Exercise 5E

S xy 100 100
1 r= = = = 0.985 (3 s.f.)
S xx S yy 92 ´ 112 101.50862

 x
2
367 ´ 367
2 S xx =  x 2  = 33845  = 33845  22 448.166K = 11396.833K
n 6
 y 
2
270 ´ 270
S yy =  y 2
 = 12976  = 12976  12150 = 826
n 6

S xy =  xy 
 x y 367 ´ 270
= 17135  = 17135  16515 = 620
n 6
S xy 620 620
r= = = = 0.202 (3 s.f.)
S xx S yy 11396.833 ´ 826 3068.189

 a
2
115 ´ 115
3 a S aa =  a 2  = 1899  = 9.7142K = 9.71 (3 s.f.)
n 7

Sah 72.1
b r= = = 0.96774K = 0.968 (3 s.f.)
Saa Shh 9.7142... ´ 571.4

c There is positive correlation. The greater the age of the person, the taller the person.

4 a Calculating the summary statistics gives:

 L = 26.8 L 2
= 150.02  T = 47.4 T 2
= 399.58  LT = 237.07
26.8 ´ 26.8
S LL = 150.02  = 150.02  119.7066K = 30.3133K = 30.3 (3 s.f.)
6
47.4 ´ 47.4
STT = 399.58  = 399.58  374.46 = 25.12
6
26.8 ´ 47.4
S LT = 237.06  = 237.07  211.72 = 25.35
6

S LT 25.35 25.35
b r= = = = 0.91865K = 0.919 (3 s.f.)
S LL STT 30.3133K ´ 25.12 27.5947K

c The data in the scatter graph appear to be linear, and the correlation coefficient found in part b is
close to 1. Therefore, a linear regression model is suitable to model the data.

© Pearson Education Ltd 2019. Copying permitted for purchasing institution only. This material is not copyright free. 1
  x
2
973 ´ 973
5 a S xx =  x 2  = 120 123 = 120 123 118 341.125 = 1781.875
n 8
 y
2
490 ´ 490
S yy =  y 2  = 33 000 
= 33 000  30 012.5 = 2987.5
n 8

S xy =  xy 
 x y = 61 595  973 ´ 490 = 61 595  59 596.25 = 1998.75
n 8
S xy 1998.75 1998.75
r= = = = 0.86629K = 0.866 (3 s.f.)
S xx S yy 1781.875 ´ 2987.5 2307.2389

b The correlation is positive. The higher the IQ, the higher the mark gained in the general knowledge
test. (Alternatively, the higher the mark gained in the intelligence test, the higher the IQ.)

6 The coding is linear, so the product moment correlation coefficient will be unaffected by the coding.
So the product moment correlation coefficient between x and y is 0.973.

7 a This is the coded data set:

p 0 5 3 2 1
q 0 17 12 10 6

b Calculating summary statistics for the coded data gives:

 p = 11  p = 39  q = 45  q
2 2
= 569  pq = 147
  p  = 39  11´11 = 14.8
2

S pp =  p 2

n 5
 q
2
45 ´ 45
S qq =  q 2  = 569 
= 164
n 5

S pq =  pq 
 p q = 147  11´ 45 = 48
n 5
S pq 48
r= = = 0.97429K = 0.974 (3 s.f.)
S pp S qq 14.8 ´ 164

c The coding is linear. The product moment correlation coefficient is independent of the linear
coding, hence it is 0.974 (3 s.f.).

© Pearson Education Ltd 2019. Copying permitted for purchasing institution only. This material is not copyright free. 2
8 a This is the coded data set:

p 10 8 11 9 12
t 4 3 5 4 6

Calculating summary statistics for the coded data gives:

 p = 50  p = 510  t = 22  t
2 2
= 102  pt = 227
  p  = 510  50 ´ 50 = 10
2

S pp =  p 2 
n 5
t 
2
22 ´ 22
Stt =  t 2  = 102 
= 5.2
n 5

S pt =  pt 
 p t = 227  50 ´ 22 = 7
n 5

S pt 7 7
b r= = = = 0.97072K = 0.971 (3 s.f.)
S pp Stt 10 ´ 5.2 7.2111K

c The coding is linear. The product moment correlation coefficient is independent of the linear
coding, hence it is 0.971 (3 s.f.).

9 a This is the coded data set:

x 15 37 5 0 45 27 20
y 30 13 34 43 20 14 0

Calculating summary statistics for the coded data gives:

 x = 149 x 2
= 4773  y = 154 y 2
= 4670  xy = 2379
 x
2
149 ´149
S xx =  x 2
 = 4773  = 1601.4285K = 1601 (4 s.f.)
n 7
 y
2
154 ´ 154
S yy =  y 2
 = 4670  = 1282
n 7

S xy =  xy 
 x y = 2379 
149 ´ 154
= 899
n 7

S xy 899 899
b r= = = = 0.62742K = 0.627 (3 s.f.)
S xx S yy 1601.4285 ´1282 1432.84K

c The shopkeeper is not correct. There is negative correlation, so as the newspaper sales go up the
sweet sales go down.

© Pearson Education Ltd 2019. Copying permitted for purchasing institution only. This material is not copyright free. 3
  f   10 x     x 
2 2 2

10 a S ff =  f 2
 ==  (10 x) 2
 = 10   x 
2 2 
n 8  8 
 
= 100S xx = 100 ´ 111.48 = 11148

 g    5( y + 10) 5 y + 50 ´ n
2 2 2

b S gg =  g 2  = 74 458.75  = 74 458.75 
n n n
(5 ´ 70.9 + 50 ´ 8) 2
= 74 458.75  = 3299.97
8
S fg 5667.5
r= = = 0.934 (3 s.f.)
S ff S gg 11148 ´ 3299.97

c The product moment correlation coefficient shows strong linear correlation. However, the scatter
diagram suggests a non-linear fit.

 x
2
122
11 a S xx =  x 2
 = 22.02  = 1.44857 K
n 7
 y 
2
97.7 2
S yy =  y 2  = 1491.69 
= 128.077K
n 7

S xy =  xy 
 x  y = 180.37  12 ´ 97.7 = 12.8842K
n 7
S xy 12.884K
r= = = 0.946 (3 s.f.)
S xx S yy 1.4485K ´128.077K

b This table sets out the residuals for each data point:

x y y = –1.2905 + 8.8945x
1.1 6.2 8.49345 –2.29345
1.3 10.5 10.27235 0.22765
1.4 12 11.1618 0.8382
1.7 15 13.83015 1.16985
1.9 17 15.60905 1.39095
2.1 18 17.38795 0.61205
2.5 19 20.94575 –1.94575

c The linear model might not be a good model for this data, as the residuals do not appear to be
randomly scattered about zero.

© Pearson Education Ltd 2019. Copying permitted for purchasing institution only. This material is not copyright free. 4