Measure of Dispersion
The measure of dispersion or variability is a
measure to determine how observed values differ from each other. Variability
is a charateristic of a distribution of data indicating how spread out
or homogeneous the data is.
1. Know the charateristics of and be able to
compute the range, interquartile range , and semi-interquartile range *
The range is the difference between
the the highest and lowest values in a distribution of data.
The interquartile range is the difference between the 1st and
3rd quartiles (25th and 75th percentile) of a distribution of data. It
ignores the most extreme observations and contain 50% of the middle observation
values.
The semi-interquartile range is half the interquartile range.
2. Know the meaning of sum
of squares.
In measures of average deviations the following intermediate statistics
are computed:
Difference: X - mean (sample or population), where X
is the set of sample observations.
When one value or number is substracted from another this is call a
deivation.
When the deviations are squared and added together, we have a sum
of square. Sometimes called a sum of square deviation. Example
if you have a series of errors, e and you find the average, ,
then the sum of square of the errors (SSE) is .
When you take the sum of square and divide it by the sample size you
get the mean of the squared deviations.
3. Know the meaning of the variance in terms
of the mean sum of squared
deviations.
The variance is the average sum of squared deviations from the
mean or the mean of the squared deviations.
The population variance is denoted by the Greek symbol, sigma
squared
or
.
The sample variance is denoted by the symbol, s2
.
4. Know how to determine the standard
deviation.
The standard deviation is the square root of the variance.
5. Know how to compute the sample and population
variances and standard deviations.
These are the formulas for the population and sample variances and standard
deviations:
Population variance, :
,
where
X are the data values, mu,
is the mean and N is the size of the population.
Population standard deviation, :
|
Sample variance, s2:
,
where
X are the data values, x-bar,
is the mean and n is the size of the sample.
Sample standard deviation, s:
, |
Group
Variance and
,
where
X are the data values, x-bar,
is the group mean and n is the size of the sample and f
is the class frequency..
Group Standard Deviations
, |
Alternate Formulas for
Variance and standard deviation
,
,
and |
Example: The age of officers
on and a local committee are: 55, 62, 73, 72, 62, 48,
50
If this are the total ages of committe members, find the committee average
age and the variation in their ages according to the formulas above.
Data arrange from smallest to largest:
48, 50, 55, 62, 62, 72, 73, N=7
and
Population Mean,
,
,
So standard deviation, |
Table of Values
,
N
= 7, N2 =49
|
Example: A group of boys
and men gathered each night to play basketball at the neighborhood court,
one night their age were recorded as: 12, 14, 13, 11,
23, 15, 19
Use this sample the get and ideal about the variance of the age of the
boys playing basket ball on any given evening.
12, 14, 13, 11,
23, 15, 19, n=7 and
Sample Mean,
,
,
So standard deviation, |
Table of Values
,
n-1
= 6
|
Workshop Problem: Find the variance
and standard deviation for the following catehorized data:
Data
Class Midpoint |
Frequency |
25 |
5 |
75 |
16 |
125 |
117 |
175 |
236 |
225 |
331 |
275 |
78 |
325 |
27 |
375 |
8 |
|
Solution:(click to see solution)
|
Use worksheet below:
Class
Index, i |
Class
Midpoint
X |
Frequency
f |
|
|
X2 |
|
1 |
|
|
|
|
|
|
2 |
|
|
|
|
|
|
3 |
|
|
|
|
|
|
4 |
|
|
|
|
|
|
5 |
|
|
|
|
|
|
6 |
|
|
|
|
|
|
7 |
|
|
|
|
|
|
8 |
|
|
|
|
|
|
9 |
|
|
|
|
|
|
10 |
|
|
|
|
|
|
n= |
|
|
|
|
|
|
Group mean
= |
Group Variance:
=
Group standard deviation: |
|