Chapter three

Numerical Descriptive Measures

Objectives

 Describe data using measures of central tendency, such as

the mean, median, mode, and midrange.
 Summarize data using measures of variation, such as the
range, variance, and standard deviation.
 Determine the position of a data value in a data set using
various measures of position, such as percentiles, deciles
and quartiles.
 A. Measure of central
tendency
A measure of central tendency is very important
tool that refer to the center of a histogram or a
frequency distribution curve.
Such measures are the mean, the median, and the
mode for the two cases (grouped and ungrouped
data sets).
 The mean
◦ The most commonly used measure of central tendency is
called mean (or the average).
• Also known as arithmetic average: it is the most common
measure.
• Calculated by adding all the values in the group & then
dividing by the number of values.
• Helps to summarizing the essential features and enables
comparison.

 Cont…
 Mean is the sum of the values divided by the
number of values. The mean of a set of
numbers x1, x2... xn is typically denoted by " ".
 This mean is a type of arithmetic mean.
 Itis the "standard" average, often simply
called the "mean".
The mean for an ungrouped data is obtained
by dividing the sum of all values by the
number of values in that data set.
 Cont…
The Mean for Ungrouped Data
calculated as

Mean for population data:

Mean for sample data: x̄ =
Example; Find the mean score of 10
students in a midterm exam in a
class if their scores are
 Cont…
25 27 30 23 16 27 29 14
20 28
=
Example2. According to example 1,
if we take a sample of 4 students
from the class and find their scores
to be: 23, 27, 16, and 29. Find the
mean of this scores.
 x̄ =
 x̄==95/4=23.75
ii. Weighted Mean

 If 𝑥1 , 𝑥2 , … , 𝑥𝑛represent values of the and 𝑤1 , 𝑤2 , … , 𝑤𝑛 the

itemsthen the weighted mean, (𝑥ҧ𝑤 ) is
arecorresponding weights,
given by
w
xw  1 1x  w x
2 2  w x
n n 
wi xi 
w1  w2  wn wi 
Example: A student’s final mark in Mathematics, Physics, Chemistry and Biology are A,
B, D and C respectively. If the respective credits received for these courses are 4, 4, 3
and 2, determine the approximate average mark the student has got for the course.
Solution:

𝑥𝑖 4 3 1 2
𝑤𝑖 4 4 3 2
𝑥 𝑖 𝑤𝑖 16 12 3 4

x w  w1 x1  w2 x2  wn xn  w x i i

𝟏𝟔+𝟏𝟐+𝟑+𝟒
w1  w2  wn w i

= 𝟑𝟓 𝟏𝟑 = = 2.69. That is, Average mark of the student is

𝟏𝟑 2.69.
 iii. Combined mean
 When a set of observations is divided into k groups and x̄ 1𝑛1 is the mean
of n1
& group 1, x̄ 2𝑛2 is the mean of n2 & group2, …, x̄ k𝑛k is the mean of nk &
group k, then the combined mean, denoted by x̄c, of all

x̄ 1𝑛1 + x̄ 2𝑛2 + ⋯ + x̄ 𝑘 𝑛𝑘
observations taken together is given by

𝑛1 + 𝑛 2 + ⋯
X̄c =
+ 𝑛𝑘
Example: There are two classes, Class A and Class B. Class A has 30 students
with an average score of 70 on a test. Class B has 20 students with an average
score of 80. What is the combined average score for both classes?
Solution:

X̄c = =3700/50 = 74.

The combined mean of the entire students will be

74.
Note:

If a constant c is added to or subtracted from every value in the data set, the
mean increases or decreases by that constant:

 New Mean=Old Mean + c, for added;

 New Mean=Old Mean - c, for subtracted

If each value in the data set is multiplied by a constant k, the mean is also
multiplied by k: New Mean=k × Old Mean.

Question 1: If the mean of a data set is 50, what will the new mean be if a constant
value of 5 is added to every value in the data set?
Given mean = 50 and constant = 5; New mean = 50 + 5 = 55.
 The mid range

The midrange (MR) is defined as the sum of the lowest and highest
values in the data set divided by 2.
MR = Lowest value + Highest value
2
Example: Find the midrange (MR) for the following data:
11, 13, 20, 30, 9, 4, 15
Solution: The lowest value is 4, and the highest value is 30, then

MR = 4 + 30 = 34 = 17
2 2
Note that, this measure (MR) is weak as a measure of central ten-
dency since it is depends only on two values among of all values in
the data set.
 Mean for Grouped data
 If data are given in the form of continuous frequency distribution,
the
sample𝑘mean can be computed as
= , 𝑥𝑖 𝑓𝑖 - is the product of mid-
σ𝑖=1 𝑥 𝑖 𝑓 𝑖 𝑥 1 𝑓1 +𝑥 2 𝑓2 + …
𝑘
σ𝑖=1+𝑥
�𝑖 𝑘 𝑓 𝑘 𝑓1 +𝑓2 + …+
x̄
𝑓𝑘
= freq. � point &

Class boundary 60-62 62-64 64-66 66-68 68-70 70-72 Total

Frequency (fi) 5 18 42 20 8 7 100

xi 61 63 65 67 69 71

xifi 305 1134 2730 1340 552 497

Solution:
σ𝑖=1 f i𝑖
𝑘 𝑥𝑖
The formula to be used for the
σ𝑖=1
x̄
𝑓
mean is as follows:
𝑘
σ𝑖=1 f i𝑖 655
=
𝑘 𝑥 8100
σ𝑖=1 𝑖
x̄ = x̄ = 65.58.
𝑓
 Median

• Is the value of the middle item of series

when it is arranged in ascending or
descending order.
• It divides the series into two half.
• It is positional average.
• It is the middle value of the distribution when
all items are arranged in either
th ascending or
 n  1
descending order  terms ofvalue
Med in value.
 Where n is odd  2 

03/08/2025 By: Menberu T. 1

2
 Cont…
 Example: Find the median for the data set:

312, 257, 421, 289, 526, 374, 497

 Solution: First, the data set after we have ranked in
increasing order is:

x1 x2 x3 x4 x5 x6 x7

257 289 312 374 421 497 526

Median=374
 Since there are 7 values in this data set, so the fourth
term a 7+ 1 = 4k in the ranked data is the median.
Therefore the median is
 median = ( )th item= = 4th item = 374
 Cont…
 Median of Even Numbers
 Step 1: Arrange the data either in ascending or in descending
order.
 Step 2: If the number of observations (say n) are even, then
identify (n/2)th and [(n/2) + 1]th observations.
 Step 3: The average of the above two observations (which
are identified in step 2) is the median of the given data.
 Cont…
Example: Find the median for the data
set:
8, 12, 7, 17, 14, 45, 10, 13, 17, 13, 9, 11

x Solution:
x1 x x2 x 3
First,
x x x we
4 x x rank
5 x x6
the data in
7 8 9 10 11 12

7
increasing order:
8 9 10 11 12 13 13 14 17 17 45

Since there are 12 values in this data

set, the median is given by the average
of the two middle values whose ranks
are
 Median for grouped data
 For grouped data, the median is
obtained
by the following formula.
Median=L+()h
Where L= lower limit of the median class
n= number of observation
f=frequency of the median class
cf=cumulative frequency of the class
preceding the median class
h=class width
~ Example: Water percentage in the body of species of Fish is given below.
x Calculate the median.
Class interval 15-24 25-34 35-44 45-54 55-64 Total

Frequency 7 17 16 6 4 50

Solution: Construct the less than cumulative frequency distribution,

then:
Class Interval 15-24 25-34 35-44 45-54 55-64 Total
Frequency 7 17 16 6 4 50
Cumulative Freq. 7 24 40 46 50

 Since n = 50, 50/2 = 25

 l=35
 f=16
 h=9
 Cf=24
 Median=L+()h = =35+()9 = 35.56
 The mode
 The mode is another measure of central tendency and it is
known as the most common value in a data set.
 Data set with none mode: In such data set each value
occurring only once.
 Data set with one mode: In such data set only one value
occurring with the highest frequency. The data set in this
case is called unimodal.
 Data set with two modes: In such data set two values that
occur with the same (highest) frequency. The distribution, in
this case, is said to be bimodal.
 Data set with more than two modes: In such data set more
than two values occurs with the same (highest) frequency,
then the data set contains more than two modes and it is said
to be multimodal.
 Cont…
 Example: Find the mode for the given data set:
 22, 19, 21, 19, 27, 21, 29, 22, 19, 25, 21, 22, 25
 Solution: Since each of the three values, 19 (occur three
times), 21 (occur three times), and 22 (occur three times)
occurs with a highest frequency in their neighborhoods,
therefore, each of these is a mode, that is the modes for this
data set are: 19, 21, and 22.
 Mode for grouped data
The formula for calculating the mode of
grouped data is:

In this formula, the variables are:

• L: The lower limit of the modal class
• h: The size of the class interval
• f1: The frequency of the modal class
• f0: The frequency of the class preceding
the modal class
• f2: The frequency of the class succeeding
the modal class
 Example : The following table shows the distribution of scores obtained by
students in an exam:
Score Range Number of Students (Frequency)
50 – 60 8
60 – 70 12
70 – 80 25
80 – 90 10
90 - 100 5

What is the mode of the exam

scores?
Answer:
• L = lower boundary of the modal class = 70
• f1 = frequency of the modal class = 25
• f0 = frequency of the class before the modal class =
12
• f2​= frequency of the class after the modal class = 10 =75
• h = class width = 10
• Using formula: Mode = 75.
Relationships Between Mean, Median and Mode:
The relationships between mean, median & mode is defined
as Mode is
equal to the difference between 3 times the median & 2 times the
mean.
That is, Mean – Mode = 3 (Mean – Median) OR;
Mode = 3 Median – 2 Mean.
Example : If the difference between mean and mode of a
population is 48 and the median is 12, then the mean is
Solution:
 Mean – Mode = 3(Mean – Median);

 48 = 3(Mean – 12);
 16 = Mean – 12;
 Mean = 28.
 B. Measures of dispersion
• An average can represent a series only as best as a
single figure can, but it certainly cannot reveal
the entire story of any phenomenon under study
• It shows the degree by which numerical data tend
to spread around an average value/mean .
• Averages do not tell anything about the scatterness
of observations within the distribution.
• In order to measure the degree of scatter, the
statistical device called measures of dispersion are
calculated.
03/08/2025 By: Menberu T. 23
  Range = highest value – lowest value
 It shows the difference b/n the highest value
and the lowest value, hence it is the weakest
measure of dispersion
 Variance
 First calculate the mean, then deduct the
mean from each value in the group square
the result and divide the result by the
number of values.
 The variance is used as a measure of how far a
set of numbers are spread out from each other.
 It describes how far the numbers lie from the
mean (expected value).
03/08/2025 By: Menberu T. 24
 n 2

 ( x  x)
i 1
i
Var ( x) 
n
 Standard deviation
 The most reliable measurement of the degree
to which the data is spread around the mean
 Putting the variance in square root

03/08/2025 By: Menberu T. 25

Example: please, find the mean, median, mode,
range, variance and standard deviation for the
ID
following row Age of respondent
data?
1 53
2 44
3 56
4 70
5 45
6 62
7 36
8 23
9 56
10 55
03/08/2025 By: Menberu T. 26
Solution: A) Mean

= ∑xi/n = 53 + 44 + 56 +70 + 45 + 62 + 36 +23 + 56 + 55/10

= 500/10 = 50

B) Median, first we should arrange the raw data in ascending or

descending order as follow:

23, 36, 44, 45, 53, 55, 56, 56, 62, 70, since n is order, therefore

Median = 53 + 55/2 = 54

C) Mode, we find the most frequently occur, 56 is the mode of

the given data since it is more frequently occur and It is uni-
modal.

D) Range = largest value-lowest value = 70-23 =47

E) Variance = ∑(xi- )2/n

03/08/2025 By: Menberu T. 27
 ID
xi xi- (xi- )2
1
53 3 9
2
44 -6 36
3
56 6 36
4
70 20 400
5
45 -5 25
6
62 12 144
7
36 -14 196
8
23 -27 729
9
56 6 36
10
55 5 25
∑(xi- )2 =1636 variance = ∑(xi- )2/n = 1636/10
= 163.6
F) SD=variance = 163.6 = 12.79
03/08/2025 By: Menberu T. 28
 Measure of dispersion for
Grouped Data

• Sample Variance Formula for Grouped Data (σ2) = ∑ f(mi – x̄ )2/(n-1)

• Population Variance Formula for Grouped Data (σ2) = ∑ f(mi – x̄ )2/n

• where,

• f is the frequency of each interval

• mi is the midpoint of the ith interval

• x̄ is the mean of the grouped data

03/08/2025 By: Menberu T. 29

 Cont…

• Find the variance and the standard deviation for the following frequency dist
of a sample:
Class Frequency fm

5–9 2

10 – 14 4
15 – 19 7
20 – 24 3
25 – 29 1

30 – 34 3

Total 20

03/08/2025 By: Menberu T. 30

Cont…

3.5 12.25

03/08/2025 By: Menberu T. 31

Cont…

• Variance= = 1105/19=58.158

• Standard deviation=7.626

03/08/2025 By: Menberu T. 32

 C. Measures of relationship
1. Coefficient of variance
It (CV) is a normalized measure of dispersion.
It is also known as unitized risk or the variation coefficient.
It is defined as the ratio of the standard deviation to the
mean.
CV is a relative measure of dispersion, V, defined by:

 SD 
CV   
 Mean 

03/08/2025 By: Menberu T. 33

 Example: If the standard deviation of a given
distribution is 0.20 and the mean is 0.50, what is
the coefficient of variation (CV)?

CV = (0.20/0.50)*100% = 40%
2. Covariance
Covariance between X and Y refers to a measure
of how much two variables change together.
Covariance indicates how two variables are
related.
A positive covariance means the variables are
positively related, while a negative covariance
means the variables are inversely related.
The formula for calculating covariance of sample
data is shown below.
03/08/2025 By: Menberu T. 34
 Note: for population (N) and
for sample (n-1)
 Often the numbers have no
meaning. Thus we focus on the
sign.

03/08/2025 By: Menberu T. 35

3. correlation
Covariance only shows the direction. It has no
upper and lower bound.
Correlation tells the degree to which the variables
tend to move together.
The most familiar measure of dependence between
two quantities is the "Pearson's correlation."
It is obtained by dividing the covariance of the two
variables by the product of their standard deviations
.
The Pearson correlation is defined only if both of the
standard deviations are finite ፥ፍልሕ፡ህ፡and both of
them are nonzero.
The correlation coefficient is symmetric: corr(X,
Y) = corr(Y, X).
03/08/2025 By: Menberu T. 36
The Pearson correlation is +1 if there is perfect
positive linear relationship, −1 if there is
perfect negative linear relationship.
If the variables are independent, Pearson's
correlation coefficient is 0.
The sample correlation coefficient is written

03/08/2025 By: Menberu T. 37

The correlation between two random
variables, X and Y, is a measure of the
degree of linear association between the
two variables.
The population correlation, denoted by , can
take on any value from -1 to 1.
  indicates a perfect negative linear relationship
-1 <  < 0 indicates a negative linear relationship
   indicates no linear relationship
0 <  < 1 indicates a positive linear relationship
   indicates a perfect positive linear relationship
The absolute value of  indicates the strength or exactness of the
relationship.

03/08/2025 By: Menberu T. 38

 Example: find covariance and Pearson
correlation following hypothetical row data?
xi yi xi- yi- (xi- )(Yi- ) (xi- )2 (Yi- )2

10 18 -4 6 -24 16 36
30 6 16 -6 -96 256 36
8 12 -6 0 0 36 0
16 15 2 3 6 4 9
6 9 -8 -3 24 64 9
Cov (X,Y)= ∑(xi- )(Yi- )/n= -90/5 = - ∑(xi- )2∑
18 = 376 (Yi- )2
= 90
r (x, y) = ∑(xi- )(Yi- )/∑(xi- )2∑ (Yi- )2 = -90/33, 840 =
-90/183 = -0.49
03/08/2025 By: Menberu T. 39
 D. Shape of Frequency
Distribution
Skewness
 It refers to symmetry or asymmetry of the
distribution.
 A distribution is symmetric if its left half is a
mirror image of its right half.
 The skewness value can be positive or
negative.
 A symmetric distribution with a single peak and
a bell shape is known as a normal
distribution.

03/08/2025 By: Menberu T. 40

Kurtosis:
 It refers to peakedness/flatness of the
distribution.
 Higher kurtosis means more of the
variance is the result of infrequent extreme
deviation. n 4

 The
 ( x  x)
i
fourth standardized
KU  i 1 moment is
(n  1) S 4
defined as

03/08/2025 By: Menberu T. 41