OTHER MEASURES OF LOCATION
6.1 MEASURES OF LOCATION Percentile values divide a set of observations into 100 equal
There are several other measures of location that describe or parts. These values, denoted by P1, P2, …, P99, are such that 1% of
locate the position of certain non-central pieces of data relative to the data falls below P1, 2% falls below P2 , …, and 99% falls
the entire set of data. These measures, often referred to as below P99.
quartiles or fractiles are values which a specific fraction or
Ungrouped Data: For ungrouped data sets, the kth percentile (k =
percentage of the observations in a given set must fall.
1, 2, 3, … ,99 ) can be determined using the formula given by
6.2 QUARTILES Pk =x
Quartiles are values that divide a set of observations into 4 ( 100k ) N where k = 1, 2, 3, … , 99
equal parts. These values, denoted by Q 1, Q2, and Q3, are such that Grouped Data: For grouped data sets, on the other hand, the kth
25% of the data falls below Q1, 50% falls below Q2 and 75% falls percentile can be determined using the formula given by
below Q3. k
Ungrouped Data: For ungrouped data sets, the kth quartile (first
and third quartiles) can be determined using the formula given by
Qk =x where
[
Pk =LB Pk +
100
N −CF
f Pk
i ]
( k4 ) N where k = 1, 2, 3
LBPk = lower boundary of the kth percentile class
N = total frequency
Grouped Data: For grouped data sets, the kth quartile can be CF =<CF immediately preceding the kth percentile class
determined using the formula given by fPk = frequency of the kth percentile class
k i = class size
where:
Q k =LBQk +
4
N−CF
f Qk[ ]
i Notes:
1. Q2, D5, and P50 are all equal to the median.
2. The subscript refers to the position of the quantile when all the
LBQk = lower boundary of the kth quartile class data points are arranged in an ascending order. If the subscript is
N = total frequency not a whole number, it should be rounded up to the next whole
CF = <CF immediately preceding the kth quartile class number.
fQk = frequency of the kth quartile class
i = class size 6.5 INTERQUARTILE RANGE (IQR)
IQR=Q3 −Q 1∨P75−P25
The kth quartile class, similar to the median class, is the class
k
interval whose <CF is the first to be greater than 4
N 6.6 PROCEDURE TO CHECK FOR OUTLIERS
(starting from the lowest interval). An outlier is an extremely small or extremely large data value
when compared with the rest of the data values.
6.3 DECILES Step 1: Arrange the data in order from smallest to largest.
Deciles are values that divide a set of observations into 10 Step 2: Find the interquartile range (IQR).
equal parts. These values, denoted by D 1, D2, …, D9, are such that Step 3: Compute Q 1−1.5 x IQR and Q 3 +1.5 x IQR .
10% of the data falls below D 1, 20% falls below D2 , …, and 90%
falls below D9. Step 4: Let x be the data value that is being checked to determine
whether it is an outlier.
Ungrouped Data: For ungrouped data sets, the kth decile (k = 1,
(a) If the value of x is smaller than Q 1−1.5 x IQR , then x is
2,…,9) can be determined using the formula given by
an outlier.
Dk = x
( 10k )N where k = 1, 2, 3, 4, 5, 6, 7, 8, 9 (b) If the value of x is larger than Q 3 +1.5 x IQR ., then x is an
Grouped Data: For grouped data sets, the kth decile can be outlier.
determined using the formula given by
k Example: Find the quartiles, interquartile range, deciles, 12th
where:
[ ]
Dk =LB Dk +
10
N −CF
f Dk
i
percentile, 24th percentile, 36th percentile, 48th percentile, 60th
percentile, 72nd percentile, 84th percentile, and 96th percentile for
the following examination scores. Display the data using a box
plot.
LBDk = lower boundary of the kth decile class Exam Scores
N = total frequency 4 |5,6,8
CF = <CF immediately preceding the kth decile class 5 |3,4,5,6, 9
fDk = frequency of the kth decile class 6 |2,3,5,6, 6, 9, 9
i= class size 7 |0,1,1,3,3,4,5,5,5,7,8
8 | 1,2,3,6, 9
Same as the median class and the k th quartile class, the kth
decile class is the interval whose <CF is the first to be greater
k
than 10
N (starting from the lowest interval).
6.4 PERCENTILES
Page 1 of 1
