Chapter 10

Descriptive Statistics

10.1 Introduction

Statistics is concerned with the scientific method by which information

is collected, organised, analysed and interpreted for the purpose of de-
scription and decision making.
Examples of firms using statistics are; RTSA, ZRA, CSO, Insurance
companies etc.
There are two subdivisions of statistical methods.
(a) Descriptive Statistics - It deals with the presentation of numerical
facts, or data, in either tables or graphs, and with the methodology of
analysing the data.
(b) Inferential Statistics - It involves techniques for making inferences
about the whole population on the basis of observations obtained from
samples.

10.2 Some Basic Definitions

Definition 10.2.1. Population

A population is the group from which data are to be collected. For
example, Cars, Animals, Trees, etc.

Definition 10.2.2. Sample

A sample is a subset of a population. For example, Saloons, Cows,
Mukula trees, etc.
Definition 10.2.3. Discrete Data
A discrete data is one which no value may be assumed between two
given values, for example, number of children in a family.
Discrete data are countable in a finite amount of time. Thus, discrete
data takes only exact values. Cars, tomatoes etc.

Definition 10.2.4. Continuous Data

A continuous data is one which any value may be assumed between
two given values.
Continuous data cannot take exact values but can be given only within
a speficified range or measured to a specified degree of accuracy. For
example, the time taken for 100-meter race, height, speed of a vehicle
etc. run.

10.3 Frequency Distribution

Statistical data obtained by means of census, sample surveys or experi-

ments usually consist of raw, unorganized sets of numerical values. Be-
fore these data can be used as a basis for inferences about the phe-
nomenon under investigation or as a basis for decision making, they
must be summarized and the pertinent information must be extracted.
A traffic inspector has counted the number of automobiles passing a
certain point in 100 successive 20-minute time periods. The observations
are listed below.
 23 20 16 18 30 22 26 15 5 18
14 17 11 37 21 6 10 20 22 25
19 19 19 20 12 23 24 17 18 16
27 16 28 26 15 29 19 35 20 17
12 30 21 22 20 15 18 16 23 24
15 24 28 19 24 22 17 19 8 18
17 18 23 21 25 19 20 22 21 21
16 20 19 11 23 17 23 13 17 26
26 14 15 16 27 18 21 24 33 20
21 27 18 22 17 20 14 21 22 19

A useful method for summarizing a set of such data is the construction

of a frequency table, or a frequency distribution. That is, we divide the
overall range of values into a number of classes and count the number
of observations that fall into each of these classes or intervals.
The general rules for constructing a frequency distribution table are;
i) There should not be too few or too many classes.
ii) Generally speaking, equal class intervals are preferred. But the first
and last classes can be open-ended to cater for extreme values.
iii) Each class should have a class mark to represent the classes. It
is also named as the class midpoint of the ith class. It can be found by
taking simple average of the class boundaries or the class limits of the
same class.

1. Setting up the classes for the above data.

Choose a class width of 5 for each class, then we have seven classes
going from 5 to 9, from 10 to 14, â€ę, and from 35 to 39.

2. Tallying and counting.

 Classes Tally Marks Count
5-9 3
10 â€Ș 14 9
15 â€Ș 19 36
20 â€Ș 24 35
25 â€Ș 29 12
30 â€Ș 34 3
35 â€Ș 39 2
3. Illustrating the data in tabular form .
Frequency Distribution for the Traffic Data
Number of autos per period Number of periods
5-9 3
10 â€Ș 14 9
15 â€Ș 19 36
20 â€Ș 24 35
25 â€Ș 29 12
30 â€Ș 34 3
35 â€Ș 39 2
Total 100
In this example, the class marks of the traffic-count distribution are
7, 12, 17, â€ę, 32 and 37.

Example 10.3.1. The data below shows the mass of 40 students in a class.
The measurement is to the nearest kg.
55 70 57 73 55 59 64 72
60 48 58 54 69 51 63 78
75 64 65 57 71 78 76 62
49 66 62 76 61 63 63 76
52 76 71 61 53 56 67 71

Construct a frequency table for the data using an appropriate scale.

 Solution.
Step 1: Find the range.
The range of a set of numbers is the difference between the least
number and the greatest number in the set
In this example, the greatest mass is 78 and the smallest mass is 48.
The range of the masses is then 78 â€Ș 48 = 30. The scale of the
frequency table must contain the range of masses.
Step 2: Find the intervals.
The intervals separate the scale into equal parts.
We could choose intervals of 5. We then begin the scale with 45 and
end with 79.
Step 3: Draw the frequency table using the selected scale and intervals.
Mass (kg) Frequency
45 â€Ș 49 2
50 â€Ș 54 4
55 â€Ș 59 7
60 â€Ș 64 10
65 â€Ș 69 4
70 â€Ș 74 6
75 â€Ș 79 7
Total 40

Frequency Distribution for continuous data.

The following data were obtained in a survey of the heights of 20
students in a sports club. Each height was measured to the nearest cm.
133 136 120 138 133 131 127 141 127 143
130 131 125 144 128 134 135 137 133 129
To form a frequency distribution of the heights of the 20 students, we
group the information into classes or intervals.
The following are the different ways of writing the same set of inter-
vals.
 Height (cm)
119.5 ≤ b < 124.5
124.5 ≤ b < 129.5
129.5 ≤ b < 134.5
134.5 ≤ b < 139.5
139.5 ≤ b < 144.5

Height (cm)
119.5 − 124.5
124.5 − 129.5
129.5 − 134.5
134.5 − 139.5
139.5 − 144.5

Height (cm)
120 − 124
125 − 129
130 − 134
135 − 139
140 − 144
The values 119.5, 124,5, 129.5, . . . are called class boundaries or inter-
val boundaries. The upper class boundary (u.c.b) of one interval is the
lower class boundary (l.c.b) of the next interval.

10.3.2 Width of an interval

The width of an interval is the difference between the boundaries. Width

of an interval = Upper class boundary - lower class boundary. Often
intervals with equal widths are chosen, like in the above illustrations in
which each width is 5cm.
To group the data, it helps to use a tally column, entering the numbers
in the first row, and the second row etc.
The frequence distribution table would read as follows:
Height (cm) Tally Frequency
119.5 ≤ b < 124.5 1
124.5 ≤ b < 129.5 5
129.5 ≤ b < 134.5 7
134.5 ≤ b < 139.5 4
139.5 ≤ b < 144.5 3
Total 20

10.4 Histograms

A histogram is usually used to present frequency distributions graphi-

cally. This is constructed by drawing rectangles over each class. The
area of each rectangle should be proportional to its frequency.
Notes:
1. The vertical lines of a histogram should be the class boundaries.
2. If the smallest observation is far away from zero, then a break sign
∼ should be introduced in the horizontal axis.
3. The vertical axis will be labelled frequency density, where

.
frequency
frequency density =
interval width
Example 10.4.1. The table below shows the distribution of ages of pas-
sengers on a flight from Lusaka to Ndola.
Age, x years 0 ≤ x < 20 20 ≤ x < 40 40 ≤ x < 50 50 ≤ x < 70 70 ≤
Frequency 4 44 36 28

Construct a histogram for the above data.

Solutions
We first begin by completing the frequency table.
 Age, x years Interval width Frequency Frequency density
0 ≤ x < 20 20 4 0.2
20 ≤ x < 40 20 44 2.2
40 ≤ x < 50 10 36 3.6
50 ≤ x < 70 20 28 1.4
70 ≤ x < 100 30 6 0.2

Age of passengers against frequency density.

The histogram will look like.

10.4.2 Modal class

The highest bar in the histogram represents the interval 40 ≤ x < 50.
This is the modal class. In a grouped frequency distribution, the modal
class is the interval with the greatest frequency density. That is the
interval represented by the highest bar in the histogram provided that
the intervals have the same width.

10.5 Frequency Polygons

A grouped frequency distribution can be displayed as a frequency poly-

gon. To construct a frequency polygon, for each interval, plot the fre-
quency density against the midpoint value, where

midpoint value = (l.c.b + u.c.b).

1
2
Then join the points with straight lines.

Example 10.5.1. Draw a frequency polygon to illustrate the following

frequency distribution which gives the time taken by 31 students to finish
a quiz test.
Time, t (min) 25 ≤ x < 30 30 ≤ x < 35 35 ≤ x < 40 40 ≤ x < 50 50
Frequency 4 12 8 4

Solution
We begin by completing the frequency distribution table as follows.

Time, t (min) Mid interval value Interval width Frequency Frequenc

25 ≤ x < 30 27.5 5 4 0
30 ≤ x < 35 32.5 5 12 2
35 ≤ x < 40 37.5 5 8 1
40 ≤ x < 50 45 10 4 0
50 ≤ x < 65 57.5 15 3 0
Below is the frequency polygon.

10.6 Cummulative Frequncy Curves (Ogives)

We recal that the frequency is the number of times an event occurs

within a given scenario. Cumulative frequency is defined as the run-
ning total of frequencies. It is the sum of all the previous frequencies up
to the current point. It is easily understandable through a Cumulative
Frequency Table shown below
 Marks Frequency (number of students) Cummulative Frequency
0−5 2 2
5 − 10 10 12
10 − 15 5 17
15 − 20 5 22

Cumulative Frequency is an important tool in Statistics to tabulate data

in an organized manner. Whenever you wish to find out the popularity
of a certain type of data, or the likelihood that a given event will fall
within certain frequency distribution, a cumulative frequency table can
be most useful. Say, for example, the Census department has collected
data and wants to find out all residents in the city aged below 45. In this
given case, a cumulative frequency table will be helpful.

Example 10.6.1. Draw a cummulative frequency curve for the following

frequency distribution.
Height h (cm) 150 ≤ h < 155 155 ≤ h < 160 160 ≤ h < 165 165 ≤ h
Frequency 4 22 56 32

Solution As before, we begin by completing the frequency table.

Height h (cm) < 150 < 155 < 160 < 165 < 170 < 175
Cummulative Frequency 0 4 26 82 114 119
10.7 Measures of Central Tendency

Measures of central tendency are simply averages. The most commonly

used are the arithmetic mean, mode and median.

10.7.1 The arithmetic mean (mean)

The mean of a set of observations x1, x2 , . . . , xn is the sum of observations

divided by the number of observations.
Pn
xi
mean = i=1 .
n
It follows that the mean of the values x1, x2 , . . . , xn with frequencies
f1, f2 , . . . , fn is
n n
1X X
xi fi , where N = fi
N
mean =
i=1 i=1

Example 10.7.2. 1. Find the arithmetic mean for the scores 3, 8, 4, 6

and 7.

2. Find the mean of the following table showing the number of children
per family.
No. Children (x) Frequency (f)
0 5
1 8
2 12
3 18
4 9
5 9

10.7.3 Assumed mean

If a guess is made at the mean, say A, and the difference between the
guess A and xi is di , then
 P
di
x̄ = A + ,
n
where di = xi − A or P
fi di
x̄ = A + ,
N
Example 10.7.4. Use a suitable guess to the arithmetic mean x̄ of the
numbers 485, 513, 498, 501, 512, 497.
Solution Let A = 500. Then the values of di are −15, 13, −2, 1, 12, −3.
Thus
X
di = 6.

Consequently, P
di
x̄ = A +
6
n
= 500 + = 501.
6
The value of this technique does not become apparent until large
groups of data are handled.

For the grouped frequency distribution, we use the midinterval values

to compute the mean.

10.8 The median (ungrouped data)

The median of a set off numbers is the middle value or arithmetic mean
of the two middle values, if they are arranged in increasing order.

Example 10.8.1. Find the median of the following data.

1. 3, 0, 4, 8, −4, 6, 12, 5, 7

2. 0, 0, 4, −3, 8, 12, 14, −3, 7, 26

10.8.2 Median for grouped data

The median for grouped data is calculated using the formula

P !
2 (n + 1) − ( f)1
1
Median = L1 + c,
f median
where L1 is the lower boundary of the class containing the median, n is
P
the total frequency, ( f)1 is the sum of all the frequencies in the classes
below the median classs, f median is the frequency of the class containing
the median, c is the class width of the median class.
Note Classes do not all have to be the same width.

Example 10.8.3. The following frequency distribution is the summary

of travel time to work for some employees
Time to travel to work (x) Frequency (f) Cummulative Frequency
1 − 10 8 8
11 − 20 14 22
21 − 30 12 34
31 − 40 9 43
41 − 50 7 50
Compute the median.
P
Solution L1 = 20.5 n = 50, ( f)1 = 22, f median = 12 c = 10. Conse-
quently
P !
1
2 (n) − ( f)1
Median = L1 + c,
f median
!
1
− 22
2 (50) 25 − 22
Median = 20.5 + 10 = 20.5 + 10 = 23
12 12
25 persons take less than 23 minutes to travel to work and another 25
persons take more than 23 minutes to travel to work.

10.9 The mode

The mode for a set of numbers is that value or values which occur most
often. A set of numbers with one mode is called unimodal. A set of
numbers with two modes is called bimodal.
10.9.1 Mode for grouped data

For groups data, the mode is defined by

 
Mode = L1 + c,
∆1
∆1 + ∆ 2
where
L1 is the lower boundary of the class containing the mode, c is the class
width of the modal class, ∆1 is the difference between the frequency of
the modal class and the frequency of the previous lower class, ∆2 is the
difference between the frequency of the modal class and the frequency
of thenext upper class.
or  
fmode − f0
Mode = L1 + c,
2fmode − f0 − f2
From the previous example,
L1 = 10.5, c = 10, ∆1 = 14 − 8 = 6, ∆2 = 14 − 12 = 2.
Consequently,
× 10 = 18.
6
Mode = 10.5 +
8

10.10 Measures of Dispersion

10.10.1 Range

The measure of dispersion which is easiest to understand and easiest to

calculate is the range. Range is defined as: Range = Largest observation
â€Ș Smallest observation.

10.10.2 Mean deviation

For n observation x1, x2 , . . . , xn , the mean deviation about their mean x̄

is given by Pn
i=1 |xi − x̄|
M.D(x̄) = .
n
Example 10.10.3. Find the Mean deviation about the mean (x̄) for the
following data: 57, 64, 43, 67, 49, 59, 44, 47, 61, 59.
 The mean deviation about the median (M) is given by
Pn
|xi − M|
M.D(M) = i=1 .
n
For grouped data, the above formulae become
Pn
fi |xi − x̄|
M.D(x̄) = i=1
n
and Pn
i=1 fi |xi − M|
M.D(M) = .
n
10.10.4 Standard Deviation

The standard deviation s of a set of values {x1, x2 , . . . , xn } is defined by

r Pn
i=1 (xi − x̄)
2
s= .
n
For grouped frequency, the formula becomes
r Pn
i=1 fi (xi − x̄)
2
s= .
N
Example 10.10.5. Find the standard deviation for the numbers 4, 5, 8, 9, 10.
Solution x̄ = 7.2
xi xi − x̄ (xi − x̄)2
4 -3.2 10.24
5 -2.2 4.84
8 0.8 0.64
9 1.8 3.24
10 2.8 7.84
Total 5 26.8
Hence, r
s=
26.8
= 2.315.
5
10.10.6 Variance

In many situations in statistcs, it is the square of the standard deviation,

s2 , which is important. We call s2 the variance.
From the previous example,

s2 = 5.36.
NOTE: THESE LECTURE NOTES ON INTEGRAL CALCULUS ARE MEANT FOR MAT1110
(2020/2021) ACADEMIC YEAR ONLY.