Percentile Score and Percentile Rank

Percentile Score

Percentiles are the points which divide the entire scale of measurement into 100 equal parts.
These are denoted by P0, P1, P2, P3, P4, P5, ......, P99, P100

According to - 

The first percentile may be defined as that point in a frequency distribution below which lie 1 percent of the total measures or scores. 

P0 lies at the beginning of the distribution and P100 at the end of the distribution. 

Importance of the Percentile Score 

Computation of Percentile 

Px = L + [(xN/100 - cfb)/f] × i
where, Px = xth percentile of data
N is the total of all frequencies
cfb = total number of scores upto interval below the percentile class
f = frequency of percentile class
i = Class interval
L = Actual lower limit of the percentile class

Let's understand it with the help of an example -

i) 10th Percentile (P10) 

10N/100 = (10×24)/100 = 2.4, it lies in the class 10-19
Thus, 10th percentile (P10) class is 10-19

Now, P10 = L + [(10N/100 - cfb)/f] × i
= 9.5 + [(2.4 - 2)/3] × 10
= 9.5 + 4/3
= 9.5 + 1.33
= 10.83

ii) 25th Percentile (P25)

25N/100 = (25×24)/100 = 6, it lies in the class 20-29
Thus, 25th percentile (P25) class is 20-29 

Now, P25 = L + [(25N/100 - cfb)/f] × i
= 19.5 + [(6 - 5)/4] × 10
= 19.5 + 10/4
= 19.5 + 2.5
= 22

Percentile Rank

Percentile Rank (PR) is the point in the distribution below which a given percentage of scores falls. 

According to - 

If the 80th percentile rank is a score of 65, then it means that 80% of the scores falls below 65.

Importance of Percentile Ranks 

A. Computation of percentile rank in ungrouped data

PR = 100 - (100R - 50)/N
where, R = Rank/Position in the given data
N = Total number of cases. 

Let's understand it with the help of an example - 

In a class of 40 students, Tarun obtained 15th rank in English class test. Find out Tarun's percentile rank -
PR = 100 - (100R - 50)/N
= 100 - (100 × 15 - 50)/40
= 100 - (1500 - 50)/40
= 100 - 36.25
= 63.75 

B. Computation of percentile rank in grouped data 

PR = [cfb + (f / i)(x - L)]/N × 100
where, X = Given score
L = Actual lower limit of the score's class
f = frequency of score's class
cfb = total number of scores upto interval below the score's class
N = Total no. of frequencies
i = Class interval

Let's understand it with the help of an example -

i) Percentile Rank of the score 42 

PR = [cfb + (f / i)(x - L)]/N × 100
= [5 + (5 / 5)(42 - 39.5)]/50 × 100
= [5 + 2.5]/50 × 100
= 7.5/50 × 100
= 7.5 × 2
= 15

ii) Percentile Rank of the score 62

PR = [cfb + (f / i)(x - L)]/N × 100
= [32 + (6 / 5)(62 - 59.5)]/50 × 100
= [32 + (6 / 5) × 2.5]/50 × 100
= [32 + 3]/50 × 100
= 35/50 × 100
= 35 × 2
= 70

Percentile Ranks are always whole numbers -
i.e., a PR of 67.34 ≈ 67
and, a PR of 73.84 ≈ 74