Module 4A
MEASURES
OF LOCATION
WILFREDO P MARINO
Associate Professor 5
[email protected]
Measures of
Location
A Measure of Location summarizes a
data set by giving a “typical value”
within the range of the data values
that describes its location relative to
entire data set.
Common Measures:
Minimum, Maximum
Central Tendency
Percentiles, Deciles, Qua iles
Maximum and
Minimum
Minimum is the smallest value in the data
set, denoted as MIN.
Maximum is the largest value in the data
set, denoted as MAX.
Maximum and
Minimum
Given the following set of data (age
of graduate school students)
obtained by random sampling
30, 42, 24, 54, 29, 24, 31
Min = 24
Max = 54
Measures of Central
Tendency
A single value that is used to
identify the “center” of the data
◦it is thought of as a typical value
of the distribution
◦most representative value of the
data
◦Mean, Median, Mode
Mean
Most common measure of central
tendencies
Also known as arithmetic average
Populatio
n Mean
Sample
Mean
Mean
Given the following set of data (age of
graduate school students) obtained by
random sampling
30, 42, 24, 54, 29, 24, 31
LEVELS OF
MEASUREMENT
1. NOMINAL-
sex (1=male,2=fem
2. ORDINAL
rank
3. INTERVAL- no absolute zero
4. RATIO- absolute zero
8
Proper ties of the Mean
• may not be an actual obse ation in
the data set
• can be applied in at least inte al
level
• easy to compute
• eve obse ation contributes to the
value of the mean
• subgroup means can be combined to
come up with a group mean
• easily a ected by extreme values
Median
Given the following set of
data 8, 14, 10, 15, 26, 28
Proper ties of a Median
• may not be an actual obse ation in
the data set
• can be applied in at least ordinal
level
• a positional measure; not a ected
by extreme values
Mode
• occurs most frequently
• nominal average
• may or may not exist
Given the following set of data (age
of graduate school students)
obtained by random sampling
30, 30, 42, 24, 54, 29, 24, 31
Proper ties of a Mode
• can be used for qualitative as
well as quantitative data
• may not be unique
• not a ected by extreme values
• can be computed for ungrouped
and grouped data
Mean, Median & Mode
Use the mean when:
• sampling stability is desired
• other measures are to be
computed
Mean, Median & Mode
Use the median when:
• the exact midpoint of the
distribution is desired
• there are extreme
obse ations
Mean, Median & Mode
• Use the mode when:
• when the "typical" value is
desired
• when the dataset is
measured on a nominal scale
Percentiles
Numerical measures that give
the relative position of a data
value relative to the entire
data set.
Divide an array (raw data
arranged in increasing or
decreasing order of magnitude)
into 100 equal pa s.
e jth percentile, denoted as
Pj, is the data value in the data
set that separates the bottom
EXAMPLE
Suppose Vinh was told that
relative to the other scores on
a ce ain test, his score was
the 95th percentile.
is means that 95% of
those who took the test had
scores less than or equal to
Vinh’s score, while 5% had
scores higher than Vinh’s.
Deciles
Divide an array into ten
equal pa s, each pa
having ten percent of
the distribution of the
data values, denoted by
Dj.
e 1st decile is the 10th
percentile; the 2nd
decile is the 20th
percentile…..
Quar tiles
Divide an array into four equal
pa s, each pa having 25% of
the distribution of the data
values, denoted by Qj.
e 1st qua ile is the 25th
percentile; the 2nd qua ile is the
50th percentile, also the median
and the 3rd qua ile is the 75th
percentile.
Steps in computing
Percentile, Decile and
Quar tile
Step 1: Arrange the data from lowest
to highest
Step 2: Determine the position of the
lower and upper value
P = j(N+1) /100 , D= j(N+1) /10 , Q =
j j j
j(N+1) /4
Step 3: Solve using interpolation
L= Lower Value + (Decimal) ( Upper Value –
j
Lower Value)
23
Example
Given the following set of
data 8, 14, 10, 15, 26, 28
Find P20
24
Given the following set of
data 8, 14, 10, 15, 26, 28
Find P20
Step 1: Arrange the data from lowest
to highest
8, 10, 14, 15, 26, 28
1st 2nd 3rd 4th 5th 6th
8, 10, 14, 15, 26, 28
1 2 3 4 5 6 N=6
st nd rd th th th
Find P20
Step 2: Determine the position of the
lower and upper value of P 20
P = j(N+1) /100
j
P = 20( 6+1) /100
20
P = 20(7) / 100 = 140/100 = 1.4
20
1. 4 is between 1 and 2 or
Lower Value = 1 = 8 st
Upper Value = 2 = 10 nd
26
Position Value = 1.4
Decimal = 0.4
Lower Value = 1 = 8
st
Upper Value = 2 = 10
nd
Step 3: Use interpolation method
P = Lower Value + (Decimal) ( Upper Value – Lower Valu
j
P20 = 8+ (0.4) ( 10– 8)
= 8+(0.4)(2)
= 8 + 0.8
= 8.8
27
Given the following set of
data 6, 8, 14, 10, 15, 26, 29,
30
Find D7
Step 1: Arrange the data from lowest
to highest
6, 8, 10, 14, 15, 26, 29,
30
1st 2nd 3rd 4th 5th 6th 7th
8th
28
6, 8, 10, 14, 15, 26, 29,
30
1st 2nd 3rd 4th 5th 6th 7th
8th N=8
Find D7
Step 2: Determine the position of the lowe
and upper value of D
7
D= j(N+1) /10
j
D = 7( 8+1) /10
7
D = 7(9) / 10 = 63/100 = 6.3
7
6.3 is between 6 and 7 or
Lower Value = 6 = 26
th
Upper Value = 7 = 29
th
29
Position Value = 6.3
Decimal = 0.3
Lower Value = 6 = 26
th
Upper Value = 7 = 29
th
Step 3: Use interpolation method
D = Lower Value + (Decimal) ( Upper Value – Lower Valu
j
D7 = 26+ (0.3) ( 29– 26)
= 26+(0.3)(3)
= 26+ 0.9
= 26.9
30
Let’s Tr y
Given the following set of
data
30, 42, 2, 5, 9, 14, 21
Find Q1 , Q3 , D7, P32
31
12,8,16,20,25,30,34
FIND
1. qua ile 3
2. decile 4
3. percentile 34
32
