Measures of

Variability
Chapter 4

This multimedia product and its contents are protected under copyright law. The following are prohibited by law:
• Any public performance or display, including any transmission of any image over a network;
• Preparation of any derivative work, including the extraction, in whole or in part, of any images;
• Any rental, lease, or lending of the program.
Introduction to Measures of Variability
• Alone, measures of central tendency yield
an incomplete picture of a set of data
EXAMPLE
Mean daily temperature in a year

Hawaii Arizona
75 degrees 75 degrees
(vary from (vary from
70-80) 40-100)
Introduction to Measures of Variability
• Alone, measures of central tendency yield
an incomplete picture of a set of data
EXAMPLE
Prison sentences of convicted defendants

Judge A Judge B
24 months 24 months
(always (vary from
24 months) 6 months to 6 years)
Introduction to Measures of Variability
• Alone, measures of central tendency yield
an incomplete picture of a set of data
EXAMPLE
Average grade in long exams

Jack Jill
60 points 60 points
(vary from (vary from
30-100) 40-70)
Introduction to Measures of Variability
• We need an index of how scores are
scattered around the center of the
distribution?
• How do scores vary?
• How spread out is a distribution of scores?
Introduction to Measures of Variability
• A measure of variability is the spread,
width, or dispersion of scores
• Are the scores spread out or clustered
together?
• Do the scores scatter over a wide range of
values or do they cluster close together?
The Range
• Range = difference between the highest and lowest
scores in a distribution
• In equation form:
RH L
R = range
H = highest score in a distribution
L = lowest score in a distribution
• For example, 100 – 40 = 60 degrees
The Range
• Advantage: quick-and-easy computation
• Disadvantage: totally dependent on only 2 score
values (2 extreme values) and ignores other scores
in the distribution
• Not a precise (or accurate) measure of variability
because it is so much affected by the score of only
one case
• Only a preliminary or rough index
The Mean Deviation

• Takes into account every score in a distribution

• Step 1: Take absolute deviation or distance of each
score from the mean = /(X – X)/
• Step 2: Add the deviations =  /(X – X)/
• Absolute values! (or else zero!)
• Step 3: Divide the sum by the number of scores N
=  /(X – X)/ / N
• Average or mean absolute deviations
Example: The Mean Deviation
• A researcher wanted to figure out how
long it took fresh grads to find jobs (in
terms of months). Using a sample of 6
alumni, these were the responses:

1, 4, 9, 8, 2, 6

Copyright © Pearson Education, Inc., Allyn & Bacon 2009

Example: The Mean Deviation

X (X – X) /(X – X)/
9
8
6
4
2
1
Example: The Mean Deviation

X (X – X) /(X – X)/
9 +4 4 X=5
8 +3 3 months
6 +1 1 jobless
4 -1 1
2 -3 3 MD = 2.67
1 -4 4 months
30 0 16
The Mean Deviation

• Can now make comparisons as to which distribution

has greater or less variability
• However, it is no longer widely used given that
absolute values are not useful in more advanced
statistical analyses
• To solve the problem with absolute values of the
mean deviation, we SQUARE the deviations from
the mean, add them together, and divide the sum by
the number of scores
The Variance (s2)
• Variance = average/mean squared deviations
• In equation form:

S 2

 (X  X ) 2

N
s 2 = variance
X  X 
2
= sum of the squared deviations from the mean
N = total number of scores
The Variance (s2)

X (X – X) /(X – X)/ (X – X)2

9 X=5
8 months
6 jobless
4
2
1
The Variance (s2)

X (X – X) /(X – X)/ (X – X)2

9 +4 4 16 X=5
8 +3 3 9 months
6 +1 1 1 jobless
4 -1 1 1
2 -3 3 9 s2 = 8.67
1 -4 4 16 squared
30 0 16 52 months???
The Variance (s2)

• Advantage: gives appropriate emphasis to extreme

scores or is more sensitive to the degree of deviation
• Disadvantage: because it squares, it is difficult to
interpret – the square of what??? (e.g. squared
months, squared degrees, squared IQ units)
• To solve the problem with meaningless squares, we
return to the original unit of measurement by taking
the SQUARE ROOT OF THE VARIANCE
The Standard Deviation (s)
• Step 1: find the mean
• Step 2: subtract the mean from each raw score to get the
deviation
• Step 3: Square each deviation and get the sum of the
squared deviations
• Step 4: Divide by the total number of scores N
• Step 5: Get the square root
• In equation form:

s
 ( X  X ) 2

N
The Standard Deviation (s)

X (X – X) /(X – X)/ (X – X)2

9 +4 4 16 X=5
8 +3 3 9
6 +1 1 1 s2 = 8.67
4 -1 1 1
2 -3 3 9 s = 2.94
1 -4 4 16 months
30 0 16 52
The Meaning of the Standard Deviation
• The standard deviation represents the average
variability in a distribution or the average
deviation from the mean
• The larger the s, the greater the variability = the
scores are more scattered or spread out
• The smaller the s, the lesser the variability = the
scores are more clustered close together
Interpreting the Standard Deviation
EXAMPLE
Prison sentences for robbery

Judge A Judge B
x = 33.0 months x = 33.0 months
s2 = 4.0 s2 = 65.0
s = 2.0 s = 8.1
The Meaning of the Standard Deviation
• Interpret based on the data given or relate to the
conceptual meaning of the data being analyzed
(e.g. measurement used for temperature, prison
sentence, grades)
• Useful for measuring the degree of variability in a
distribution or for comparing variability in
different distributions
• Also useful for calibrating the relative standing of
individual scores within a distribution – a standard
against which we assess the placement of one score
within the whole
Summary
• Measures of variability allow distributions
of data to be described more completely
• Three choices discussed:
• Range
• Variance
• Standard deviation
 Practice
• Get the range, variance, and standard
deviation
Municipality Suicide Rate

A 15
B 13
C 13
D 18
E 20
F 13

Copyright © Pearson Education, Inc., Allyn & Bacon 2009

 Practice
Jeff and Alice both received their scores on 5 quizzes. The table below
shows the results of their quizzes over 10 (Yeah, I know – sad. lol).
Use this data set to answer the questions in this practice quiz

Jeff Alice
4 6
9 5
3 7
8 5
9 6

Copyright © Pearson Education, Inc., Allyn & Bacon 2009

 Practice
On a 20-item measure of self-esteem
(higher scores reflect greater self-esteem),
five teenagers scored as follows:

16 5 18 9 11

Calculate the range, variance, and standard

deviation
Copyright © Pearson Education, Inc., Allyn & Bacon 2009