Introduction to Statistics

Measures of Central Tendency

 Two Types of Statistics
• Descriptive statistics of a POPULATION
• Relevant notation (Greek):
  mean
– N population size
  sum

• Inferential statistics of SAMPLES from a

population.
– Assumptions are made that the sample reflects
the population in an unbiased form. Roman
Notation:
– X mean
– n sample size
  sum
• Be careful though because you may
want to use inferential statistics even
when you are dealing with a whole
population.

• Measurement error or missing data may

mean that if we treated a population as
complete that we may have inefficient
estimates.
– It depends on the type of data and project.
– Example of Democratic Peace.
• Also, be careful about the phrase
“descriptive statistics”. It is used
generically in place of measures of
central tendency and dispersion for
inferential statistics.

• Another name is “summary statistics”,

which are univariate:
– Mean, Median, Mode, Range, Standard
Deviation, Variance, Min, Max, etc.
Measures of Central Tendency
• These measures tap into the average
distribution of a set of scores or values in
the data.
– Mean
– Median
– Mode
 What do you “Mean”?
The “mean” of some data is the average
score or value, such as the average
age of an MPA student or average
weight of professors that like to eat
donuts.

Inferential mean of a sample: X=(X)/n

Mean of a population: =(X)/N
 Problem of being “mean”
• The main problem associated with the
mean value of some data is that it is
sensitive to outliers.

• Example, the average weight of political

science professors might be affected if
there was one in the department that
weighed 600 pounds.
 Donut-Eating Professors
Professor Weight Weight

Schmuggles 165 165

Bopsey 213 213
Pallitto 189 410
Homer 187 610
Schnickerson 165 165
Levin 148 148
Honkey-Doorey 251 251
Zingers 308 308
Boehmer 151 151
Queenie 132 132
Googles-Boop 199 199
Calzone 227 227
194.6 248.3
The Median (not the cement in the middle of
the road)

• Because the mean average can be

sensitive to extreme values, the median is
sometimes useful and more accurate.

• The median is simply the middle value

among some scores of a variable. (no
standard formula for its computation)
 What is the Median?
Professor Weight Weight

Rank order
Schmuggles 165 132
and choose
Bopsey 213 148
middle value.
Pallitto 189 151
Homer 187
If even then 165
Schnickerson 165
average 165
Levin 148
between two 187
Honkey-Doorey 251
in the middle 189
Zingers 308
Boehmer 151 199
Queenie 132 213
Googles-Boop 199 227
Calzone 227 251
194.6 308
 Percentiles
• If we know the median, then we can go up
or down and rank the data as being above
or below certain thresholds.

• You may be familiar with standardized

tests. 90th percentile, your score was
higher than 90% of the rest of the sample.
 The Mode (hold the pie and the ala)
(What does ‘ala’ taste like anyway??)

• The most frequent response or value

for a variable.

• Multiple modes are possible: bimodal

or multimodal.
 Figuring the Mode
Professor Weight

What is the mode?

Schmuggles 165
Bopsey 213
Pallitto 189
Homer 187 Answer: 165
Schnickerson 165
Levin 148 Important descriptive
Honkey-Doorey 251 information that may help
Zingers 308 inform your research and
Boehmer 151 diagnose problems like lack
Queenie 132
of variability.
Googles-Boop 199
Calzone 227
Measures of Dispersion (not something
you cast…)

• Measures of dispersion tell us about

variability in the data. Also univariate.

• Basic question: how much do values differ

for a variable from the min to max, and
distance among scores in between. We
use:
– Range
– Standard Deviation
– Variance
• Remember that we said in order to glean
information from data, i.e. to make an
inference, we need to see variability in
our variables.

• Measures of dispersion give us

information about how much our
variables vary from the mean, because if
they don’t it makes it difficult infer
anything from the data. Dispersion is
also known as the spread or range of
variability.
 The Range (no Buffalo roaming!!)
• r=h–l
– Where h is high and l is low

• In other words, the range gives us the

value between the minimum and maximum
values of a variable.

• Understanding this statistic is important in

understanding your data, especially for
management and diagnostic purposes.
 The Standard Deviation
• A standardized measure of distance from
the mean.

• Very useful and something you do read

about when making predictions or other
statements about the data.
Formula for Standard Deviation

S = ( X  X ) 2

(n - 1)
=square root
=sum (sigma)
X=score for each point in data
_
X=mean of scores for the variable
n=sample size (number of
observations or cases
 X X- mean x-mean squared
Smuggle 165 -29.6 875.2
Bopsey 213 18.4 339.2
Pallitto 189 -5.6 31.2
Homer 187 -7.6 57.5
Schnickerson 165 -29.6 875.2
Levin 148 -46.6 2170.0
Honkey-Doorey 251 56.4 3182.8
Zingers 308 113.4 12863.3
Boehmer 151 -43.6 1899.5
Queeny 132 -62.6 3916.7
Googles-boop 199 4.4 19.5
Calzone 227 32.4 1050.8
Mean 194.6 2480.1 49.8
We can see that the Standard Deviation equals 165.2
pounds. The weight of Zinger is still likely skewing this
calculation (indirectly through the mean).
 Example of S in use
• Boehmer- Sobek paper.
– One standard deviation increase in
the value of X variable increases the
Probability of Y occurring by some
amount.
 Table 2: Development and Relative Risk of Territorial Claim

Probability* % Change

Baseline 0.0401
development 0.0024 -94.3

pop density 0.0332 -17.3

pop growth 0.0469 16.8
Capability 0.0813 102.5
Openness 0.0393 -2
Capability and pop growth 0.0942 134.8

Change in prob after 1 sd change in given x variable, holding others at their means
 Let’s go to computers!
• Type in data in the Excel sheet.
 Variance

( X  X ) 2
2=
S (n - 1)
• Note that this is the same equation except for
no square root taken.

• Its use is not often directly reported in research

but instead is a building block for other statistical
methods
Organizing and Graphing
Data
 Goal of Graphing?

1. Presentation of Descriptive Statistics

2. Presentation of Evidence

3. Some people understand subject

matter better with visual aids

4. Provide a sense of the underlying

data generating process (scatter-
plots)
 What is the Distribution?
• Gives us a picture of
the variability and
central tendency.

• Can also show the

amount of skewness
and Kurtosis.
Graphing Data: Types
 Creating Frequencies
• We create frequencies by sorting data
by value or category and then
summing the cases that fall into those
values.

• How often do certain scores occur?

This is a basic descriptive data
question.
Ranking of Donut-eating Profs.
(most to least)
Zingers 308
Honkey-Doorey 251
Calzone 227
Bopsey 213
Googles-boop 199
Pallitto 189
Homer 187
Schnickerson 165
Smuggle 165
Boehmer 151
Levin 148
Queeny 132
 Here we have placed the Professors into
weight classes and depict with a histogram in
columns.
Weight Class Intervals of Donut-Munching Professors

3.5
3
2.5
2
Number
1.5
1
0.5
0
130-150 151-185 186-210 211-240 241-270 271-310 311+
Here it is another histogram depicted
as a bar graph.

Weight Class Intervals of Donut-Munching Professors

311+
271-310
241-270
211-240 Number
186-210
151-185
130-150

0 0.5 1 1.5 2 2.5 3 3.5

 Pie Charts:
Proportions of Donut-Eating Professors by Weight Class

130-150
151-185
186-210
211-240
241-270
271-310
311+
Actually, why not use a donut
graph. Duh!
Proportions of Donut-Eating Professors by Weight Class

130-150
151-185
186-210
211-240
241-270
271-310
311+

See Excel for other options!!!!

 Line Graphs: A Time Series
100

90

80
Approval

70
Approval

60

50

40

30

Economic approval
20

10

Month
 Scatter Plot (Two variable)

Presidential Approval and Unemployment

100

80
Approval

60
Approve
40

20

0
0 2 4 6 8 10 12
Unemployment