20 of 20 people found the following review helpful:
1.0 out of 5 stars
A disappointing book, November 12, 2004
This review is from: Data Analysis with Excel®: An Introduction for Physical Scientists (Paperback)
This is an introductory book on Excel for physical scientists. Cambridge University Press deserves a compliment for a beautifully produced volume. Unfortunately, its contents are disappointing, because the text contains serious errors and omissions. The most obvious error is the statement, on page 308, that Excel does not provide built in facilities for fitting equations to data using nonlinear least squares. Excel does provide these, in the form of Solver, but the reader will look in vain for any mention of Solver in this book. (Figure 9.1 on page 365 shows that the author indeed has not bothered to activate the Solver Add-in.) The most serious omission is that the existence of user-definable functions and macros is not mentioned either. This leaves out two of the most powerful features of Excel: nonlinear least squares, and user programmability.
Another major problem with this book is that it doesn't show the reader how to use the spreadsheet effectively, but often goes out of its way to make easy things difficult. The almost exclusive emphasis in this book is on least squares methods, yet these are handled quite clumsily. On page 244, e.g., the linear correlation coefficient is computed from its formula by calculating the necessary sums, rather than by taking advantage of the fact that Linest, Regression, and Trendline all provide this parameter or its square. On page 284 the reader is shown the matrix algebra for fitting data to a parabola, and then told that "The built in matrix functions of Excel are well suited to estimating parameters in linear least squares problems", as if Linest, Regression, and Trendline are not there to take care of such tedious data manipulations. Likewise, on page 290, the user is not informed that Linest and Regression can also do multivariate analysis, but instead is instructed to do this the hard way, again by setting up and solving matrix equations. It is as if the author hasn't quite figured out yet that the spreadsheet has several built-in facilities specifically designed to make such least squares problems user-friendly.
In comparison with other books vying for the scientific spreadsheet market it is difficult to come up with any area in which Kirkup's book has the edge over its competitors: Billo (2nd ed., Wiley, 2001), Bloch (2nd ed., Wiley, 2003), de Levie (Oxford, 2004), Gottfried (2nd ed., McGraw-Hill, 2002), Liengme (3rd ed., Newnes, 2002), and Orvis (2nd ed., Sybex, 1996) all provide much more useful information, and don't make their readers jump through unnecessary hoops either.
Help other customers find the most helpful reviews
Was this review helpful to you? Yes
No
4.0 out of 5 stars
Mostly about data analysis and introductory excel usage, August 27, 2008
This review is from: Data Analysis with Excel®: An Introduction for Physical Scientists (Paperback)
It seems Kirkup's book revolves around data analysis mainly and does the spreadsheet stuff by the book (i.e. using primarily built-in tools that comes with Excel and thus the hard way) rather than using more advanced tool such as the Solver. It is rather rudimentary and is best suited for introductory laboratory courses in the physical sciences and engineering. I think the rating of 1 star as was given by one reviewer is a bit harsh; perhaps it deserves 4 stars out of 5 if viewed from the angle that the book covers mostly basic data analysis material and should be left at the purpose of introducing basic concepts only. Of course, there is always the option of approaching the same coverage of material using a book solely on data analysis such as John R. Taylor's "An Introduction to Error Analysis" or Louis Lyons's "A Practical Guide to Data Analysis for Physical Science Students" along with the Microsoft manual of the version of Excel used. Should be adequate as a supplementary for a second-year laboratory course (full-year) of a 4-year BSc Physics program based on the North American college curriculum. Not for advanced students.
Help other customers find the most helpful reviews
Was this review helpful to you? Yes
No
1 of 4 people found the following review helpful:
5.0 out of 5 stars
A must for engineering statistics, October 19, 2003
By A Customer
This review is from: Data Analysis with Excel®: An Introduction for Physical Scientists (Paperback)
If you are doing an engineering statistics course this book is of a hell of a lot more value than Engineering Statistics by Hubele, Montgomery and Runger.
This book teaches you how to do statistics using excel.
Should be aplicable for most statistics but is of great
assistance if your doing Engineering statistics and get stuck without much support.
Help other customers find the most helpful reviews
Was this review helpful to you? Yes
No
