954/3 STPM (PCS3 2020)

Section A (45 marks)

Answer all questions

1. The following are the annual amounts of money spent on clothes, to the nearest RM10, by
27 people.
10 40 60 80 100 130 140 140 140
150 150 150 160 160 160 160 170 180
180 200 210 250 270 280 310 450 570
(a) Construct a stem-and-leaf diagram for the data. [3
marks]
(b) Find the median and the interquartile range of the data. [3
marks]
An ‘outlier’ is defined as any data value which is more than 1.5 times the interquartile range
above the upper quartile, or more than 1.5 times the interquartile range below the lower
quartile.
(c) List the outliers. [3
marks]

2. The probability that a car will be stolen if it had been fitted with a security alarm system is
0.05 whereas the probability that it will be stolen if it had not been fitted with a security
alarm system is 0.15. It is known that 60% of the cars have been fitted with security alarm
systems.
(a) Find the probability that a randomly chosen car will be stolen. [2
marks]
(b) If a car had been reported stolen, find the probability that it had been fitted with a
security alarm system. [3
marks]

1
3. The random variable X has a binomial distribution with parameters n = 500 and p= .
2
Using a suitable approximation distribution, find P (|X −E ( X )|≤ 25 ) . [7 marks]

4. The weight of a female student in a particular school is denoted by x kilograms. The

weight of a random sample of 130 female students are summarized by
∑ x =7280∧∑ ( x−56 )2 =5005.
(a) Calculate unbiased estimates of the mean and variance of X. [2
marks]
(b) Obtain a 95% confidence interval for the population mean weight of the female
students in the school. [4
marks]
The principal claimed that the mean weight of a female student in the school is 55
kilograms.
(c) Using the result in part (b), determine whether the principal is understating the mean
weight of a female student. [2
marks]

5. A survey carried out by a manufacturer of handphones finds that 136 out of 400 shops sell
the handphones at price less than the recommended prices.
(a) Determine the smallest sample size required so that the estimated proportion of shops
 selling the handphones at prices less than the recommended prices is within 2% of the
actual proportion at the 90 % symmetrical confidence interval level. [4 marks]
(b) Calculate the probability that more than 60% of a random sample of 500 shops sell
the handphones at price less than the recommended prices. [3 marks]
6. A large international company employs university graduates who have Class I, II or III
degrees. After one year’s employment, the performance of 250 graduates was assessed. It was
found that 32 graduates were graded A (high), 140 graduates were graded B (average) and the
rest were graded C (below average). Of these 250 graduates, 30 had class I degrees, 150 had
Class II degrees and the rest had Class III degrees. Of the graduates with Class I degrees, 8
were graded A and 14 were graded B. Of those with Class II degrees, 21 were graded A and
70 were graded B. The rest were each graded C. A table of the results is shown below.

Class
Total
I II III
A 8 21 3 32
Performance B 14 70 56 140
C 8 59 11 78
Total 30 150 70 250

Perform a chi-squared test, at the 5% significance level, to determine whether there is an

association between performance grade and degree class. [9 marks]

Section B [15 marks]

Answer one question only in this section.

7 (a) A random sample of n children is chosen from the town, if the 95% symmetric
confidence interval for the proportion of children like McDonald in the town is
(0.445, 0.235), find the value of n. [4]

(b) A teacher devises a short multiple choice test. The test has 5 questions and for each
question, a student has to choose one of the three possible answers, only one of which is
correct. For each question the possible answers intended to be such that, to a student who
has limited knowledge of the subject, each of the three seems equally sensible. The
teacher tries this test on a sample of 100 students who have limited knowledge of the
subject. The numbers of correct answer achieved by these students are given in the table.
Answer correct 0 1 2 3 4 5
Frequency 6 30 34 23 5 2

Test, all 5% significance level, the hypothesis that, for these questions, the 3
possible answers are equally likely to be chosen. [11 marks]

8. Forty three of n randomly selected houses in a housing area are found to have installed
CCTVs. A symmetric k % confidence interval for the proportion of houses with CCTVs
is (0.0654, 0.1066). Given that n > 30.
(a) Determine the distribution of the sample proportion. [2 marks]
(b) Determine the value of n . [3 marks]
(c) Show that k  90. [6 marks]
(d) Estimate the size of the required random sample so that a symmetric 90% confidence
interval has a width of 0.014. [4 marks]

END OF QUESTION PAPER

 Q Solution Marks
1(a)
B1
Stem leaf
0 1468
1 034445556666788
2 01578
B1
3 1
4 5
5 7
B1
Key : 1  4 represents RM 140
1(b) Median = x14 = 160 B1
IQ range = UQ – LQ = 210 – 140 M1
= 70 A1
1(c) Lower boundary = LQ – 1.5(IQ range) = 140 – 1.5(70) M1
= 35
Upper boundary = UQ + 1.5(IQ range) = 210 + 1.5(70)
= 315 A1
Outliers are 10, 450, 570 A1

2(a) P(S) = P(F ¿ S) + P(F’ ¿ S) = 0.6(0.05) + 0.4(0.15) M1

= 0.09 A1
0.6(0.05)
2(b) P(F|S) = M1M1
0.09
1
= A1
3

3 1
(
X B ( n , p )  X B 500 ,
2 )
1 M1
μ= E ( X )=np=500 × =250
2
1 1
σ 2=npq=500× × =125
2 2
1 M1
Since, n>50, p= and np> 5, then the normal approximation is used.
2
X N ( 250 ,125 ) M1
P (|X −E ( X )|≤ 25 )
¿ P (|X −250|≤25 )
¿ P (−25 ≤ X −250 ≤−25 )
= P(225 ≤ X ≤ 275) P(224.5 < X < 275.5) [Taking continuity correction] M1
224.5−250 X−250 275.5−250 M1
¿P ( √125
<
√ 125
<
√ 125 )
= P(–2.281 < Z < 2.281) M1
= 0.977 A1

4(a) 7280
^μ =
130
= 56 A1
 5005
σ^ 2 =
129
= 38.8 A1
4(b) 6.229 B1M1
(
95% confidence interval for  = 56 ± 1.96
√130 )
= (54.93, 57.07) A1
4(c) 55 kg is in the 95% confidence interval, therefore do not reject H0, B1
that means the principal does not understate the mean weight. A1

5(a) ^p = 0.34 B1
0.34(0.66)
1.645
√
n ≥ 1518.08
n
≤ 0.02 M1
A1
The smallest size is 1519 A1
5(b) 0.2244
^p ~ N(0.34, )
500
P( ^p > 0.6)  P( ^p ≥ 0.65)
0.65−0.34
= P(^p ≥ 0.2244 ) M1

p ≥
√
= P(^ 14.63)
500

=0 A1

6 H0 : There is no association between the performance grade and the degree

class
B1
H1 : There is an association between the performance grade and the degree
class
Observed Frequency
Class
Total
I II III
A 8 21 3 32
Performance B 14 70 56 140
C 8 59 11 78
Total 30 150 70 250
Expected Frequency
Class
Total
I II III
A 3.84 19.2 8.92 32
Performance B 16.80 84.0 39.20 140 M1
C 9.36 46.8 21.84 78 A1
Total 30 150 70 250
Combine the data : Grade A with Grade B performance.
Degree of freedom, v = (2 – 1)(3 – 1) = 2 B1
If 2 ≥ 5.991, reject H0 B1
After Combination, M1A1
(Oi−E i)2
Oi Ei
Ei
22 20.64 0.0896
8 9.36 0.1976
 91 103.2 1.4422
59 46.8 3.1803
59 48.16 2.4399
11 21.84 5.3803
2 = 12.7299
Since 2 = 12.7299 > 5.991 M1
There is sufficient evidence to reject H0, i.e., there is an association between
A1
the performance grade and the class degree.

7(a) psqs
(
95% CI = p s ± z 0.05
2 √ ) n
=( 0.235 , 0.445 ) B1

p s .(1− ps ) M1
(
¿ p s ±1.96
√ n )
0.235+ 0.445
ps = =0.34 ; M1
2
p s .(1−p s )
1.96
√ n
p s .(1− ps ) 0.105
=0.445−0.34=0.105

√ n
p s .(1− ps )
=
1.96
=0.05357
A1
=0.00287
n
0.34 .(1−0.34)
=0.00287
n
n=78.19 ≈78

7(b)
H 0 :3 possible answers are equally likely ¿ be chosen B1
5
1 x 2 5
H 1 :3 possible answers are not equally likely ¿ be chosen
❑
C❑x ( )( )
3 3
Answer Freq, P( X =x) Ei=¿ np¿ (Oi−E i) 2

correct O i Ei M1
0 6 0.131687 13.16872 3.9024 A1- all correct
1 30 0.329218 32.92181 0.2593 for P( X =x)
M1
2 34 0.329218 32.92181 0.0353
3 23 0.164609 16.46091 A1-(all corr Ei )
4 5 0.041152 4.115226 20.988 3.8696 M1
A1(all corr
5 2 0.004115 0.411523
(Oi−E i)2
)
Ei

2 (O i−E i)2
❑ =∑ =8.0666 A1
Ei
If 2 > 7.815, reject H0 B1
Since 8.0666 > 7.815, H0 is rejected M1
There is significance evidence to show that the 3 possible answers A1
are not equally like to be chosen
8.(a) Since the sample size is large, by using the Central Limit Theorem
(CLT) , the sampling distribution of ps is approximately normal. B1 for CLT
43 43
ps N
43 n
, ( 1−
n ( )( ) ) B1
n n
8.(b) ps × qs
[ √ ]
ps ± Z α
2
n
= [ 0.0654 , 0.1066 ]

ps × qs
√
ps + Z α
2

ps × qs
n
=0.1066  Equation 1

√
ps Z α
2
n
=0.0654  Equation 2

( equation 1 ) +( equation 2 ) B1
2 p = 0.1066 + 0.0654
p = 0.086
43 M1
=0.086
n
 n = 500 A1
8.(c) pq
[ 0.086 ± Z α
2√ 0.086× 0.914
500 ]
= [ 0.0654 , 0.1066 ] B1 for
M1
√ n

0.086× 0.914
0.086+ Z α
2 500 √ =0.1066

0.086 × 0.914
Zα
2 √
Z α =1.643
500
=0.0206

B1
2
α B1
1− =0.9498
2
 = 0.1004
k % = ( 1−α ) 100 %
k % = ( 1−0.1004 ) 100 % M1
k % = ( 0.8996 ) 100 %
k % = 89.96 %
k = 89.96
 k  90 A1
8.(d) 0.086 × 0.914 B1 for 1.645
1.645
√ n
=0.007 B1 for 0.007
M1
1.6452 (
n= 0.086 ×0.914 )
0.0072
n = 4340.9059
Sample size = 4341 A1