BKHFSTAT.RVW   20081203

"Head First Statistics", Dawn Griffiths, 2009, 978-0-596-52758-7,
U$34.99/C$34.99
%A   Dawn Griffiths
%C   103 Morris Street, Suite A, Sebastopol, CA   95472
%D   2009
%G   978-0-596-52758-7 0-596-52758-6
%I   O'Reilly & Associates, Inc.
%O   U$34.99/C$34.99 800-998-9938 707-829-0515 nuts@ora.com
%O  http://www.amazon.com/exec/obidos/ASIN/0596527586/robsladesinterne
  http://www.amazon.co.uk/exec/obidos/ASIN/0596527586/robsladesinte-21
%O   http://www.amazon.ca/exec/obidos/ASIN/0596527586/robsladesin03-20
%O   Audience i- Tech 2 Writing 1 (see revfaq.htm for explanation)
%P   677 p.
%S   Head First
%T   "Head First Statistics"

As with other similarly introductory books, there is a summary table
of contents followed by a second, more detailed one.  In this work,
there is also an explanatory paragraph, for each chapter, in the
itemized table.  In the introduction, the author states that the
reader of this text will learn to use statistics, for whatever purpose
they may wish, in a stimulating and easy manner.  (Most of the
introduction is a promotion for the visual, conversational, and
emotional style of the Head First series.)

Chapter one champions the idea of visualizing information.  Charts and
graphs are explained, but the examples are sometimes forced. 
Determining the central tendency can be problematic, and chapter two
raises the issues of concern, but doesn't really resolve them. 
(Confidence in the book is not helped by arithmetic errors: 10,000
multiplied by 1.1 does not equal 12,000.)  Range, quartiles, box
plots, standard deviation, and standard score are all means of
measuring variability, but the conversational style of the material
does not help once we get into the more advanced topics, and the
content starts to become confusing at this point.  Chapter four does
fine on calculation of the basic probabilities, but, again, the move
into set theory and dependencies strains this format.  (As a security
analyst I was particularly interested to see how Bayesian functions
were handled, but this section was far too terse to be useful.)  The
formulae for discrete probability distribution are presented fairly in
chapter five, and the inclusion of permutation is handled in six, but
when we get to geometric, binomial, and poisson distribution in seven,
the style is once again impeding the explanation.

Chapters eight and nine, using the normal distribution, are really
starting to get into calculus.  Statistical sampling, in chapter ten,
is primarily involved in issues of sample choice, rather than
mathematics.  Chapter eleven does use some of the calculations
introduced previously to predict, based on random populations.  Using
material from chapters three and four, chapter twelve examines issues
of confidence in our figures.  Testing of statistically-based claims,
as well as standard error types, is covered in chapter thirteen. 
Chapter fourteen introduces the chi-squared distribution for goodness-
of-fit and independence.  Correlation and regression are dealt with in
chapter fifteen.

On the title page is a photograph of a teenaged girl with a thought
balloon reading, "Wouldn't it be dreamy if there was a statistics book
that was more fun than an overdue trip to the dentist?  But it's
probably just a fantasy ..."  Unfortunately, for this book, that wish
does seem to be relegated to the realms of the fantastic.  There are
parts of the book that are fun.  There are parts that explain
statistics.  However, they aren't the same parts ...

copyright Robert M. Slade, 2009    BKHFSTAT.RVW   20081203