Measures of Central Tendency Grouped Data
Measures of Central Tendency Grouped Data
Quantiles are values that divide the data (distribution) into a given number of equal
parts. Like median, the quantiles are also “positional measures” Some of the quantiles
are:
a.) Quartiles – values that divide the distribution into 4 equal parts.
▪ Q1 - ( 1st quartile ) which is 25 % or less of the given distribution
▪ Q2 - ( 2nd quartile ) which is 50 % or less of the given distribution
▪ Q3 - ( 3rd quartile ) which is 75 % or less of the given distribution
▪ Q4 - ( 4th quartile ) which is 100 % or less of the given distribution
b.) Deciles – values that divide the distribution into ten (ten) equal parts.
▪ D1 - ( 1st decile ) which is 10 % or less of the given distribution
▪ D2 - ( 2nd decile ) which is 20 % or less of the given distribution
▪ D3 - ( 3rd decile ) which is 30 % or less of the given distribution
▪ D4 - ( 4th decile ) which is 40 % or less of the given distribution ….
▪ D10 - ( 10th decile) which is 100 % or less of the given distribution
c.) Percentiles – values that divide the distribution into 100 equal parts
▪ P1 - ( 1st percentile ) which is 1 % or less of the given distribution
▪ P2 - ( 2nd percentile ) which is 2 % or less of the given distribution
▪ P3 - ( 3rd percentile ) which is 3 % or less of the given distribution
▪ P4 - ( 4th percentile ) which is 4 % or less of the given distribution …….
▪ P100 - (100th percentile) which is 100 % or less of the given distribution
Ex. 1.) Find the 20th percentile or P20 of the following scores 40, 45, 42, 37, 36,
32, 28, 26, 25.
Solution:
1. Array the scores in decreasing order of magnitude.
45 42 40 37 36 32 28 26 25
2. Locate the position of the score corresponding to the 20 th percentile using the
formula
P ( n + 1 ) = 20 ( 9 + 1 ) = 2
100 100
3. Locate the second score from the lowest. The answer is 26. Thus, P 20 = 26.
This means that 20 % falls below 26.
1
Example 2. Find the 5th decile ( or D5 ) of the following scores : 19, 25, 38, 45,
65, 81.
Solution:
1. Arrange the scores in an increasing order of magnitude.
19 25 38 45 65 81
2. Locate the position of D5 using the formula: P (n + 1)/10 = 5 (6 + 1)/10 = 3.5
Since 3.5 is not an exact number, we have to interpolate using the following
steps:
a. Get the difference between the third and fourth scores from the lowest score,
since 3.5 is between the 3rd and 4th scores. 45 – 38 = 7.
b. Multiply the difference obtained by the decimal in no. 2. That is 7 x 0.5 = 3.5.
c. Add the product to the lower score (38) to obtain D 5. Therefore, D5 = 38 + 3.5
= 41.5
This means that 50 % of the scores lie below 41.5.
Grouped Data – refer to sets of data presented in the forms of frequency distributions.
These are data
grouped or classified into categories for better presentation and
analysis.
1. Long method:
X̅ = _∑ Xi fi_
n
where :
X̅ = mean
Xi, X1, X2, …. Xn = are the class midpoints
fi, f1,f2, …. fn = are the corresponding frequencies
2. Short Method :
A shorter method of finding the mean of grouped data is with the use of CODING
.
The coded formula for the mean is :
X̅ = X0 + [ _∑ ui fi_] x C
2
[ n ]
where :
X0 = assumed mean or coded mean
C = class size
Instead of using the actual class midpoints or classmarks, codes (denoted by u)
are used which is composed of consecutive integers assigned to each class.
The coding technique is as follows:
1. Choose one of the class midpoints, preferably at or near the center of the
distribution or choose the class interval with the highest frequency as the
assumed mean denoted by X0.
2. Under the new column u, write the zero value code opposite of X 0, and assign
positive integers whole numbers) to the classes higher in value than the
class with the assumed mean, and consecutive negative integers to those
classes lower in value.
3. Multiply the coded values with their corresponding frequencies and compute
for the algebraic sum.
4. Substitute the given coded formula and compute for the mean.
Example:
The following is the distribution of the wages of 50 workers of HGW
Manufacturing Co. taken during a particular week last May.
3
X̅ = ∑ Xi fi
n
= 48545
50
= P970.90
Short Method:
X̅ = X0 + [ _∑ ui fi_] x C
[ n ]
X̅ = 974.5 + [- 6_ x 30 ]
[ 50 ]
= 974.5 - 3.6
= P970.90
4
( fi )
870 – 899 4 4
900 – 929 6 10
930 – 959 10 20 F≤Me
960 – 989 13 33
Me class
990 – 1019 8 41
1020 – 1049 7 48
1050 – 1079 2 50
TOTAL 50
Then set up the cumulative less than (F≤) frequency and locate where the 25 th
item is in the distribution. The median class is the 960 – 989 class interval where
the 25th items fall.
= 959.5 + [ 50/2 - 20 ] x ( 30 )
13
= 959.5 + [ 25 - 20 ] x ( 30 )
13
= 959.5 + 11.54
Me = P971.04
5
The formula in finding the mode is:
M0 = LM0 + [ ___d1____ ] x C
[ d 1 + d2 ]
where :
LMo = lower boundary of the modal class.
d1 = difference between the frequency of the modal class and the
frequency of the
class next lower in value
d2 = difference between the frequency of the modal class and the
frequency of the
class next higher in value
C = class size
Solution:
M0 = LM0 + [ ___d1____ ] x C
[ d 1 + d2 ]
= 959.5 + [ __3__ ] x 30
6
[ 3+5 ]
= 959.5 + _90_
8
M0 = P970.75
Activity:
The distribution of the hourly rates of 100 employees in a certain company are
shown below:
7
Hourly Rate ( in peso ) Number of Employees ( f )
55 – 64 8
65 – 74 10
75 – 84 15
85 – 94 28
95 – 104 19
105 – 114 13
115 - 124 7
Determine the:
a. Arithmetic mean using the two methods ( long method and
coding method )
b. Median
c. Mode