Book Review: Planung und Auswertung von Versuchen und Erhebungen. By D. Rasch, L. R. Verdooren and J. I. Govers
✍ Scribed by Wolfgang Näther
- Book ID
- 101717422
- Publisher
- John Wiley and Sons
- Year
- 2007
- Tongue
- English
- Weight
- 35 KB
- Volume
- 49
- Category
- Article
- ISSN
- 0323-3847
No coin nor oath required. For personal study only.
✦ Synopsis
This is the German version of the extended and revised 2nd edition of the bilingual textbook by Rasch et al. (1999). A parallel version in English entitled "The Design and Analysis of Experiments and Surveys" is available from the same publisher. The authors present basic methods for statistical analysis of designed experiments and surveys. The level of presentation is intended for students and experimental scientists in all disciplines, motivation and examples, however, usually have a biometrical background. Only elementary statistical knowledge is presumed which is summarized briefly in appendix B.
The authors philosophy is: Before experimentation think about the following question: What would you like to know and how exact would you like to know this? This question leads to the problem of a suitable design of experiments, which is answered in this book usually with an appropriate choice of sample sizes. Sometimesfor regression modelsfactorial designs and optimal designs are used. All the presented methods and models are illustrated with an example, sometimes rather (too?) detailed on pocket calculator level but usually by software, CADEMO or SPSS.
This book is divided into four main parts. The first part (25 pages) introduces into basic notions of experimental design like multiple measurements, principle of replication, principles of blocking and stratification, randomisation, block designs, factorial designs. The second part (67 pages) presents classical statistics (point estimation, confidence intervals, hypothesis testing) usually of the parameters of the normal distribution. The third part (48 pages) deals with different models of Analysis of Variance, e.g. one-way analysis in model I and II, several variants of two-way analysis in model I, model II and mixed models, multiple comparisons of means. Finally, the fourth main part (56 pages) focusses on Regression Analysis, especially on scatter plots, model I and model II, method of least squares, simple and multiple linear regression, simple polynomial and multiple quadratic regression, non-linear regression, optimal designs.
In the context of the general title "Design and Analysis of Experiments", the authors focus only on parametric methods. The interested reader cannot find any short remarks and additional references on the importance of nonparametric methods, bootstrap methods and simulation methods.
In summary, however, this book is a welcome introductory teaching tool at undergraduate and graduate levels in statistics and other disciplines which require statistical applications.