Statistical Measure of Data Chapter 08

Measure of Central Tendency:

Arithmetic Mean:
Ungrouped Data:
∑
̅
Grouped Data:
∑
̅
∑
Properties of Arithmetic Mean:
 Quantitative data only
 Uses all values of the data
 Affected by extreme values
 Mean of constant values is constant
 Sum of deviation from mean is always zero i.e. ∑( ̅)
Example: Find mean of the below given salaries

Salary 5--10 10--15 15--20 20--25 25—30 30--35

No. of People 2 15 18 12 2 1

Weighted Mean: It is calculated where all the values of the data are of different importance.
̅ ∑
∑
Example: Find Mean of the following data:

Subject Maths Physics Chemistry English

Weighting 5 4 2 0
Marks 20 50 80 7
(42)
Combined Arithmetic Mean:
( ) ( )
̅c

Example 1:
Group A Group B
No. of Students 1200 400
Mean score 60 70
(62.5)
Example 2: Mean of 35 values is 68. If mean of 15 values is 54 then find the mean of remaining 20 values.
(78.5)
Find missing frequency of following if mean is 2.

X 0 1 2 3 4 5
F 40 ? 50 30 20 10
(20)

M. SHOAIB (M.PHIL) 1
 Statistical Measure of Data Chapter 08

Correction in Mean: The mean of a data of 100 observations were found to be 104. Later error was
detected as below. Find correct mean.
Correct Figure 151 78 98
Figure Taken 115 87 89
Solution:
∑ ( ) = 104 x 100 = 10400
∑ ( ) = 10400-115-87-89+151+78+98 = 10436
So,
Mean = 10436/100 = 104.36

Geometric Mean:
 G.M is the nth root of product of all values.
 G.M is impossible to calculate if any value in the data is zero or negative.
 Percentages calculated by G.M
 Price changes, population growth rate, depreciation rate etc. are calculated by G.M.
Ungrouped Data:
 √
∑
 ( )
Example: If a property increases its value by 10% in year 1, 50% in year 2 and 30 percent in year 3. Find
percentage growth rate over three year period.
Sol.
√
So, 29% growth rate over three year time period.

Grouped Data:
∑
 ( ∑
)
Example: Find G.M of the following.
Diameter 5 15 25 35
No. of Screws 5 8 3 4
(14.58)

M. SHOAIB (M.PHIL) 2
 Statistical Measure of Data Chapter 08

Harmonic Mean:
 It is used to average the speed of an object over equal intervals.
 It is impossible to find H.M if any value of the data is zero
Ungrouped data:

∑
Example: A bus move from Bahawalpur to Multan at an average speed of 60 km/h and back at average
speed of 40 km/h. Find H.M
(48 km/hr.)

Grouped Data:
∑
∑( )

Example: Find H.M

Class 2---6 6---10 10---14

F 10 12 18
(7.27)
Median:
 Median is the middle most value of the arranged data.
 It is the average of middle two values if number of observations is even.
 It is not influenced by extreme values
Ungrouped Data (Raw):
 Arrange the data
 Find position of median using ( ) observation.
 If the position is in fraction e.g. 12.5th value we will use 12th value + 0.5(13th value – 12th value)
Example: Find median of 3, 5, 4, 6, 2, 8, 9. (5)
Example: Find median of 2, 3, 4, 5, 6, 7, 8, 9. (5.5)
Ungrouped Frequency Distribution:
 Identify median class using ( )
 Locate the answer in cumulative frequency to identify median class
 Value of x against that frequency.
Example:
Find median.
X 1 2 3 4
F 2 3 5 5
C.F 2 5 10 15
Median class = ( ) = 16/2 = 8th observation.
So, Median = 3

M. SHOAIB (M.PHIL) 3
 Statistical Measure of Data Chapter 08

Grouped Frequency Distribution:

 Identify median class using ( ) .
 Locate the answer in cumulative frequency to identify median class and then apply the following
formula.
( )
Where,

L = Lower class boundary of median class

h = Class interval of median class
f = Frequency of median Class
n = Total number of observations in the data that is ∑
c = Cumulative frequency of previous class to the median class.
Example:
Find median of the given data

Class 1---5 6---10 11---15 16---20

F 2 4 8 6

C.F 2 6 14 20

Median class = n/2 = 20/2 = 10 c

( ) 13 Median Class

Finding Median Using Ogives (Cumulative Frequency Polygon):

M. SHOAIB (M.PHIL) 4
 Statistical Measure of Data Chapter 08

Mode: Most repeated value in the data.

Ungrouped Raw data: Most repeated value in the data.
Ungrouped Frequency Distribution: Value of X with the highest frequency.
Grouped Frequency Distribution:
 Find modal class. (Class with the highest frequency)
 Apply the formula to find Mode.
( )
( ) ( )
L = Lower class boundary of model class
= Frequency of modal class
= Frequency of previous class to modal class
= Frequency of next class to modal class
= Class interval of modal class

Finding Mode using Histogram:

Quartiles: There are three quartiles in the data Q1, Q2 and Q3. Quartiles divide the data into 4 four equal
parts each part contains 25% of the data.

Ungrouped Raw Data:

 Arrange the data
 ( )
 ( )
 ( )

M. SHOAIB (M.PHIL) 5
 Statistical Measure of Data Chapter 08

 If the position is in fraction e.g. 12.25th value we will use 12th value + 0.25(13th value – 12th
value)
Ungrouped Frequency Distribution:
 Identify quartile class using ( ) where k is quartile number, say 1, 2 and 3
 Locate the answer in cumulative frequency to identify quartile class
 Value of x against that frequency in quartile class
Grouped Frequency Distribution:
 Identify quartile class using ( ), where k is quartile number, say 1, 2 and 3
 Locate the answer in cumulative frequency to identify quartile class
 Apply formula
( )
L = Lower class boundary of quartile class
h = Class interval of quartile class
f = Frequency of quartile Class
n = Total number of observations in the data that is ∑
c = Cumulative frequency of previous class to the quartile class
k = 1 for first quartile, 2 for second and 3 for third quartile
Deciles:
 Divide the arranged data in ten equal parts
 There are 9 deciles starting from D1 and ending at D9.
 Each decile represents 10% of the data

Ungrouped Raw Data:

 Arrange the data
 ( )
 ( )
 ( ) . Where k = 1, 2, 3, 4, 5, 6, 7, 8, 9
 If the position is in fraction e.g. 12.70th value we will use 12th value + 0.70(13th value – 12th value)

Ungrouped Frequency Distribution:

 Identify decile class using ( ) where k is decile number, say 1, 2, 3 up to 9
 Locate the answer in cumulative frequency to identify quartile class
 Value of x against that frequency in decile class

Grouped Frequency Distribution:

 Identify decile class using ( ), where k is decile number, say 1, 2, 3 up to 9
 Locate the answer in cumulative frequency to identify decile class

M. SHOAIB (M.PHIL) 6
 Statistical Measure of Data Chapter 08

 Apply formula
( )
L = Lower class boundary of decile class
h = Class interval of decile class
f = Frequency of decile class
n = Total number of observations in the data that is ∑
c = Cumulative frequency of previous class to the decile class
k = 1 for first decile, 2 for second, 3 for third decile and so on up to 9.

Percentiles:
 Divide the arranged data in 100 equal parts
 There are 99 percentiles starting from P1 and ending at P99.
 Each percentile represents 1% of the data

Ungrouped Raw Data:

 Arrange the data
 ( )
 ( )
 ( ) . Where k = 1, 2, 3, 4 up to 99
 If the position is in fraction e.g. 12.77th value we will use 12th value + 0.77(13th value – 12th value)

Ungrouped Frequency Distribution:

 Identify percentile class using ( ) where k is percentile number, say 1, 2, 3 up to 99
 Locate the answer in cumulative frequency to identify percentile class
 Value of x against that frequency in percentile class

Grouped Frequency Distribution:

 Identify percentile class using ( ), where k is decile number, say 1, 2, 3 up to 99
 Locate the answer in cumulative frequency to identify percentile class
 Apply formula
( )
L = Lower class boundary of percentile class
h = Class interval of percentile class
f = Frequency of percentile class
n = Total number of observations in the data that is ∑
c = Cumulative frequency of previous class to the percentile class
k = 1 for first percentile 2 for second, 3 for third percentile and so on up to 99.

M. SHOAIB (M.PHIL) 7
 Statistical Measure of Data Chapter 08

Graphical representation of Quartiles, Deciles, percentiles and Median using Ogive:

Remember that the Value of X represents your

answers rather than the value of frequency.

Note:
 Median = Q2 = D5 = P50
 Q2= P50
 Q3 = P75
 D1= P10
 D2 = P20
 D3 = P30 and so on.
Change of Origin:
 Change of origin means adding or subtracting a number in all data values
 All measure of central tendency (mean, median, mode, quartiles etc.) are affected by change of
origin.
 If 2 is added or subtracted to all values of the data then the answer must be changed and increase
or decrease by adding or subtracting 2 respectively.
Change of Scale:
 Change of scale means multiplying or dividing all data values by a number
 All measure of central tendency (mean, median, mode, quartiles etc.) are affected by change of
scale.
 If all values of the data are multiplied or divided by 2 then the answer will also be multiplied or
divided by 2 respectively.
Note:
 Mode = 3median – 2mean
 A.M G.M H.M ( For distinct data)
 A.M = G.M = H.M ( For constant data)

M. SHOAIB (M.PHIL) 8
 Statistical Measure of Data Chapter 08

Box and Whisker:

Measure of Dispersion:
 All measure of dispersion give positive answer
 Higher dispersion means less consistency and vice versa
 For constant data all measure of dispersion will be zero
 Measure of dispersion contains Range, Skewness, Standard Deviation, Variance and Co-efficient of
variation. Coefficient of variation is categories in relative dispersion while all others are absolute
dispersion.

M. SHOAIB (M.PHIL) 9
 Statistical Measure of Data Chapter 08

Range:
Ungrouped data: Range = Maximum value – Minimum value
Grouped Data:
Range = Upper class boundary/limit of highest class – Lower class boundary/limit of first class
Formulae to Remember:
 Interquartile range =
 Semi-interquartile range =
 Co-efficient of semi interquartile range =
Variance (S2):
Ungrouped Data:
∑( ̅)

∑ ∑
 ( )
∑
 ( ̅)
Grouped Data:
∑ ( ̅)

∑
∑ ∑

∑
(∑ )
∑
 ∑
(̅ )
Standard Deviation: (S)
Ungrouped Data:
∑( ̅)
 √
∑ ∑
 √ ( )
∑
 √ ( ̅)
Grouped Data:
∑ ( ̅)
 √ ∑
∑ ∑
 √ ∑
(∑ )
∑
 √ ∑
(̅ )
Co-efficient of Variation:

M. SHOAIB (M.PHIL) 10
 Statistical Measure of Data Chapter 08

 C.V is commonly used for comparison of consistency between two. Less C.V means higher
consistency and vice versa.
Example 1: For data 50, 51, 48, 51 49, 51. Find mean, variance, standard deviation and co-efficient of
variation.
Example 2: Find mean, variance, standard deviation and co-efficient of variation of the following data

X 0 1 2 3 4 5 6
F 2 3 1 2 2 3 1
Example 3:
From the set of test which test have relatively more dispersion?

Test 1 Test 2

Mean 12 64

S.D 3 10

Example 4: Which of the following is more consistent?

Babar 20 22 26 19 30

Rizwan 13 12 96 20 10

Note:
 There is no effect of change of origin in data on S2 and S.
 Both are affected by change of scale in following manners
 If S.D of x = 5 then S.D of 3x = 3 x 5 = 15
 If Var. of x = 5 then Var. of 2x = 22x 5 = 20
Points to Remember:
( )
 Pearson’s co-efficient of skewness =

 Bowley’s measure of skewness =

 Positive answer means positively skewed data
 Negative answer means negatively skewed
 Zero answer means the data is symmetrical
Skewness:
Symmetrically skewed data:
 Mean = Median = Mode

 Equally distributed data

M. SHOAIB (M.PHIL) 11
 Statistical Measure of Data Chapter 08

Positively skewed data:

 Mode < Median < Mean


Negatively Skewed Data:

 Mode > Median > Mean


M. SHOAIB (M.PHIL) 12
 Statistical Measure of Data Chapter 08

M. SHOAIB (M.PHIL) 13