
35 0.23 0.0910 0.5910 0.6667 0.0757
50 0.90 0.3159 0.8159 0.8333 0.0174
60 1.34 0.4099 0.9099 1,0000 0.0901

Lobs = 0.1700
6) Critical area
Ltable : Lα ; n = L 0.05 ; 6
= 0.3190
Accept Ho if Lobs < Ltable
0.1700< 0.3190
7) Test decision
Thank Ho
8) Conclusion
The sample comes from a normally distributed population

c. Homogeneity test
To test whether models X and Y have the same variation (homogeneous). At
random, 6 models X and 6 models Y were taken. The data were:
X: 12571012
Y: 41020355060
By taking α = 5%, are the two population variances homogeneous or not?
Solution:

1) Ho: σ =σ
2 2
1 2 ( variance of both populations is homogeneous/same)

H 1: σ σ 2 2 (inhomogeneous population variance)

2
1

2) α = 0.05
3) test statistics

Fobs = ~ F(n 1 – 1, n 2 – 1 )
 4) computing
it is obtained that: s 1 = 4.36 and S 2 = 22.36

Fobs =

Fobs =

Fobs =

5) critical area
F table = F 0.05 ; (n 1 -1, n 2 -1) = F 0.05 ; 5.5 = 5.05
Ho is accepted if F obs < F table
0.0378 < 5.05
6) test decision
Thank Ho
7) conclusion
The variances of the two populations are homogeneous / the same
d. regression and correlation
In accordance with information which states that increasing rice production
depends on the area of the rice fields. So, these assumptions are then examined and
the results are as follows:
No X Y XY X2 Y2
1 1 4 4 1 16
2 2 10 20 4 100
3 5 20 100 25 400
4 7 35 245 49 1225
5 10 50 500 100 2500
6 12 60 720 144 3600

amount 37 179 1589 323 7841

1) Determine the linear regression equation

solution:
Everything is normal
 ∑ y = na + b∑ x................ 1)
∑ xy = a ∑x + b ∑ x 2 ............2)
I didn't know anything about it
179 = 6a + 37b
1589 = 37a + 323b

179 = 6a + 37b |x 37| 6623 = 222a + 1369b

1589 = 37a + 323b |x 6 | 9534 = 222a + 1938b _ -
-2911 = -569b
b = 5.1159

Substituting this information 1) some water:

179 = 6a + 37(5,1159)
179 = 6a + 189,28
179 – 189.28 = 6a
-10.28 = 6a
a = -1.713

So the regression model equation is y = -1.713 + 5.1159x

2) What is the magnitude of the influence (contribution) of rice field area on rice
production?

r=
√

r=
√

r=
√

r=
√

r=
√
 r=

r = 0.99
the correlation coefficient (r) is 0.99
coefficient of determination (R) = r 2 = (0.99) 2 = 0.98
conclusion:
The influence of rice field area on rice production is very strong. The
contribution of rice field area to production results is 98%

2. Data on study time and student GPA are known as follows.

Time GPA
Study (hours) Got it
5 3
3 2.5
3 3.6
5 3.3
2 2.5
3 3.5
4 3.6
1 2

With α = 5%

a. Normality test for study time

Solution :
1) Ho: the sample comes from a normally distributed population
H 1 : the sample does not come from a normally distributed population

2) α : 0.05
3) test statistics
Lobs = Max | F(Zi) – S(Zi) |

4) computing
Based on the data above, it is obtained:
 ̅ = 3.25
= 1.38
Table to find Lmax
Xi Zi Z table F(Zi) S(Zi) |F(Zi)-S(Zi)|
1 -1.63 0.4484 0.0516 0.1250 0.0734
2 -0.90 0.3159 0,1841 0,2500 0,0659
3 -0,18 0,0714 0,4286 0,6250 0,1964
3 -0,18 0,0714 0,4286 0,6250 0,1964
3 -0,18 0,0714 0,4286 0,6250 0,1964 *
4 0,54 0,2054 0,7054 0,7500 0,0446
5 1,26 0,3962 0,8962 1,0000 0.1038
5 1.26 0.3962 0.8962 1,0000 0.1038

Lobs = 0.1964
5) Critical area:
L table = L α ; n = L 0.05 ; 8
= 0.2850
Ho is accepted if Lobs < Ltable
0.1964 < 0.2850
6) Test decision
Thank Ho
7) Conclusion
The sample comes from a normally distributed population

b. Homogeneity test
With α = 5%
Solution :
1) Ho: = (variance of both homogeneous populations)
H 1: ≠ (the variance of the two populations is not homogeneous)
2) α = 0.05

3) test statistics used

 Fobs = ~ F(n 1 – 1, n 2 – 1 )

4) computing
it is obtained that: s 1 = 1.38 and S 2 = 0.60

Fobs =

Fobs =

Fobs =

5) critical area
F table = F 0.05 ; (n 1 -1, n 2 -1) = F 0.05 ; 7.7 = 3.79
Ho is accepted if F obs < F table
5.2777 > 3.79
6) test decision
Reject Ho
7) conclusion
The variances of the two populations are not homogeneous / the same

c. The regression equation

X = Length of time studying
Y = GPA

NO X Y XY X2 Y2
1 5 3 15 25 9
2 3 2,5 7,5 9 6,25
3 3 3,6 10,8 9 12,96
4 5 3,3 16,5 25 10,89
5 2 2,5 5 4 6,25
6 3 3,5 10,5 9 12.25
7 4 3.6 14.4 16 12.96
8 1 2 2 1 4
AMOUNT 26 24 81.7 98 74.56

Solution:
 From the normal equation
∑ y = na + b∑ x................ 1)
∑ xy = a ∑x + b ∑ x 2 ............2)
So we get the equation
24 = 8a + 26b
81.7 = 26a + 98b

24 = 8a + 26b |x 26| 624= 208a + 676b

81.7 = 26a + 98b |x 8 | 653.6 = 208a + 784b _ -
-29.6 = -108 b
b = 0.2740
Substituting into equation 1) then we get:
24 = 8a + 26(0.2740 )
24 = 8 a + 7.124
24 – 7.124 = 8 a
16,876 = 8 a
a = 2.1095

so the regression model equation is y = 2.1095 + 0.2740 x

d. If a student wants to get a GPA of 3.8 then how much per day should he study?
From the regression equation model above, it is obtained
y = 2.1095 + 0.2740 x
if y = 3.8 then obtained
3.8 = 2.1095 + 0.2740 x
3.8 – 2.1095 = 0.2740 x
1.6905 = 0.2740 x
x = 6.16
So to get a GPA of 3.8 a student must study for 6.16 hours a day.

e. How big is the influence (contribution) of time on the student's GPA?

r=
√
r=
√

r=
√

r=
√

r=
√

r=

r = 0.62
the correlation coefficient (r) is 0.62
coefficient of determination (R) = r 2 = (0.62) 2 = 0.38
conclusion:
The influence of the length of study hours on GPA is not strong. The contribution of
length of study to GPA is only 38%