Transcript Chapter 2: Visual Displays of Information (an introduction)

Intro to Probability and Statistics
i.e…“The stuff from the first bit of your book that
you need to know right now.”
Data is information that is collected.
Qualitative (attribute) data refers to non –
numerical data.
Quantitative (numerical) data represents
numerical observations. It falls into two
classes:
Data that is counted is called discrete.
Data that is measured is called continuous.
Descriptive statistics describes data that has
been collected.
2012 US Presidential Election electoral
results (by state)
Barack Obama (50.6% popular, 61.7% electoral)
Mitt Romney (47.8% popular, 38.3% electoral)
Descriptive statistics describes data that has
been collected.
2012 US Presidential Election electoral
results (by state, weighted by population)
Wyoming – 600,000
Delaware – 900,000
Barack Obama (50.6% popular, 61.7% electoral)
Mitt Romney (47.8% popular, 38.3% electoral)
Descriptive statistics describes data that has
been collected.
2012 US Presidential Election electoral
results (by county)
Barack Obama (50.6% popular, 61.7% electoral)
Mitt Romney (47.8% popular, 38.3% electoral)
Descriptive statistics describes data that has
been collected.
2012 US Presidential Election electoral
results (by county, weighted by population)
Barack Obama (50.6% popular, 61.7% electoral)
Mitt Romney (47.8% popular, 38.3% electoral)
Descriptive statistics describes data that has
been collected.
2012 US Presidential Election electoral
results (by county, “shades of purple”)
Barack Obama (50.6% popular, 61.7% electoral)
Mitt Romney (47.8% popular, 38.3% electoral)
Descriptive statistics describes data that has
been collected.
2012 US Presidential Election electoral results
(by county, “weighted shades of purple”)
Barack Obama (50.6% popular, 61.7% electoral)
Mitt Romney (47.8% popular, 38.3% electoral)
Inferential statistics makes predictions,
based on a random sample taken from a
population.
Inferential statistics makes predictions,
based on a random sample taken from a
population.
To approximate a parameter (a measure of a
population) we measure a statistic (the related
characteristic of a sample) with its
corresponding margin of error.
How should we present our data?
A company hires a statistician who samples 20
employees randomly, and records how many
products each produces in one week:
10 16 19 30 17 16 35 29 15 18
21 20 26 14 18 28 30 39 19 21
I.
Frequency Distribution
 Left column
 possible data values
 Broken into equally
spaced classes
 Right column
 frequency of
occurrence of
classes
10 16 19 30 17 16 35 29 15 18
21 20 26 14 18 28 30 39 19 21
Visual Displays of Information (an introduction)
II. Frequency Histogram
 A picture of the
frequency distribution
Visual Displays of Information (an introduction)
Other Types of Displays
Pareto Diagram (or Pareto Chart)
“finger”
“refused”
“dedicated”
My father was slaughtered by a six fingered man. He was
a great sword maker, my father. When the six fingered
man appeared and requested a special sword, my father
took the job. He slaved a year before it was done. The six
fingered man returned and demanded it, but at one tenth
his promised price. My father refused. Without a word,
the six fingered man slashed him through the heart. I
loved my father. So naturally, I challenged his murderer to
a duel. I failed. The six fingered man left me alive, but he
gave me these. I was eleven years old, and when I was
strong enough, I dedicated my life to the study of fencing.
So the next time we meet, I will not fail. I will go up to the
six fingered man and say, "Hello. My name is Inigo
Montoya. You killed my father. Prepare to die."
Visual Displays of Information (an introduction)
Other Types of Displays
Pareto Diagram (or Chart)
Stem and Leaf Plot (Stem Plot)
Dot Plot
Others?
One million dollars (100 of the previous packets)
One hundred million dollars (100 of the previous stacks...
on a pallet, for easy maneuverability!)
One billion dollars (10 of the previous palleted stacks)
One trillion dollars (make sure you see the guy over
there...and the pallets are stacked two high)...
...and here’s our national debt as of March 2009.
Chris Jordan: Skull With
Cigarette, 2007
...based on a painting by
Van Gogh...
200,000 packs of
cigarettes, equal to the
number of Americans who
die from cigarette smoking
every six months.
www.chrisjordan.com
Visual Displays of Information (an introduction)
One Last Thing...
How many keys do you have with you today?
(I’ll bring a histogram next time)
Homework for Intro To
Probability and Statistics