Chapter Four

Numerical Descriptive Techniques

4.1
Introduction…
Recall Chapter 2, where we used graphical techniques to
describe data:

While this histogram provides some new insight, other

interesting questions (e.g. what is the class average? what is
the mark spread?) go unanswered.
4.2
Numerical Descriptive Techniques…
Measures of Central Location
Mean, Median, Mode

Measures of Variability
Range, Standard Deviation, Variance, Coefficient of Variation

Measures of Relative Standing

Percentiles, Quartiles

Measures of Linear Relationship

Covariance, Correlation, Least Squares Line

4.3
Measures of Central Location…
The arithmetic mean, a.k.a. average, shortened to mean, is
the most popular & useful measure of central location.

It is computed by simply adding up all the observations and

dividing by the total number of observations:

Sum of the observations

Mean =
Number of observations

4.4
Notation…
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 arithmetic mean for a population is denoted with Greek

letter “mu”:

The arithmetic mean for a sample is denoted with an

“x-bar”:
4.5
Statistics is a pattern language…

Population Sample

Size N n

Mean

4.6
Arithmetic Mean…

Population Mean Sample Mean

4.7
Statistics is a pattern language…

Population Sample

Size N n

Mean

4.8
The Arithmetic Mean…
…is appropriate for describing measurement data, e.g.
heights of people, marks of student papers, etc.

…is seriously affected by extreme values called “outliers”.

E.g. as soon as a billionaire moves into a neighborhood, the
average household income increases beyond what it was
previously!

4.9
Measures of Central Location…
The median is calculated by placing all the observations in
order; the observation that falls in the middle is the median.

Data: {0, 7, 12, 5, 14, 8, 0, 9, 22} N=9 (odd)

Sort them bottom to top, find the middle:
0 0 5 7 8 9 12 14 22

Data: {0, 7, 12, 5, 14, 8, 0, 9, 22, 33} N=10 (even)

Sort them bottom to top, the middle is the

simple average between 8 & 9:
0 0 5 7 8 9 12 14 22 33
median = (8+9)÷2 = 8.5
Sample and population medians are computed the same way.
4.10
Measures of Central Location…
The mode of a set of observations is the value that occurs
most frequently.

A set of data may have one mode (or modal class), or two, or
more modes.

Mode is a useful for all data types, though mainly used for
nominal data.

For large data sets the modal class is much more relevant
than a single-value mode.

Sample and population modes are computed the same way.

4.11
Mode…
E.g. Data: {0, 7, 12, 5, 14, 8, 0, 9, 22, 33} N=10

Which observation appears most often?

The mode for this data set is 0. How is this a measure of
“central” location?

A modal class
Frequency

Variable

4.12
=MODE(range) in Excel…
Note: if you are using Excel for your data analysis and your
data is multi-modal (i.e. there is more than one mode), Excel
only calculates the smallest one.

You will have to use other techniques (i.e. histogram) to

determine if your data is bimodal, trimodal, etc.

4.13
Mean, Median, Mode…
If a distribution is symmetrical,
the mean, median and mode may coincide…
median
mode

mean

4.14
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

4.15
Mean, Median, Mode…
If data are symmetric, the mean, median, and mode will be
approximately the same.

If data are multimodal, report the mean, median and/or mode

for each subgroup.

If data are skewed, report the median.

4.16
Mean, Median, & Modes for Ordinal & Nominal Data…

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 but not “central location”.

4.17
Geometric Mean…
The geometric mean is used when the variable is a growth
rate or rate of change, such as the value of an investment
over periods of time.

If Ri denotes the rate of return in period i (i = 1, 2, …, n),

then
The geometric mean Rg of the returns R1, R2, … Rn is defined
such that:

Solving for Rg we produce the following formula:

4.18
Finance Example…
Suppose a 2-year investment of $1,000 grows by 100% to $2,000 in
the first year, but loses 50% from $2,000 back to the original $1,000 in
the second year. What is your average return?

Using the arithmetic mean, we have

This would indicate we should have $1,250 at the end of our

investment, not $1,000.

Solving for the geometric mean yields a rate of 0%, which is correct.

The upper case Greek Letter “Pi” represents a product of terms…

4.19
Measures of Central Location • Summary…
Compute the Mean to
• Describe the central location of a single set of interval data

Compute the Median to

• Describe the central location of a single set of interval or ordinal
data

Compute the Mode to

• Describe a single set of nominal data

Compute the Geometric Mean to

• Describe a single set of interval data based on growth rates

4.20
Measures of Variability…
Measures of central location fail to tell the whole story about
the distribution; that is, how much are the observations
spread out around the mean value?

For example, two sets of class

grades are shown. The mean
(=50) is the same in each case…

But, the red class has greater

variability than the blue class.

4.21
Range…
The range is the simplest measure of variability, calculated
as:

Range = Largest observation – Smallest observation

E.g.
Data: {4, 4, 4, 4, 50} Range = 46
Data: {4, 8, 15, 24, 39, 50} Range = 46
The range is the same in both cases,
but the data sets have very different distributions…

4.22
Range…
Its major advantage is the ease with which it can be
computed.

Its major shortcoming is its failure to provide information on

the dispersion of the observations between the two end
points.

Hence we need a measure of variability that incorporates

all the data and not just two observations. Hence…

4.23
Variance…
Variance and its related measure, standard deviation, are
arguably the most important statistics. Used to measure
variability, they also play a vital role in almost all statistical
inference procedures.

Population variance is denoted by

(Lower case Greek letter “sigma” squared)

Sample variance is denoted by

(Lower case “S” squared)

4.24
Statistics is a pattern language…
Population Sample

Size N n

Mean

Variance

4.25
Variance…
population mean

The variance of a population is:

population size
sample mean

The variance of a sample is:

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

4.26
Variance…
As you can see, you have to calculate the sample mean (x-
bar) in order to calculate the sample variance.

Alternatively, there is a short-cut formulation to calculate

sample variance directly from the data without the
intermediate step of calculating the mean. Its given by:

4.27
Application…
Example 4.7. The following sample consists of the number
of jobs six students applied for: 17, 15, 23, 7, 9, 13.
Finds its mean and variance.

What are we looking to calculate?

The following sample consists of the number of jobs six

students applied for: 17, 15, 23, 7, 9, 13.
Finds its mean and variance.

…as opposed to  or 2
4.28
Sample Mean & Variance…
Sample Mean

Sample Variance

Sample Variance (shortcut method)

4.29
Standard Deviation…
The standard deviation is simply the square root of the
variance, thus:

Population standard deviation:

Sample standard deviation:

4.30
Statistics is a pattern language…
Population Sample

Size N n

Mean

Variance

Standard
Deviation

4.31
Standard Deviation…
Consider Example 4.8 where a golf club manufacturer has
designed a new club and wants to determine if it is hit more
consistently (i.e. with less variability) than with an old club.
Using Tools > Data Analysis… > Descriptive Statistics in Excel,
we produce the following tables for interpretation…

You get more

consistent
distance with the
new club.

4.32
Interpreting Standard Deviation…
The standard deviation can be used to compare the variability of
several distributions and make a statement about the general shape
of a distribution. If the histogram is bell shaped, we can use the
Empirical Rule, which states:

1) Approximately 68% of all observations fall within one standard

deviation of the mean.
2) Approximately 95% of all observations fall within two standard
deviations of the mean.
3) Approximately 99.7% of all observations fall within three standard
deviations of the mean.

4.33
 The Empirical Rule…
Approximately 68% of all observations fall
within one standard deviation of the mean.

Approximately 95% of all observations fall

within two standard deviations of the mean.

Approximately 99.7% of all observations fall

within three standard deviations of the mean.

4.34
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 =

4.35
Statistics is a pattern language…
Population Sample

Size N n
Mean

Variance

Standard
S
Deviation
Coefficient of
Variation CV cv
4.36
Coefficient of Variation…
This coefficient provides a
proportionate measure of variation, e.g.

A standard deviation of 10 may be perceived as large when

the mean value is 100, but only moderately large when the
mean value is 500.

4.37
Measures of Variability…
If data are symmetric, with no serious outliers, 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 interval data.

4.38
Measures of Relative Standing & Box Plots
Measures of relative standing are designed to provide
information about the position of particular values relative to
the entire data set.

Percentile: the Pth percentile is the value for which P percent

are less than that value and (100-P)% are greater than that
value.
Suppose you scored in the 60th percentile on the GMAT, that
means 60% of the other scores were below yours, while 40% of
scores were above yours.

4.39
Quartiles…
We have special names for the 25th, 50th, and 75th percentiles,
namely quartiles.

The first or lower quartile is labeled Q1 = 25th percentile.

The second quartile, Q2 = 50th percentile (which is also the

median).

The third or upper quartile, Q3 = 75th percentile.

We can also convert percentiles into quintiles (fifths) and deciles (tenths).

4.40
Commonly Used Percentiles…
First (lower) decile = 10th percentile
First (lower) quartile, Q1, = 25th percentile
Second (middle)quartile,Q2, = 50th percentile
Third quartile, Q3, = 75th percentile
Ninth (upper) decile = 90th percentile

Note: If your exam mark places you in the 80th percentile,

that doesn’t mean you scored 80% on the exam – it means
that 80% of your peers scored lower than you on the exam;
its about your position relative to others.

4.41
Location of Percentiles…
The following formula allows us to approximate the location
of any percentile:

4.42
 Location of Percentiles…
Recall the data from example 4.1:
0 0 5 7 8 9 12 14 22 33

Where is the location of the 25th percentile? That is, at

which point are 25% of the values lower and 75% of the
values higher?
0 0 5 7 8 9 12 14 22 33
L25 = (10+1)(25/100) = 2.75

The 25th percentile is three-quarters of the distance between the second

(which is 0) and the third (which is 5) observations. Three-quarters of the
distance is: (.75)(5 – 0) = 3.75
Because the second observation is 0, the 25th percentile is 0 + 3.75 = 3.75

4.43
 Location of Percentiles…
What about the upper quartile?

L75 = (10+1)(75/100) = 8.25

0 0 5 7 8 9 12 14 22 33

It is located one-quarter of the distance between the eighth and the ninth
observations, which are 14 and 22, respectively. One-quarter of the distance
is: (.25)(22 - 14) = 2, which means the 75th percentile is at: 14 + 2 = 16

4.44
 Location of Percentiles…
Please remember…

position
2.75 16

0 0 | 5 7 8 9 12 14 | 22 33
position
3.75 8.25

Lp determines the position in the data set where the percentile value lies,
not the value of the percentile itself.

4.45
Interquartile Range…
The quartiles can be used to create another measure of
variability, the interquartile range, which is defined as
follows:

Interquartile Range = Q3 – Q1

The interquartile range measures the spread of the middle

50% of the observations.
Large values of this statistic mean that the 1st and 3rd
quartiles are far apart indicating a high level of variability.

4.46
 Box Plots…
The box plot is a technique that graphs five statistics:
• the minimum and maximum observations, and

Whisker

Whisker (1.5*(Q3–Q1))

• the first, second, and third quartiles.

The lines extending to the left and right are called whiskers. Any points that lie outside
the whiskers are called outliers. The whiskers extend outward to the smaller of 1.5
times the interquartile range or to the most extreme point that is not an outlier.
4.47
Box Plots…
These box plots are based on
data in Xm04-15.

Wendy’s service time is

shortest and least variable.

Hardee’s has the greatest

variability, while Jack-in-
the-Box has the longest
service times.

4.48
Measures of Linear Relationship…
We now present two numerical measures of linear
relationship that provide information as to the strength &
direction of a linear relationship between two variables (if
one exists).

They are the covariance and the coefficient of correlation.

Covariance - is there any pattern to the way two variables
move together?
Coefficient of correlation - how strong is the linear
relationship between two variables?

4.49
Covariance…
population mean of variable X, variable Y

sample mean of variable X, variable Y

Note: divisor is n-1, not n as you may expect.

4.50
Covariance…
In much the same way there was a “shortcut” for
calculating sample variance without having to calculate the
sample mean, there is also a shortcut for calculating sample
covariance without having to first calculate the mean:

4.51
Statistics is a pattern language…
Population Sample

Size N n
Mean

Variance S2
Standard
Deviation S
Coefficient of
Variation CV cv
Covariance Sxy
4.52
 Covariance Illustrated…
Consider the following three sets of data (textbook §4.5)…

In each set, the values of X are the same, and the value for Y are the same;
the only thing that’s changed is the order of the Y’s.
In set #1, as X increases so does Y; Sxy is large & positive
In set #2, as X increases, Y decreases; Sxy is large & negative
In set #3, as X increases, Y doesn’t move in any particular way; Sxy is “small”
4.53
Covariance… (Generally speaking)
When two variables move in the same direction (both increase or both
decrease), the covariance will be a large positive number.

When two variables move in opposite directions, the covariance is a

large negative number.

When there is no particular pattern, the covariance is a small number.

4.54
Coefficient of Correlation…
The coefficient of correlation is defined as the covariance
divided by the standard deviations of the variables:

Greek letter
“rho”

This coefficient answers the question:

How strong is the association between X and Y?
4.55
Statistics is a pattern language…
Population Sample
Size N n
Mean

Variance S2
Standard
Deviation S
Coefficient of
Variation CV cv
Covariance Sxy
Coefficient of
Correlation r
4.56
Coefficient of Correlation…
The advantage of the coefficient of correlation over covariance is that
it has fixed range from -1 to +1, thus:

If the two variables are very strongly positively related, the coefficient
value is close to +1 (strong positive linear relationship).

If the two variables are very strongly negatively related, the coefficient
value is close to -1 (strong negative linear relationship).

No straight line relationship is indicated by a coefficient close to zero.

4.57
Coefficient of Correlation…

+1 Strong positive linear relationship

 or r = 0 No linear relationship

-1 Strong negative linear relationship

4.58
Coefficient of Correlation (Application)
Consider Example 4.16, where MBA grade point averages
are compared with GMAT scores. Is the GMAT score a good
predictor of MBA success?

Excel:
Tools > Data Analysis… > Covariance
Tools > Data Analysis… > Correlation
4.59
GMAT & GPA Interpretation…

The covariance is 26.16 and the coefficient of correlation is .5365.

These two statistics tell us that there is a positive linear relationship
between GMAT score and GPA.

The coefficient of correlation tells us that the linear relationship is

“moderately” strong.

4.60
Least Squares Method…
Recall, the slope-intercept equation for a line is expressed in these
terms:

y = mx + b

Where:
m is the slope of the line
b is the y-intercept.

If we’ve determined there is a linear relationship between two

variables with covariance and the coefficient of correlation, can we
determine a linear function of the relationship?

4.61
The Least Squares Method…
…produces a straight line drawn through the points so that
the sum of squared deviations between the points and the
line is minimized. This line is represented by the equation:

b0 (“b” naught) is the y-intercept,

b1 is the slope, and
(“y” hat) is the value of y determined by the line.

4.62
The Least Squares Method…
The coefficients b0 and b1 are given by:

4.63
Least Squares Line…
Example 4.17 :: find the least squares line for the previous
example (e.g. MBA / GMAT).

4.64
Least Squares Line…

4.65