Randomness, by Deborah J. Bennett (Hardcover, 2011)|
(You can print this review in landscape mode, if you want a hardcopy)
Reviewer: Mark Lamendola, author of over 6,000 articles.
In all probability, you will like this book.
Professor Bennett took a subject that most people consider dry and made it
interesting. Randomness, in case you don't know, is a concept fundamental to the
application of statistical science. Along with other statistical concepts, it's
commonly misunderstood. Professor Bennett made a solid contribution to the
literature, by making an understanding of randomness accessible to the lay
It's a fairly short book, and didn't take me long to read it. No dense text,
no complex math, no mind-numbing problems to work out. The book has a small form
factor (8.5H x 6.5W) and runs only 188 pages. So if you've had a fear of
statistics, don't worry. You can dive right into this book and gain an
understanding of randomness without breaking a sweat.
Professor Bennett writes in a conversational style. I think that, with this
particular topic, it's a struggle to write in a way that eliminates all possible
"eyes glazed over" moments for every possible reader. But Professor Bennett
seems to have met that challenge.
The real meat of this book doesn't take 188 pages. What I mean by that is
most of the book doesn't directly explain randomness. Most of it provides an
historical overview of how mathematicians and others contributed to, or
sidetracked, our modern understanding of the subject. Personally, I find this
interesting. The history does help show how you can arrive at wrong conclusions,
and that additional depth in the book just reinforces the much smaller content
that explains what randomness is.
So this book on randomness spends most of its time explaining what randomness
is not. And that's probably just as important as understanding what it is, if
you want a solid understanding rather than a superficial one. The bulk of this
pertains to gaming (mostly games of dice).
I did say there's no complex math and there are no mind-numbing problems to
work out. But there is math and there are problems. These are mostly to
illustrate the points being made. I think a book on a statistical science topic
would be confusing if it didn't use math to show you how the concepts actually
work. So, fairly simply math plus some "Think about it" kinds of problems help
make the concepts clear.
It's good that Professor Bennett explains related concepts, such as
probability, the theory of errors, uncertainty, random numbers, and calculating
odds. I would have liked to see more about Chaos Theory, as it's related to
randomness. But she didn't go into it except for touching on its edges in
Chapter 7 (Order in Apparent Chaos).
This book consists of 10 chapters. It has extensive notes and an extensive
bibliography. It makes a good foundation stone for anyone considering a course
of study that involves anything to do with statistical science.
I think it is also good for anyone who tries to draw conclusions from
anything that presents any sort of data. The rampantly poor use and presentation
of data is, to me appalling. People make non-existent associations, see random
sequences as patterns, see patterns as random sequences, see correlations that
do not exist, and see causal relationships where it's actually randomness they
are looking at. As I neared completion of this book, it dawned on me that these
problems exist because people don't understand what they are looking at because
they don't have exposure to the concepts involved.
The kinds of errors I just described lead to bad decisions. They are,
effectively, detuning a person's brainpower by a large number of IQ points. When
you aren't misled by poor presentations of data, it's like having an IQ boost of
a few dozen points. This book can help bring that kind of boost about, though I
also recommend grabbing a few of the titles from the bibliography so you can get
a primer on the related topics.
Add it to your collection. It's a tiny investment that will pay for itself
many times over, plus you will just seem smarter to everyone around you when
your interpretation of events or data or news or whatever is on a more solid
foundation than theirs and you can explain "what is wrong with this picture."
Why I like stats:
When I was obtaining my MBA, one of the professors held a PhD in statistical
analysis. He had honed his abilities in the Navy, where this kind of thing finds
important uses in seemingly impossible situations. It's hard to wrap your head
around some statistical concepts, while others are just obvious. And some things
that seem obvious simply are not true. Of the professors I had, this professor
was my favorite. We stayed in touch for several years after I graduated. His
statistical background gave him a wonderful way of clearly viewing the world. A
good understanding of statistical concepts, if not the math and its application,
can help you avoid becoming a victim of obfuscation.