Statistics
provides of us with a power set of tools for describing things in our world and
making inferences about them. They can also rely on math and logic that seems counterintuitive and they
are subject to other pitfalls. Economist Charles Wheelan provides and accessible introduction
into how we can use, misuse and abuse statistics in Naked Statistics.
Data is everywhere. In my life time, the falling prices and
increased interconnectivity of computers have massively increased the
collection of data. It can be overwhelming. At the same time, my experience as
an engineer and government employee have left me frustrated with
lack of data on some issue and wonder what inferences I might draw and how much
I can rely on them. Statistics provides
us tools for dealing with these issues.
For instance, statistics
provides us a way to summarize lots of data with a simple number such as an
average (many people are familiar with sports statistics that summarize a
performance of a play or team over a game, season or even a whole career).
Statistics can help us find trends and estimate how much various factors may be
contributing toward those trends. Even in the case where there is little data,
statistics can help us evaluate the reliability of your conclusions (statistics
can’t prove something definitively, but it can quantify how likely you are to
be wrong).
“Statistics cannot prove anything with certainty.”-Charles Wheelan, Naked Statistics
Though he doesn’t delve
too deeply into the mathematics of statistics, he shows that the math is often
the easy part. Getting good data, designing experiments, constructing
reasonable hypotheses, and avoiding bias present many stumbling blocks that can
turn statistics into nonsense.
Not only that, people
can take advantage of the weaknesses of statistics to provide persuasive
support for wrong conclusions. Not everyone throwing around statistics intends
to deceive, but a few do. A few just make mistakes, too. Wheelan describes many
of the common mistakes people make while using statistics. This can help people
new (or not new to statistics) avoid them. Possibly more important, it can help
users of statistics recognize possible problems in how the statistics they use
are developed or interpreted.
“Statistics cannot be any smarter than the people who use them.”-Charles Wheelan, Naked Statistics
This is not a statistics
textbook. Wheelan does not delve into the details, but he does provide
intuitive explanations of the concepts and simple examples. A student of
statistics might find this book helpful in getting over some of the conceptual
hurdles that may get in the way of understanding the rest of the material.
If you’re interested in
this book, you may also be interested in
Wheelan, Charles. Naked Statistics: Stripping the Dread from
Data. New York: W. W. Norton & Company, 2013.
No comments:
Post a Comment