University of Illinois at
Chicago
College of Business Administration
Department of Information & Decision
Sciences
IDS
270 Business
Statistics I
Textbook McClave, Benson
& Sincich (MBS)
Instructor Prof. Stanley L. Sclove
Commentary
to Accompany MBS Chapter 1:
Statistics,
Data, and Statistical Thinking
These notes Copyright ©
2001 Stanley Louis Sclove
These
notes, to accompany many of the chapters of the textbook, are closely
keyed to the book (as commentary) and are meant to be read right along
with them.
Data are numbers with a context.
The purpose of Statistics is the transformation of data into information.
The purpose of information in business is decision-making.
The purpose of decision-making is action.
The purpose of action is improvement of the general welfare.
Data
are numbers with a context.
Data
illuminate
We hope that
knowledge of facts changes incorrect attitudes.
Data beat
Anecdotes
Statistical
data are better than anecdotes and stories because the data view a whole
situation rather than just a few incidents.
Beware the Lurking
Variable
Almost all
relationships between variables are influenced by other variables lurking in
the background.
Where the Data
Come from is Important
Polls and
experiments must be statistically designed to yield unbiased information.
Variation is
Everywhere
Any number, to
be meaningful, must be accompanied by a range that describes our actual
knowledge.
Conclusions
are not Certain
Variation is everywhere, and conclusions are uncertain.
_______________________________________________________________________
What is Statistics?
What are
Statistics and What is Statistics?
Statistics
(plural) are derived (computed) from data.
Statistics
(singular) is the body of methods used to deal with data, by
·
computing
and interpreting Statistics (plural) and thus
·
transforming
data into information.
In
the first meeting we view the first program , "What is Statistics?"
from the video series Against All Odds: Inside Statistics. This first program shows examples of the uses of statistics in business
and other areas. Some of these examples will be expanded upon
later in the video series and the course.
The reasons
Managers need Statistics are the same as why everyone needs statistics:
·
Variation
is everywhere.
·
Conclusions
are not certain.
Most businesses now
are organized around a computer-based information system:
Five
types of resources are brought to bear in Information Management:
Information
is used to make decisions which manage systems. A system is a
group of elements, integrated with the common purpose of achieving an
objective.
A
typical system has an input element, transformation element and output
element.
The
systems view:
1.
Assists
the manager in seeing "the forest instead of the trees."
2.
Recognizes
the necessity of good objectives ("Efficiency is doing the job right;
effectiveness is doing the right job.")
3.
Emphasizes
the importance of parts working together.
4.
Acknowledges
internal and external connections
5.
Places
appropriate high value on feedback, achievable only by a closed-loop system.
A
massive dataset can hide the real information it contains. Underlying
patterns must be found to reveal the essence of what is there. This is the purpose of
Statistics.
We believe that
correct information, gleaned from data, leads somehow to the truth.
Contrasting data and information (truth). About the two major newspapers in Russia, there is this saying: In Izvestya ("Information"), no truth; in Pravda ("Truth"), no information.
information
processor
decision
(incl.
stat'l
analysis)
analysis
Data ---------------------> Information ---------------> Decisions
------------> Action
Data are numbers with a context.
The purpose of Statistics is the transformation of data into information.
The purpose of information in business is decision-making.
The purpose of decision-making is action.
The purpose of action is improvement of the general welfare.
Let's
start with two or three quotes that we'll try to dispell by teaching otherwise
during the course.
There are three kinds of lies: lies, damned lies, and statistics. -- British Prime Minister Disraeli
Round numbers are always false. -- Samuel Johnson
Statistics is the art of proceeding in a straight line from
an unwarranted assumption to a foregone conclusion. -- Source unknown
Now
let's consider some serious quotes.
Statistical thinking will one day be as necessary for
efficient citizenship as the ability to read and write. -- historical and science-fiction author H.G. Wells
It ain't so much the things we don't know that get us in
trouble. It's the things we know
that ain't so. -- Artemus Ward
(pen-name of
Charles Farrar Browne, American humorist, 1834-1867)
I have a great subject [statistics] to write upon, but feel
keenly my literary incapacity to make it easily intelligible without
sacrificing accuracy and thoroughness. -- Sir Francis Galton
And
we close our list of quotes with a paragraph from the preface of a book on
statistics by the well-known Norwegian statistician Erling Sverdrup:
It is perhaps not uncommon to
imagine a statistician as a person with a ravenous appetite for figures. Ths is perhaps correct, but it
certainly does not mean that the statistician has a desire to throw himself on
figures as soon as they are available.
On the contrary he will approach them with a certain caution; perhaps he
will in the introductory phase of the investigation not be interested in seeing
the data at all. Instead he will
want information about the assumptions, the experimental conditions or the
mechanism which were present when the figures were produced, and he would be
interested in the a priori insight which professionals within the field might
possess about the subject-matter which the data are supposed to throw light
on. Finally he will of course be
interested in what professionals do not know and might want to find out from
the observations. This is the
background for the figures, and it is an essential part of the task of a
statistician, as a statistician, to clarify this background and give it a
mathematical formulation (the model).
When this is done he can devise the method to be applied to the
data. The statistical method will
be derived from the a priori knowledge, and the reliability of the inferences
will depend both on the background for generating the data and on the method
that is applied. It is the first
duty of a statistician to clarify these relationships. When this has been done it is time to
look at the actual observations and perhaps look for more and other kinds of
observations.
Moore, The Basic Practice
of Statistics, 2nd ed. W. H. Freeman & Co.
Moore and McCabe, Introduction
to the Practice of Statistics, 3rd ed. W. H. Freeman & Co.
Created 1998: Dec 29 Updated 2001: Aug 22