EXERCISES [MAI 4.

4]
LINEAR REGRESSION
SOLUTIONS
Compiled by: Christos Nikolaidis

A. Paper 1 questions (SHORT)

1. (a) 0.965
(b) strong positive
(c) y = 2.2x – 0.5
(d) whenever x increases by 1 unit, y increases by 2.2 units.
(e) The value of y corresponding to 0 units of x.

2. (a) y = 2.2x – 0.5  y + 0.5 = 2.2x  x = 0.455 y + 0.227

(b) x = 0.423y + 0.385
(c) The relation in (a) is in fact the inverse function of the line y = 2.2x – 0.5
If y is given, the answer in (c) gives a more reliable estimation of x.

3. (a) rs = 1.
(b) It describes the monotonic relationship of the data. When x increases y also increases.
4. (a) II
(b) V
(c) III
(d) I

5. (a) (i) 1 (ii) 3

(b) (i) 1 0.04 2 –0.20 3 –0.85

(ii) 1.60 A product–moment correlation coefficient cannot be > 1.
0.90 There is no diagram with a strong positive correlation.

6. (a) (see diagram)

(b)
Relationship between leaf length and width

70
60
50 M
Width
(mm) 40
30
20
10

0 20 40 60 80 100 120 140 160

Length (mm)

(c) leaf length and leaf width are positively correlated

7. (a) r = 0.76
(b) Fairly strong positive correlation between high school grades and university grades
(c) y = 0.052x – 1.29 (3 s.f.)

1
8. (a) High positive or high or positive or good correlation etc.
(b) M(29, 31)
(c) Suitable line which should pass through the candidate’s M and
have nearly as many crosses (plotted points) below it as above it.
(d) value (including non-itegers) obtained using candidate’s line of best fit.
(Follow through from part (c).)

9. (a) The scores are negatively correlated

(b)

100
90
80
70
Game 60
50 ×
×
40 M
30
20
10
00 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95
Mathematics score
Line must be drawn straight.
It must pass through (0, 69).
It must pass through the mean point M = (56.9, 45.9).
(c) 51 is closest. A1low 50 or 52 (ft from candidate’s graph).

10. (a) On the graph

(b) x = 34 and y = 40
y

60

50
number
of 40
situps
30

20

10

O 10 20 30 40 50 60 x
number of pushups
(c) 50 situps (allow ±2) (ft from candidate’s graph)

2
11. (a) r = 0.6399706…  0.64 (2 d.p.)
(b) This indicates that there is a degree of positive correlation between scores in Mathematics
and scores in English.
Therefore those who do well in Mathematics are likely to do well in English also. (Or
equivalent statements.)
12. (a) r = 0.838
(b) it is the average of the ranks 4,5,6,7
(c)
Rank x Rank y
(x) (y )
original modified original modified
10 100 1 1 1 1
20 110 2 2 2 2
30 130 3 3 4 4.5
40 120 4 5.5 3 3
40 130 5 5.5 5 4.5
40 140 6 5.5 6 6.5
40 150 7 5.5 8 8
50 140 8 8 7 6.5

Spearman rank correlation coefficient rs = 0.761

B. Paper 2 questions (LONG)

13. (a) r = 0.929

(b) strong positive
(c) y = 0.929 x +10.3
(d) x = 0.929 y – 9
(e) ranks
x y
1 1
2 3
3 2
4 4
5 6
6 5
rs = 0.886

(f) r = 0.929 indicates the degree of linear relationship between x and y (strong positive)
rs = 0.886 indicates the degree of monotonic relationship between x and y (in what
extent y increases when x increases)

3
14. (a)
60
55
50
45
40
35
FUEL
30
25
20
15
10
5
0
0 100 200 300 400 500 600 700 800 900
DISTANCE
2 marks for all 3 points correct. Only 1 mark for 2 points correct
(b) Straight line with negative gradient passing through the mean
intercept on y-axis between 50 and 55
(c) 32 (read answer from candidate’s line)
(d) r = – 0.978
(e) ranks
x 1 2 3 4 5 6
y 6 5 4 3 2 1

rs = –1
(f) r = – 0.978 indicates the degree of linear relationship between x and y (strong negative)
rs = –1 indicates a perfect negative monotonic relationship between x and y
(y decreases when x increases)

182  173 162  178 190 73  68  60  66  75

15. (a) (i) = 177cm (ii) = 68.4 kg
5 5
(Or directly by GDC)
(b)

190

185

180
M
175
height
(cm) 170

165

160

0
60 65 70 75
weight (kg)
(c) r = 0.943
(d) ranks
162 60 1 1
173 68 2 3
178 66 3 2
182 73 4 4
190 75 5 5
rs = 0.9
(e) Because the rank of y for 173 is larger than the rank of y for 178 (no perfect
monotonic relationship)

4
16. (a) y = 0.070x – 3.22
Accept 0.07x.
(b) for each cm of height the shoe size increases by 0.070
(c) (i) y = 0.070 × 162 – 3.22 = 8.12
Therefore shoe size 8 or 9 (8.12).
OR y = 8 or 9
(ii) interpolation since a.62 is withing the range of values of x.
(d) r = 0.681
(e) Moderately strong, positive correlation.

17. (a)
x 155 161 173 150 182 165 170 185 175 145
y 50 75 80 46 81 79 64 92 74 108

100

90

80

70
weight
(kg) 60

50

40

30

20

10

0 x
20 40 60 80 100 120 140 160 180 200
Height (cm)

(b) Mean height = 166.1 = 166 (3 s.f.)

Mean weight = 74.9 (3 s.f.)
(c) (i) y = 0.276x + 29.1
(ii) Line on graph. y-intercept at 29.1, straight line through (166, 74.9).
(d) (i) y = 0.276 × 190 + 29.1= 81.5 kg
(ii) 72 = 0.276x + 29.1
72  29.1
x= = 155 cm.
0.276
OR From the graph
(i) y = 81 (1)
(ii) x = 155 (1)
(e) The ‘line of best fit’ becomes closer to the remaining points.
OR
Gradient becomes steeper and the line is more accurate ‘best fit’.
OR
Any reasonable explanation. (Line becomes y = 1.10x – 113)

5
18. (a) (i) r = 0.935 (3 s.f.)
(ii) it suggests a strong positive correlation between the two variables.
(b) y = 0.291x + 1.56
(c) (i) y = 0.291 × 30 + 1.56 = 10.29 = 10 hours
(ii) 8 = 0.291x + 1.56
x = 22.13 = 22 years
(d) x = 3y – 0.846, For y = 8, x = 23.154 = 23

19. (a) (i) 1992 mean = $1.59, Sd = $0.727 (or 0.73)

(ii) 2002 mean = $1.98, Sd = $0.635 (or 0.64)
(b) (i) r = 0.672
(ii) There is a weak positive correlation
(c) y = 0.588x + 1.05
(d) y = 0.588 × 2.60 + 1.05 = $2.58
(e) Coffee – because it is the only item to go down in price.
OR
Rolls – because the price increased significantly.

20. (a)

90
80 M

70
60
Science
Scores 50
40
30
20
10

0 10 20 30 40 50 60 70 80 90
Maths Scores
(b) Point M(73,78)
(c) y = 0.359x + 51.8
(d) going through M, y intercept anywhere from 50 to 54
(e) y = 0.359 × 88 + 51.8 = 83
OR
y = 83 (±2) if read from the graph and method is shown.

6
21. (a) y = 0.290x + 56.5
(b) y = 0.770x + 14.3
(c) r = 0.472
(d)
Mathematics (x) Science (y) Rank x Rank y

38 60 1 1
56 80 2 6
58 78 3 5
73 65 4 4.5 2
73 90 5 4.5 9
80 85 6 7 7.5
90 70 7 8 3
90 71 8 8 4
90 85 9 8 8 7.5
95 96 10 10
rs = 0.444