1

Unit Three
Measures of Central Tendency
[email protected] 4/20/2023
2 objectives

 To comprehend the data easily.

 To get a single value that represent(describe)
characteristics of the entire data
 To summarizing/reducing the volume of the data
 To facilitating comparison within one group or between
groups of data
 To enable further statistical analysis

[email protected] 4/20/2023
3 Recall Properties of Summation
 𝑛
1𝑐 = 𝑛𝑐 𝑛
1 𝑐𝑥𝑖 =𝑐 𝑛
1 𝑥𝑖 c is constant
 𝑛
𝑖=1(𝑎 + 𝑏 × 𝑥𝑖 ) = n × a + b × 𝑖=1 𝑥𝑖 ,
𝑛
a and b are constants
 𝑛
𝑖=1(𝑥𝑖 ± 𝑦𝑖 ) = 𝑛
𝑖=1(𝑥𝑖 ) ± 𝑛
𝑖=1(𝑦𝑖 )
𝑛 𝑛 𝑛
 𝑖=1(𝑥𝑖 × 𝑦𝑖 ) ≠ 𝑖=1(𝑥𝑖 ) × 𝑖=1(𝑦𝑖 )

 Example: If 12
𝑖=1 𝑥𝑖 = 26, 12
𝑖=1 𝑦𝑖 = 17, 12
𝑖=1 𝑥𝑖
2 = 484, 12
𝑖=1 𝑦𝑖
2 = 362

𝑛 𝑛
Find 𝑖=1(4𝑥𝑖 + 3𝑦𝑖 ) 𝑖=1 2𝑥𝑖 (𝑥𝑖 − 7)

[email protected] 4/20/2023
4 Mean
 Arithmetic mean (𝑿):
 Definition
𝒏
𝒊=𝟏 𝒙𝒊
, 𝒇𝒐𝒓 𝒓𝒂𝒘 𝒅𝒂𝒕𝒂 𝒙𝟏 , 𝒙𝟐 , … , 𝒙𝒏
𝒏
𝒌
𝒊=𝟏 𝒇𝒊 × 𝒙𝒊
𝒙= 𝒌
, 𝑭𝒐𝒓 𝒇𝒓𝒆𝒒𝒖𝒆𝒏𝒄𝒚 𝒅𝒊𝒔𝒕𝒓𝒊𝒃𝒖𝒕𝒊𝒐𝒏
𝒊=𝟏 𝒇𝒊
𝒌
𝒊=𝟏 𝒇𝒊 × 𝒎𝒊
𝒌
, 𝒇𝒐𝒓 𝒈𝒓𝒐𝒖𝒑𝒆𝒅 𝑭𝑫, 𝒘𝒊𝒕𝒉 𝒎𝒊 = 𝒄𝒍𝒂𝒔𝒔 𝒎𝒂𝒓𝒌
𝒊=𝟏 𝒇𝒊
 Properties of Athematic Mean
𝑛 𝒌 𝒌
 𝑖=1 𝑥𝑖 − 𝑥 = 0; 𝒊=𝟏 𝒇𝒊 × 𝒙𝒊 − 𝒙 = 𝟎 ; 𝒊=𝟏 𝒇𝒊 × 𝒎𝒊 − 𝒙 = 𝟎

 For 𝑥1 , 𝑥2 , …, 𝑥𝑛 𝑤𝑖𝑡ℎ 𝑚𝑒𝑎𝑛 𝒙

𝑖𝑓 𝑥1 ± 𝑘, 𝑥2 ± 𝑘, …, 𝑥𝑛 ± 𝑘, then the mean becomes 𝒙 ± 𝒌
𝑖𝑓 𝑘𝑥1 , 𝑘𝑥2 , …, 𝑘𝑥𝑛 , then the mean becomes k𝒙
 Combined mean

 If 𝑥1 is mean for 𝑛1 𝑜𝑏𝑠. 𝑎𝑛𝑑 𝑛1 ×𝑥1 +𝑛2 ×𝑥2

=> combined mean = 𝑛1 +𝑛2
 𝑥2 is mean for 𝑛2 𝑜𝑏𝑠.
[email protected] 4/20/2023
5 Mean …
 Weighted Mean:
𝒌
𝒊=𝟏 𝒘𝒊 ×𝒙𝒊
 𝒙𝒘 = 𝒌 𝒘 , 𝒘𝒉𝒆𝒓𝒆 𝒙𝟏 𝒘𝒊𝒕𝒉 𝒘𝒉𝒆𝒊𝒈𝒉𝒕 𝒘𝟏 , … . 𝒙𝒌 𝒘𝒊𝒕𝒉 𝒘𝒉𝒆𝒊𝒈𝒉𝒕 𝒘𝒌
𝒊=𝟏 𝒊

 Geometric Mean:
 For averages of %, Growth rates, proportions
𝑁
 GM = 𝒙𝟏 𝒇 𝟏 × 𝒙2 𝒇 2 × ⋯ × 𝒙𝑛 𝒇 𝑛 ,
𝒙𝟏 𝒙𝟐 … 𝒙𝒏
 Harmonic Mean: For
𝒇𝟏 𝒇𝟐 … 𝒇𝒏
 For averages of rate, Price/Earnings with 𝑁 = 𝑛
𝑖=1 𝑓𝑖
𝑁
 𝐻𝑀 = 𝑛 𝑓𝑖
𝑖=1𝑥
𝑖

Relationships:
1. 𝑥 ≥ 𝐺𝑀 ≥ 𝐻𝑀
2. For two observations 𝑥 × 𝐺𝑀 = 𝐻𝑀
3. For constant observations: 𝑥 = 𝐺𝑀 = 𝐻𝑀
[email protected] 4/20/2023
6 Example
𝑁 8
 For the data: 2, 8,6,4,10,6,8,4 𝐻𝑀 = =1 2 2 2 1
𝑛 𝑓𝑖 + + + +
 Find 𝑖=1𝑥
𝑖
2 4 6 8 10

 Athematic mean
8/1.68 = 4.7525

 Geometric mean and

Example: Suppose you get an increment of 50%,
 Harmonic Mean
40% and 70% in your salary at the end of first, 2nd
 Sol: Values 2 4 6 8 10 Total and 3rd year respectively. If your initial salary is
Frequency 1 2 2 2 1 8=N 100USD,
a) what is the average increase in your salary?
𝒌
𝒊=𝟏 𝒇𝒊 × 𝒙𝒊 𝟐 + 𝟐 × 𝟒 + 𝟐 × 𝟔 + 𝟐 × 𝟖 + 𝟏𝟎 b) what is your salary at the end of 3rd year salary?
𝒙= 𝒌
=
𝒊=𝟏 𝒇𝒊
𝟖
=48/8 =6
𝑁
Example: A motorist travels 1440km in 3 days.
GM = 𝒙𝟏 × 𝒙2 × ⋯ × 𝒙𝑛
𝒇𝟏 𝒇2 𝒇𝑛
She travels for 10 hours at rate of 48km/hr on 1st
8
= 21 × 42 × 62 × 82 × 101 day, for 12 hours at rate of 40km/hr on the 2nd
= 8 737,280 day and for 15 hours at rate of 32km/hr on the
=5.4132 3rd day. What is her average speed?
[email protected] 4/20/2023
 7 Median
 Middle Value - Sorting is needed
(𝑛+1) 𝑡ℎ 𝑜𝑏𝑠. 𝑖𝑓 𝑛 𝑖𝑠 𝑜𝑑𝑑
2
 𝐹𝑜𝑟 𝑛 𝑜𝑟𝑑𝑒𝑟𝑑 𝑣𝑎𝑙𝑢𝑒𝑠 𝑥1 , 𝑥2 , … , 𝑥𝑛 : 𝑥 = 1 𝑛 𝑡ℎ 𝑛+2 𝑡ℎ
𝑜𝑏𝑠. + 2 𝑜𝑏𝑠. 𝑖𝑓 𝑛 𝑖𝑠 𝑒𝑣𝑒𝑛
2 2
(Ungrouped data)

𝑛
Example: Find the median for
 𝑋 = 𝐿𝑚𝑒𝑑 + 𝑊 2
−𝐹
(For Grouped data) 1. At rest pulse rates for 16 athletes at a meet
𝑓𝑚𝑒𝑑
are: 53, 54, 56, 57, 58, 56, 54, 64, 67, 57, 54,
55, 57, 68, 60 and 58. 57
2. The production of butter fat during 7
𝐿𝑚𝑒𝑑 = LCB for median class consecutive days was recorded for 75 cows
W = class width and the frequency is given below. (54.16)
𝑓𝑚𝑒𝑑 = 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑜𝑓 𝑚𝑒𝑑𝑖𝑎𝑛 𝑐𝑙𝑎𝑠𝑠 Butter Fats in kgs 40-44 45-49 50-54 55-59 60-64 65-69 70-74
F = cumulative frequency of the class that
Number of cows 7 10 22 15 12 6 3
precedes the median class

[email protected] 4/20/2023
 8 Mode
 Most Frequently Occurring Value/or values
 For ungrouped data-take the value with highest frequency by inspection
 For Grouped data
𝐿𝑚𝑜𝑑 = 𝐿𝐶𝐵 𝑜𝑓 𝑀𝑜𝑑𝑎𝑙 𝑐𝑙𝑎𝑠𝑠
∆1
𝑓ℎ = ℎ𝑔𝑒𝑠𝑡 𝐹𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦
 𝑀𝑜𝑑𝑒 = 𝐿𝑚𝑜𝑑 + 𝑤 × 𝑓ℎ−1 = 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑐𝑙𝑎𝑠𝑠 𝑡ℎ𝑎𝑡 𝑝𝑟𝑒𝑐𝑒𝑑𝑠 𝑚𝑜𝑑𝑎𝑙 𝑐𝑙𝑎𝑠𝑠
∆1 +∆2
𝑓ℎ−2 = 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑜𝑓 𝑡ℎ𝑒 𝑐𝑙𝑎𝑠𝑠 𝑡ℎ𝑎𝑡 𝑝𝑟𝑒𝑐𝑒𝑑𝑠 𝑚𝑜𝑑𝑎𝑙 𝑐𝑙𝑎𝑠𝑠
∆1 = 𝑓ℎ -𝑓ℎ−1
∆2 =𝑓ℎ -𝑓ℎ−2
 Example: Find the mode for Butter Fats in kgs 40-44 45-49 50-54 55-59 60-64 65-69 70-74
Number of cows 7 10 22 15 12 6 3
 (53.03)

[email protected] 4/20/2023
9
Important Characteristics of a Good MCT
 It Should be
 Easy to calculate and understand (interpret).
 Based on all the observations during computation.
 Rigidly defined. Clear and Unambiguous
 Representative of the data based on random sample.
 Resistant to sampling fluctuation sampling stability.
 Resistant to extreme value.
 Capable for further statistical analysis and /or algebraic manipulation.

[email protected] 4/20/2023
10 Quantiles
 Quartiles: Partitions the data in to 4 equal parts – Needs ordered data
𝑡ℎ
𝑗
𝑄𝑗 = 𝑛+1 𝑜𝑏𝑠. for ungrouped data
4
𝑛
−𝐹
=𝐿𝑄𝑗 + 𝑊 4
𝑓𝑚𝑒𝑑
(For Grouped data)j
j=1,2,3

 Deciles Example: Find Q3, D6 and P85 values for

1. Data set 70, 120, 110, 101, 88, 83, 95, 98, 107, and 100
 Percentiles
2. Frequency Distribution
Butter Fats in kgs 40-44 45-49 50-54 55-59 60-64 65-69 70-74
Number of cows 7 10 22 15 12 6 3

Note: Median = Q2 = D5 =P50

[email protected] 4/20/2023
11

[email protected] 4/20/2023