H2 Mathematics (9758) Topic 23: Hypothesis Testing Tutorial Questions

Download as pdf or txt
Download as pdf or txt
You are on page 1of 19
At a glance
Powered by AI
Hypothesis testing is used to test claims about population parameters by analyzing sample data. The null and alternative hypotheses are formulated and a test statistic such as the z-score is computed to obtain a p-value which is compared to the significance level to determine whether to reject or fail to reject the null hypothesis.

There is insufficient evidence to conclude that the manufacturer's claim overstates the value of the mean length of string.

There is sufficient evidence to conclude that the machine produces overweight bags.

Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

H2 Mathematics (9758) Suggested


discussion
Topic 23: Hypothesis Testing
questions:
Tutorial Questions Q2, 6,7, 8, 12
Section 1: Practice Questions (Students are to attempt all questions on their
own)

1 The length of string in the balls of string made by a particular manufacturer has mean
 m and variance 27.4 m2. The manufacturer claims that   300 . A random sample of
100 balls of string is taken and the sample mean is found to be 299.2 m. Test whether
this provides significant evidence, at the 3% level, that the manufacturer's claim
overstates the value of  . 0.0632

Concepts/ Skills required


Formulate hypothesis.
Find p-value of 1-tail z-test.

Let X m be the length of string in a ball of string made by that manufacturer.


To test H0 :  = 300 vs H1 :  < 300
Perform a 1–tail test at 3% level of significance

 27.4 
Under H0 , X ~ N  300,  approximately, by Central Limit Theorem
 100 

Using Z-test, p  value  P( X  299.2)  0.063 216

Since p-value = 0.0632 > 0.03, H0 is not rejected at the 3% level of significance. So
there is insufficient evidence to conclude that the manufacturer's claim overstates the
value of 

2 A machine packs flour into bags. A random sample of eleven filled bags was taken and
the masses of the bags to the nearest 0.1 g were:
Tutorial

1506.8, 1506.6, 1506.7, 1507.2, 1506.9, 1506.8,


1506.6, 1507.0, 1507.5, 1506.3, 1506.4.
Testing

Obtain the mean and the variance of this sample. 1506.8, 0.109
Filled bags are supposed to have a mass of 1506.5 g. Assuming that the mass of a bag
has normal distribution with variance 0.16 g2, test whether the sample provides
Hypothesis

significant evidence at the 5% level that the machine produces overweight bags.
0.006 43
Explain what you understand by “5% level of significance” in the context of this
23:

question.
Topic

Teachers’ Copy Page 1 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

Concepts/ Skills required


Use GC to find mean and variance of data.
Formulate hypothesis.
Find p-value of 1-tail z-test.
Define “level of significance”.

Using GC,
Sample mean, x = 1506.8
and sample variance = 0.330 292 = 0.109
Let X g be the mass of a filled bag.
To test H0 :  = 1506.5 vs H1 :  > 1506.5 GC skills:
Perform a 1–tail test at 5% level of significance
[VARS] [5:Statistics] [4: x]
 0.16 
Under H0 , X ~ N 1506.5, 
 11 

Using Z-test, p-value = 0.0064 328


Since p-value = 0.006 43 < 0.05, H0 is rejected at the 5% level of significance. So there
is sufficient evidence to conclude that the machine produces overweight bags.
“5% level of significance” means there is a 5% chance that the test concludes that the
machine produces overweight bags when it actually does not.

3 Under usual conditions a certain type of melon seed produces melons with mean mass
1.85 kg and standard deviation 0.12 kg. A crop of these melons is grown under new
conditions and each melon in a random sample of 60 of the melons produced is weighed.
The mean mass of the sample is 1.877 kg.
(i) Assuming a normal distribution of masses, test at the 5% significance level whether
the data indicate that the mean mass of this crop differs from 1.85 kg. 0.081 4
(ii) If the test had been for an increase in the mean at the 5% significance level, what
would the result of the test have been? 0.040 7
Tutorial

(iii) State, giving a reason, whether it is necessary for the masses to have a normal
distribution for the test to be valid.
Concepts/ Skills required
Testing

Formulate hypothesis.
Find p-value of 2-tail and 1-tail z-tests.
Know when Central Limit Theore is necessary.
Hypothesis

(i) Let X kg be the mass of a melon which grows under new conditions and
23:
Topic

Teachers’ Copy Page 2 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

To Test H0 :  = 1.85 vs H1 :   1.85


Perform a 2–tail test at 5% level of significance

 0.122 
Under H0 , X ~ N 1.85, 
 60 

Using z-test, p-value = 0.081 361


Since p-value = 0.081 4 > 0.05, H0 is not rejected at the 5% level of significance.
So there is insufficient evidence to conclude that the mean mass of this crop differs
from 1.85 kg.
(ii) If the test had been for an increase in the mean,
To Test H0 :  = 1.85 vs H1 :  >1.85
1
Using z-test with n = 60, p-value = (0.081 361) = 0.040 681
2
Since p-value = 0.040 7< 0.05, H0 is rejected at the 5% level of significance. So
there is sufficient evidence that the mean mass of this crop has increased.
(iii) No, it is not necessary for the masses to have a normal distribution for the test to
be valid because the sample size (n = 60) is sufficiently large, so by Central Limit
Theorem, X is approximately normal.

4 Bottles of a particular brand of washing-up liquid are said to contain 500 ml. A random
sample of 50 bottles is taken and the volumes of liquid in the bottles are measured. The
volumes, x ml, are summarised by
 ( x  500)  35.8 and  ( x  500)  150.5 .
2

(i) Find unbiased estimates of the population mean and variance. 499, 2.55
(ii) Assuming a normal distribution, test at the 5% significance level whether the
population mean volume is less than 500 ml. 0.000 758
(iii) State, giving a reason, whether it is necessary to assume a normal distribution for
the test to be valid.
Tutorial

Concepts/ Skills required


Find unbiased estimates of population mean and variance from ∑ (x – h) and ∑ (x – h)2.
Formulate hypothesis.
Testing

Find p-value of 1-tail z-tests.


Know when Central Limit Theore is necessary.
Hypothesis

(i) Let y  x  500 , we get  y  35.8 ,  y 2


 150.5 ,
23:
Topic

Teachers’ Copy Page 3 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

Unbiased estimate of the population mean = x  y  500 


 y  500
n
35.8
  500  499.284  499 (3sf)
50

Unbiased estimate of the population variance = sx2 = s y2

1    y  1 
2
 35.8  
2


n 1 
 y  n   49 150.5  50   2.548310204  2.55 (3sf)
2

   

(ii) Let  ml be the actual population mean volume of liquid per bottle.

To test H0 :   500 vs H1 :   500

Perform a 1–tail test at 5% level of significance

 2.5483 
Under H0 , X ~ N  500, 
 50 

Using z-test, pvalue = 0.000 758 20.

Since p-value = 0.000 758 < 0.05, H0 is rejected at the 5% level of significance. So
there is sufficient evidence to show that the population mean volume is less than
500 ml.

(iii) No, it is not necessary to assume a normal distribution for the test to be valid.

Since n = 50 is large, by Central Limit Theorem, X is approximately normal. Tutorial

5 A company sells balls of yarn. A manager claims that the average length of yarn is at
least 300 m. To test this claim, a random sample of 100 balls of yarn is checked and the
lengths of yarn per ball, x m, are summarised by
 ( x  300)  60 and  ( x  300)  1240 .
Testing

1204
(i) Find unbiased estimates of the population mean and variance. 299.4,
99
Hypothesis

(ii) Test at the 5% significance level whether the manager’s claim is valid. 0.042 7
Concepts/ Skills required
Find unbiased estimates of population mean and variance from ∑ (x – h) and ∑ (x – h)2.
23:
Topic

Teachers’ Copy Page 4 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

Formulate hypothesis.
Find p-value of 1-tail z-tests.

(i) Let u  x  300 . Then  u  60,  u 2  1240


60
Unbiased estimate of population mean is x  u  300   300  299.4
100
1   60   1204
2

Unbiased estimate of population variance is sx  su  1240 


2 2

99  100  99
 
(ii) Let X be the length of yarn in a ball.
To test H 0 :   300 vs H1 :   300 [the manager’s claim is not valid]

Perform a 1–tail test at 5% level of significance

 1204 
Under H0 , X ~ N  300, 99  approximately, by Central Limit Theorem
 100 
 
Using z-test, p-value = 0.042 670

Since p-value 0.042 7 < 0.05, H 0 is rejected at 5% level of significance. So there is


sufficient evidence at that the manager’s claim is not valid.

6 A random sample of 90 batteries, used in a particular model of mobile phone, is tested


and the standby-time, in hours, is measured. The sample mean and sample standard
deviation of the standby-time are found to be 33.787 hours and 12.035 hours
respectively. Test, at the 2% significance level, whether the mean standby-time is less
than 36.0 hours. 0.041 4
This type of battery is advertised as having a ‘talk-time’ of less than 5 hours. In a test at
the 5% significance level, it is found that there is significant evidence that the population
mean talk-time is less than 5 hours. Using only this information, and giving a reason in
each case, state whether each of the following statements is (i) necessarily true, (ii)
necessarily false, or (iii) neither necessarily true nor necessarily false.
Tutorial

(a) There is significant evidence at the 10% significance level that the population mean
talk-time is less than 5 hours. (i)
Testing

(b) There is significant evidence at the 5% significance level that the population mean
talk-time is not 5 hours. (ii)
Concepts/ Skills required
Hypothesis

Formulate hypothesis.
Find p-value of 1-tail z-tests.
Know the effect of changing the significance level.
Know the effect of changing the alternative hypothesis.
23:
Topic

Teachers’ Copy Page 5 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

n 90
Unbiased estimate of population variance, s 2  (Sample Variance)  (12.0352 )
n 1 89
Let X h denote the standby time of a randomly chosen battery.
To test H0 :   36.0 vs H1 :   36.0

Perform a 1–tail test at 2% level of significance

 90 
 (12.0352 ) 
Under H0 , X ~ N  36, 89  approximately, by Central Limit Theorem
 90 
 

Using z-test, pvalue = 0.041 395.


Since pvalue = 0.041 4 > 0.02, H0. is rejected at the 1% level of significance. Hence, the
insufficient evidence at 1% level of significance to show that the mean standby-time is
than 36.0 h.

Let p1 be the p-value of the following one-tail test:

H0 :  = 5
H1 :  < 5
Given that H0 is rejected at the 5% level of
significance,
we have p1  0.05

(i) Clearly, p1  0.10 Note that the critical values (or


calculated values) are not
 H0 is also rejected at the 10% level of
changed because there was no
significance
change made to the null
and (i) is necessarily true. hypothesis and the same
sample was used throughout.

(ii) Let p2 be the p-value of the following two-tail test:

H0 :  = 5
Tutorial

H1 :   5

Clearly p2 = 2 p1
Testing

So p2  0.10 but we do not know if p2  0.05 or 0.05  p2  0.10

 (ii) is neither necessarily true nor necessarily false.


Hypothesis

[Further clarifications:
23:
Topic

Teachers’ Copy Page 6 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

Case (1) if p1 < 0.025  2 p1 < 0.05  p2 < 0.05


H0 is now rejected at the 5% level of significance
Case (2) if 0.025 < p1 < 0.05  0.05 < 2 p1 < 0.10  0.05 < p2 < 0.10
H0 is now not rejected at the 5% level of significance]

7 From the claims received by the company, over a long period of time, a random sample
of 120 is taken. The values of the claims, $x, are summarised by
2
 ( x  1000)  5320 ,  ( x  1000)  8 282 000 .

(i) Find, correct to 2 decimal places, the unbiased estimates of the population mean
and variance. 1044.33, 67 614.68
(ii) What do you understand by the term ‘unbiased estimate’?
(iii) The population mean is denoted by $  . Using the sample data, a significance test
of the null hypothesis   1000 against the alternative hypothesis   1000 is
carried out at the  % level of significance. Find the set of values of  for which
the null hypothesis will be rejected.    :   6.18
Concepts/ Skills required
Find unbiased estimates of population mean and variance from ∑ (x – h) and ∑ (x – h)2.
Define “unbiased estimate”.
Find range of values of α from critical region of 2-tail z-tests.

(i) Let y  x  1000 , we get  y  5320 ,  y 2


 8282000

Unbiased estimate of the population mean,

x  y  1000    1000
y
n
5320 1
  1000  1044  1044.33 (2dp)
120 3
Unbiased estimate of the population variance
Tutorial

1    y 
2

 y  n 
2
s s  2
y
2

n 1 
Testing

 

1   5320  
2

 8282000    67614.67787  67614.68 (2dp)


Hypothesis

119  120 
23:
Topic

Teachers’ Copy Page 7 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

(ii) An unbiased estimate is an estimate of an unknown population parameter  (eg


mean, variance) where the expected value of the estimator = .

(iii) Let X be the claim received by the company


To test H0 :   1000 vs H1 :   1000

Perform a 2–tail test at  % level of significance

 67614.68 
Under H0 , X ~ N 1000,  approximately, by Central Limit Theorem
 120 

Using z-test, pvalue = 0.061 807 7



For the null hypothesis to be rejected, 0.0618077 
100

Set of values of  is    :   6.18

8 Mass-produced washers have thicknesses which are normally distributed with mean
3 mm and standard deviation 0.2 mm. During a check on the manufacturing process a
random sample of 25 washers is taken from production and the mean thickness mm is
calculated. Find the interval in which the value of must lie in order that the hypothesis
that the production mean thickness is 3 mm will not be rejected in favour of the
hypothesis that the mean thickness is not 3 mm at the 5% level of significance.
2.92 < x < 3.08
Concepts/ Skills required
Formulate hypothesis.
Find range of values of x from critical region of 2-tail z-tests.

Let X be the thickness of washer from production.


To test H0 :  = 3 vs H1 :   3
Perform a 2–tail test at 5% level of significance
Tutorial

 0.22 
Under H0 , X ~ N  3, 
 25 
Testing

Region at 5% level of significance is:


| Z | 1.95996
Hypothesis

Using z-test 0
-1.95996 1.95996
23:
Topic

Teachers’ Copy Page 8 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

x 3
z -value 
0.2 
 25 x  3 
5
For H0 not to be rejected,

 
1.95996  25 x  3  1.95996

2.9216  x  3.07839
2.92  x  3.08

9 Spec 15/II/10 (adapted)


The average discharge time of a new rechargeable battery is 13 hours. An alternative
manufacturing process is trialled, and the discharge time, t hours, of each of 50 randomly
chosen new batteries using the alternative process is recorded. The results are
summarised as follows.
n = 50 ∑t = 656.9 ∑t 2 = 8648.64
The Production Manager wishes to test whether the the average discharge time is
different using the alternative process by carrying out a hypothesis test.
(i) Explain whether the Production Manager should use a 1-tail test or a 2-tail test.
(ii) Explain why the Production Manager is able to carry out a hypothesis test without
knowing anything about the distribution of the discharge times of the new batteries.
(iii) Find unbiased estimates of the population mean and variance and carry out the test
at the 10% level of significance for the Production Manager. 13.1, 0.373, 0.110
(iv) Suggest a reason why the Production Manager might be prepared to use an
alternatie process that gives a shorter average discharge time than the original
process.
The Marketing Manager wishes to test whether the average discharge time is longer
using the alternative process. The Marketing Manager finds the average discharge time
of 40 randomly chosen new batteries using the alternative process is 13.2 hours. He
carries out a hypothesis test at the 10% level of significance.
Tutorial

(v) Explain, with justification, how the population variance of the discharge times will
affect the conclusion made by the Marketing Manager. σ < 0.987
Testing

Concepts/ Skills required


Decide between 1-tail and 2-tail tests.
Know when Central Limit Theorem is necessary.
Hypothesis

Find unbiased estimates of population mean and variance.


Formulate hypothesis.
Find p-value using 2-tail z-test.
23:
Topic

Teachers’ Copy Page 9 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

Give contextual reasons for preference of a certain mean value.


Find range of values of σ from critical region of 1-tail z-tests.

(i) A 2-tail test because the Production Manager wants to test whether there is a
change in the average discharge time using the alternative process.
(ii) Central Limit Theorem states that the sample means will have a normal
distribution when the sample size is sufficiently large.
656.9
(iii) unbiased estimate of population mean, t   13.138  13.1
50
unbiased estimate of population variance,
1  656.82 
s2  8648.64    0.37322  0.373
49  50 

Let X be the discharge time of a new battery using the alternative process.

To test H0: µ = 13 vs H1: µ ≠ 13


Perform a 2–tail test at 10% level of significance
 0.37322 
Under H0 , X ~ N 13,  approximately, by Central Limit Theorem
 50 

Using z-test, , p-value = 0.110 20


Since p-value 0.110 > 0.10, H0 is not rejected at the 10% level of significance. So
there is insufficient evidence that the average discharge time of the new batteries
produced using the alternative process is different.
(iv) The batteries produced may be of better quality OR may be safer OR may
require fewer resources.
(v) To test H0: µ = 13 vs H1: µ > 13
Perform a 1–tail test at 10% level of significance
 2 
Under H0 , X ~ N 13,  approximately, by Central Limit Theorem
 40 
Critical Region: {Z > 1.2816}
To reject H0,
Tutorial

13.2  13
z-value = 
 1.2816
40
Testing

0.2
 40  0.98702
1.2816
So the Marketing Manager will be able to conclude that the average discharge
Hypothesis

time is longer if the population variance is less than 0.9872.


23:
Topic

Teachers’ Copy Page 10 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

10 A consumer association is testing the lifetime of a particular type of battery that is


claimed to have a lifetime of 150 hours. A random sample of 70 batteries of this type is
tested and the lifetime, x hours, of each battery is measured. The results are summarised
 x  10317 ,  x  1540 231 .
2
by

The population mean lifetime is denoted by  hours. The null hypothesis   150 is to
be tested against the alternative hypothesis   150 . Find the p-value of the test and state
the meaning of this p-value in the context of the question. 0.0975
A second random sample of 50 batteries of this type is tested and the lifetime, y hours,
of each battery is measured, with results summarised by
 y  7331 ,  y  1100565 .
2

Combining the two samples into a single sample, carry out a test, at the 10% significance
level of the same null and alternative hypotheses. 0.0499
Concepts/ Skills required
Find unbiased estimates of population mean and variance.
Find p-value using 1-tail z-test.
Find unbiased estimates of population mean and variance from combined data.

10317 1   x   1 
2
10317 2 
x  x 
2
 147.38571 s =2
 =  1540231    284.82008
70 n 1  n  69  70 
 

H0 :  = 150
H1 :  < 150
Perform a 1–tail test.
X  150
Under H0 ,Test statistic: Z  approximately, by Central Limit Theorem
S/ n

Using z-test with n = 70, p-value = 0.0975


The p-value is the probability that the mean battery lifetime of a sample is less than
147.39 h when the actual population mean battery lifetime is 150 h.
Tutorial

OR
9.75% is the smallest probability that we wrongly support the claim that the mean
lifetime of the battery is less than 150 h.
Testing

Let W h be the lifetime of the battery in the combined sample.


Hypothesis

n  70  50  120 ,  w   x   y  10317  7331  17648


2 2 2
 w   x   y  1540231  1100565  2640796
23:
Topic

Teachers’ Copy Page 11 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

17648 1  176482 
w  147.06667 ; s 
2
w  2640796  120   381.2056
120 119  
H0:   150

H1:   150

Perform a 1–tail test at 10% level of significance


W  150
Under H0 , Test statistic: Z  approximately, by Central Limit Theorem
Sw
n
Using z-test with n = 120, p -value  0.0499

Since p-value = 0.0499 < 0.10, H0 is rejected at the 10% level of significance. To there
is sufficient evidence that the mean lifetime of the batteries is less than 150 hours.

11 Bags of salted cashew nuts display net contents 100 g. The manufacturer knows that the
standard deviation of the population is 1.6 g.
A customer claimed that the bags have been lighter in recent purchases, so the factory
quality control manager decided to investigate. He sampled n bags and found that their
mean weight was 99.4 g. He performed a 1% significance test and concluded that the
customer’s claim was not valid.
Find the largest possible value of n. 38
Concepts/ Skills required
Find range of values of n.

Let μ g be the actual mean weight of a bag of salted cashew nuts.

H0 :  = 100
H1 :  < 100
Perform a 1–tail test at 1% level of significance
X  100
Under Ho, Test statistic: Z  1.6
Tutorial

Crictical region: Z < −2.3263


Testing

For H0 to be not rejected,


z  2.3263
Hypothesis

99.4  100
1.6
 2.3263
n
23:
Topic

Teachers’ Copy Page 12 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

n  6.2036
n  38.484
Largest n is 38

12 An alpaca breeder wants to produce fleece which is extremely fine. In 2008, his herd
had mean fineness 22.3 microns with standard deviation σ microns. The standard
deviation remains relatively constant over time. In 2012, a sample of 80 alpacas from
the herd was randomly selected, and the mean fineness was 21.2 microns.
Find the range of values of the standard deviation which will indicate a significant
change in mean fineness at the 5% two-tailed significance test. 2.89
Concepts/ Skills required
Find range of values of σ.

Let μ microns be the actual mean fineness of fleece in 2012 and σ micron be its standard
deviation.

H0 :  = 22.3
H1 :  ≠ 22.3

Perform a 2–tail test at 1% level of significance


X  22.3
Under Ho,Test statistic: Z  
n

Crictical region: |Z| > 1.95996


For H0 to be rejected,
| z | 1.95996

21.2  22.3

 1.95996
80
Tutorial

|  | 5.0198

Since   0 ,   5.02
Testing

Section 2: Supplementary Questions (For students to practice after going


Hypothesis

through tutorial for extra practice)

13 A farmer grows a new variety of potato. A random sample of eighty roots of the new
variety is taken and the crop, x kg per root, is measured. It is found that the mean value
23:
Topic

Teachers’ Copy Page 13 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

of x is 3.20. Based on the publicity for the new variety and his knowledge of his land, the
farmer had thought that the mean crop would be 3.30 kg per root. Assuming that the
population standard deviation is 0.50 kg per root, test, at the 4% level, whether the sample
provides significant evidence that the farmer’s hypothesis about the crop from the new
variety is incorrect. 0.0736

Let X kg be the mass of a potato of the new variety,


and  kg be the actual population mean mass of a potato of the new variety.
H0 :   3.3

H1 :   3.30 (2-tail test)

Perform a 2–tail test at 4% level of significance


X  3.3
Under Ho Test statistic: Z .
0.5
n

Using z-test with n = 80, pvalue = 0.073 638


Since pvalue = 0.0736  0.04 , H0 is not rejected at the 4%
level of significance. So there is insufficient evidence to show that the farmer’s
hypothesis about the crop from the new variety is incorrect.

14 A supermarket manager investigated the lengths of time that customers spent shopping
in the store. The time, in minutes, spent by each of a random sample of 150 customers
was measured, and the sample mean and variance of the time spent were found to be
19.140 minutes and 33.854 squared minutes respectively. Test at the 5% level of
significance, the hypothesis that the mean time spent shopping by the customers is
20 minutes, against the alternative that it is less than this. 0.0356

Let X min be the length of time that customer spent shopping in the store
and  min be the actual population mean length of time that customer spent shopping in the
store.

Unbiased estimate of population variance, p 1


Tutorial

150
 (33.854)  34.08 (2dp)
149
Testing

H0 :   20

H1 :   20 (1-tail test)
Hypothesis

Level of Significance: 5 %
23:
Topic

Teachers’ Copy Page 14 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

X  20
Test statistic: Z 
S
n

Using z-test with n = 150, pvalue = 0.035 599


Since pvalue = 0.0356 < 0.05, H0 is rejected at the 5% level of significance. So there is
sufficient evidence to show that the mean time spent shopping by the customers is less than
20 minutes.

15 A horticulturist takes a random sample of 80 bean seeds of a particular variety, and


sows them under standard laboratory conditions. The total crop weight, x kg, of each of
the 64 plants is measured, with the following summarized results:
 x = 303.4,  x2 = 1615.96.
Find unbiased estimates of the population mean and variance of crop weight per plant.
The horticulturist wishes to test the hypothesis that the mean crop weight per plant is 5 kg
against the alternative hypothesis that the mean crop weight is less than 5 kg. Carry out
the test at the 10% level of significance. Find the largest level of significance for which
the test would not result in rejection of the null hypothesis. 4.74, 2.82, 10.8%

Unbiased estimate of population mean, x 


 x  303.4  4.740625  4.74 (3sf)
n 64

1    x 
2

Unbiased estimate of population variance, s 2

n 1 
 x  n 
2

 

1  303.42 
 1615.96 
63  64 

 2.819910714  2.82 (3sf)

Let  kg be the actual population mean crop weight.

H0 :   5
Tutorial

H1 :   5 (1-tail test)

Perform a 1–tail test at 10% level of significance


Testing

X 5
Under Ho, Test statistic: Z
S
n
Hypothesis

Using z-test with n = 64, pvalue = 0.010 829


Since pvalue = 0.108 > 0.10, H0 is not rejected at the 10% level of significance. So there is
insufficient evidence to show that the mean crop weight is less than 5kg
23:
Topic

Teachers’ Copy Page 15 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

In order not to reject H0, largest level of significance = 10.8% (3sf)

16 A manufacturer claims that the average life of his electric light bulbs is 2000 hours. A
random sample of 64 bulbs is tested and the life x (in hours) recorded. The results
obtained are as follows:
 x  127 808 ,  ( x  x )  9 694.6 .
2

Is there sufficient evidence, at the 2% level of significance, that the manufacturer is over-
estimating the length of life of his light bulbs? 0.0265

Unbiased estimate of the population mean, x 


 x  127808  1997
n 64
1
 
2
Unbiased estimate of the population variance, s 2 
n 1
 xx
1
 9 694.6  153.8825397  154
63
Let  h be the actual population mean life of light bulbs.

H0 :   2000

H1 :   2000 (1-tail test)

Perform a 1–tail test at 2% level of significance


X  2000
Under Ho, Test statistic: Z
S
n

Using z-test with n = 64, pvalue = 0.026513


Since pvalue = 0.0265  0.02 , H0 is not rejected at the 2% level of significance. So there
is insufficient evidence to show that the manufacturer is over-estimating the length of life of
his light bulbs. Tutorial

17 N07/II/7
A large number of students in a college have completed a geography project. The time,
x hours, taken by a student to complete the project is noted for a random sample of 150
Testing

students. The results are summarised by


 x  4626 ,  x  147691 .
2
Hypothesis

Find, correct to 2 decimal places, unbiased estimates of the population mean and
variance. 30.84, 33.73
Test, at the 5% significance level, whether the population mean time for a student to
complete the project exceeds 30 hours. 0.0382
23:
Topic

Teachers’ Copy Page 16 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

State, giving a reason, whether any assumptions about the population are needed in order
for the test to be valid.

Unbiased estimate of the population mean, x 


 x  4626  30.84 (2dp)
n 150

1    x 
2

Unbiased estimate of the population variance, s 2

n 1 
 x  n 
2

 

1  46262 
 147691   33.72590604  33.73 (2dp)
149  150 

Let  hours be the actual population mean time taken for a student to complete the project.

H0 :   30

H1 :   30 (1-tail test)

Perform a 1–tail test at 5% level of significance


X  30
Under Ho,Test statistic: Z
S
n

Using z-test with n = 150, pvalue = 0.038 238


Since pvalue = 0.0382 < 0.05, H0 is rejected at the 5% level of
significance. So there is sufficient evidence to show that the
population mean time for a student to complete the project
exceeds 30 hours.
No assumptions about the population are needed in order for the test to be valid.

X  30
[Since  2 is unknown and n = 150 is large, by Central Limit Theorem, ~ N  0,1
S
n
approximately.]
Tutorial

18 Salt is packed in bags which the manufacturer claims contains 25 kg each, on average.
A random sample of 80 bags is examined and the mass, x kg, of the contents of each
Testing

bag is determined. It is found that


 ( x  25)  27.2 and  ( x  25)  85.1 .
2
Hypothesis

(i) Estimate the population mean and variance of the mass of contents of a bag.
25.3, 0.960
(ii) Test, at the 10% level of significance, whether the manufacturer is understating the
average mass of the contents of a bag.
23:
Topic

Teachers’ Copy Page 17 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

Let X kg be the mass of salt in a bag and


 kg be the actual population mean mass of salt in a bag.

Let t = x  25, we have  t  27.2 and  t 2  85.1 .

27.2
(i) Unbiased estimate of population mean, x = t  25   25  25.34
80

1  27.7 2 
Unbiased estimate of population variance, sx  st = 85.1    0.96015
2 2

79  80 

(ii) H0 :  = 25

H1 :  > 25 (The manufacturer has understated the average mass)


Perform a 1–tail test at 10% level of significance
X  25
Under Ho, Test statistic: Z 
S/ n
Using z-test with n = 80, p-value = 0.000 956 25
Since p-value = 0.000 956 < 0.10, H0 is rejected at the 10% level of significance. So
there is sufficient evidence that the manufacturer has understated the average mass of
contents of a bag.

19 N12/2/6
On a remote island a zoologist measures the tail lengths of a random sample of 20
squirrels. In a species of squirrels known to her, the tail lengths have mean 14.0 cm. She
carries out a test, at the 5% significance level, of whether squirrels on the island have the
same mean tail length as the species known to her. She assumes that the tail lengths of
squirrels on the island are normally distributed with standard deviation 3.8 cm.
(i) State appropriate hypotheses for the test.
The sample mean tail length is denoted by x cm.
(ii) Use an algebraic method to calculate the set of values of x for which the null
hypothesis would not be rejected. (Answers obtained by trial and improvement
from a calculator will obtain no marks.) 12.3< <15.7
Tutorial

(iii) State the conclusion of the test in the case where x  15.8 .
Let X cm be the tail length of a squirrel on the island.
Testing

(i) Let  cm be the mean tail length of a squirrel on the island.


H0 :  = 14.0
Hypothesis

H1 :   14.0
23:
Topic

Teachers’ Copy Page 18 of 19


Eunoia Junior College 2018 JC2 H2 Mathematics Tutorial

(ii) Perform a 2–tail test at 5% level of significance


X  14
Under Ho Test statistic: Z 
3.8 / n

Critical region: |Z| > 1.959 96


x  14.0
For H0 to be not rejected, 1.95996   1.95996
3.8 / 20
 12.335  x  15.665
 12.3  x  15.7 (3 s f)

(iii) From (ii) result, for the case when x = 15.8, H0 is rejected at the 5% level of
significance. So there is sufficient evidence to conclude that the squirrels on
the island do not have the same mean tail length as the species known to her.

Tutorial
Testing
Hypothesis
23:
Topic

Teachers’ Copy Page 19 of 19

You might also like