Measures of Position

Objectives
At the end of this lesson, you should be able to:
• Gain deeper understanding about the z-score.
• Identify the position of a data value in a data set, using
various measures of position, such as percentiles, deciles,
and quartiles.
• Interpret the analysis reported by box-and-whisker plots
Measures of Position
In addition to the measures of central tendency and
measures of variation, there are measures of position or
location. These measures include standard scores, percentiles,
deciles, and quartiles. They are used to locate the relative
position of a data value in the data set.
Standard Score (z score)
Case A Case B

𝜇1 = 𝜇2 𝜇1 ≠ 𝜇2
𝜎1 = 𝜎2 𝜎1 = 𝜎2

Case C Case D

𝜇1 = 𝜇2 𝜇1 ≠ 𝜇2
𝜎1 ≠ 𝜎2 𝜎1 ≠ 𝜎2
Standard Score
A standard score or z score tells how many standard
deviations a data value is above or below the mean for a
specific distribution of values. If a standard score is zero, then
the data value is the same as the mean.
Standard Score
Example
Charlie got a mark of 85 on a math test which had a
mean of 75 and a standard deviation of 5. Daisy got a mark of
75 on an English test which had a mean of 69 and a standard
deviation of 2. Relative to their respective mean and standard
deviation, who got the better grade?
Solution
Using the z score formula, we calculate the z scores for
each student and find:
Charlie
𝑥−𝑥ҧ 85−75 10
𝑧= = = =2
𝑠 5 5
Daisy
75−69 6
𝑧= = =3
2 2
Solution
The results tell us that Charlie got a test mark 2 standard
deviations higher than the mean of the class, while Daisy got a
mark that is 3 standard deviations higher than the mean in her
class. Therefore, proportionally speaking, Daisy did better
within her class in comparison to Charlie.
Example
Test Scores. A student scored 65 on a calculus test that
had a mean of 50 and a standard deviation of 10; she scored
30 on a history test with a mean of 25 and a standard
deviation of 5. Compare her relative positions on the two
tests.
Solution
Calculus
𝑥−𝑥ҧ 65−50 15
𝑧= = = = 1.5
𝑠 10 10
History
30−25 5
𝑧= = =1
5 5

Since the z score for calculus is larger, her relative position in the
calculus class is higher than her relative position in the history class.
Note: A z-score specifies the precise location of each X value
within a distribution. The sign of the z-score (+ or –) signifies
whether the score is above the mean (positive) or below the
mean (negative). The numerical value of the z-score specifies
the distance from the mean by counting the number of
standard deviations between X and 𝜇. When all data for a
variable are transformed into z-scores, the resulting
distribution will have a mean of 0 and a standard deviation of
1.
Example
Test Scores. Find the z score for each test, and state
which is higher.
Test A x = 38 𝑥ҧ = 40 s=5
Test B x = 94 𝑥ҧ = 100 s = 10
Solution
Test A
𝑥−𝑥ҧ 38−40 −2
𝑧= = = = −0.4
𝑠 5 5
Test B
94−100 −6
𝑧= = = −0.6
10 10

The score for test A is relatively higher than the score for test B.
Percentiles, Deciles, and
Quartiles
 To locate a specific point in any distribution, percentiles,
quartiles and deciles are the tools that can be used. These
parameters percentiles, deciles, and quartiles are called
quantiles.
Percentiles

The relative position of the raw score can be described precisely

by converting it into a percentile. When quantiles are distributed over
100 equally sized intervals, they are referred to as percentiles.
Percentiles are symbolized by 𝑃1 , 𝑃2 , 𝑃3 , . . . , 𝑃99 .
Percentiles
Percentile (Ungrouped Data)
The formula for finding a data value corresponding to a
kth percentile is
𝑘(𝑛+1) 𝑡ℎ
𝑃𝑘 = observation
100
where
n – total number of observations
k = 1, 2, 3, …, 99
Example
Test Scores. A teacher gives a 20-point test to 10
students. The scores are shown here. Find the values
corresponding to the 20th and 75th percentile.
18, 15, 12, 6, 8, 2, 3, 5, 20, 10
Solution
Arrange the data in order from lowest to highest.
2, 3, 5, 6, 8, 10, 12, 15, 18, 20
Then substitute into the formula. For 20th percentile,
20(10+1) 𝑡ℎ
𝑃20 = observation
100
20(11) 𝑡ℎ
𝑃20 = observation
100
𝑃20 = (2.2)𝑡ℎ observation
𝑃20 = 2𝑛𝑑 observation + 0.2(3rd – 2nd)
𝑃20 = 3 + 0.2(5 – 3)
𝑃20 = 3.4
Solution
For 75th percentile,
75(10+1) 𝑡ℎ
𝑃75 = observation
100
75(11) 𝑡ℎ
𝑃75 = observation
100
𝑃75 = (8.25)𝑡ℎ observation
𝑃75 = 8𝑡ℎ observation + 0.25(9th – 8th)
𝑃75 = 15 + 0.25(18 – 15)
𝑃75 = 15.75
Example
Calculate fifteenth, thirty-seventh and sixty-fourth
percentile from the following data array;
20 28 29 30 36 37 39 42 53 54
55 58 61 67 68 70 74 81 82 93
Solution
For 15th percentile,
15(20+1) 𝑡ℎ
𝑃15 = observation
100
15(21) 𝑡ℎ
𝑃15 = observation
100
𝑃15 = (3.15)𝑡ℎ observation
𝑃15 = 3𝑟𝑑 observation + 0.15(4th – 3rd)
𝑃15 = 29 + 0.15(30 – 29)
𝑃15 = 29.15
Solution
For 37th percentile,
37(20+1) 𝑡ℎ
𝑃37 = observation
100
37(21) 𝑡ℎ
𝑃37 = observation
100
𝑃37 = (7.77)𝑡ℎ observation
𝑃37 = 7𝑡ℎ observation + 0.77(8th – 7th)
𝑃37 = 39 + 0.77(42 – 39)
𝑃37 = 41.31
Solution
For 64th percentile,
64(20+1) 𝑡ℎ
𝑃64 = observation
100
64(21) 𝑡ℎ
𝑃64 = observation
100
𝑃64 = (13.44)𝑡ℎ observation
𝑃64 = 13𝑡ℎ observation + 0.44(14th – 13th)
𝑃64 = 61 + 0.44(67 – 61)
𝑃64 = 63.64
Percentile (Grouped Data)
Percentiles can also be calculated for grouped data which is done
with the help of following formula
𝑘(𝑛)
−𝑐𝑢𝑚𝑓𝑏
100
𝑃𝑘 = 𝑙𝑏𝑃𝑘 + 𝑐
𝑓𝑃𝑘

where
𝑙𝑏𝑃𝑘 – lower boundary of the kth percentile class
𝑐𝑢𝑚𝑓𝑏 – cumulative frequency before the kth percentile class
𝑓𝑃𝑘 – frequency of the kth percentile class
𝑐 – class interval
Example
Calculate thirty-seventh, forty-fifth and ninetieth percentile
from the frequency distribution of weights of 120 students, by using
the this table.
 Solution (37th Percentile)
𝑘(𝑛) 37(120)
= = 44.4
100 100

𝑃37 𝑙𝑏𝑃𝑘 = 139.5

𝑐𝑢𝑚𝑓𝑏 = 22
𝑓𝑃𝑘 = 28
𝑐 = 10

44.4−22
𝑃37 = 139.5 + 10 = 147.5 pounds
28
 Solution (45th Percentile)
𝑘(𝑛) 45(120)
= = 54
100 100
𝑙𝑏𝑃𝑘 = 149.5
𝑃45
𝑐𝑢𝑚𝑓𝑏 = 50
𝑓𝑃𝑘 = 25
𝑐 = 10

54−50
𝑃45 = 149.5 + 10 = 151.1 pounds
25
 Solution (90th Percentile)
𝑘(𝑛) 90(120)
= = 108
100 100
𝑙𝑏𝑃𝑘 = 179.5
𝑐𝑢𝑚𝑓𝑏 = 106
𝑃90 𝑓𝑃𝑘 = 6
𝑐 = 10

108−106
𝑃90 = 179.5 + 10 = 182.83 pounds
6
 The results of P37, P45 and P90 indicates that 37% of the
students weigh 147.5 pounds or less. Similarly, 45% of the
students weigh 151.1 pounds or less and 90% weigh 182.83
pounds or less.
Deciles
Deciles divide the distribution into 10 groups. They are
denoted by 𝐷1 , 𝐷2 , 𝐷3 , . . . , 𝐷9 .
Decile (Ungrouped Data)
The formula for finding a data value corresponding to a
kth decile is
𝑘(𝑛+1) 𝑡ℎ
𝐷𝑘 = observation
10
where
n – total number of observations
k = 1, 2, 3, …, 9
Example
The Shell company has a number of outlets in the
Metropolitan Manila area. The numbers of changes at the
EDSA Heritage outlet in the past 20 days are:
65 98 55 62 79 79 59 51 90 72
56 70 62 66 80 94 63 73 71 85
Calculate the 3rd decile, 5th decile, and 8th decile.
Solution
Arrange the data in order from lowest to highest.
51, 55, 56, 59, 62, 62, 63, 65, 66, 70, 71, 72, 73, 79, 79, 80, 85 90, 94, 98.
Then substitute into the formula. For 3rd decile,
3(20+1) 𝑡ℎ
𝐷3 = observation
10
3(21) 𝑡ℎ
𝐷3 = observation
10
𝐷3 = (6.3)𝑡ℎ observation
𝐷3 = 6𝑡ℎ observation + 0.3(7th – 6th)
𝐷3 = 62 + 0.3(63 – 62)
𝐷3 = 62.3
Solution
For 5th decile,
5(20+1) 𝑡ℎ
𝐷5 = observation
10
5(21) 𝑡ℎ
𝐷5 = observation
10
𝐷5 = (10.5)𝑡ℎ observation
𝐷5 = 10𝑡ℎ observation + 0.5(11th – 10th)
𝐷5 = 70 + 0.5(71 – 70)
𝐷5 = 70.5
Solution
For 8th decile,
8(20+1) 𝑡ℎ
𝐷8 = observation
10
8(21) 𝑡ℎ
𝐷8 = observation
10
𝐷8 = (16.8)𝑡ℎ observation
𝐷8 = 16𝑡ℎ observation + 0.8(17th – 16th)
𝐷8 = 80 + 0.8(85 – 80)
𝐷8 = 84
Example
The following data array are the test scores of 12
students in a statistics class:
56 62 65 70 73 77 79 82 85 87 92 99
Find the 1st and 7th decile.
Solution
For 1st decile,
1(12+1) 𝑡ℎ
𝐷1 = observation
10
1(13) 𝑡ℎ
𝐷1 = observation
10
𝐷1 = (1.3)𝑡ℎ observation
𝐷1 = 1𝑠𝑡 observation + 0.3(2nd – 1st)
𝐷1 = 56 + 0.3(62 – 56)
𝐷1 = 57.8
Solution
For 7th decile,
7(12+1) 𝑡ℎ
𝐷7 = observation
10
7(13) 𝑡ℎ
𝐷7 = observation
10
𝐷7 = (9.1)𝑡ℎ observation
𝐷7 = 9𝑡ℎ observation + 0.1(10th – 9th)
𝐷7 = 85 + 0.1(87 – 85)
𝐷7 = 85.2
Decile (Grouped Data)
Deciles can also be calculated for grouped data which is done
with the help of following formula
𝑘(𝑛)
−𝑐𝑢𝑚𝑓𝑏
10
𝐷𝑘 = 𝑙𝑏𝐷𝑘 + 𝑐
𝑓𝐷𝑘

where
𝑙𝑏𝐷𝑘 – lower boundary of the kth decile class
𝑐𝑢𝑚𝑓𝑏 – cumulative frequency before the kth decile class
𝑓𝐷𝑘 – frequency of the kth decile class
𝑐 – class interval
Example
Calculate third, fifth and ninth decile from the frequency
distribution of weights of 120 students, by using the this table.
 Solution (3rd Decile)
𝑘(𝑛) 3(120)
= = 36
10 10

𝐷3 𝑙𝑏𝐷𝑘 = 139.5
𝑐𝑢𝑚𝑓𝑏 = 22
𝑓𝐷𝑘 = 28
𝑐 = 10

36−22
𝐷3 = 139.5 + 10 = 144.5 pounds
28
 Solution (5th Decile)
𝑘(𝑛) 5(120)
= = 60
10 10
𝑙𝑏𝐷𝑘 = 149.5
𝐷5
𝑐𝑢𝑚𝑓𝑏 = 50
𝑓𝐷𝑘 = 25
𝑐 = 10

60−50
𝐷5 = 149.5 + 10 = 153.5 pounds
25
 Solution (9th Decile)
𝑘(𝑛) 9(120)
= = 108
10 10
𝑙𝑏𝐷𝑘 = 179.5
𝑐𝑢𝑚𝑓𝑏 = 106
𝐷9 𝑓𝐷𝑘 = 6
𝑐 = 10

108−106
𝐷9 = 179.5 + 10 = 182.83 pounds
6
 The results of D3, D5 and D9 means that 30% of the
students weigh 144.5 pounds or less. Similarly, 50% of the
students weigh 153.5 pounds or less and 90% weigh 182.83
pounds or less.
Quartiles
Quartiles divide the distribution into four groups,
separated by 𝑄1 , 𝑄2 , 𝑄3 .
Quartile (Ungrouped Data)
The formula for finding a data value corresponding to a
kth quartile is
𝑘(𝑛+1) 𝑡ℎ
𝑄𝑘 = observation
4
where
n – total number of observations
k = 1, 2, 3
Example
The Shell company has a number of outlets in the
Metropolitan Manila area. The numbers of changes at the
EDSA Heritage outlet in the past 20 days are:
65 98 55 62 79 79 59 51 90 72
56 70 62 66 80 94 63 73 71 85
Calculate the 1st quartile, 2nd quartile, and 3rd quartile.
Solution
Arrange the data in order from lowest to highest.
51, 55, 56, 59, 62, 62, 63, 65, 66, 70, 71, 72, 73, 79, 79, 80, 85 90, 94, 98.
Then substitute into the formula. For 1st quartile,
1(20+1) 𝑡ℎ
𝑄1 = observation
4
1(21) 𝑡ℎ
𝑄1 = observation
4
𝑄1 = (5.25)𝑡ℎ observation
𝑄1 = 5𝑡ℎ observation + 0.25(6th – 5th)
𝑄1 = 62 + 0.25(62 – 62)
𝑄1 = 62
Solution
For 2nd quartile,
2(20+1) 𝑡ℎ
𝑄2 = observation
4
2(21) 𝑡ℎ
𝑄2 = observation
4
𝑄2 = (10.5)𝑡ℎ observation
𝑄2 = 10𝑡ℎ observation + 0.5(11th – 10th)
𝑄2 = 70 + 0.5(71 – 70)
𝑄2 = 70.5
Solution
For 3rd quartile,
3(20+1) 𝑡ℎ
𝑄3 = observation
4
3(21) 𝑡ℎ
𝑄3 = observation
4
𝑄3 = (15.75)𝑡ℎ observation
𝑄3 = 15𝑡ℎ observation + 0.75(16th – 15th)
𝑄3 = 79 + 0.75(80 – 79)
𝑄3 = 79.75
Example
Calculate first, second and third quartile from the
following data array;
20 28 29 30 36 37 39 42 53 54
55 58 61 67 68 70 74 81 82 93
Solution
For 1st quartile,
1(20+1) 𝑡ℎ
𝑄1 = observation
4
1(21) 𝑡ℎ
𝑄1 = observation
4
𝑄1 = (5.25)𝑡ℎ observation
𝑄1 = 5𝑡ℎ observation + 0.25(6th – 5th)
𝑄1 = 36 + 0.25(37 – 36)
𝑄1 = 36.25
Solution
For 2nd quartile,
2(20+1) 𝑡ℎ
𝑄2 = observation
4
2(21) 𝑡ℎ
𝑄2 = observation
4
𝑄2 = (10.5)𝑡ℎ observation
𝑄2 = 10𝑡ℎ observation + 0.5(11th – 10th)
𝑄2 = 54 + 0.5(55 – 54)
𝑄2 = 54.5
Solution
For 3rd quartile,
3(20+1) 𝑡ℎ
𝑄3 = observation
4
3(21) 𝑡ℎ
𝑄3 = observation
4
𝑄3 = (15.75)𝑡ℎ observation
𝑄3 = 15𝑡ℎ observation + 0.75(16th – 15th)
𝑄3 = 68 + 0.75(70 – 68)
𝑄3 = 69.5
Quartile (Grouped Data)
Quartiles can also be calculated for grouped data which is done
with the help of following formula
𝑘(𝑛)
−𝑐𝑢𝑚𝑓𝑏
4
𝑄𝑘 = 𝑙𝑏𝑄𝑘 + 𝑐
𝑓𝑄𝑘

where
𝑙𝑏𝑄𝑘 – lower boundary of the kth quartile class
𝑐𝑢𝑚𝑓𝑏 – cumulative frequency before the kth quartile class
𝑓𝑄𝑘 – frequency of the kth quartile class
𝑐 – class interval
Example
Calculate first, second and third quartile from the frequency
distribution of weights of 120 students, by using the this table.
 Solution (1st Quartile)
𝑘(𝑛) 1(120)
= = 30
4 4

𝑄1 𝑙𝑏𝑄𝑘 = 139.5
𝑐𝑢𝑚𝑓𝑏 = 22
𝑓𝑄𝑘 = 28
𝑐 = 10

30−22
𝑄1 = 139.5 + 10 = 142.36 pounds
28
 Solution (2nd Quartile)
𝑘(𝑛) 2(120)
= = 60
4 4
𝑙𝑏𝑄𝑘 = 149.5
𝑄2
𝑐𝑢𝑚𝑓𝑏 = 50
𝑓𝑄𝑘 = 25
𝑐 = 10

60−50
𝑄2 = 149.5 + 10 = 153.5 pounds
25
 Solution (3rd Quartile)
𝑘(𝑛) 3(120)
= = 90
4 4
𝑙𝑏𝑄𝑘 = 159.5
𝑄3 𝑐𝑢𝑚𝑓𝑏 = 75
𝑓𝑄𝑘 = 18
𝑐 = 10

90−75
𝑄3 = 159.5 + 10 = 167.83 pounds
18
 The results from Q1, Q2 and Q3 indicates that 25% of the
students weigh 142.36 pounds or less. Similarly, 50% of the
students weigh 153.5 pounds or less and 75% weigh 167.83
pounds or less.
Relationships among Percentiles, Deciles, and Quartiles

• 𝐷1 , 𝐷2 , 𝐷3 , . . . , 𝐷9 correspond to 𝑃10 , 𝑃20 , 𝑃30 , . . . , 𝑃90 .

• Quartiles 𝑄1 , 𝑄2 , 𝑄3 correspond to 𝑃25 , 𝑃50 , 𝑃75 .
• The median is the same as 𝑃50 or 𝐷5 or 𝑄2 .
Interquartile Range
Interquartile Range
The interquartile range (IQR) is defined as the difference
between 𝑄1 and 𝑄3 and is the range of the middle 50% of the
data.
𝐼𝑄𝑅 = 𝑄3 − 𝑄1
Example
Check the following data set for outliers.
5, 6, 12, 13, 15, 18, 22, 50
Solution
The data value 50 is extremely suspect. These are the steps in
checking for an outlier.
Step 1 Find 𝑄1 and 𝑄3 .
1(8+1) 𝑡ℎ 3(8+1) 𝑡ℎ
𝑄1 = observation 𝑄3 = observation
4 4
𝑡ℎ 𝑡ℎ
𝑄1 = 2.25 observation 𝑄3 = 6.75 observation
𝑄1 = 7.5 𝑄3 = 21
Solution
Step 2 Find the interquartile range (IQR), which is 𝑄3 − 𝑄1 .
𝐼𝑄𝑅 = 𝑄3 − 𝑄1 = 21 − 7.5 = 13.5
Step 3 Multiply this value by 1.5.
1.5(13.5) = 20.25
Step 4 Subtract the value obtained in step 3 from 𝑄1 , and add the
value obtained in step 3 to 𝑄3 .
7.5 – 20.25 = –12.75 and 21 + 20.25 = 41.25
Solution
Step 5 Check the data set for any data values that fall outside
the interval from –12.75 to 41.25. The value 50 is outside this
interval; hence, it can be considered an outlier.