JSMath6 Part3

4. Give 5 different numbers such that their average is 21.
The numbers are:

I found these numbers by:
5. A median of a set of scores is the value in the middle when the scores are
placed in order. In case there is an even number of scores there is not a
single middle score. In that case you must take the middle two scores
and calculate their average.
Four students scored the following results for a test: 19, 20, 17, 11. Find
the median.
Median =
This is how I found my answer:
6. a) If you add a number to a set of numbers, the mean changes
ALWAYS / SOMETIMES / NEVER (circle the correct answer)
Reason:
b) If you add a number to a set of numbers, the median changes
ALWAYS /SOMETIMES / NEVER (circle the correct answer)
Reason:
c) If you add a number to a set of numbers, the mode changes
ALWAYS /SOMETIMES / NEVER (circle the correct answer)
Reason:
Lesson outline 1: Mean, median, mode

Time: 80 minutes
Prerequisite knowledge: Pupils have met mean, median and mode before and
know the arithmetic involved in computing these measures of central
tendency.
Objectives: Pupils should be able to
a) find the mean, median and mode of a set of data in context.
b) make statements about the effect on mean / median / mode if values are
added to the data set (adding zero value, adding two values with equal but
opposite deviation from the central measure, adding values equal to the
central value).
Review of mean, median and mode
Exposition - discussion strategy (15 minutes)
Teacher presents the question:
A test was scored out of 20 (only whole marks were given) and 12 pupils
scored: 19, 20, 17, 11, 19, 19, 15, 8, 15, 20, 17 and 18.
Module 6: Unit 4 121 Measures of central tendency

Pupils are asked to compute the mean, median and mode. (Give sufficient
time to pupils to do the working.)
Answers: mean 16.5 / mode 19 / median 17.5
Teacher calls on pupils to explain how they obtained their results.
Expected to result in:
Mean = (sum of all scores) (number of pupils).
Illustrate the mean as the balancing point:
Forces (deviations) at one side balance the forces (deviations) at the other
side.
Mode is the score with the highest frequency (the value that is in fashion,
the most popular).
Median is the value in the middle when the scores are placed in order (if odd
number of observations) OR average of the middle two scores (if even
number of observations).
Half the number of observations are to the right of the median the other half
to the left.
Questions: Which of these threemean, median or modedo you feel can
be used best to represent the set of scores? Justify your answer.
DO NOT ANSWER the question at this stage; only make an inventory of the
pupils opinions and their reasons, without further comment.
Write the results on the chalkboard:
Best measure to use number of because
students in favour
mean
median
mode
Inform pupils that they are going to investigate how mean, mode and median
behave, so as to make a decision on which measure might be best used in a
certain context.
Investigating (40 minutes)
The following are covered in the pupils worksheets (Worksheet for pupils is
on a following page seven pages ahead)

a) Is mean, median, mode necessarily a value belonging to the set and/or a
value that could be taken in reality?
b) The effect on mean, median, mode of adding a zero value to the value set.
c) The effect on mean / median/ mode of adding two values with equal but
opposite deviations or unequal deviations from mean, mode, median.
d) The effect on mean / median / mode of adding values equal to mean /
median / mode.
Pupils activity
Teacher gives worksheets to pupils.
In small groups pupils are to answer the questions individually, then next
compare and discuss the following questions.
scored: 19, 20, 17, 11, 19, 19, 15, 8, 15, 20, 17 and 18.
Using the above or other pupils scores (using real scores obtained by the
class for example) answer the following question:
Q1) Must the mean / median / mode be a score attained by one of the pupils
in the class?
Justify your answer. Illustrate with examples and non examples.
Note to the teacher:
(i) mean
The mean represents the scores but need not be one of scores itself, it
might even be a score that is impossible ever to get.
(ii) median
The median will be a score of one of the pupils if number of scores is
odd. If the number of scores is even the median will be a score
nobody did get or even nobody ever can get. If the median is half way
between 16 and 18, then 17 is a possible score although nobody did
score 17; if the median is between 17 and 18 the median is 17.5, a
score nobody can ever get as it is not a whole number.
(iii) mode
The mode is necessarily a score attained by several pupils. If all
scores are different there is no mode. If certain scores have the same
frequency a set of scores can have more than one mode (bimodal ,
trimodal, etc., distribution).
Q2) Investigate how the mean / median / mode changes when a zero score is
added to the following set of scores.
a) 19, 20, 17, 11, 19, 19, 15, 8, 15, 20, 17 and 18.
Mean is 16.5; median is 17.5 and mode is 19.
b) 19, 20, 17, 11, 19, 19, 15, 8, 15, 20, 17, 18 and 11
Mean is 16.1; median is 17; mode is 19
c) 0, 19, 20, 17, 11, 0, 19, 19, 8, 15, 20, 17 and 18.
Mean is 14.1; median 17; mode 19

Make a correct statement:
1. If to a set of scores a zero score is added the mean changes
ALWAYS / SOMETIMES / NEVER
2. If to a set of scores a zero score is added the median changes
3. If to a set of scores a zero score is added the mode changes
Does the size of the number of observations matter? Are the changes (if
any) the same whether you considered 20 observations or 2000?
(Answer: Mean / mode / median all change sometimes. If a large number
of observations is involved, the change in the mean is very small (the first
decimal place might not change at all) or when the mean is zero, adding a
zero will not change the mean. Median changes are likely to be smaller in a
large population than in a smaller, but even there changes are generally
small. The nature of the observations (do observations have close to the
same frequency) determines whether or not changes in mode occur.)
Q3) A set of scores has a mean of 16.
Without calculating the new mean state how the mean changes if two
more scores are to be taken into account.
a) the two scores are 14 and 18
b) the two scores are 15 and 17
c) the two scores are 14 and 17
d) the two scores are 12 and 20
e) the two scores are 12 and 19
Make a general statement about when the mean will change and when it
will not change.
(Answer: the mean will not change if two values with equal but opposite
deviations from the mean are added, or if the added value equals the
mean; otherwise it will change.)
Q4) A set has a median score of 16.
Without calculating the new median state how the median changes if two
f) the two scores are 18 and 19
Make a general statement: when will the median change, when will it not
change?
(Answer: median will not change whatever values are added as long as
one is to the left and one to the right of the median; if the added values are
both to the right or the left the median might change.)

Q5) A set has a mode score of 16.
Without calculating the new mode state how the mode changes if two
Make a general statement: when will the mode change, when will it not
change?
(Answer: No statement can be made as the added values might make the
distribution bimodal or trimodal. For example 14, 14, 16, 16, 16, 18 has
mode 16 adding 14 and 18 makes it a bimodal distribution with modes 14
and 16. If the original set was 14, 14, 16, 16, 16, 18, 18 the adding of 14
and 18 makes it a trimodal distribution with modes 14, 16 and 18.)
Q6) A set of scores has a mean of 16. Without calculating the new mean state
how the mean changes if two more scores equal to the mean are added.
Q7) A set of scores has a median of 16. Without calculating the new median
state how the median changes if two more scores equal to the median are
added.
Q8) A set of scores has a mode of 16. Without calculating the new mode state
how the mode changes if two more scores equal to the mode are added.
Q9) Answer question 6, 7 and 8 if only ONE value equal to mean /median
/mode respectively were added.
(Answer Q6/ Q7/ Q8/ NO changes in mean, median and mode; Q9/ only
the median might change.)
Q10) Write down a data set of the ages of 12 people travelling in a bus with
a) mean 24
b) median 24
c) mode 24
Compare the data sets each member in your group has written down.
Are all the same?
Why are there differences? How can different data sets have the same
mean (median / mode)?
Which set is the best? Why?
A baby is born in the bus, making now 21 passengers the last one with
age 0.
Each pupil is to compute the change in mean / median / mode of her /his
data set.
The grand-grand parents (age 90 and 94) of the newborn enter the bus,
making up a total of 23 passengers.

Each pupil is to compute the change in mean / median / mode of her /his
data set.
The 0 and the 90 / 94 are called outliersthey are far from the mean /
mode/ median.
Comparing your results how do outliers affect the mean / median / mode?
1. If to a set of ages outliers are added the mean changes
2. If to a set of ages outliers are added the median changes
3. If to a set of ages outliers are added the mode changes
Which of the three measures is most affected? (Answer: In general the
mean is most affected by outliers as compared to median and mode.)
N.B. The above outlined activity would be more powerful if carried out
on a computer using spreadsheets. In the summary the teacher could use a
computer (provided the screen can be projected) to illustrate the effect of
certain changes on both large and small data sets.
Reporting, summarising of findings, setting assignment (25 minutes)
Groups report / discuss / agree. Teacher summarises in table (outline already
on the (back) of board before start of lesson).
a) Mean and median need not be observed values (values included in the
observation set). They might even have a value that can never be an
observed value. The mode (if it exists) always is an observed value.
b) Effect on mean / median / mode if one or two observations are to be
included.
CHANGE EFFECT ON
MEAN MEDIAN MODE
Adding zero value(s) S S S
Adding two values with equal but
N N S
opposite deviations
Adding two values with opposite unequal
A N S
deviations
Adding two values with deviations both
A S S
positive (negative)
Adding two values equal in value to the
N N N
central measure at the top of each column
Adding one value equal in value to the
N S N
A indicates will always change

S indicates will sometimes change
N will never change

c) Effect of the number of observations involved (small sample or large
sample)
In the case of a large number of observations, adding of observations (not
equal to the central measure) will ALWAYS change the meanbut the
change will be (very) small. Outliers have a great impact on the mean of a
small data set, but very little on a very large data set.
The median and mode are more likely to remain the same in the case of
large numbers of observations, but can change.
Now come back to the original question:
Questions: Which of these threemean, median or modedo you feel
can be used best to represent the set of scores? Justify your answer.
The tabulated answers of the pupils.
Best measure to use number of because
students in favour
mean
median
mode
Ask whether or not pupils want to change their previous opinion based on
the increased insight on behaviour of the measures. If a pupil wants to
change he/she is to justify the decision.
The discussion should lead to the decision that the median is most
appropriate: half of the pupils scored below / above 17.5. The mean is less
appropriate as it does not give any information as to how many pupils
scored above / below the average of 16.5 (as mean is affected by outliers).
Pupils assignment
(or take some questions for discussion in class if time permits)
In each of the following cases decide, giving your reasons, whether the mean,
median or mode is the best to represent the data.
1. Mr. Taku wants to stock his shoe shop with shoes for primary school
children. In a nearby primary school he collects the shoe sizes of all the
200 pupils (one class group from class 1 to class 7). Will he be interested
in the mean size, median size or modal size?
Answer: mode
2. In a small business 2 cleaners earn P340 each, the 6 persons handling the
machinery earn P600 each, the manager earns P1500 and the director
P3500 per month.Which measuremean, median or mode best
represents these data?
Answer: mode
3. An inspector visits a school and want to get an impression of how well
form 2X is performing. Will she ask the form teacher for mean, median or
mode?
Answer: median

4. A pupil did 4 small projects in mathematics on the topic of number
patterns during the term scoring (out of 20) in order : 4, 16, 15 and 16.
Which represents best the overall attainment level of the pupil on project
work on number patternsmean, median or mode?
Answer: median/mode
Discuss: Is using the mean score to represent the work done in
mathematics during a term a fair measure for the attainment of the pupil?
5. A house building company wanting to find out what type of houses they
should build most often in a region carried out a survey in that region to
find out the number of people in a family. Will they use mean, median or
mode to decide what type of houses should be build most?
Answer: mode
6. A car battery factory wants to give a guarantee to their customers as to the
lifetime of their batteries, i.e., they want to tell the customer if you have a
problem with the battery in the next ??? months we will replace your
battery with a new one. They checked the lifetime of 100 batteries. Will
they use mean, median or mode to decide on the number of months to
guarantee their batteries?
Answer: mean

Worksheet for Pupils
Investigation
Question 1:
scored: 19, 20, 17, 11, 19, 19, 15, 8, 15, 20, 17 and 18.
Using the above or other pupils scores (using real scores obtained by the
class for example) answer the following question:
Has the mean, median or mode to be a score attained by one of the pupils in
the class?
Justify your answer. Illustrate with examples and non examples.
Question set 2:
Investigate how the mean, median and mode change when a zero score is
added to the following set of scores.
a) 19, 20, 17, 11, 19, 19, 15, 8, 15, 20, 17 and 18.
b) 19, 20, 17, 11, 19, 19, 15, 8, 15, 20, 17, 18 and 11
c) 0, 19, 20, 17, 11, 0, 19, 19, 8, 15, 20, 17 and 18.
1. If to a set of scores a zero score is added the mean changes
2. If to a set of scores a zero score is added the median changes
3. If to a set of scores a zero score is added the mode changes
Question 3:
A set of scores has a mean of 16. Without calculating the new mean state how
the mean changes if two more scores are to be taken into account.
Make a general statement about when the mean will change and when it will
not change.

Question 4:
A set has a median score of 16. Without calculating the new median state
how the median changes if two more scores are added to the set as follows:
f) the two scores are 18 and 19
Make a general statement about when will the median change, and when will
it not change.
Question 5:
A set has a mode score of 16. Without calculating the new mode state how
the mode changes if two more scores are added to the set as follows:
Make a general statement about when will the mode change, and when will it
not change.
Question 6:
A set of scores has a mean of 16. Without calculating the new mean, state
how the mean changes if two more scores equal to the mean are added.
Question 7:
A set of scores has a median of 16. Without calculating the new median, state
how the median changes if two more scores equal to the median are added.
Question 8:
A set of scores has a mode of 16. Without calculating the new mode, state
how the mode changes if two more scores equal to the mode are added.
Question 9:
Answer question 6, 7 and 8 if only ONE value equal to mean, median and
mode respectively were added.

Question 10:
Write down a data set of the ages of 12 people travelling in a bus with
a) mean 24
b) median 24
c) mode 24
Compare the data sets each member in your group has written down.
Are all the same?
Why are there differences? How can different data sets have the same mean
(median / mode)?
Which set is the best? Why?
A baby is born in the bus, making now 21 passengers the last one with age 0.
Compute the change in mean / median / mode of your data set.
The grand-grand parents (age 90 and 94) of the newborn enter the bus making
up a total of 23 passengers.
Compute the change in mean / median / mode of your data set.
The 0 and the 90 / 94 are called outliersthey are far from the mean /
mode/ median.
Comparing your results how do outliers affect the mean / median / mode?
1. If to a set of ages outliers are added the mean changes
2. If to a set of ages outliers are added the median changes
3. If to a set of ages outliers are added the mode changes
Which of the three measures is most affected?

Recording sheet for students
Effect on mean, median and mode if one or two observations are to be added.
(Place an A, S, or N in each empty box. A indicates will always change, S
indicates will sometimes change, N will never change.)
CHANGE EFFECT ON
MEAN MEDIAN MODE
Adding zero value(s)
Adding two values with equal but opposite
deviations
Adding two values with opposite unequal
deviations
Adding two values with deviations both
positive (negative)
Adding two values equal in value to the
Adding one value equal in value to the
DISCUSS IN YOUR GROUP

1. If you collect data on the pupils in your class will the data collected
always have a mode? a median? a mean? Justify your answer, illustrate
with examples and non examples.
2. Averages are meant to be representative for the data. List advantages and
disadvantages of the mode, median and mean and find examples when
you would choose one rather than another.
3. How real are averages? Consider the following:
The average number of children in a family is 2.58, so each child in the
family will always have 1.58 other children to play with.
If you have no money you join a group of 4 friends who have P2.50
each. Now you are a group of 5 and on average each person in the group
has P2.00. You suddenly have P2.00.
4. A bar graph illustrates the number of brothers and sisters of a group of
students.
From the bar graph find the mean, median and mode. Which of the three
measures is easiest to find?

Lesson outline 2: Mean, median, mode
A look at the average wage
The scenario for this lesson idea is a manufacturing and marketing company,
in which the notion of average wage is considered from different points of
view. The purpose is to show how a selection of the mean, the median, or the
mode gives different answers to the same question.
Materials
The question A look at average wage below can either be photocopied or
copied onto the blackboard.
Students are allowed to use a calculator.
Classroom organisation
Students can work individually, in pairs or in small groups.
The following information is to be given to pupils:
Question: A look at average wage
The Head of the Union Mr. Motswiri in the Matongo Manufacturing and
Marketing Company was negotiating with Ms. Kelebogile Matongo, the
president of the company. He said, The cost of living is going up. Our
workers need more money. No one in our union earns more than P9000.- a
year.
Ms. Matongo replied, Its true that costs are going up. Its the same for
uswe have to pay higher prices for materials, so we get lower profits.
Besides, the average salary in our company is over P11000.-. I dont see how
we can afford a wage increase at this time.
That night the union official conducted the monthly union meeting. A sales
clerk spoke up. We sales clerks make only P5000.- a year. Most workers in
the union make P7500.- a year. We want our pay increased at least to that
level.
The union official decided to take a careful look at the salary information. He
went to the salary administration. They told him that they had all the salary
information on a spreadsheet in the computer, and printed off this table:
Type of job Number employed Salary Union member
President 1 P125 000 No
Vice president 2 P65 000 No
Plant Manager 3 P27 500 No
Foreman 12 P9 000 Yes
Workman 30 P7 500 Yes
Payroll clerk 3 P6 750 Yes
Secretary 6 P6 000 Yes
Sales Clerk 10 P5 000 Yes
Security officer 5 P4 000 Yes
TOTAL 72 P796 750 -
The union official calculated the mean:

P796 750
MEAN = P11 065,97
72
Hmmmm, Mr. Motswiri thought, Miss Matongo is right, but the mean
salary is pulled up by those high executive salaries. It doesnt give a really
good picture of the typical workers salary.
Then he thought, The salary clerk is sort of right. Each of the thirty
workmen makes P7500.- That is the most common salarythe mode.
However, there are thirty-six union members who dont make P7500.- and of
those, twenty-four make less.
Finally, the union head said to himself, I wonder what the middle salary is?
He thought of the employees as being lined up in order of salary, low to high.
The middle salary (its called the median) is midway between employee 36
and employee 37. He said, employee 36 and employee 37 each make
P7500.- , so the middle salary is also P7500.-.
Questions:
1. If the twenty-four lowest salaried workers were all moved up to P7500.-,
what would be
a) the new median?
b) the new mean?
c) the new mode?
2. What salary position do you support, and why?
Activity 1: Presentation of the scenario
Review with students the problem setting and the salary information. Ask
students to identify how the mean, median, and the mode are used in the
problem description. Use question 1 to review one way in which pay raises
may be distributed.
Answers: a) New median P7 500
b) New mean P11 812.50
c) New mode: P7 500
Pose these questions:
Which measures of central tendency stayed the same?
Which measures of central tendency changed? Why?
If you changed only one or two salaries, which measure of central
tendency will be sure to change? [The mean, since its calculation includes
all values.]
If you changed only one or two salaries, which measure of central
tendency will be most likely to stay the same? [The mode is most likely to
stay the same, because it is the most frequently occurring salary, and only
one or two salaries are being changed.]
If you change only one or two salaries, how likely is the median to
change? [It depends. If the median is embedded in the middle of several

salaries that are the same, it wont change. If the median is close to a
different level of salary, it is not likely to change.]
Activity 2: Using a Spreadsheet (optional)
Having students enter the employees salaries into a spreadsheet on a
computer or demonstrating the use of a spreadsheets to students may clarify
the role of a computer in solving real life problems. Column A could list the
number of employees of each type, and column B could list the salary of that
type of employee. Display the mean salary for all employees in a cell at the
bottom of the spreadsheet labelled mean salary. [Define the cell as the total
salary value (payroll) divided by the total number of employees.] After using
the spreadsheet to display the new salaries and calculating the new mean for
each situation, pose the following inquiries:
Predict the mean if the twenty-four lowest paid employees have their
salaries increased to P7 500. Make the changes in the spreadsheet to find
the actual mean.
The president gave himself a raise that resulted in increasing the mean
salary by P500. Predict what you think his new salary was. Use the
spreadsheet to experiment and find the new salary.
Two new employees were hired by the company: a plant manager and a
foreman. Predict whether the mean salary will increase, decrease, or stay
the same. Explain your prediction. Check it out with the spreadsheet.
Activity 3: Developing an argument
Use question 2 to initiate a discussion on drawing conclusions from the
information. Small groups of students can develop position statements and
report back to the class. There is no single correct answer to the discussion
question. Management would naturally favour the mean; the union leader, the
median; and the lower-paid members the mode.
Evaluation: This problem has more than one reasonable solution. However,
many students expect problems in the mathematics class to have only one
correct solution. Teachers can promote student consideration of multiple
solutions by asking students to write up or present at least two reasonable
alternatives. At first, students may simply take ideas from one another
without much reflection, but if the teacher continues to value creative,
reasonable alternatives, students will begin to enjoy actively looking for
multiple solutions.
Practice task 2
1. Try out the lesson outline 2 in your class: A look at the average wage
2a) Write an evaluative report on the lesson. Questions to consider are: Did
pupils meet difficulties? Were pupils well motivated to work on the
activity? Were the objectives achieved? Did you meet some specific
difficulties in preparing the lesson or during the lesson? Was discussion
among pupils enhanced?
b) Present the lesson plan and report to your supervisor.

Section F: Mean, median and mode for grouped discrete
data
A 60 item multiple choice test was tried in a class with 43 pupils. The results
are represented in the following frequency distribution.
No. of correct answers 1-10 11-20 21-30 31-40 41-50 51-60

Frequency 4 5 11 9 8 6
A mode or median cannot be obtained from this frequency table. You can
only read off the class interval that contains the mode and the class interval
that contains the median. The class interval with the highest frequency is
21 - 30: the modal class. The median is the 22nd observation, i.e., the score
of the 22nd student; that falls in the class interval 31 - 40.
If the number of data is large but discrete (for example the scores of 2000
pupils in an examination marked out of 60) or continuous (the time taken to
run 100 m) data is best placed in groups or intervals.
The scores could be grouped in five intervals: 1 - 10, 11 - 20, 21 - 30, 31 - 40,
41 - 50, and 51 - 60 as in the distribution table above for 43 students only. By
grouping data some information is lost. For example in the class
1 - 10 there are 4 students, and we no longer can see what their actual scores
were (did all score 10?).
For calculation purposes the mid-interval value (average of lower bound
and upper bound value of the interval) is used.
From a grouped frequency table you can find:
the modal class (the class with the highest frequency)
the interval in which the median is found
an estimate of the mean
A calculation of the mean using mid-interval values.
sum of [mid interval value frequency]
Estimate of the mean =
sum of the frequencies
Example
The table gives the end of year examination mark of 200 students (maximum
mark was 50) and the calculation to obtain an estimate of the mean.
Mark Frequency Mid-value Mid-value Frequency

1-10 10 5.5 55
11-20 20 15.5 310
21-30 60 25.5 1530
31-40 90 35.5 3195
41-50 20 45.5 910
TOTAL 200 6000

6000
An estimate for the mean is = 30. The estimated mean is 30.
200
The modal class is 31 - 40. Most students scored in the range 31 - 40.
1 1
The median is the (200 + 1) = 100 th term, i.e., the average of the 100th
2 2
and 101st term. As the actual value of these terms is unknown you can only
give the interval in which these values are: the interval 31 - 40.
Calculator
Your calculator can help you with these and longer calculations in statistics.
Actual procedures vary with the brand of calculator, but if yours has statistics
capability, the general way to use it is as follows. For more details, consult
the instructions that came with the calculator.
1. Place it in Statistics Mode (if it has such a mode).
2. Clear out any previously stored statistical data from the memory registers.
3. Now enter individual data values:
a) for single values key in each value, followed by a press of the
DATA or ENTER or + key.
b) for grouped values some calculators allow the entry of grouped data
by having separate entry keys for the group frequency (or count) and
for that groups average value. Consult your documentation.
4. Once all data points have been entered, a press of the X key (or
equivalent) will display the mean of all the entered data. In most
calculators there are other keys for the sample and population standard
deviations.

Self mark exercise 3
1. The histogram illustrates the mass of fish caught in the Sashe dam one
day.
a) Make the grouped frequency table.

b) How many fish were caught in all?
c) What is the modal class?
d) In what class interval is the median mass?
e) Calculate an estimate for the mean mass of fish caught (use your
frequency table).
2. The diameters of a batch of oranges were measured and the results
displayed in a histogram.
Continued on next page

a) How many orange had a diameter d cm in the range 8 d < 9?
b) Make the grouped frequency table corresponding to the histogram.
c) How many oranges in total were measured?
d) What is the modal class interval?
e) Calculate an estimate of the mean length of the diameter of the
oranges (use your frequency table).
3. The height of fruit trees was measured one year after planting. The result
is displayed in the histogram below.
As the data, length of the trees, is continuous and taken to the nearest cm
the first class 100 - 109 has as class boundaries 99.5 and 109.5. The class
boundaries are taken half-way between the class intervals.
a) How many trees had a height h cm in the range 110 - 119 cm?
b) Make the grouped frequency table corresponding to the histogram.The
first class interval is 100 - 109, the second 110 - 119, etc.
c) How many fruit trees in total were measured?
d) What is the modal class interval?
e) Calculate an estimate of the mean height of the fruit trees from the
grouped frequency table.
4. The ages of pupils in Sefhare CJSS were represented in a histogram.
Although age can be considered to be continuous, you are 14 from the day
you turn 14 until the day you turn 15. The class interval has as lower
bound 14 and as upper bound 15.

a) What are the modal classes? (a bi-modal distribution)
b) How many pupils are in their 16th year?
c) Make a grouped frequency table, first class 11 age < 12.
d) Calculate an estimate of the mean age of the pupils in the school (use
the frequency table).
5. The number of water pipes of different lengths made in a factory during a
month are shown in the histogram below.
a) How many pipes with length L m, 14 L < 18 were made?

b) What is the modal class?
c) In which class is the median length of pipe?
d) How many pipes were produced during the month?
e) Make an estimate for the total length of pipe produced.
f) Make an estimate for the mean length of pipe produced.

6. The histogram illustrates the mass of eggs collected on a poultry farm
during a day.
a) What is the modal class interval?

b) Complete the grouped frequency table
Mass (g) 30 m < 50
Frequency
c) Calculate an estimate of the mean mass of an egg.
7. The monthly salary of the workers in a factory are illustrated in the
histogram.
a) In what interval falls the income of most workers?
b) Complete the grouped income table.
Income (P) 250 I < 500
Frequency
0.004
Density
Frequency

c) In which class is the median monthly income?
d) Calculate an estimate for the mean income.
e) Which of the three averages, modal class, median class or estimated
mean, best represents the data? Justify your choice.
8. The height of pupils in a class was distributed as follows
Height (cm) Frequency
151 -155 4
156 - 160 10
161 - 165 16
166 - 170 22
171 - 175 26
176 - 180 15
181 - 185 2
a) Draw a histogram to represent these data.
b) Calculate an estimate of the mean height.
9. The ages of participants in a fund raising walk were distributed as
follows:
Age Frequency
10 -14 28
15 - 19 65
20 - 24 82
25 - 34 76
35 - 44 54
45 - 59 43
60 - 74 12
a) Draw a histogram to represent these data.
b) Calculate an estimate of the mean age.
Suggested answers are at the end of this unit.

Section G: Estimation of the median, quartiles and
percentiles
Median, quartiles and percentiles can be estimated. Estimations are based on
some assumptions. In this case the assumption is that the data is evenly
distributed over the class interval. There are three methods considered:
estimation using the cumulative frequency curve (section G1), estimation by
using linear interpolation (section G2), and estimation of the median using the
histogram (section G3).
Section G1: Estimation of the median, quartiles and
percentiles from cumulative frequency curves
Making a cumulative frequency table and plotting the cumulative frequency
curve:
A factory producing batteries might be interested in what percent of their
batteries last up to 3 hours, what percent last more than 3.5 hours, etc. To
obtain this information the data from the experiment on the time batteries
lasted is to be represented in a cumulative frequency table and curve.
Cumulative frequency table
Battery life time t hours Frequency Cumulative frequency
0<t<1 5 5
1 < t < 1.5 12 17
1.5 < t < 2 32 49
2 < t < 2.5 40 89
2.5 < t < 3 16 105
3<t<4 9 114
4<t<5 6 120
TOTAL 120

To plot the cumulative frequency curve the cumulative frequency is plotted at
the end of each class. For example (1, 5), (1.5, 17), (2, 49) are points on the
curve.
Reading from cumulative frequency curves

To obtain an estimate of the median battery life, start at the cumulative
frequency axis at the 60th observation and follow the arrows to reach the time
axis. (Half of 120, strictly speaking you are to take the average of the 60th
and the 61st observation. However for larger number of observation generally
for the median is just taken as half of the total).
The estimate of the median is 2.1 hours. 50% of the batteries last for more
than 2.1 hours (and 50% last less or equal to 2.1 hours).
Similarly you can read from the cumulative frequency curve the lower
quartile LQ (at 30th observation being one-quarter of 120, giving an
estimated 1.8 hours. Check this!) and the upper quartile UQ at 90th
observation, being three-quarters of 120, giving an estimated value of 2.5
hours.
The interquartile range is defined as UQ - LQ. In the above example
2.5 - 1.8 = 0.7 h. The middle 50% of batteries last between 1.8 h and 2.5 h.

1. The cumulative frequency curve gives information on the height of 200
students.
a) From the cumulative frequency graph obtain:

(i) an estimate of the median
(ii) an estimate of the lower and upper quartile
b) Calculate an estimate of the interquartile range.
c) Estimate the number of students that are between 126 cm and 136 cm
tall.
d) Estimate the number of student taller than 138 cm.
e) Complete the grouped frequency table.
Height (cm) 120-124 125-129
Number of students
f) Draw a histogram and a frequency polygon to represent the data.
g) The frequency table, histogram, frequency polygon and cumulative
frequency curve all display the same data.
(i) What are the advantages and disadvantages of each form of
display?
(ii) When will you use which format?
(iii) Is one of the forms more useful than the others?

2. The cumulative frequency curve was made by students of Form 5 who
kept record of the number of minutes their mathematics teacher came late
to class on the 60 days of a term.
a) From the cumulative frequency graph obtain:

(i) an estimate of the median
(ii) an estimate of the lower and upper quartile
b) Calculate an estimate of the interquartile range.
c) How many times was the teacher between 5 and 10 minutes late?
d) Complete the grouped frequency table.
Number of 0<t<2 2<t<4
minutes late
Number of
days
e) Draw a histogram and a frequency polygon to represent the data.
f) Calculate an estimate of the mean number of minutes the teacher is
late for class.
3. The table shows the length of time, in minutes, cars stayed in a parking lot
in front of an office.
Time t 0< t < 20 20< t < 40 40 < t < 60 60 < t < 90 90 < t < 120
Frequency 12 42 78 22 6
a) Make a cumulative frequency table.
b) Draw a cumulative frequency curve.

c) Use your curve to obtain an estimate of
(i) the median (ii) the lower quartile (iii) the upper quartile
4. The table shows the times, in minutes, it took for patients to be treated in
a clinic.
Time t (min) 0< t < 10 10< t < 20 20 < t < 30 30 < t < 40 40 < t < 50
Frequency 32 60 54 36 18
a) Make a cumulative frequency table.
b) Draw a cumulative frequency curve.
c) Use your curve to obtain an estimate of:
(i) the median (ii) the lower quartile (iii) the upper quartile
Percentiles
Quartiles divided the data into quarters, similarly percentiles divide the data
into hundred parts.
1
The median is the 50th percentile, the ( n + 1) th value in the ordered
2
sequence of the n values.
1
The lower quartile is the 25th percentile, the ( n + 1) th value.
4
3
The upper quartile is the 75th percentile, the ( n + 1) th value.
4
The pth percentile can be estimated from a cumulative frequency curve by
p
taking (as an approximation) the of the total number of values: p% n.
100
p
More exact it is the ( n + 1) th value.
100

Example
The cumulative frequency curve shows the height of 60 bean plants.
Estimate the height of the tallest 10% of plants.

If 10% is taller than h cm, 90% will be below h cm. You are looking for the
90th percentile.
90% of (60 + 1) is 55. Following the arrows in the diagram: the 90th
percentile is 16 cm. The tallest 10% of plants is between 16 cm and 20.5 cm.
Section G2: Estimation of median, quartiles and

percentiles by linear interpolation
A multiple choice test was tried with 200 students. The number of correct
responses are tabulated:
Number of
1 10 11 20 21 30 31 40 41 50
correct answers
Number of
12 43 71 49 25
students
From this grouped frequency table mean, median and mode cannot be
calculated exactly.
You have seen that from the table you can obtain:
(i) an estimate of the mean (assuming all the values in the interval take the
mid interval value)
(ii) the modal class: 21 - 30 correct answers

1
(200 + 1) th value, the average
(iii) The interval that contains the median, the
2
of the 100th and 101st valueboth are in the interval with 21 - 30 correct
1
answers. Generally you just take the 200 which is the 100th value.
2
The cumulative frequency curve makes it possible to give an estimate of
the median.
It is also possible to calculate an estimate of the median from the table by
assuming that all the data values in the intervals are evenly (linear)
distributed.
The class boundaries for the interval containing the median are 21 and 30
(discrete variable).
At the beginning of the interval you have covered already 12 + 43 = 55

values. So the interval starts at the point with co-ordinates (21, 55).
By the end of the interval you have covered the 71 values in the interval, so
you end at the 55 + 71 = 126th value. Co-ordinates (30, 126).
Linear interpolation means that you assume these two points to be connected
by a line segment.
You want the value M, corresponding with the 100th value i.e., 45 more than
at the beginning of the interval. As the interval contains 71 values, you have
45
to go of the way along the interval (which is 10 long).
71
45
So an estimate of the median is 20 + 10 = 26 to the nearest whole
71
number.
The median number of correct questions is 26.
The process of estimation of the median from a grouped frequency table is
called linear interpolation.
Quartiles and percentiles can be estimated by linear interpolation in a similar
way.

1. The number of letters in the words in a newspaper article were counted.
The result was
Number of letters 1 3 4 6 79 10 12 13 - 15
Frequency 57 46 28 6 3
a) Calculate an estimate of the median word length.
b) Calculate an estimate of the lower quartile and upper quartile word
length.
2. The table gives the heights, to the nearest cm, of boys and girls in a class.
Height 151- 156 - 161 - 166 - 171 - 176 - 181- 186- 191-
(cm) 155 160 165 170 175 180 185 190 195
Girls 3 8 9 16 12 2
Boys 1 2 7 10 14 14 5 5 2
a) Calculate an estimate of the median height of boys and girls.
b) What height is exceeded by 80% of the (i) girls (ii) boys?
3. A multiple choice test was tried with 200 students. The number of correct
responses are tabulated:
Number of Correct answers 1 10 11 20 21 30 31 40 41 - 50
Number of students 12 43 71 49 25
a) Calculate an estimate of the number of correct answers of the top 10%
of the pupils.
b) Calculate an estimate of the number of correct answers of the bottom
10% of the pupils.
4. Use the following raw data of the length (mm) of nails found in packets of
assorted nails.
11 48 53 32 28 15 17 45 37 41
55 31 23 36 42 27 19 16 46 39
41 28 43 36 21 51 37 44 33 40
15 38 54 16 46 47 20 18 48 29
31 41 53 18 24 25 20 44 13 45
a) Using the raw data calculate mean and median.
b) Make a grouped frequency table taking class intervals 10 -14,
15 - 19, etc. Using the grouped frequency table calculate the estimate
of the mean and the median.
c) Make a grouped frequency table taking class intervals 10 - 19,
20 -29, etc. Using the grouped frequency table calculate the estimate
of the mean and the median.
d) What is the effect of changing class width on the estimate of
(i) mean (ii) median?

Section G3: Estimation of the median from a histogram
In a histogram the area is proportional to the frequency. The median is the
value in the middle and will therefore divide the area under the histogram into
two equal parts.
For example the histogram illustrates the height of 100 fruit trees 1 year after
being planted.
Altogether there are 100 units of area contained in the histogram. We are
looking for a line that will divide the area such that 50 units of area are to the
left of the line and 50 units of area to the right.
The line is to be drawn somewhere in class 119.5-129.5 (containing 30 units
of area). To the left of this class there are 10 + 20 = 30 units of area. We need
20 more units of area to make up the 50.
We need therefore to divide the 30 units of area of the class 119.5-129.5 in
the ratio 20: 10 = 2 : 1.
To divide AB in the ratio 20 : 10 = 2 : 1 you first locate the point P on AD 20

unit in vertical direction.
Now AP : PD = 2 : 1.
Draw the diagonal AC. This diagonal meets the line PR at Q. Drop from Q a
vertical to AB. The foot of this vertical (M) is the estimate for the median.
This can be proved as follows using the similar triangles APQ and ACD. This
implies AP : AC = PQ : CD = 2 : 3

But PQ = AM and AB = DC so also AM : AB = 2 : 3 or AM : MB = 2 : 1.
The estimate for the median is therefore read at the point M.

1. The times (to nearest tenths of a second) taken by pupils to run 50 m is
tabulated in the following frequency distribution table.
Time(s) Number of pupils
9.0-9.9 1
10.0-10.9 4
11.0-11.9 6
12.0-12.9 7
13.0-13.9 12
14.0-14.9 11
15.0-15.9 6
16.0-16.9 3
a) Calculate an estimate for the median.

b) Represent the data in a histogram.
c) Use your histogram to obtain an estimate of the median.
2. Obtain an estimate for the median
(i) by calculation
(ii) from a cumulative frequency curve
(iii) from a histogram
The time taken by 110 pupils to complete a mathematics assignment (to
nearest minute) is represented in the following frequency distribution
table.
Time (min) Number of pupils
5 - 14 10
15 - 24 14
25 - 34 40
35 - 44 31
45 - 54 5

Section H: Estimation of the mode from grouped data
When data has been grouped into classes the class with the highest frequency
can easily be identified: the modal class. An estimate of the mode can be
made from the modal class.
1. Geometrical estimate of the mode from a histogram:
The length of plants (cm) on a plot one month after planting showed the
following distribution.
Length 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90

Frequency 20 40 80 100 50 20 10 10
Representing the data in a histogram gives the following:
The modal class is 41 -50 cm.

An estimate of the mode can be found from the histogram by drawing the
lines as illustrated in the diagram. This gives as estimated mode 43.
2. Estimation of the mode by calculation:
The modal class contains 20 more than the class below and 50 more than the
class above the modal class. We therefore assume that the modal class is
divided by the estimated mode in the ratio 20 : 50 = 2 : 5.
2 6
The calculated estimate is therefore 40.5 + (10) = 40.5 + 2 43.4
7 7
The modal length of the plants is 43.4 cm.
3. Estimate of the mode from the cumulative frequency curve:
As the modal class is the class with the highest frequency on a cumulative
frequency curve the cumulative frequency will increase fastest at the mode
(the rate of increase of the cumulative frequency is highest at the mode).
Calculus teaches us that in the case of a cumulative frequency type of curve

the mode must be located at the point of inflexion of the curve (the point
where the direction of the curvature changes from concave to convexor the
other way round). Locating as best as possible the point of inflexion on a
cumulative frequency curve will give an estimate of the mode. The tangent at
the point of inflexion passes through the curve.

1. For the calculation you need the modal class and the class below and
above of it. Using the diagram below show that the calculated estimate of
d1
the mode is given by (lower class boundary L) + (class width)
d1 + d 2
2. As calculated estimate of the mode is frequently used:

calculated estimate of mode =
3 calculated estimate of median - 2 calculated estimate of mean
Investigate the validity of this relation.
3. For the plant height data used above draw a cumulative frequency curve
and use it to find an estimate of the mode.
The length of plants (cm) on a plot one month after planting showed the
following distribution.
Length 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90
Frequency 20 40 80 100 50 20 10 10

4. The length cars remained in a parking lot of a supermarket was recorded
during a day to the nearest minute. The results are tabulated below.
Length of
6-25 26-45 46-65 66-85 86-105 106-125 126-145
stay (min)
Frequency 60 70 90 120 80 50 40
a) Represent the data in a histogram.

b) Use the histogram to obtain an estimate for the mode.
c) Calculate an estimate of the mode.
d) Make a cumulative frequency table and draw the cumulative
frequency curve.
e) Use the cumulative frequency curve to obtain an estimate for the
mode.
Section I: Boxplots or box and whisker diagrams

An average summarises all the collected data in a single value (mode, median
or mean). This obviously leads to loss of information as conveyed by the
original data. Reducing the data to five numbers chosen from across the range
of values is more informative. A five number summary gives the minimum
and maximum values (the extremities) together with the lower quartile, the
median and the upper quartile. These five values can be illustrated in a box
plot, also called box and whisker plot or diagram.
The stem-leaf diagram illustrated the marks scored in a maths test.
1 67
2 1368
3 1122335589
4 01122345678
5 0012667899
6 234566
7 01379
8 05
9 2
n = 50 1 | 6 represent 16 marks.
To represent this data in a boxplot first find the five measures.
Minimum score 16
Maximum score 92

46 + 47
Median is mean of 25th and 26th score: = 46.5 (Q2)
2
The lower quartile is the median of the lower 25 observation, i.e., the 13th
which is 35 (Q1).
The upper quartile is the median of the upper 25 observations, i.e., the 38th
which is 62 (Q3).
Represented in a box plot:
Note that 25% of the pupils scored less than Q1 = 35 (represented by the
lower whisker).
50% of the pupils scored between Q1 = 35 and Q3 = 62 (represented by the
box).
25% of the pupils scored more than Q3 = 62 (represented by the upper
whisker).
The diagram also illustrates the lowest (16) and the highest score (92).
The box illustrates that of the middle 50% of the pupils, 255 scored between
35 and 46.5 (the lower part of the box) and 25% between 46.5 and 62 (the
upper part of the box).
The diagram not only illustrates the measures of central tendency but simple
measures of the amount of spread (variability) can be obtained from the
diagram:
the range (maximum - minimum)

the interquartile range (IQR) Q3 - Q 1
1
semi-interquartile range (semi-IQR) (Q3 - Q1)
2

1. On a stretch of road with a 60 km/h limit the following speeds of cars
were measured (in km/h).
57 53 53 71 73 54 69 56 58 49
56 53 52 82 62 61 60 71 75 60
57 61 58 78 64.
a) Represent the data in a stem-leaf plot.
b) Use your stem-leaf plot to obtain the median speed, the upper quartile
and the lower quartile speed.
c) Represent the data calculated in a box plot.
d) What does it imply that one whisker is longer than the other?
e) Explain why the median is not in the centre of the box.
f) What percent of the drivers was speeding over the limit?
Practice task 3
1. Discuss what you consider the most effective method to facilitate the
learning of data handling. Illustrate with example activities.
2. a) Collect test data on the same topic from two parallel classes.
b) Represent the data in (i) grouped frequency table (ii) histogram (iii)
frequency polygon (both sets of data on the same axes) (iv) double
stem-leaf plot.
c) Which of the representations do you feel best represents the data?
Justify your choice.
d) Calculate (i) the exact value of the mean (ii) an estimate of the mean
from the grouped frequency table (iii) the percent error in the
estimated value.
e) Which of the three averages, mean, mode or median, best represents
the data? Explain.
f) Represent the data of both classes in a box plot.
g) What conclusions can you safely draw from the data?
3. a) Collect data for your school on the ages of the students by gender.
b) Present your data in two frequency polygons, one for the girls and one
for the boys, using the same axes.
c) Calculate an estimate of the mean age of (i) boys (ii) girls in your
school.

d) Comparing the data for boys and girls comment on any differences
and try to find an explanation for the differences.
4. Obtain the height of all pupils in your class. Investigate the effect of
choosing different class intervals on the estimated mean height. Compare
with the actual mean obtained from the raw data. What conclusion can
you reach?
Summary
This unit began with a project-based approach to the teaching of central
tendency. It ended with a number of self-marking exercises to teach you some
lesser-known ways of representing quantitative data. It is hoped that you, and
eventually your students, will benefit from this practical approach to
statistics. Remember this caution from the Introduction to the unit: no set of
projects could ever teach your students all, or even most, of the techniques we
have covered. But since your students will benefit more (in later life) from the
projects, a wise teacher omits many techniques of statistics in order to leave
room for completion of interesting projects.

Unit 4: Answers to the self mark exercises

1. Mean height 1.61 cm (2 dp)
2. 11.4 pips (1 dp)
3. Mean 2.75 h, median 2.5 h, mode 2 h
Use mean (majority of pupils 15 spend between 1 3 hours) or median
(half of the pupils spend less than 2.5 h, half more)
4. Mean 2.6 days, median 2 days, mode 1 day.
Half of the pupils that were absent were absent for one day. Most
common is that if a pupil is absent it is just for one day. Mode best to use.
5. Girls: Mean height 160.0 cm (1 dp), Mode trimodal: 153 cm, 154 cm and
162 cm, median height 159.5 cm.
Boys: Mean height 165.4 cm, multi modal 162 cm, 165 cm, 166 cm, 168
cm, median height 165 cm.
Generally boys are taller than girls (higher mean and median). There is no
most common height (the distributions have no single mode).
6. a) Mean 2.3 persons, mode 1 and median 2

b) A bar or bar line graph would be appropriate (discrete data,
ungrouped).

7. a) Mean mass 30.8 g, mode 30 g, median 31 g
b) Frequency
8. a) Team A: mean 12.86 s, bimodal 13.3 s & 12.8 s, median 12.9 s

Team B: mean 12.91 s, no mode, median 12.7 s
b) Median or mean time for team A (two 1 dp they are the same)
Median for team B 12.7 s.
c) Team B with the lower median, half their runners run below 12.7 s, in
team A this is 12.9 s.

1a. Mode, giving the most common case.

b & c) Median (half of pupils names are longer, half shorter than that
number) or mean if there are no outliers.
d) Mode, the most common size will be of interest to traders.
e) Mode, the subject liked by most pupils. Median, mean non existing.
f) None of them might be very useful as most pupils will not have been
absent. It makes sense to leave out those never absent (say 80%) and
use the mode for those who were absent for one or more days.
g) Mode, giving the most common method. Mean and median non
existing.
h) Mode is the only average available for this type of data.
2. (i) Mean 1.75, mode 1, median 1
Mode / median as mean is influenced by outlier 6.
(ii) Mean 59.9 % (1 dp), mode 75%, median 68%
Median might be best reflection of pupils attainment. Mean is influenced
by outliers. For few data mode does not make much sense.
3. Arrangements Examples
mode<median<mean 1, 1, 4, 6 mode 1<median 2.5<mean 3
mean<median<mode 1, 3, 6, 9, 1 mean 6<median 8<mode 9
mode<mean<median 1, 1, 7, 8, 13 mode 1<mean 6<median 7
median<mean<mode 0, 1, 4, 10, 10 median 4< mean 5<mode 10
median<mode<mean 0, 3, 7, 8, 8, 28 median 7.5<mode 8<mean 9
mean<mode<median -28, -8, -8, -7, -3, 0 mean -9<mode -8<median -7.5
4. Mode
Advantages
Simple to understand
Not affected by extreme values (outliers)
Only one that can be used for qualitative data
Is an actual observation data
Disadvantages
Cannot be used in calculations or combined with mode of similar
distributions
Might not exist or distributions might be multiple modal
Mean
Advantages
Includes all the values of the distribution
Allows use in further calculations (e.g. SD)
Allows combining with results from other similar groups
Disadvantages
Sensitive to outliers in the distribution. This might give a distorted
picture.

Median
Advantages
Easy to understand
Not affected by extreme values
Disadvantages
Cannot be used in further calculations or combined with median of
similar distributions
5. a) If p and q are positive integers:
(p q)2 0
p2 2pq + q2 0
p2 2pq + q2 + 4pq 4pq
p2 + 2pq + q2 4pq
(p + q)2 4pq
p + q 2 (pq)
p+ q
(pq)
2
Arithmetic mean of p and q Geometric mean of p and q
b) Arithmetic mean A=
p+ q
2A = p + q
2
Geometric mean = pq G2 = pq
1 1 1 1 p+ q 1 pq G
= + = = =
H p q H pq H p + q 2A

1. a)
Mass(g) 45m<50 50m<55 55m<60 60m<65 65m<70 70m<75
Frequency 4 8 12 16 5 2
b) 47
c) modal class 60 m < 65
d) Median in class 55 m < 60
e) Estimate of mean 59.2 g
2. a) 95
b)
Diameter (cm) 5d<6 6d<7 7d<8 8d<9 9d<10
Frequency 20 50 120 95 15

c) 3000 oranges d) Modal class 7 d < 8. e) 7.6 cm
3. a) 20
b)
Height (cm) 100h<109 110h<119 120d<129 130d<139 140d<149
Frequency 10 20 30 35 5
c) 100 d) 130 h < 139 e) 125 cm

4. a) 14 a < 15 and 15 a < 16
b) 80
c)
Age (years) 11a<12 12a<13 13a<14 14a<15 15m<16 16a<17
Frequency 10 90 100 150 150 80
Age (years) 17a<18 18a<19

Frequency 40 20
d) 14.8 years
5. a) 24 b) 14 length < 18
c)
Length (m) 10 l <12 12 l < 13 30 l < 14 14 l< 18 18 l < 20
Median (26th observation) in class 14 length < 18
d) 51 e) 743.5 m f) 14.6 m
6. a) Modal class 70 m < 100
b)
Mass (g) 30 m <50 50 m < 60 60 d < 70 70 d< 100
Frequency 80 80 70 120
c) 62.1 g
7. a) 250 I < 500
b)
Income (P) 250 I <500 500 I < 1000 1000 d < 2000 2000 d< 5000
Frequency
0.3 0.1 0.04 0.004
density
Frequency 75 50 40 12
c) 500 I < 1000
d) P947
e) Mode, the salary earned by most people

8. a)
b) 168.7 cm
9. a) Boundaries of bars 10, 15, 20, 25, 35, 45, 60 and 75
Frequency
Density 5.6 13 16.4 7.6 5.4 2.9 0.8
9. b) 30.1 years

1. a) (i) 133 cm (ii) LQ 129 cm, UQ = 137.5 cm
b) IQR 8.5 cm
c) 120
d) 40

e) Height 120 124 125 129 130 134 135 139 140 144
Frequency 10 40 70 65 15
f) Frequency polygon
g) (i) & (ii) (listed under advantages)

Frequency table
Advantages: Overview of data
Needed to draw various graphical representations
Calculations can be based on the table
Allows to obtain mean, median and mode (if grouped estimates of
these measures can be obtained by linear interpolation procedures)
Disadvantage: Difficult to get an overall idea of the distribution.
Histograms
Advantages: For display of continuous grouped data (also used for
discrete grouped data), especially if classes over of unequal width.
Graphical estimates of median and mode can be obtained from the
histogram.
Disadvantage: Difficult for pupils: where to take the class
boundaries?
Frequency polygon:
Advantages: Gives impression of the distribution
Useful for comparison: more than one frequency polygon on the same
axes (e.g. height of boys and girls).
Easy to plot using points with co-ordinates (midpoint of interval,
frequency)
Disadvantage: No use for calculation of statistics
Cumulative frequency polygon
Advantage: Useful for obtaining estimates of median, quartiles,
percentiles and mode
Disadvantage: Rather time consuming to draw
(iii) All have their specific use depending on what one wants to illustrate /
calculate or estimate
2. a) (i) 4.8 minutes (ii) LQ 2.9 minutes, UQ 6.0 minutes
b) 3.1 minutes
c) 25

d)
Number of 0 t < 2 2 t < 4 4 t < 6 6 t < 8 8 t < 10 10 t < 12
minutes late
Number of 10 14 21 10 3 2
days
e)
f) 4.6 minutes
3. a)
Time t 0< t 20 20< t 40 40 < t 60 60 < t 90 90 < t 120
CF 12 54 132 154 160
b)
c) (i) 50 (ii) 35 (iii) 58

4. a)
Time t (min) 0< t 10 10< t 20 20 < t 30 30 < t 40 40 < t 50
CF 32 92 146 182 200

b)
c) (i) 22 (ii) 13 (iii) 30

1. a) 4.8
b) LQ 2.8 UQ 7.2
2. a) boys 174,1 cm girls 167.1 cm
b) boys LQ 168.3 cm, UQ 179.6 cm girls LQ 161.6 cm, UQ 171.3
c) (i) 172.2 cm (ii) 180 5 cm
3. a) 43 or more b) less then 13
4. a) Mean 33.6 mm Median 36 mm
4b Class interval Frequency 4c Class interval Frequency
10 l 14 2 10 l 19 10
15 l 19 8
20 l 24 5 20 l 29 10
25 l 29 5
30 l 34 4 30 l 39 10
35 l 39 6
40 l 44 8 40 l 49 15
45 l 49 7
50 l 54 4 50 l 59 5
54 l 59 1
b) Mean 34.9 mm. Median 35.7 mm
c) Mean 33.5 mm. Median 35.0 mm
d) In this example both have reduced with increased class width. (It is a
good investigation to find out whether or not that is always true!)

1. a) 13.53 s.
c) Median in class with boundaries 12.95 13.95.
Total area 50 units, to be divided into two.
Left of the class is already 18 units, required 7 more.
Draw accurate diagram in your histogram as illustrated above. The

median should be close to the calculated value of 13.53.
2. (i) 29.4 minutes

Hint. The class width is divided in the ratio d1 : d2.
3. Mode 45.
4. a) Use as class boundaries 5.5, 25.5, 45.5, etc. Class width is 20.
Modal class 66 85. Use construction as illustrated in question 1.
Estimated mode 74.1.
d, e) Read length on horizontal axis at the point of inflexion as illustrated
in question 3.

1. a)
4 9
5 23334
5 667788
6 001124
6 9
7 113
7 58
8 2
n = 25 6 | 7 represent 67 km/h
Median 60 km/h, LQ 55 km/h, UQ 70 km/h
c)
d) More drivers had a speed beyond the upper quartile speed than driver
with a speed below the lower quartile speed.
e) The 25% of drivers with speed above the median speed of 60 km/h
had a wider spread (60 to 70 km/h) than the 25% of drivers with a
speed below the median speed (range 55 to 60 km/h).
f) 50%

Unit 5: Measures of dispersion
Introduction to Unit 5
In the previous unit you learned how to describe data sets using measures of
central tendency. However a measure of central tendency cannot describe the
data in sufficient detail. Look at the following two sets of data representing
marks (out of 25) of two pupils on three different tests.
Pupil 1 scored 11, 12, 13 and pupil 2 scored 1, 12, 23. Both pupils have the
same mean (12) and the same median (12), but can we say that they
performed equally well? Pupil 1s marks are all very close together, while
the marks of pupil 2 are widely spread (from 1 to 23). You could say that
pupil 1 is more consistent in performance than pupil 2. It is the range or
spread of marks which gives us this information. In this unit you are going to
look at different measures of spread or dispersion.
Purpose of Unit 5
The main aim of this unit is to look at some basic measures of spread: how to
calculate them and how to interpret them. This unit covers range, inter
quartile range, variance and standard deviation. Box plotsas covered in
Unit 4are a useful graphical aid to visualise spread of data.
Objectives
When you have completed this unit you should be able to:
calculate the range and inter quartile range of ungrouped data
obtain an estimate of the inter quartile range of grouped data
calculate the standard deviation and variance of ungrouped data
calculate an estimate of the standard deviation and variance of grouped
data
use measures of central tendency and of spread to compare sets of similar
data
Time
To study this unit will take you about five hours.
Module 6: Unit 5 170 Measures of dispersion

Unit 5: Measures of dispersion
Section A: Interquartile range for ungrouped data
The simplest measure to describe the spread or dispersion of values is the
range. The range is the difference between the lowest and the highest values.
The problem with the range is that only two values are used and so it can
give a wrong impression if one (or both) of the values is very high or very
low.
The problem of distortion by extreme values can be overcome by calculating
the range of the central half (middle 50%) of the values. This is called the
interquartile range.
Example 1
7 students estimated the length of a book to the nearest cm. In order their
estimates were: 25 cm 28 cm 30 cm 31 cm 32 cm 34 cm and 37 cm.
The range is 37 25 = 12 cm.
7+1
The median is the middle value, = 4th value, which is 31 cm.
2
1
The lower quartile (LQ), the value away in the list of values, so the value
4
7+1
of the = 2nd term, which is 28 cm.
4
3
The upper quartile (UQ), the value away in the list of values, this is the
4
3
(7 + 1) = 6th term, which is 34 cm.
4
The interquartile range is (UQ) - (LQ) = 34 - 28 = 6 cm.
Example 2
The estimate of the length of the book by 10 students was in order to the
nearest cm: 24 24 26 28 29 31 32 33 34 36.
The range is 36 24 = 12 cm.
10 + 1
As = 5.5, the median is the average of the 5th and 6th term in the
2
29 + 31
ordered sequence. Median is = 30 cm.
2

10 + 1
As = 2.75, the lower quartile is the value of the third term which is
4
26 cm.
Similarly the upper quartile value is the value of the 8th which is 33.
The interquartile range is 33 - 26 = 7 cm.

1. During a term a pupil obtained the following percent marks.
Setswana 56 49 63 58 52 50 57 61
English 61 70 53 60 57 52 48 79 65
Science 68 56 58 73 39 47 55 76
Maths 45 46 42 48 40 45 44 41 47
a) Find for each subject the range and interquartile range.
b) In which subject is the pupil most consistent. Explain.
c) What is the pupils best subject? Explain.
Section B: Interquartile range for grouped data

If the data is grouped, you use either:
(i) a cumulative frequency curve to obtain estimates of the lower and upper
quartile (see section G1on the cumulative frequency curve in unit 4)
(ii) linear interpolation to obtain estimates of the lower and upper quartiles
(See section G2 on linear interpolation in unit 4)
From these an estimate for the inter quartile range can be obtained.

1. The height of a group of pupils is distributed as in the table below.
Height (cm) 151-155 156-160 161-165 166-170 171-175

a) Make a cumulative frequency table and draw a cumulative frequency
curve of the data.
b) Use the cumulative frequency curve to obtain an estimate for the
2. A machine is to produce nails of 7 cm length. A sample is taken and
measured to the nearest 0.1 cm. The results are tabulated:
Length of nail 6.7-6.8 6.8-6.9 6.9-7.0 7.0 - 7.1 7.1-7.2
curve of the data.
3. The ages of people attending a football match were distributed as in the
following table.
Age Frequency
5-9 8
10-14 26
15-19 74
20-24 90
25-29 124
30-34 142
35-39 86
40-44 54
45-59 26
60-74 15
curve of the data.
4. Two types of batteries were tested on the number of hours they lasted.
Number of hours 5-10 10-15 15-20 20-25
Type A 8 37 43 12
Type B 16 30 32 22
a) Calculate estimates for the quartiles and obtain an estimate for the
interquartile range of each type of battery.
b) Which type would you recommend a school to buy? Justify your
answer.

Section C: Standard deviation of ungrouped data
There are three commonly used measures of dispersion. You know already
two of them: range and interquartile range.
The disadvantage of both of these measures is that not all data are used. For
the range you use only the two extreme values: the highest and the lowest
and this can be misleading.
Using the interquartile range ignores the top and bottom quarter of the
values.
A measure of spread using all the data is the standard deviation. It uses the
differences (deviations) of the data from the mean.
To calculate the standard deviation you calculate
(i) the mean of the data
(ii) the deviations of the data from the mean
(iii) the mean of the squares of the deviations (which is called the variance)
(iv) the square root of the variance which is the standard deviation
Calculation of the standard deviation is best done using a table or an
electronic device (calculator or computer). (Only the electronic method is
worth remembering! Do not assess your students on their ability to calculate
a standard deviation by hand.)
Disadvantage of using the standard deviation is that it is difficult to
understand intuitively.
Example
The length of six leaves from a certain tree were measured to the nearest cm.
7 cm, 9 cm, 11 cm, 12 cm, 12 cm, 14 cm.
x x x ( x x)2
7 -4 16
9 -2 4
11 0 0
12 1 1
12 1 1
15 4 16
x = 66 38
x 66
x== = 11
6 6
x means sum of all the data x. x 2
means sum all the squares of the
data x.
Variance = s2 =
(x x) 2
=
38
n 6

38
The standard deviation is s.d. = = 2.5 cm (1 dp)
6
A formula for the standard deviation is s.d. =

( x x) 2
At times the form s.d. =

x 2
x 2 is more convenient to use.

n
Practice task 1
Derive the form

x 2
x from
2 ( x x) 2
. Remember that
x = x
n n n
Suggested answer at the end of this unit.
Section D: Standard deviation of grouped data

The standard deviation of a set of data occurring with given frequencies can
be found as illustrated in the following example.
Example
A sample of 60 batteries was tested as to how long (in hours) they lasted (the
life span of the battery).
The results are in the table.
Battery life span (h) 12 13 14 15 16 17 18
Frequency 6 8 10 14 15 5 2
The working can be laid out as follows.
x x2 f fx fx2
12 144 6 72 864
13 169 8 104 1352
14 196 10 140 1960
15 225 14 210 3150
16 256 15 240 3840
17 289 5 85 1445
18 324 2 36 648
x= x2= f= fx= fx2=
105 1603 60 887 13259

The mean is
fx = x = 887 = 14.78 h (2 dp)
f 60
The standard deviation s.d. =

fx 2
x2 =
13259
(14.78) 2 = 1.6 h (1 dp)
f 60
If the data is grouped or continuous an estimate of the standard deviation

can be computed by assuming that all the data in a particular class interval
has the value of the mid-point of the class interval. If the mid-interval value
is indicated by m, the relation for mean and standard deviation becomes:
fm = m and s.d. =
fm 2
m2
f f

1. In a test (maximum marks 50) the distribution of the marks was as
follows.
Marks 1 - 10 11 - 20 21 - 30 31 - 40 41 - 50
Frequency 2 14 22 26 16
a) Copy the table below. Complete the column for the mid-interval
values and the other columns.
Mid-interval
Marks Frequency f m2 fm fm2
values m
1 10 2 5.5 30.25 11 60.5
11 20 14
21 30 22
31 40 26
41 50 16
f = fm = fm2 =
b) Calculate an estimate of the mean mark and the standard deviation.

c) Explain why mean and standard deviation are estimates and not the
exact value.
2. The time spent by customers in a supermarket was measured and the
distribution was as shown:
Time (min) 0 < t 10 10 < t 20 20 < t 30 30 < t 40 40 < t 50
Frequency 22 86 62 21 9
Calculate an estimate of the mean time spent by customers in the
supermarket, and the standard deviation.

3. The height of 12-year-old boys and girls in a school was measured. The
heights were distributed as in the table below.
130 < h 135- 140- 145- 150- 155- 160- 165- 170- 175-
Height
135 140 145 150 155 160 165 170 175 180
Frequency
0 2 5 17 24 31 36 28 6 1
for boys
Frequency
1 6 8 20 32 32 28 18 4 1
for girls
a) Using the same axes draw a frequency polygon for the heights of the
boys and of the girls.
b) Use the frequency polygon to compare the heights of boys and girls.
c) Calculate an estimate of the mean height and standard deviation for
both boys and girls.
d) Compare the height of boys and girls using your estimated values of
mean and standard deviation.
4. Two novels were compared with each other by counting the number of
words in the sentences in a section of the novel.
Number of words 5 - 9 10 - 14 15 - 19 20 - 24 25 - 29 30 - 34 35 - 39
Novel A 12 15 10 26 14 8 4
Novel B 8 18 12 31 18 4 2
a) Calculate for each novel an estimate for the mean number of words in
a sentence and the standard deviation.
b) Compare the results of a. Which novel do you think is easier to read?
Justify your answer.
5. The number of hours spent on sports by two groups of pupils, group A
and group B, are represented in the histograms below.
Which of the data represented in the histograms has a greater standard
deviation? Justify your answer.

6. Investigate what happens to mean and standard deviation of a set of
data if
(i) to each value the same number N is added
(ii) each value is multiplied by the same number k
7. a) Find the mean and standard deviation of 12, 15, 16, 14, 17, 13.
Using the result of 5(i) write down the mean and standard deviation of
b) 22, 25, 26, 24, 27, 23
c) 83, 86, 87, 85, 88, 84
d) 8, 11, 12, 10, 13, 9
8. a) Find the mean and standard deviation of 52, 61, 73, 68, 49, 67.
Using your result from 5(ii) write down the mean and standard
deviation of
b) 5.2, 6.1, 7.3, 6.8, 4.9, 6.7
c) 26, 30.5, 36.5, 34, 24.5, 33.5
d) 208, 244, 292, 272, 196, 268

Practice task 2
1. a) Collect data on the height of the boys and girls in one or two of your
classes.
b) Calculate mean and standard deviation for boys and girls separately
from the raw data.
c) Make a grouped frequency table separating boys and girls. Use class
widths of 5 cm and 10 cm.
d) Use the grouped frequency tables to calculate an estimate for the
mean and the standard deviation for boys and girls separately for
both class widths.
e) Calculate an estimate of the interquartile range from both the grouped
frequency tables.
f) Using your calculated data, compare and make some valid
statements.
g) Comparing the exact values of mean and standard deviation with the
estimated values from the grouped frequency tables (with width 5
cm and width 10 cm), which of these three do you consider to
represent the data best?

Unit 5: Answers to the self mark exercises

1a)
Subject Range IQR
Setswana 63 49 = 14 59.5 51 = 8.5
English 79 48 = 31 67.5 52.5 = 15.0
Science 76 39 = 37 70.5 51 = 19.5
Maths 48 40 = 8 46. 5 41. 5 = 5.0
b) Mathematics, smallest range and IQR

c) English with median 60%

1a)
Height 151 - 155 156 - 160 161 - 165 166 - 170 171 - 175
Cumulative
6 15 29 52 60
Frequency
b) Plot the points (155.5,6), (160.5, 15), (165.5, 29), (170.5, 52), (175.5, 60)
c) IQR 168.5 160.0 = 8.5 cm
2a) Plot (6.85, 4), (6.95, 15), (7.05, 51), (7.15, 95), (7.25, 100)
b) IQR 7.09 6.99 = 0. 1 cm
3a) Plot (10, 8), (15, 34), (20, 108), (25, 198), (30, 322), (35, 464), (40, 550),
(45, 604), (60, 630), (75, 645)
b) IQR 36 23 =13 years.
4a) Use linear interpolation for the estimates (to 2 dp)
LQ UQ IQR Median
Type A 12.30 18.49 6.19 15.58
Type B 11.50 19.53 8.03 15.63
b) Type A , smaller IQR and more consistent in performance [as both types
have the same median (to 1 dp)]

Practice task 1
( x x) 2
=
( x 2
2 xx + x 2 )
=
x 2

2 nx x
+
nx 2
n n n n n
=
x 2
2x2 + x2 =
x 2
x2
n n

1b) mean 30.5 SD 10.7
c) It is not known how the frequencies are distributed over each interval.
Mid interval values were used, i.e., assuming that the mid interval value
had the indicated frequencies.
2. Mean 20.5 minutes (1 dp) SD 9.7 minutes
3c) Boys Mean 158.6 cm, SD 8.0 cm (1 dp)
Girls Mean 156.0 cm, SD 8.6 cm
d) Boys are on average taller than girls. The heights of the girls have a
wider spread around their mean than is the case for the boys.
4a) Novel A : Mean number of words per sentence 20.1, SD = 8.3 (1 dp)
Novel B : Mean number of words per sentence 19.9, SD = 7.2 (1 dp)
b) Novel B lower mean and less spread about the mean.
5. Group A, wider spread than group B
6. (i) Mean increases by N, SD remains unchanged.
(ii) Mean and standard deviation change by factor k.
7a) Mean 14. 5, SD 1.7 (1 dp)
b) Mean 14.5 + 10 = 24.5, no change in SD
c) Mean 14. 5 + 71 = 85.5, no change in SD
d) Mean 14.5 4 = 10.5, no change in SD
8a) Mean 61.67, SD = 8.67 (2 dp)
b) Mean and SD of 8a divided by 10: Mean 6.17, SD 0.87 (1 dp)
c) Mean and SD of 8a divided by 2: Mean 30.8, SD 4.3 (1 dp)
d) Mean and SD of 8a multiplied by 4: Mean 246.7, SD 34.7 (1 dp)

References
Cockcroft W. H., Mathematics Counts, 1982, HMSO London
Additional References
In preparing the materials included in this module we have borrowed ideas
extensively from other sources and in some cases used activities almost
intact as examples of good practice. As we have been using several of the
ideas, included in this module, in teacher training over the past 5 years the
original source of the ideas cannot be traced in some cases. The main sources
are listed below.
Mathematics Teacher, Journal of the National Council of Teachers of
Mathematics
Mathematics in School, Journal of the Association of Teachers of
Mathematics
NCTM, Dealing with Data and Change, 1991, ISBN 087 353 3216
NCTM, Data Analysis and Statistics, 1992, ISBN 087 353 3291
Owens, D. T., 1993, Research Ideas for the Classroom,. Middle Grades
Mathematics, NCTM ISBN 002 895 7954
Further reading
Bank, T. et al. 1999, Mathematics for SEG GCSE Intermediate Tier,
Causeway Press Limited, ISBN 187 392 9870.
Maths in Action, 1999, Intermediate 1, Nelson 017 431 4973
Maths in Action, 1999, Intermediate 2, Nelson 017 431 4949
Maths in Action Statistics for Higher Mathematics, Nelson 017 431 4965
Module 6 182 References

Glossary
Census collection of data on the whole population
Class interval the width of the groups used in grouped frequency
tables
Continuous data data that can take any value within a certain range
(e.g., height / mass of persons)
Data facts, numbers, measures collected on a population or
sample
Descriptive Statistics branch of statistics covering collecting, representing
and analysing of data
Discrete data data that can take only specific values (e.g., shoe
size) or falls in specific categories (e.g., sex)
Estimation theory theory that describes how the statistics obtained on a
sample can be used to estimate the parameters of the
population
Experimental data data collected using a scientific experimental design,
frequently in the form of an experimental group and a
control group
Frequency table a way of collating the information recorded on a data
collection sheet
Hypothesis a statement which may or may not be true
Inferential Statistics branch of statistics dealing with drawing conclusions
from data, testing hypotheses, etc.
Nominal classification into categories using words
/descriptions. Non numerical data
Ordinal data data that can be placed in an order, e.g., taste of
oranges from very sweet to sour
Parameter a single fact (numerical or nominal) for the whole
population
Population the entire collection of objects with at least one
similar characteristic
also: set of all the possible observations
Qualitative data data which can only be described in words
Quantitative data data that has a numerical value
Questionnaire a set of questions used to collect data in a survey
Random sampling method of obtaining a sample such that each member
of the population has an equal chance of being
included
Module 6 183 Glossary

Sample portion of the entire collection of objects of similar
characteristics
also: collection of data from a subset of the
population
Simulation method of collecting data using random number to
model a real life situation
Statistics a single fact (numerical or nominal) obtained from a
sample
Survey method of collecting data using, e.g., questionnaires,
interviews, tests, observations, secondary sources
Tally a way of recording each item of data on a data
collection sheet
Variable characteristic that varies over the population
Module 6 184 Glossary

JSMath6 Part3

Uploaded by

Copyright:

Available Formats

JSMath6 Part3

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

JSMath6 Part3

Uploaded by

Copyright:

Available Formats

4. Give 5 different numbers such that their average is 21.

The numbers are:

Lesson outline 1: Mean, median, mode

Module 6: Unit 4 121 Measures of central tendency

Module 6: Unit 4 122 Measures of central tendency

Module 6: Unit 4 123 Measures of central tendency

Module 6: Unit 4 124 Measures of central tendency

Module 6: Unit 4 125 Measures of central tendency

A indicates will always change

Module 6: Unit 4 126 Measures of central tendency

Module 6: Unit 4 127 Measures of central tendency

Module 6: Unit 4 128 Measures of central tendency

Module 6: Unit 4 129 Measures of central tendency

Module 6: Unit 4 130 Measures of central tendency

Module 6: Unit 4 131 Measures of central tendency

DISCUSS IN YOUR GROUP

Module 6: Unit 4 132 Measures of central tendency

Module 6: Unit 4 133 Measures of central tendency

Module 6: Unit 4 134 Measures of central tendency

Module 6: Unit 4 135 Measures of central tendency

No. of correct answers 1-10 11-20 21-30 31-40 41-50 51-60

Mark Frequency Mid-value Mid-value Frequency

Module 6: Unit 4 136 Measures of central tendency

Module 6: Unit 4 137 Measures of central tendency

a) Make the grouped frequency table.

Continued on next page

Module 6: Unit 4 138 Measures of central tendency

Module 6: Unit 4 139 Measures of central tendency

a) How many pipes with length L m, 14 L < 18 were made?

Module 6: Unit 4 140 Measures of central tendency

a) What is the modal class interval?

Continued on next page

Module 6: Unit 4 141 Measures of central tendency

Module 6: Unit 4 142 Measures of central tendency

Module 6: Unit 4 143 Measures of central tendency

Reading from cumulative frequency curves

Module 6: Unit 4 144 Measures of central tendency

a) From the cumulative frequency graph obtain:

Module 6: Unit 4 145 Measures of central tendency

a) From the cumulative frequency graph obtain:

Module 6: Unit 4 146 Measures of central tendency

Module 6: Unit 4 147 Measures of central tendency

Estimate the height of the tallest 10% of plants.

Section G2: Estimation of median, quartiles and

Module 6: Unit 4 148 Measures of central tendency

At the beginning of the interval you have covered already 12 + 43 = 55

Module 6: Unit 4 149 Measures of central tendency

Module 6: Unit 4 150 Measures of central tendency

To divide AB in the ratio 20 : 10 = 2 : 1 you first locate the point P on AD 20

Module 6: Unit 4 151 Measures of central tendency

Self mark exercise 6

a) Calculate an estimate for the median.

Suggested answers are at the end of this unit.

Module 6: Unit 4 152 Measures of central tendency

Length 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90

The modal class is 41 -50 cm.

Module 6: Unit 4 153 Measures of central tendency

Self mark exercise 7