Session 5

Chapter Three
Measures of Central Tendency
Describing Data Numerically
Describing Data Numerically

Central Tendency Variation

Arithmetic Mean Range

Median Interquartile Range

Mode Variance

Harmonic Mean Standard Deviation

Coefficient of Variation
Geometric Mean
 Chapter Three (Session 5)
Measures of Central Tendency

Chapter Four
Measures of Variation
 Objectives of Measures of Central Tendency

• To determine a single value around which the

other data will concentrate
• To summarize/reduce the volume of the data
• To facilitate comparison within one group or
between groups of data
 Example

Age and Height of Broadleaf Trees in Forests

Forest Height of Trees Age of Trees
Forest I 60, 55, 70, 66, 80, 45, 35, 41, 38, 44, 40, 50, 30, 26,
49, 56, 64, 60, 54, 80, 74, 30, 35, 40, 39, 33, 52, 47,
62, 30, 44, 100, 50, 42 42, 24, 32, 79, 40, 29
Forest II 95, 78, 66, 110, 88, 99, 70, 55, 50, 80, 59, 73, 80,
112, 101, 105, 77, 125, 77, 79, 67, 70, 60, 80, 45,
56, 98, 87, 95, 110, 105, 74, 57, 70, 78, 68, 79
100, 99, 105
Forest III 45, 40, 55, 56, 49, 50, 44, 22, 20, 30, 30, 29, 30, 21,
46, 55, 60, 59, 50, 44, 61, 23, 31, 35, 30, 29, 30, 23,
54, 50, 49, 59, 53, 60 36, 32, 24, 37, 22, 20
Forest IV 83, 80, 60, 99, 88, 100, 68, 52, 51, 83, 63, 63, 79,
110, 88, 81, 74, 120, 66, 80, 84, 66, 77, 72, 90, 39,
88, 78, 85, 93, 70, 98, 79, 80, 67, 68, 60, 59, 50
65
 Questions

❖How can we determine a single value?

❖How can we summarize data?
❖How can we compare data observed from various sources?

Use Measures of Central Tendency

 Desirable properties of measures of
central tendency
• Simple to understand
• Easy to calculate/interpret
• Unique (?)
• Defined by mathematical formula
• Based on all observations
• Not be seriously affected by extreme observations
• Support further statistical analysis
 The Summation Notation (∑)

• General Notation

• Some Properties
Measures of Central Tendency
i) Mean:
i) Arithmetic Mean
ii) Geometric Mean
(Appl.: growth rates; interest rates of
a financial investment over time, etc.)
iii) Harmonic Mean
ii) Median
iii) Mode
 Arithmetic Mean
Definition:

x
x1 + x2 +  + xn i =1 i
A.M = x = =
n n
 Arithmetic Mean
Properties:
i) sum of the deviations of the items from their
arithmetic mean is zero.

ii)
 Properties Cont’d

iii) Arithmetic Mean addresses the

following aspect:
"if all the quantities had the same value,
what would that value have to be to get
the same total obtained from all
observations?"
 Example: Arithmetic Mean
(Broadleaf trees in forest)

Forest Average Height of Trees Average Age of Trees

(Arithmetic Mean) (Arithmetic Mean)

Forest I 58.8 39.6

Forest II 95.6 68.5

Forest III 52.0 27.7

Forest IV 85.3 67.6

- Forest II has trees with tall & old trees

- Forest III: has young trees with average age of 27.7 years
 Example: Weighted Mean
(Weighted average of scores of a student)
Exam Weight Score Weighted
Component Score
Group .05 90% 4.5%
Assignment I

Test I .20 80% 16%

Mid Exam .25 75% 18.75%

Final Exam .50 80% 40%

Arithmetic Mean 81.25%

Weighted Average 79.25%

Arithmetic Mean for Frequency Distribution

For frequency distribution based on observations

f1 x1 + f 2 x2 +  + f n xn fx i i
A.M ( fd ) = = i =1
f1 + f 2 +  + f n n

f
i =1
i

For grouped frequency distribution we use the same formula by

replacing observations by mid-points of classes
Geometric Mean
 Geometric Mean

Application: Data on
❑ growth rates;
❑ interest rates of a financial investment over
time, etc.
Geometric Mean addresses the following
aspect:
"if all the quantities had the same value,
what would that value have to be in order
to obtain the same product?
 Example
Suppose an investment group generated profit at the rate of 10% the
first year, 12% the second year, 14.8% the third year, 3.8% the fourth
year, and 16% the fifth year. What is the average profit over the
period of 5 years?
Geometric mean for frequency
Distribution
 Drawbacks of Geometric Mean

i. The Geometric Mean is not defined even if a single

observation is negative (negative growth rate, etc.)

ii. If the value of any observation is zero (no growth), then

the value of the geometric mean becomes zero
 Harmonic Mean

Application:
Data on repeatedly applied fixed
quantity:
Example: average speed of
automobiles given in a duration of
various trips.
So Harmonic Mean gives averages of
rates and ratios
Definition: Harmonic Mean
Harmonic Mean (H.M) is the reciprocal of
the arithmetic mean of the reciprocal of
each observation
n n
H .M = = n
1 1 1 1
+ ++
x1 x2 xn 
i =1 xi
 Example: Harmonic Mean
Find the Harmonic mean of the speed per
hour of a car in three trips: 40, 60, 50

3 3
H .M = = = 48.65
1 1 1 1
+ +
40 60 50
x
i

The Arithmetic Mean = 50

is not recommended
Drawback of Harmonic Mean

If the value of any observation is zero, then it is

not defined. But most of the time ratios and
speed of automobiles are positive.
 Median
• In an ordered list of observations, the median
is the “middle” number (50% above, 50% below)

0 1 2 3 4 5 6 7 8 9 10 0 1 2 3 4 5 6 7 8 9 10

Median = 3 Median = 3

• Not affected by extreme values

 Example: Median
Height of broadleaf trees:
a) Odd number of observations (11)
60, 55, 70, 66, 80, 45, 35, 49, 56, 64, 60
b) Even number of observations (12)
60, 55, 70, 66, 80, 45, 35, 49, 56, 64, 60, 54

Question: Calculate the Median.

a) List in ascending order

35, 45, 49, 55, 56, 60, 60, 64, 66, 70, 80
Median= 60
b) List in ascending order
35, 45, 49, 54, 55, 56, 60, 60, 64, 66, 70, 80
Median = average of 56 and 60 = 58
 Median for grouped Data
• Median for grouped data can be computed by using
the following formula:

• Where, l1 is the lower class boundary of the median class, n is the

total number of observations (sum of all frequencies), C is the sum of
frequencies of all classes lower than the median class, W is the width
of the median class.
 Median Cont’d

Example: Weight of Primary School STUDENTS

Interval Frequency CD Below

20 but less than 25 5 5
25 but less than 30 6 11
30 but less than 35 12 23
35 but less than 40 2 25
40 but less than 45 1 26
45 but less than 50 2 28
Total 28
Steps: 1) construct cumulative distribution
2)Find n and divide by 2, to get the median class: (n=28; (n/2)=14. The
cumulative distribution indicates that 14 is in the interval [30,35).
The median class is [30,35). Lower class limit is 30; Class size is 5.
3) Frequency of the median class is 12
4) Cumulative frequency of the class preceding the median class 11
 Calculation
Unknowns in the formula:

l1 = 30; n = 28; C = 11;

w = 5; f Median = 12

 28 
 − 11 
~
 2   14 − 11  3
X = 35 + * 5 = 30 +   * 5 = 30 +   * 5 = 31.25
 12   12   12 
 
 
 Mode
• A measure of central tendency
• Value that occurs most often
• Not affected by extreme values
• Used for either numerical or categorical data
• There may be no mode
• There may be several modes
 Mode

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 0 1 2 3 4 5 6

No Mode
Mode = 10
 MODE for Frequency Distribution
Formula
Let
l = the lower class boundary of the modal class,
fo = the frequency of the class just preceding the modal class,
f1 = frequency of the modal class,
F2 = frequency of the class just following the modal class, and
W is the class width.
The modal class is the class with the highest frequency.
 MODE for Frequency Distribution Cont’d

Example: Weight of Primary School STUDENTS

Interval Frequency CD Below

20 but less than 25 5 5
25 but less than 30 6 11
30 but less than 35 12 23

35 but less than 40 2 25

40 but less than 45 1 26

45 but less than 50 2 28
Total 28

Unknowns in the Formula:

modal class= [30,35); l=30; f0=6;
f1=12; f2=2; W=5.
Mode=30.4
 Which measure of location
is preferable?

• Mean is generally used, unless extreme

values (outliers) exist.
• Mode: not recommended for highly
dispersed values, even though not affected
by extreme value.
• Then median is often used, since the
median is not sensitive to extreme values.
 Quantiles
Types of Quantiles:

1. Quartiles

2. Deciles and

3. Percentiles
 Quartiles
• Quartiles divide a given set of data into four equal
parts.
• The corresponding values to these divisions, denoted
by Q1 , Q2 and Q3 are called the first (lower), second
and third (upper) quartile respectively.
• The second quartile is the median.
Quartile position:
- First quartile Q1 =0.25(n+1)
- Second quartile Q2 =0.5(n+1)
- Third quartile Q3 =0.75(n+1)
 Example
• Compute first, second and third quartile for the
following series?

• Solution
 Data in ascending order (a):
6, 12, 12, 14, 32, 32, 34

Pos-1 Pos-7

Positions: 0.25*8=2nd ;
0.5*8=4th ; 0.75*8=6th
 Data in ascending order (b):
23, 23, 25, 27, 31, 34, 35, 38, 45, 46, 60

Pos-11
Pos-1

• Q1-position= 0.25*12=3; Q1=25

• Q2-position=0.5*12=6; Q2=34
• Q3-position=0.75*12=9, Q3=45
 Quartiles for Grouped data
• For grouped frequency distribution, the quartiles
are computed as follows: (remember median)

• Qk is the kth quartile, fQk is the frequency of the kth quartile class, l
is the lower class boundary of the kth quartile class, C is the sum of
frequencies of all classes lower than the kth quartile class
 Deciles
• Deciles divide a given set of data in to ten equal
parts and the kth decile of a grouped frequency
distribution is computed by:
 Percentiles

• Percentiles divide a given set of data in to

hundred equal parts. The kth percentile for a
grouped frequency distribution is computed by:
 Example
• Compute the 60th percentile
Solution: