Q & ATHE PERFECT STATISTICS GIFT?
Michael R. Frey
Bucknell University
Newsletter for the Section on Statistical Education
Volume 2, Number 1 (Winter 1996)
We invite contributions to future Q & A articles. If you have
a question about teaching statistics, send it to us and we can
solicit answers. If you have both a question and answers,
write a short article and send it to us. Send correspondence
to Tom Moore. (Eds.)
In September of this year a man asked me to suggest a gift
suitable for his son who was enthusiastically studying
statistics at college. I made a few suggestions but was
satisfied with none of them. I teach statistics at Bucknell
University, a small school in central Pennsylvania, and have
few colleagues with whom to share these sorts of questions.
So I turned to the internet and sent a message to a distribution
list of about fifty statisticians, each like myself, professionally
isolated. This list, maintained by Jeff Witmer at Oberlin
College, can be reached at "isostat@oberlin.edu." Here is the
message I sent to the group:
Hello fellow isolated statisticians! I was asked this
morning by a student's father to recommend a statistics/
probability book to be given as a gift to his son. The son
is a thirdyear student at William and Mary, loves
statistics (who doesn't?!), and plans to go on to graduate
study in statistics. The father asked me to suggest a book
that I find indispensable and that his son would
appreciate. OK everyone, help me out. I thought to
suggest The Handbook of Small Data Sets by Hand et al.
or maybe Counterexamples in Probability. Of course, a
gift ASA membership might also work. Does anybody
have other suggestions? Thanks, Mike Frey.
This question evidently sparked some interest because of the
number and rapidity of the responses. Included were several
requests that I assemble the results and share them back with
the group. After brief editing this is what I sent back to the
group:
Hello isolated statisticians! I recently asked for
suggestions for a book gift that might be appreciated
by an undergraduate student of statistics. Here's an
edited, cutandpaste summary of your recommendations.
Thanks to everyone  Mike Frey, Bucknell University.

Statistics for Experimenters by Box, Hunter &
Hunter or The History of Statistics by Stigler.
These books were the most often recommended.

Edward Tufte's two books on graphics make nice gifts,
since they are so artistically put togetherone is The
Visual Display of Quantitative Information and the
other is Envisioning Information. These books are
truly works of art and contain information that
beginning statisticians should be required to learn.
Moreover, they are not books that a student is likely to
buy for a course. These two books received several
recommendations.

A useful reference set is the Johnson and Kotz series
on distributionsfour volumes. A couple of them are
available in new editions; these would be the ones to
purchase if one doesn't want to lay out for all four.
Two recommendations for this set.

ASA gift membership including subscriptions to The American Statistician
, Stats, and Chance.

Statistics: A Guide to the Unknown, by Tanur et. al.
a largely nontechnical collection of applications of
prob & stats to a wide variety of areas (predicting the
chance of an earthquake in the next few years,
estimating whale population sizes, forecasting
elections, looking at the bunt strategy in baseball).
Put out by Wadsworth Brooks/Cole.

Fisher's Statistical Methods, Experimental Design
and Scientific Inference or Student (a biography of
Gossett), both published by Oxford University Press.
R.A. Fisher: The Life of a Scientist would be good
too. Historical/biographical materials regarding
statisticians are hard to come by and make valued gifts.

Thisted, The Elements of Statistical Computing;
Kendall et. al., The Advanced Theory of Statistics;
De Finetti, Theory of Probability; Kotz and Johnson,
Breakthroughs in Statistics; Mosteller and Wallace,
Applied Bayesian and Classical Inference: The Case
of the Federalist Papers; Tanner, Tools for
Statistical Inference; and Feller, An Introduction to
Probability Theory and Its Applications.

Problem Solving: A Statistician's Guide by C.
Chatfield or Counting for Something: Statistical
Principles and Personalities, by W.S. Peters
(SpringerVerlag).

Don't underestimate C.R. Rao's Linear Statistical
Inference and Its Applications or a matrixtheory
usefulforstatistics book like Graybill's or Searle's.
Everybody takes a linear models course but nobody
has had all the linear algebra that graduate school
assumes they've had.

Two recent books offer numerous datasets to
supplement an introductory statistics course. The books
differ in level and the extent to which sample analyses
for the data are supplied. One is A Casebook for a
First Course in Statistics and Data Analysis by
Chatterjee, Handcock and Simonoff (Wiley, 1995).
The other is A Handbook of Small Data Sets by
Hand et. al. (Routledge, 1994). Robert W. Hayden of
Plymouth State College has written a joint review of
these books which will soon appear in The American
Statistician.

Of course, a dream gift suitable for the newlyminted
stats Ph.D. is the multivolume Statistical
Encyclopedia. But it is ohsoexpensive!
The father, when I gave him this list, was amazed and
delighted. He's decided to give his son a student ASA
membership. Oh, and now he wants me to suggest a gift
for his wife ...
Michael Frey
Department of Mathematics
Bucknell University
Lewisburg, PA 17837
Phone: (717)5241598
FAX: (717)5243760
mfrey@bucknell.edu
Return to V2N1 Contents
Return to Newsletter Home Page