Transcript Chapter 5
Chapter 5
Measures of Variability
Golf Analogy: Which golfer is more consistent?
Each golfer hits 5
shots. The data are
distance from the
hole.
Mean
Variance
Beginner
Pro
-20
1
20
2
1
-2
-15
1
16
3
0.4
1
320.3
3.5
Variability
Variability is the spread or dispersion of a set of
scores.
Leptokurtic distribution has small dispersion.
Platykurtic distribution has large dispersion.
If the groups have a large amount of variance
the difference between the means will have to
be greater to find significance.
Interquartile Range
Difference between the 75th and 25th
percentile.
Variance is the average squared
deviations from the mean.
The sum of the deviations is ZERO
Calculation of the Variance
Standard Deviation
The SD is the square root of the variance.
The variance is in squared units
The SD is in original data units
Sigma is the population statistic
Degrees of Freedom: The number
of things that are free to vary
Assume that the mean of four values is 5
Therefore the sum must equal 20
Let 2, 3, and 7 be the first three numbers
What must the 4th value be so sum = 20?
It must be 8
In this example the first 3 numbers are FREE
TO VARY
The df for a single data set is (N-1)
[ see p 68 ]
DF and Sampling
Samples rarely contain the extreme
values found in the population.
A random sample of 100 university
men from a population of 15,000 is
unlikely to contain a subject with wt
350 or 100 although they exist in the
population.
The variability of the sample is almost
never as large as the population.
df is a correction factor so that a
statistic is not a biased estimate of the
parameter.
When N is large and the distribution is close to
normal there are 5-6 sds within the range
6 SDs