Statistics and Econometrics

Lesson 3: Central Tendency and Dispersion

Zhejin ZHAO

Statistical Analysis for Business & Economics: Spring 2011

 Introduction
Recall Lesson 2, where we used graphical techniques to
describe data:
Histogram
18

16

14

12

10
Frequency

6 Frequency
4

0
0~5 5~10 10~15 15~20 20~25 25~30 30~35
Waiting time

While this histogram provides some new insight, other

interesting questions (e.g. what is the average of waiting time?)
remain unclear.
2
 Introduction: Summarizing Distributions

3
 3 Measures of Central Tendency

Statistic Formula Excel Formula Pro Con

Average Familiar and
of all the Influenced
uses all the
Mean data =AVERAGE(Data) by extreme
data
values.
information.

Middle May not be

Robust when
value in influenced
Median =MEDIAN(Data) extreme data
sorted by extreme
values exist.
array values.

4
 3 Measures of Central Tendency (cont’d)

E.g. 1 2 2 2 2 4 7
Mean? Median? Mode?

5
 Notations

 When referring to the number of observations in a

population, we use uppercase letter N

 When referring to the number of observations in a

sample, we use lower case letter n

 The mean for a population is denoted with :

Parameter or Statistic?

 The mean for a sample is denoted with : Parameter

or Statistic?

6
 Characteristics of the Mean
Note: mean is very sensitive to
exceptionally large or small
observations called outliers
Examples
 As soon as a billionaire
moves into a neighborhood,
the average household
income increases beyond what
it was previously!

7
 Characteristics of the Mean

 Imagine if Ming Yao were in this class, what happens to

the mean height of the class.

8
 Median

 Defined as the value below which are 50% of observation

in the data set and above which are 50% of observation.

 Median separates the upper and lower half of the sorted

observations.

 If n is odd, the median is the middle observation in the

data array.

 If n is even, the median is the average of the middle two

observations in the data array.

9
 Median (cont’d)
Example 1
Consider the following n = 6 data values:
11 12 15 17 21 32

What is the median?

xn / 2  x( n / 21)
For even n, Median =
2
n/2 = 6/2 = 3 and n/2+1 = 6/2 + 1 = 4
M = (x3+x4)/2 = (15+17)/2 = 16

11 12 15 16 17 21 32
10
 Median (cont’d)
Example 2
Consider the following n = 7 data values:
12 23 23 25 27 34 41

What is the median?

For odd n, Median = x( n 1) / 2
(n+1)/2 = (7+1)/2 = 8/2 = 4
M = x4 = 25
12 23 23 25 27 34 41

11
 Median (cont’d)

• The median is insensitive to extreme data values.

• For example, consider the following quiz scores for 3 students:

Adrian’s scores:
20, 40, 70, 75, 80 Mean =57, Median = 70, Total = 285
Dustin’s scores:
60, 65, 70, 90, 95 Mean = 76, Median = 70, Total = 380
Josh’s scores:
50, 65, 70, 75, 90 Mean = 70, Median = 70, Total = 350

• In above case, is the median informative?

12
 Mode

 Defined as the most frequently occurring data value.

 May have multiple modes or no mode

cf.) Mean and median: unique to a data set

13
 Mode (cont’d)

An example

Consider the following quiz scores for 4 students:

Scarlett’s scores:
60, 70, 70, 70, 80 Mean =70, Median = 70, Mode = 70
Johan’s scores:
45, 45, 70, 90, 100 Mean = 70, Median = 70, Mode = 45
Tim’s scores:
50, 60, 70, 80, 90 Mean = 70, Median = 70, Mode = none
Ryan’s scores:
50, 50, 70, 90, 90 Mean = 70, Median = 70, Modes = 50,90

14
 Mean, Median, Mode

If a distribution is symmetric, the mean, median and

mode may coincide.
median
mode

mean

15
 Mean, Median, Mode

If a distribution is asymmetrical, say skewed to the left or

to the right, the three measures may differ. E.g.:

median
mode

mean

16
 Practice

1, 2, 3, 4, 2, 2, 3, 4, 5, 2

1. Calculate mean, median and mode of the data set.

2. Add 2 to each observation and recalculate mean,

median and mode.

3. Multiply each original observation by 2 and recalculate

mean, median and mode.

4. Describe how each measure of center changed with the

different operations.
17
 Tips: Which Measures to Use?

 For ordinal and nominal data the calculation of the mean is

not valid.
 Median is appropriate for ordinal data.
 For nominal data, a mode calculation is useful for
determining highest frequency.

18
 Outline of Dispersion

 Dispersion is the “spread” of data points about the center

of the distribution.
 Measures of dispersion:
• Range
• Variance
• Standard deviation
• Coefficient of variation

19
 Range

 The difference between the largest and smallest

observation.
Range = xmax – xmin

 An example: Tina’s homework score

85 98 87 83 84 7 86
Range = 98 – 7 = 91

 Drawback: determined by only the two extreme values

20
 Variance
N
2
 The population variance is defined as   xi   
the sum of squared deviations around 2  i 1
N
the mean divided by the population size.

 For the sample variance, we divide by n

2
n – 1 instead of n, otherwise it would tend   xi  x 
to underestimate the unknown population s 2  i 1
variance. n 1

Note! the denominator is sample size (n) minus one !

 Drawback: due to its units, hard to interpret

21
 Standard Deviation

 The square root of the variance.

 Explains how individual values in a data set vary from the
mean.
 Units of measure are the same as X.

Population N Sample n
2
standard
2
  xi    standard   xi  x 
i 1
 i 1
deviation s
deviation N n 1

22
 Standard Deviation (cont’d)

 Excel’s built in functions are

Statistic Excel population Excel sample
formula formula
Variance =VARP(Array) =VAR(Array)

Standard deviation =STDEVP(Array) =STDEV(Array)

23
 Calculating a Standard Deviation

 Consider the following five quiz scores for Stephanie.

(Table 412)

 Now, calculate the sample standard deviation:

n
2
  xi  x  2380
i 1
s   595 24.39
n 1 5 1
24
 Calculating a Standard Deviation

 The standard deviation is nonnegative because

deviations around the mean are squared.
 When every observation is exactly equal to the mean,
the standard deviation is zero.
 Standard deviations can be large or small, depending
on the units of measure.
 Compare standard deviations only for data sets
measured in the same unit.

25
 Coefficient of Variation
 The coefficient of variation of a set of observations is
the standard deviation of the observations divided by
their mean, that is:
• Population coefficient of variation = CV =

• Sample coefficient of variation = CV =

 This coefficient provides a proportionate measure of

variation, which is free of units
 It measures relative dispersion.
26
 Coefficient of Variation (cont’d)

Example 1
data sets with the same units

• Let’s compare two stocks you’re interested in:

“Penny” Stock (A) vs. Regular Stock (B)

X = $2.00 X = $100.00
s = $0.04 s = $2.00

• Question: Which stock is more unstable?

27
 Coefficient of Variation (cont’d)

• In terms of absolute dispersion, stock B seems to be more

unstable. However, to compare the dispersion, we often
adjust to the returns. i.e. using the CV, we have:

Stock A: CV = ($.04/$2)*100 = 2%
Stock B: CV = ($2 /$100)*100 = 2%

Statistic Formula Excel Pro

Coefficient Measures relative variation in

of s percent so can compare data
100  None
variation x sets even with different units.
(CV) (Unit free)

28
 Coefficient of Variation (cont’d)

Example 2
data sets with different units

• Compare variability for the following two variables.

National SAT mean = 1000 vs. GPA mean = 2.5

National SAT s.d. = 200 vs. GPA s.d. = 1.2

• Again, here CV becomes a useful tool for comparison.

Without CV, it is not possible to compare them.

29
 Measures of dispersion
Population Sample

Size N n
Mean

Variance

Standard Deviation S

Coefficient of  s
Variation CV= 
CV=
x
 Tips: Which Measures to Use?

 If data are symmetric, with no serious outliers,

can use range and standard deviation.
 If comparing variation across two data sets, use
coefficient of variation.
 The measures of variability introduced in this section
can be used only for numerical data.

31