Date: Tue, 29 Jun 2004 18:34:54 -0600

From: Gene V Glass
Content-Length: 1656

nice joke. very funny.

Leah S. Gordon wrote:
> Dear Professors Glass and Hopkins,
>
> I recently purchased your book, _Statistical Methods in Education
> and Psychology_ (3rd edition) for a Statistics course at Boston
> College.
> The book was expensive (well over $100), but I had been told that
> it was a good text. I have been dismayed to find careless errors
> in the scientific examples in the text.
>
> For example, in section 2.8, the table refers to "Running Speed" but
> gives measurements in seconds (a unit of time). In section 2.5, temperature
> is given as an example of "Interval Measurement" when measured in
> degrees Fahrenheit. While technically this is a reasonable example
> of a variable that has consistent intervals and no "zero," it is a poor
> example because in fact there is such a thing as absolute temperature
> (Kelvin), with equivalencies in Fahrenheit, where there is a ratio
> correlation of heat and temperature. (By the way, you may have made
> an editorial error in referring to degrees Celsius in section 2.9 while
> you had Fahrenheit in the preceding.)
>
> (I am happy to list other errors if you wish.)
>
> I do not want to be too cranky, but I must say that I am disappointed
> to find this kind of carelessness in such an expensive and popular book.
> It would be great if I could have at least a partial refund. Please let
> me know whom to contact, as I'm sure it is not the responsibility of
> the authors.
>
> Thank you so much for your attention to this matter.
>
> Very Truly Yours,
> Leah S. Gordon

On the other hand, this book is not totally worthless. If you are well versed in statistics and have had several advanced classes, this book serves as an actually decent review manual.

Overall, in retrospect it was worth the buy. However, I am amazed by the number of professors who actually think this book is adequate for an introductory statistics class. (The typical professor retort "the authors are respected in the field and have published prolifically" doesn't really mean a lot in the real world, where people are more interested in learning and not really impressed by the number of publications a professor has).

I have to say that students do not like the book. They think it is too hard. They want something easy and COMPLETELY non-technical (and this book is not really very technical), If students are going to learn the material, which is by its nature technical, the book has to be somewhat technical. I try to ease the tedium by supplementing the book with MANY handout and other materials. For what it is the _Cartoon Guide to Statistics_ is actually pretty good and also not full of errors. This latter book is a good supplement to G&H. The cartoons add some fun but G&H is not really much more technical.

My advice to students is to go to the library and find lots of different stats books. Everyone learns differently and find one that helps you. Be aware though that the non-technical, really simple books tend to be written by people who don't know what they are talking about and so it's the blind-leading-the-blind. The best way to survive stats is to study every day. If you don't understand something find your prof or ta and have them explain it until you get it. Good luck & persevere!