Course Title: Applied Statistics for Optometrists

Course Code: ST-326

Instructor: Ms. Rabeea Samad

CHAPTER#7

CORRELATION

Exercise Numerical Problems (Solution)

Q-1: Compute the correlation coefficient between the variables and represented in the
following table:

2 4 5 6 8 11
18 12 0 8 7 5

Solution:

2 18 36 4 324
4 12 48 16 144
5 0 0 25 0
6 8 48 36 64
8 7 56 64 49
11 5 55 121 25
∑ = 36 50 243 266 606

∑ ∑ ∑
√[ ∑ (∑ ) ][ ∑ (∑ ) ]

( )( ) ( )( )
√[( )( ) ( ) ][( )( ) ( ) ]

As it means there exists a strong negative correlation between and .

Q-2: Compute the coefficient of determination for the following data and interpret the co-
efficient:

Income ( ) (000) 10 20 30 40 50 60
Expenditure ( ) ( ) 7 21 23 34 6 53
Solution:

10 7 70 100 49
20 21 420 400 441
30 23 690 900 529
40 34 1360 1600 1156
50 6 300 2500 36
60 53 3180 3600 2809
∑ = 210 144 6020 9100 5020

∑ ∑ ∑
√[ ∑ (∑ ) ][ ∑ (∑ ) ]

( )( ) ( )( )
√[( )( ) ( ) ][( )( ) ( ) ]

As it means there exists a weak positive correlation between and .

Q-3: Compute the correlation coefficient for a sample of 20 pairs of observations, given that:

̅ ,̅ ,∑ ,∑ and ∑ .

Solution:
∑ ∑ ∑
,
√[ ∑ (∑ ) ][ ∑ (∑ ) ]

∑
̅ ∑ ̅

∑
̅ ∑ ̅

( )( ) ( )( )
√[( )( ) ( ) ][( )( ) ( ) ]

As it means there exists a weak positive correlation between and .

Q-4: The following data were collected to determine a suitable correlation coefficient relating
the length of an infant, ( ), to age, ( )(days), and weight at birth, (kg).

57.5 52.8 61.3 67.0 53.5 62.7 56.2 68.5 69.2

78 69 77 88 67 80 74 94 102
2.75 2.15 4.41 5.52 3.21 4.32 2.31 4.30 3.71

Solution:

78 2.75 57.5 214.5 6084 7.56 3306.25 4485 158.125

69 2.15 52.8 148.35 4761 4.62 2787.84 3643.2 113.52
77 4.41 61.3 339.57 5929 19.45 3757.69 4720.1 270.333
88 5.52 67.0 485.76 7744 30.47 4489 5896 369.84
67 3.21 53.5 215.07 4489 10.30 2862.25 3584.5 171.735
80 4.32 62.7 345.6 6400 18.66 3931.29 5016 270.864
74 2.31 56.2 170.94 5476 5.34 3158.44 4158.8 129.822
94 4.30 68.5 404.2 8836 18.58 4692.25 6439 294.55
102 3.71 69.2 378.42 10404 13.76 4788.64 7058.4 256.732
∑ =729 32.68 548.7 2702.41 60123 128.6602 33773.65 45001 2035.521

∑ ,∑ ,∑ ,∑ ,∑ ,
∑ ,∑ ,∑ , .
(∑ ) ( )
∑ ∑

(∑ ) ( )
∑ ∑

(∑ )(∑ ) ( )( )
∑ ∑

(∑ )(∑ ) ( )( )
∑ ∑

(∑ )(∑ ) ( )( )
∑ ∑

Now we compute the regression co-efficient and constant as follows:

(∑ )(∑ ) (∑ )(∑ ) ( )( ) ( )( )
(∑ )(∑ ) (∑ ) ( )( ) ( )

(∑ )(∑ ) (∑ )(∑ ) ( )( ) ( )( )
(∑ )(∑ ) (∑ ) ( )( ) ( )
And

̅ ̅ ̅

( )( ) ( )( )

Hence the desired multiple linear regression equation is

The coefficient of multiple determinations is

∑( ̂ ̅ )
∑( ̅)

(∑ )
∑ ∑ ∑
(∑ )
∑

( )
( )( ) ( )( ) ( )( )
( )
( )

This means that 2617% of the variability in infant is explained by its linear relationship
with age and weight at birth.