Page 1 of 7

Topic 4
Measures of Location
In Topic 2, we learned how to take samples from the population using different sampling methods and
described the data obtained from these samples using textual, tabular, and graphical methods shown
in Topic 3. In Topic 4, we will describe the data numerically first by finding the Measures of Location
(or Position) which include the Minimum (the smallest value in the data set) the Maximum (the largest
observation in the data set) and the Measures of Central Tendency (MCT). A measure of central
tendency is a value in which a set of observations tend to cluster around. It is a single value that
represents or summarizes all values in the data set. For example, a student’s General Passing Average
(GPA) summarizes his/her academic performance in a semester.

We have three MCTs: Mean, Median, and Mode.

The mean is the summation of all observations divided by the total number of observations.
(Recall your summation notation from Topic 1.)

𝑛
∑𝑁
𝑖=1 𝑥𝑖 ∑ 𝑥
Population mean: 𝜇 = Sample mean: 𝑋̅ = 𝑖=1 𝑖
𝑁 𝑛

µ (Greek mu) is read as “myoo”. 𝑋̅ is read as “X bar”.

𝑁 is the total number of observations in the 𝑛 is the total number of observations in the
population or the population size. sample or the sample size.

Example: Example:

If the size of population A is 𝑁 = 5 with We wish to get a sample of size 𝑛 = 3 from

observations 𝑥1 = 2, 𝑥2 = 5, 𝑥3 = 4, 𝑥4 = population A using SRS and suppose these
3, and 𝑥5 = 1, then observations are 2, 3, and 1, then

∑𝑁
𝑖=1 𝑥𝑖 ∑5𝑖=1 𝑥𝑖 2 + 5 + 4 + 3 + 1 ∑𝑛𝑖=1 𝑥𝑖 ∑3𝑖=1 𝑥𝑖 2 + 3 + 1
𝜇= = = 𝑋̅ = = = =2
𝑁 5 5 𝑛 3 3
15
= =3
5

Some Characteristics:

• Can be obtained from numerical data only

• Unique (a data set has only one mean)
• Easily affected by extreme values
• Always available but might not be an actual data value in the data set
• ∑𝑁𝑖=1(𝑥𝑖 − 𝜇) = 𝑜, ∑𝑛𝑖=1(𝑥𝑖 − 𝑋̅) = 𝑜
• ∑𝑁 2
𝑖=1(𝑥𝑖 − 𝑎) is at minimum when 𝑎 = 𝜇 or ∑𝑁 2 ̅
𝑖=1(𝑥𝑖 − 𝑎) is at minimum when 𝑎 = 𝑋

hvvvalle
 Page 2 of 7

The median is the middle value in an ordered set of observations. When 𝑵 (or 𝒏) is odd, the
median is the single middle value and when it is even, it is the mean of the two middle values.

Population median: 𝑀𝐷 Sample median: 𝑋̅𝐷

Example 1 Example 1

Find the median of Population A. Find the median of the random sample 2, 3, 1
from population A.

Arrange the observations in ascending order Arrange the observations in ascending order
first. first.,
1, 2, 3, 4, 5 1, 2, 3

Since 𝑁 = 5 (odd), then 𝑀𝐷 = 3. Since 𝑛 = 3 (odd), then 𝑋̅𝐷 = 2.

Example 2 Example 2

Find the median of 1, 3, 5, 5, −2, 0. Let 0, 1, −2, 5 be a sample. Find its median.
,
In ascending order: −2, 0, 1, 3, 5, 5 In ascending order: −2, 0, 1, 5

1+3 4 0+1
Since 𝑁 = 6 (even), then 𝑀𝐷 = 2
=2=2 Since 𝑛 = 4 (even), then 𝑋̅𝐷 = 2 = 0.5

Some Characteristics:

• Can be obtained from numerical data only

• Unique (a data set has only one median)
• Not easily affected by extreme values
• Always available but might not be an actual data value in the data set

The mode is the observation that occurs the most frequent of times in a data set.

Population mode: 𝑀𝑂 Sample mode: 𝑋̅𝑂

Example: Find the modes of the following sets of data.

2, 2, 3, 3, 2, 2, 5, 3 𝑀𝑂 = 2 The data set is unimodal (only 1 mode).

2, 3, 2, 5, 3, 5, 3, 2 𝑀𝑂 = 2, 3 The data set is bimodal (2 modes).
2, 1, 2, 1, 5, 4, 3, 4 𝑀𝑂 = 1, 2, 4 The data set is trimodal (3 modes).
2, 2, 2, 2, 3, 3, 3, 3 Mode does not exist (Why?)
2, 2, 2, 2, 2, 2, 2, 2 𝑀𝑂 = 2 The data set is unimodal (only 1 mode).
B, I, O, S, T, A, T, I, S, T, I, C, S 𝑀𝑂 = 𝑆, 𝑇, 𝐼 The data set is trimodal (3 modes).
Some Characteristics:

• Can be obtained from categorical and numerical data

• Not unique (data set may have more than one mode)
• Might not exist in the data set

hvvvalle
 Page 3 of 7

We can always obtain the mean, median, and mode. However, some measures are more appropriate
for a particular set of data. If the observations are more or less similar in magnitude, then the mean
is best used to summarize the data. If extreme values (very high or very low values) are present, then
the median is the preferred one to use.

Unless otherwise stated or implied, assume the data that you are working on
are population data. Thus, appropriate symbols and formulas should be used.

Practice:

1. Find the minimum, maximum, mean, median, and mode of 𝟓, 𝟔, −𝟒, 𝟎, 𝟓, −𝟑, 𝟒, 𝟐, 𝟒, 𝟕.

𝑴𝒆𝒂𝒏 = _______________ 𝑴𝒆𝒅𝒊𝒂𝒏 = _____________ 𝑴𝒐𝒅𝒆 = _____________

2. The mean of four numbers is 𝟓𝟎. If the first three numbers are 𝟒𝟎, 𝟔𝟎, 𝒂𝒏𝒅 𝟑𝟓, what is the value
of the last number? ____________

3. If ∑𝟏𝟎 ̅
𝒊=𝟏 𝒙𝒊 = 𝟔𝟎, what is the value of 𝑋? _____________

4. If the mean of a population of 𝟓𝟎 is 𝟏𝟑𝟎, what is the summation of the observations? __________

5. Find the minimum, maximum, mean, median, and mode from the data below.

Student’s score in the quiz Frequency

5 10
9 15
13 25
16 20
25 10
30 5
50 1

hvvvalle
 Page 4 of 7

Other Measures of Location

If observations are arranged in ascending order, the minimum is the first observation in the array
and the maximum is the last and the measures of central tendency are, more or less, in the middle.
If we want to know the observation in a specific location in an array which is divided into equal
number of parts, then we need to find the fractiles (percentiles, deciles, quartiles).

• Percentiles (𝑃1 , 𝑃2, … , 𝑃99 ) are values that divide a set of observations (in ascending order)
into 100 equal parts. 𝑃1 means 1st percentile, 𝑃2 (2nd percentile), up to 𝑃99 (99th percentile).
One percent (1%) of the observations are less than 𝑃1 , 2% of the observations are less than
𝑃2 , 3% of the observations are less than 𝑃3 , and so on.

1 2 3 4 100

▪ ▪ ▪
Min 𝑷𝟏 𝑷𝟐 𝑷𝟑 𝑷𝟗𝟗 Max

• Deciles (𝐷1 , 𝐷2 , … , 𝐷9 ) are values that divide a set of observations (in ascending order) into
10 equal parts. 𝐷1 means 1st decile, 𝐷2 (2nd decile), up to 𝐷9 (9th decile). Ten percent (10%)
of the observations are less than 𝐷1 , 20% of the observations are less than 𝐷2 , 30% of the
observations are less than 𝐷3 , and so on.

1 2 3 4 5 6 7 8 9 10

Min 𝑫𝟏 𝑫𝟐 𝑫𝟑 𝑫𝟒 𝑫𝟓 𝑫𝟔 𝑫𝟕 𝑫𝟖 𝑫𝟗 Max

• Quartiles (𝑄1 , 𝑄2 , 𝑄3 ) are values that divide a set of observations (in ascending order) into 4
equal parts. 𝑄1 means 1st quartile, 𝑄2 (2nd quartile), and 𝑄3 (3rd quartile). Twenty-five percent
(25%) of the observations are less than 𝑄1 , 50% of the observations are less than 𝑄2 , and
75% of the observations are less than 𝑄3 .

1 2 3 4

Min 𝑸𝟏 𝑸𝟐 𝑸𝟑 Max

How to find fractiles:

1. Arrange the values in ascending order.

2. Solve for 𝐿.

hvvvalle
 Page 5 of 7

𝑚𝑛 where 𝐿 is the position of the observation in the array; 𝑚

For percentiles: 𝐿=
100 is the subscript in the fractile you are looking for and 𝑛 is
𝑚𝑛
For deciles: 𝐿= the total number of observations.
10
𝑚𝑛
For quartiles: 𝐿=
4

3. If 𝐿 is a whole number, the mean of the 𝑳th and (𝑳 + 𝟏)th observations is the value of the desired
fractile. Do not round off the answer.

If 𝐿 is not a whole number, round it to the next higher integer (remember 𝒌 in FDT). The
observation corresponding to the 𝑳th position is the value of the desired fractile.

Example 1. The scores of students in the first biostatistics quiz are given below:

5, 2, 8, 9, 8, 9, 12, 6, 8, 9, 6, 10, 8, 9, 7, 6, 8, 9, 10, 8

Find 𝑷𝟑𝟎 , 𝑫𝟓 , 𝐚𝐧𝐝 𝑸𝟑 and interpret.

In ascending order:

2, 5, 6, 6, 6, 7, 8, 8, 8, 8, 8, 8, 9, 9, 9, 9, 9, 10, 10, 12

Take note that 𝒏 = 𝟐𝟎.

1. Find P30 .
𝒎 = 𝟑𝟎

𝑚𝑛 30(20)
𝐿 = 100 = 100
=𝟔 6𝑡ℎ observation

Find the (𝐿 + 1)th observation since 𝐿 is whole number.

(𝐿 + 1) = (6 + 1) = 7 7𝑡ℎ observation
.

Therefore
7+8
𝑃30 = 𝑚𝑒𝑎𝑛 𝑜𝑓 6𝑡ℎ 𝑎𝑛𝑑 7𝑡ℎ 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑜𝑛𝑠 = = 7.5
2

which means that 30% of the 20 observations are less than 7.5 and 70% of them are greater than
7.5.

2. Find 𝐷5 .
𝑚𝑛 5(20)
𝐿= 10
= 10
= 10 10th observation

(𝐿 + 1) = (10 + 1) = 11 11th observation

hvvvalle
 Page 6 of 7

Thus
8+8
𝐷5 = 𝑚𝑒𝑎𝑛 𝑜𝑓 10𝑡ℎ 𝑎𝑛𝑑 11𝑡ℎ 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑜𝑛𝑠 = =8
2
which means that 50% of 20 observations are less than 8 and 50% of them are greater than 8.
𝟓
[𝟏𝟎 ∗ 𝟏𝟎𝟎%] = 𝟓𝟎%

Denominator is 10 since the desired fractile is a decile.

3. Find 𝑄3 .
𝑚𝑛 3(20)
𝐿= 4
= 4
= 15 15th observation

(𝐿 + 1) = (15 + 1) = 16 16th observation

Thus
9+9
𝑄3 = 𝑚𝑒𝑎𝑛 𝑜𝑓 15𝑡ℎ 𝑎𝑛𝑑 16𝑡ℎ 𝑜𝑏𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑜𝑛𝑠 = =9
2
which means that 75% of 20 observations are less than 9 and 25% of them are greater than 9.
𝟑
[𝟒 ∗ 𝟏𝟎𝟎%] = 𝟕𝟓%

Denominator is 4 since the desired fractile is a quartile.

Example 2

Find the 19𝑡ℎ percentile and 2𝑛𝑑 quartile from the data given below. Interpret the results.

56 52 63 45 56 75 68 63 49 55 58 54 51 49 61 64 58 59 50 57 53 63 60

In ascending order:

45 49 49 50 51 50 51 54 55 56 56 57 58 58 59 60 61 63 63 63 64 68 75

1. Find 19𝑡ℎ percentile (𝑃19 ).

Take note that 𝒏 = 𝟐𝟑, 𝒎 = 𝟏𝟗

𝑚𝑛 19(23)
𝐿 = 100 = 100
= 4.37 ≈ 5 5th observation

Thus 𝑃19 = 51

which means that 19% of 23 observations are less than 51 and 81% of them are greater than 51.

hvvvalle
 Page 7 of 7

2. Find 2nd quartile (𝑄2 ).

𝑚𝑛 2(23)
𝐿= = = 11.5 ≈ 12 12th observation
4 4

Thus 𝑄2 = 57

which means that 50% of 23 observations are less than 57 and another 50% are greater
than 57.

Practice:

Find the 88th percentile, 6th decile, and 3rd quartile of the quiz scores

Student’s score in the quiz Frequency

5 10
9 15
13 25
16 20
25 10
30 5
50 1

hvvvalle