Saturday, June 13, 2009

Manga Guide to Statistics: Revisited

from
http://www.scientificcomputing.com/article-da-Manga-Guide-to-Statistics-051509.aspx

Statistics with heart-pounding excitement (well, maybe)

John A. Wass, Ph.D.

Manga Guide to Statistics
This is something new. Not to the world, but to readers of this column. For those of you not into cartoons or comic books, Manga is (are) a genre of Japanese comic books and cartoons. They are usually drawn in black and white format and cover a wide range of subject matter. Wikipedia has some good background information, in case you are interested.

The publishers at No Starch Press saw fit to send me a copy of The Manga Guide to Statistics and, although possibly way under your heads, some of you might find this useful. At first glance, it may appear to be a Dummies Guide to statistics, and it does have a few things to recommend it to those terrified by anything that appears mathematical. This particular volume teaches very basic statistics by weaving the subject matter within the story of a young Japanese girl, Rui.

Rui is smitten by a young market researcher who her father brings home one day, since they work together and were having a drink in the neighborhood. She is immediately infatuated with the handsome young man and begs her father to have him teach her statistics as a ploy to get closer to her love interest. However, instead of coming himself, Mr. Igarashi sends another colleague, Mr. Yamamoto, a seemingly geeky young man. She is crushed but agrees to learn, still hoping to get closer to her love interest.

Through 190 or so pages, the tutor instructs her in data types, understanding numerical and categorical data, using histograms to simplify data, measures of central tendency and variability, standardization, probability, correlations and hypothesis tests. He ends up with a chapter on using Excel for many of the calculations illustrated in the previous chapters.

Although logically arranged and interesting in its approach, I don’t see it as meeting the needs of its target audience. In the preface, this is defined as researchers, business analysts, those wishing to know a little bit about statistics, and those wishing to know more. Except for the third group, this is just a bit too simplistic. It would be very suitable as an introduction for grammar school students in middle school.

It would have been wise to stress to an audience at such a level the dangers in the novice use of statistics in Excel! On the plus side, the author does point out that statistics is the branch of mathematics that is “most closely related to everyday life,” and this is a great hook.

Most cartoon sections are followed by an all-too-brief single-page narrative recapping the lesson, and a single worked example. Since the level is elementary and the examples simple, the complete hand calculation is shown, something that I would love to see in all advanced-level texts! Every example, both in the cartoon and at the end of the chapter, uses real-life examples that would be of interest to the average teenager and, thus, makes the text more relevant.

Unfortunately, when we get to something really core, such as the normal distribution, the author (through Mr. Yamamoto) merely advises the student to memorize it since, in the narrative, Rui seems to freak out at the math. As I mentioned, this is something a bit different.

So, if you have any interest in seeing the book, take a look at their Web site, www.nostarch.com. (Editorial Disclosure: Lest the readers think that this editor is down on cartoons, he must admit to a lifelong infatuation with Walt Disney’s Duck clan. He was especially impressed, as a young boy, to see an indefinite integral among the 1+1=2’s on the board in Huey, Dewey, and Louie’s kindergarten class!)

Availability
The Manga Guide to Statistics by Shin Takahashi and Trend-pro Co., Ltd. No Starch Press Inc., San Francisco, CA. (November 2008), 224pp. $19.95. ISBN-10 1-59327-189-3, ISBN-13 978-1-59327-189-3.

John Wass is a statistician based in Chicago, IL. He may be reached at editor@ScientificComputing.com.


Scientific Computing
Rockaway NJ 07866

No comments: