CMM Subject Support Strand: STATISTICS Unit 4 Sampling: Text

STRAND: STATISTICS
Unit 4 Sampling

TEXT

Contents

Section

4.1 Questionnaires and Surveys

4.2 Sampling
 CMM Subject Support Strand: STATISTICS Unit 4 Sampling: Text

4 Sampling
4.1 Questionnaires and Surveys
The UK Government's census, first undertaken in 1801 and with the exception of 1941,
continuing to be held every ten years, is a vast and costly exercise but is important for
policy and planning purposes. The most recent UK census took place on 27th March
2011; data is available from the Office for National Statistic (ONS) at
[Link]
There have been suggestions that a sample (that is, a subgroup of people) rather than a
full census (everyone) might be a more efficient method of collecting the data.
In either case though, as for any worthwhile survey, the questions asked must be clear,
unambiguous and unbiased. This is the theme of this section.

Note
When designing a questionnaire to use as part of a survey, bear in mind the following
guidelines.
• The questionnaire should try to find out the information you need.
• You should know how you are going to collect your responses.
• Questions should be clear and concise with no ambiguity.
• Do not ask for information you do not need.
• Allow for all possible responses.
• Questionnaires should be fair and not biased in any way.
• The people asked to complete the questionnaire should be from a variety of
backgrounds.
It should also be noted that if you need to design a questionnaire for your project work, it
is strongly recommended that you first try it out with a small number of people. This is
called a pilot survey.

Questions can be designed in two distinct ways, namely those that require a specified
response to be chosen from a number of options or by giving a number, and those that
allow more detailed responses. The first group are often referred to as closed questions;
the second as open questions. Here are some examples.

Closed

"Did you watch the football match on TV last night?" YES / NO

"How many hours of TV did you watch last night?"

Choose from ' 0 - 1, 1 - 2, 2 - 3, 3 - 4, 4 - 5, more than 5 '

1
4.1 CMM Subject Support Strand: STATISTICS Unit 4 Sampling: Text

The second example of a closed question shows an example with multiple responses – by
which we mean more than 2! So 'YES / NO' is not regarded as a multiple response.
There is though a problem with these responses as there is potential overlap. i.e. in which
category do you place 1 hour or 2 hours, etc.? It should be made clear, i.e. 0 – 1 includes
everything up to and including 1, etc.

Open

"What sort of TV programmes do you like to watch, and why?"

"Where did you go for your holiday last year?"

The second example of open questions is good for a face-to-face discussion, as it allows a
wide range of answers. The questioner will then be able to react to these answers, and
continue the discussion. This example is bad for a questionnaire for the same reason!
Any answer is possible but the writer of the questionnaire does not have the chance to
follow up answers. It could though be modified and become a closed question with
multiple responses: for example,

"Please tick a box to show where you went on holiday last year."

UK USA France Italy

Greece Other (please say where)

Worked Example 1
Consider this questionnaire:
Are you concerned about the environment?
Are you concerned about the level of pollution caused by cars?
Do you think the health of young children is at risk due to exhaust fumes from
cars?
Is there too much congestion in the city centre?
Is public transport under-used?
Do you think cars should be banned from the city centre?
Comment on the questions given here.

Solution
This questionnaire is very biased and has been designed to lead people towards answering
'yes' to the last question. There are no questions about the advantages of cars, the
problems of using public transport or other ways of reducing pollution.

2
 CMM Subject Support Strand: STATISTICS Unit 4 Sampling: Text
4.1

Worked Example 2
What is wrong with the following survey, used to find the favourite sports of the students
in one school?

Which is your Tennis

favourite sport?
Rugby
Netball
Basketball
Swimming
Snooker
Other

Solution
The question used in this survey does not cover all the possible responses well. Many
students might have, for example, football as their favourite sport and so would tick the
box marked 'Other'. This would give unhelpful results. Also, students are asked to make
only one choice although they may well like more than one sport.
The following wording would produce more useful responses.

"Please tick one box in each row to show your preference."

I really enjoy

I've no great

I dislike this
preference

I hate this
this sport
I like this

sport

sport
sport

Tennis

Rugby

Netball

Basketball

Swimming

Snooker
Please put any
other sport that
you like here

3
4.1 CMM Subject Support Strand: STATISTICS Unit 4 Sampling: Text

Exercises
1. Consider the following two sets of questions.

Is meat a good source of protein? Is all meat obtained from dead

animals?
Is protein an important part of Do you like killing animals?
your diet?
Do you like eating meat? Do you like eating meat?

(a) Comment on the way that the last question might be answered in each case.
(b) Write a better set of questions to find out if people like eating meat.

2. Design a questionnaire that would encourage students at a school to say:

(a) that school uniform is a good policy,
(b) that school uniform should be abandoned.

3. Wendy asks her class the following question.

Which of these breakfast cereals do you like best?

Cornflakes Porridge Bran Flakes

She then says that her class's favourite breakfast cereal is Cornflakes.
(a) Is her conclusion valid?
(b) Criticise her question.
(c) Write a better question for her to use.

4. Design a questionnaire which you could use to find out:

(a) if people think they have a good public transport service,
(b) whether the music tastes of girls and boys are different,
(c) whether younger people are more likely than older people to be vegetarian,
(d) who might win the next General Election.
Use your questionnaire to collect some data and present your results.

5. You have been asked to estimate what percentage of all cars on the road have each
registration letter (M, N, etc.).
(a) Describe how you would collect the data for a survey to answer this
problem.
(b) Describe any problems that might arise as you collect data.

4
 CMM Subject Support Strand: STATISTICS Unit 4 Sampling: Text
4.1
6. Akiko did a survey to find out which colours of cars are popular.
She found that red was the most popular colour. She did not find any cars that were
purple or pink.
Think about a survey which you could do. It must not be about the colour of cars.
(a) What is your survey about?
(b) Write down two things which you might expect to find out.

7. In a survey of community life on a new housing estate the following question is

suggested.
(A) "What do you most like about living here?"
An alternative is proposed.
(B) "Tick the box which describes why you most like living here."
Design of houses
Friendliness of neighbours
More open space

Give one advantage of each form of question.

8. A survey is done to find out in which sports pupils take part.

The results of the survey are shown below.

% Boys % Girls

Athletics 37.1 33.3

Basketball 12.7 4.5
Cycling 71.7 60.5
Dancing 4.9 31.5
Football 72.0 14.5
Gymnastics 5.0 16.3
Hockey 6.5 3.5
Judo 16.0 6.7
Netball 3.3 35.0
Roller-skating 7.3 26.0
Rugby 11.8 2.5
Swimming 53.1 55.0

(a) Which two sports are the most popular with both boys and girls?
(b) The question asked in the survey was:
Which sports do you take part in?
You want to find out more about pupils' involvement in sport.
Write down another question that you could ask.

5
 CMM Subject Support Strand: STATISTICS Unit 4 Sampling: Text
4.1
9. The school governors are worried about road safety outside the school gates.
They think there ought to be a Pelican Crossing.
To decide on this, they need to find out how much traffic comes along the road at
different times of day and on different days of the school week.
Design an observation sheet to gather this information.

10. Mee Ling thinks that pupils who come to school by bus are more likely to be late
than those who do not travel by bus.
In order to test whether or not this is true, she carries out a survey on 100 pupils,
from years 7 and 8, for 5 consecutive Tuesdays.
The results are shown in the following table.

METHOD OF NUMBER OF NUMBER OF

TRAVEL STUDENT-DAYS LATES

Bus 150 40
Cycle 50 10
Car 100 22
Walk 200 25

TOTAL 500 97

(a) Do the results show that Mee Ling is correct?

Show the working on which you base your answer.
(b) Suggest 3 ways in which Mee Ling could have improved her survey.
(c) A pupil is selected at random from Year 7.
Mee Ling stated:
"The table shows that the probability that this pupil walks to school
on Tuesdays is 25 ."
Would you describe this statement as
correct, about right, or wrong?
Choose one of the three alternatives and give reasons for your choice.

11. (a) A headline in a newspaper this year stated:

Students skip Breakfast

Our survey shows that few students
are eating cereals, fruit, or bread
for breakfast.
In fact they are eating nothing at all!

You are asked to conduct a survey to find out what students eat for
breakfast.
Design an observation sheet to collect the data you need.
Invent the first 20 entries on your data sheet.

6
4.1 CMM Subject Support Strand: STATISTICS Unit 4 Sampling: Text

(b) The newspaper made the following statement about the eating habits of
teenagers.
Only one in a hundred teenagers eats fruit
and vegetables each day. Over half eat no
vegetables other than chips.

You are asked to find out whether this statement is true in your area.
Give three questions you could ask teenagers to see if what the article says is
true in your area.

12. A bus company attempted to estimate the number of people who travel on local
buses in a certain town. They telephoned 100 people in the town one evening and
asked 'Have you travelled by bus in the last week?'
Nineteen people said 'Yes'. The bus company concluded that 19% of the town's
population travel on local buses.
Give 3 criticisms of this method of estimation.

4.2 Sampling
In 1975 the Government wanted census information before the 1981 full census, so a ten
per cent census was carried out using 1 in 10 of the population. This is known as a survey
or sample. Data are obtained by asking people to fill in forms which are then given to
collectors trained to sort out any queries.
When deciding how to carry out a data collection there are several decisions to be made:
(a) What size of sample can you reasonably expect to take, given limited time, money
and resources?
(b) How are items to be used in the sample to be chosen to avoid introducing bias?
(c) How are the data to be collected to avoid bias?
The answer to question (a) clearly depends on the individual circumstances. It should be
obvious, however, that the larger the sample the more sensitive the result.
In questions (b) and (c) the key element is to eliminate possible bias. For example, if you
are a political pollster wanting to know the state of the parties, then a sample taken only
in an affluent, expensive out-of-town suburb will be biased just as a sample taken in a
run-down inner city area will be. So the sample must not be biased – but the same also
applies to the questions. Questions such as
"Should we increase spending on education?"
are biased as they do not represent the complete picture. If we spend more on education,
this means that we must spend less on something else. A less biased question would be,
"Should we spend more on education and less on other services?"

Here, we will be looking for sampling methods that lead to a representative sample, by
which we mean that the sample has similar properties to the total population. As we said
before, here we use the word 'population' not just for human or animal populations but to
represent the complete group under study, e.g. all schools in the UK, all cars in England,
etc.

7
 CMM Subject Support Strand: STATISTICS Unit 4 Sampling: Text
4.2
The main methods used for sampling in practice are described below. First, we create a
list of all members of the population; this is called the sampling frame.
(a) Random – to be truly random each individual must have an equal chance of being
chosen. This method is often used for selecting people from Electoral Registers. If
the researcher is calling at people's houses the system must be rigidly adhered to
(i.e. call back if people are out). It does not necessarily ensure a representative
sample.
(b) Systematic – taking items at regular intervals, e.g. every 4th tree when sampling in
a forest. Although this does not necessarily ensure a representative sample it
should be better than random sampling. Again the system must be rigidly adhered
to. This method is often used when sampling goods on a production line.
(c) Stratified – this is used to ensure that the sample is representative and that it has
the same proportions as the population. To do this you first need to divide the
whole of the population into appropriate categories and then, inside each category,
choose a random sample of appropriate size. This can be difficult in practice.
What is commonly used in street surveys is a quota sampling method where
interviewers are simply asked to interview a certain proportion of each category,
e.g. age groups, and these can be chosen at random.
(d) Cluster – here, because of the large size of a population, particular regions are
drawn at random, and then an appropriate number of individuals are drawn at
random from each chosen region.
In the following example you will see how these methods are used in practice.

Worked Example 1
There are 12 teachers in a small school. A sample of
1. Mrs Skirton
size 4 is to be selected from this population.
2. Mrs Pearson
(a) Create a systematic sample for the population. 3. Mrs Green
(b) Create a random sample for the population. 4. Mrs Crocker
5. Mrs Barnes
Solution 6. Mr Vinner
7. Mr Krishnan
(a) As there are 12 teachers, a systematic sample
8. Mr Gadd
could be made by selecting every third
teacher from the list. This would create a 9. Mr Hampson
sample containing: 10. Mr Thompson
11. Mr Mobey
Mrs Green, Mr Vinner, Mr Hampson and Mr Grainger.
12. Mr Grainger
80 02 86 03
23 90 18 56
84 34 73 51
(b) The list of random numbers shown here has been
31 98 73 75
generated using a spreadsheet. The numbers have
28 73 32 83 been arranged to give two digit numbers. Numbers
52 70 53 23 greater than 12 are not needed in this case so these
47 14 69 68 numbers can be deleted.
15 45 05 18
59 79 80 51
75 11 01 10

8
4.2 CMM Subject Support Strand: STATISTICS Unit 4 Sampling: Text

Beginning in the top left-hand corner and Select Teacher no. 2

working down the columns gives the result 80 02 86 03
shown opposite. 23 90 18 56
The teachers selected are: 84 34 73 51
31 98 73 75
No. 2 Mrs Pearson 28 73 32 83
No.11 Mr Mobey 52 70 53 23
No. 5 Mrs Barnes 47 14 69 68
Select Teacher no. 5
15 45 05 18
No. 1 Mrs Skirton
59 79 80 51
75 11 01 10

Select Teacher no. 11 Select Teacher no. 1

Worked Example 2
A headteacher wishes to select a stratified sample
of 50 pupils from Years 10, 11, 12 and 13. The Year Number of pupils
table shows how many students are in each year.
10 320
Solution 11 300
First find the total number of pupils in the 12 180
population. 13 150

Total = 320 + 300 + 180 + 150

= 950
320
The fraction of the population in Year 10 is .
950
So the number of pupils selected from Year 10 is given by:
320
× 50 = 16.84 , so select 17 pupils.
950
The number of pupils selected from Year 11 is given by:
300
× 50 = 15.79 , so select 16 pupils.
950
The number of pupils selected from Year 12 is given by:
180
× 50 = 9.47 , so select 9 pupils.
950
The number of pupils selected from Year 13 is given by:
150
× 50 = 7.89 , so select 8 pupils.
950
Note
Note that, having designed a questionnaire and selected a sampling method, the issue of
how to obtain the data is an important one. Although not covered here, it should be noted
that data collection might be through
• personal interviews (at home, work or in the street)
• telephone enquiry (used by many opinion pollsters)
• postal or email enquiry (cheap but potentially biased sample, e.g. open to misuse)
• electronic method (used to obtain data, for example, TV viewing of selected
viewers)
9
 CMM Subject Support Strand: STATISTICS Unit 4 Sampling: Text
4.2

Worked Example 3
The diagram below shows 57 small fish. To estimate the average fish length (that is, the
length in mm from the tip of its 'nose' to the middle of its tail, as shown here length
take random samples of size
(a) 5 (b) 10.
Which will give you the more accurate estimate?
1
4
3 5
2

6 7

10 11
12

13 14
15
16

19
17 18

23
21 22
20

27
24 25 26

29
28
30
32
31
35
33
36

34
37
39

38
40
41

44
43
42
45

46

48
47

51 49 50

53 54
52

57

55 56

10
4.2 CMM Subject Support Strand: STATISTICS Unit 4 Sampling: Text

Solution
Here there are no obvious different categories so we will take a random sample. There is
a table of random numbers on page 15.
Starting arbitrarily on row 16, combining two digits together gives numbers from 00 to
99. We use only numbers in the range 01 to 57 (and ignore any repeats), so that each fish
has an equal chance of being picked.
8 4 1 3 8 4 9 1 5 4 1 1 9 9 8 2 8 4 7 5 4 3 3 4 3 4 0 9 0 8
84 13 84 91 54 11 99 82 84 75 43 34 34 9 8
↓ ↓ ↓ ↓ ↓ ↓ ↓ ↓
13 54 11 43 34 (repeat so 9 8
do not use)

Thus our sample of size 5 consists of the

{13th, 54th, 11th, 43rd, 34th} fish.
The lengths of these fish are measured, to the nearest mm, as
2.6, 2.7, 1.8, 3.0, 3.2
giving an estimate of
(2.6 + 2.7 + 1.8 + 3.0 + 3.2) = 2.66 mm
5
for the average length of all 57 fish.

For a sample of size 10, we continue in the same way with the random number table,
choosing as our sample, the
{13th, 54th, 11th, 43rd, 34th, 9th, 8th, 20th, 32nd, 23rd } fish
and these have an average length of

(3.1 + 2.6 + 1.8 + 3.0 + 3.3 + 1.7 + 3.5 + 1.5 + 1.7 + 2.5) = 2.47 mm
10

We would expect this second estimate of 2.47 mm to be more accurate than the first one
since it is obtained from a larger sample size.

Worked Example 4
Irving is a community worker and is asked to assess the views of the community in which
he works on the introduction of a neighbourhood watch scheme. The community has
three different types of housing consisting of a new building development of 270 privately
owned homes, a large housing association estate of 450 houses and 180 flats for the
elderly.
Explain how Irving could obtain a stratified random sample of size 100.

Solution Private 270

There are three categories of housing in the community. Housing Assn. 450

Flats 180

Total 900

11
4.2 CMM Subject Support Strand: STATISTICS Unit 4 Sampling: Text

Since Irving requires a sample of size 100 there will No. No. ÷ 9
be one sample member for every 9 in the
population. Private 270 30
This gives the representative numbers as shown
Housing Assn. 450 50
on the right:
Flats 180 20

Total 900 100

Inside each category, they need to be chosen at random by numbering, for example, the
private houses 001 to 270, and using the first 30 numbers in this range from a table of
random digits, taken three at a time.

Worked Example 5
A survey is carried out to investigate the quality of new houses.
All the people in the country who have recently bought a new house are to be surveyed.
(a) Give one reason why a pilot survey might be carried out first.
(b) Give one reason why the survey should be carried out by post.
(c) Give one possible disadvantage of a postal survey.

Solution
(a) To see if there are any difficulties with any of the questions in the survey.
(b) It is far cheaper than sending people to each new house in the country.
(c) Not everyone will respond to a postal survey so the survey might not be truly
representative.

Exercises
1. A factory contains 24 identical machines which are labelled:
A, B, C, D, E, F, G, H, I, J, K, L, M, N, O, P, Q, R, S, T, U, V, W, X

(a) Create a systematic sample containing: 68 06 42 38 06 15

(i) 8 machines, (ii) 6 machines, 50 96 24 21 82 53
55 40 32 46 74 76
(iii) 4 machines.
06 57 35 48 20 92
(b) Use the random numbers in the table to create 28 74 44 73 44 25
random samples containing: 65 08 56 68 45 74
63 77 87 58 00 09
(i) 5 machines, (ii) 10 machines.
13 26 76 76 81 60
(c) Describe another way that a random 80 51 89 59 38 10
sample could be made. 27 22 87 55 80 52

12
4.2 CMM Subject Support Strand: STATISTICS Unit 4 Sampling: Text

2. The table shows the number of

students in each year group of a Year Number of students
school. 7 150
How many students should be 8 148
selected from each year group to 9 162
create a stratified sample of 80 10 154
students? 11 152
12 80
13 62

3. A company wants to form a stratified

Number employed
sample to discuss issues with the
staff. Managers 8
Supervisors 20
(a) How many of each type of
employee should be Administrators 12
included in a sample of Manual Staff 140
size 20? Delivery Staff 30

(b) If a stratified sample contains

8 manual staff, what would
be the size of the complete sample?

4. In order to form a sample a number of pupils are selected from a number of different
teaching groups. The size of the samples and the teaching groups are given in the
table below.

Group Number in Group Number in Sample

A 36 9
B 32 8
C 24 5
D 18 4

(a) Is this a stratified sample?

(b) How should it be changed to give a stratified sample?

5. For an experiment in Biology a square metre

A B C D E F G H I J
of ground has been divided into 100 squares
as shown. 1
2
You are required to create a sample of 3
10 squares.
4
(a) Describe two ways of creating a 5
systematic sample, giving the results 6
from each method. 7
8
(b) Use the following list of random
9
numbers to create a random sample.
10

13
 CMM Subject Support Strand: STATISTICS Unit 4 Sampling: Text
4.2
11 57 42 70 41 12 88 51 Create a second random sample, by
32 38 14 65 49 46 18 62 starting in a different place in the
48 20 47 44 56 65 46 36 list of random numbers.
42 59 61 41 78 42 72 11
50 42 58 78 71 78 12 37 How many squares do the two
66 08 21 84 94 61 31 30 samples have in common?
14 64 51 05 53 93 45 86
93 49 05 27 54 18 64 57
87 61 55 67 23 26 70 75
03 87 19 48 10 69 35 61
42 52 83 74 35 09 13 36
51 43 76 62 91 39 89 75

6. What problems might be encountered if samples are formed in the following ways?
(a) Selecting people at random from a telephone directory.
(b) Selecting every third person entering a shopping arcade.
(c) Selecting people at random at a football match.
(d) Selecting people leaving a golf course in red cars.

7. Describe a survey, in which each of the samples in question 6 could be a good

sample to base the survey on.

8. Describe how you might select a sample if you were asked to conduct a survey to
find out if:
(a) the parents of primary school children were happy with their schools,
(b) users of mobile phones were happy with the service provided,
(c) the local bowling green was properly maintained,
(d) the local public transport services are adequate.

9. (a) In a school there are 420 pupils in the lower school, 310 pupils in the middle
school, and 130 pupils in the upper school.
(i) How many pupils from each part of the school should be included in a
stratified random sample of size 100?
(ii) Explain why a stratified random sample should be taken rather than a
simple random sample.
(b) These 100 pupils were asked to keep a record of the number of hours of
television that they watched in one week.
The results are summarised below.

No. of hours watched 0– 5– 10– 20– 30– 40– 60– 80-100

No. of viewers 7 3 15 35 22 10 6 2

Draw a histogram to illustrate these data.

14
 CMM Subject Support Strand: STATISTICS Unit 4 Sampling: Text

Table of Random Numbers

93319 51747 56137 84026 09938 33225 09260 99111 95811 65380
66963 84712 47728 09302 97054 18325 39521 09998 13502 03175
32402 79709 90355 63399 65756 88808 03729 68890 53723 54769
54957 86171 49491 66036 18419 76886 79804 58345 95727 43667
08145 65172 33770 22459 81756 34626 12437 77989 93784 63509
68822 62237 97229 84879 85615 27975 08792 06415 97140 18104
12863 24668 57894 41004 71069 88470 10189 64518 96966 05133
61482 50377 82349 42127 20668 79952 63515 31342 86673 73307
27435 16613 17128 95452 16456 07938 38823 56349 61056 39089
90894 90120 81304 97886 30899 90879 76106 42547 61752 83702
91542 63732 52826 13498 15318 32814 17932 81000 36354 62391
61541 21834 16647 45699 68935 19547 29529 16353 73731 23654
00721 61877 13197 61565 90793 22343 47901 27820 29716 64722
58234 83176 07400 93972 12910 26171 62233 48527 22017 86412
90901 96697 93813 92525 38424 83822 12367 92113 58091 73871
84138 49154 11998 28475 43343 40908 74679 82072 32902 33021
99982 38394 12149 92320 45068 56675 15303 61472 20199 77634
55118 33697 88713 83980 35348 99632 79747 67890 66721 52867
44596 57212 94875 90894 75242 16150 80707 84160 49579 45894
69441 12716 60696 90838 22693 56043 56973 12863 04986 30644
65987 65511 94550 58344 41328 95482 31816 96121 95629 61646
30442 37490 00651 19619 69512 56027 49675 69408 54436 11548
31841 63651 72390 47059 10040 55521 86911 23854 51584 41387
25590 26741 75800 15745 69549 91253 80164 11724 45069 04435
94224 21091 04534 84401 76736 29703 52139 31717 28279 93696
52985 99052 71324 15364 26327 39095 26028 29939 23509 13451
73517 02148 31714 13119 15732 45512 68301 89041 38890 81037
22887 10718 31454 52309 06514 42507 75468 48605 06298 18200
76438 04101 75458 55270 99354 51854 74394 43987 63733 27449
14482 91534 34322 58089 96509 13719 73664 68489 82340 88210
35831 77045 25029 98535 03402 70868 04587 48599 47285 46148
87857 90657 80145 26419 21262 28733 28579 43054 61071 91889
29247 64438 86175 87703 83446 67031 54625 99997 35700 91742
47772 29667 93738 68677 99266 49739 67150 90133 30988 31886
95635 92152 39697 58015 36592 55997 19531 65129 89328 61894

50339 66056 60354 65281 29894 37775 34516 11016 69852 54617
67913 58978 35732 47891 64046 04856 97359 73264 19503 37402
80090 57366 79265 48560 14360 87736 91582 27931 38780 05247
13249 74207 05827 30621 83365 47467 32843 91140 18080 40854
76852 80223 46216 65444 64815 81402 50385 77524 73167 43195

16731 07241 55052 54549 92503 74292 68834 89765 41287 36098
98006 83605 33978 33014 52325 17769 53144 13260 34459 85508
56587 43498 48494 71779 58770 27636 32147 99218 01012 26591
83471 72666 64849 24946 63288 51123 85098 50281 64871 47266
52502 18474 08744 40914 79954 28137 63705 89423 49704 90259

03846 57786 80554 51830 19528 19318 70590 50811 50034 99810
40864 30009 89931 33920 20729 04074 14893 06818 83086 36458
37140 43492 22727 88547 56098 32953 53634 91516 82806 39158
35137 83718 60390 45365 31548 65730 64725 96106 70568 79585
26709 48705 88170 52680 26408 18142 71563 73812 87382 45203

15