The effective design of scientific experiments is critical to success, yet graduate students receive very little formal training in how to do it. Based on a well-received course taught by the author, Experimental Design for Biologists fills this gap.
Experimental Design for Biologists explains how to establish the framework for an experimental project, how to set up a system, design experiments within that system, and how to determine and use the correct set of controls. Separate chapters are devoted to negative controls, positive controls, and other categories of controls that are perhaps less recognized, such as "assumption controls," and "experimentalist controls." Furthermore, there are sections on establishing the experimental system, which include performing critical "system controls." Should all experimental plans be hypothesis driven? Is a question/answer approach more appropriate? What was the hypothesis behind the Human Genome Project? What color is the sky? How does one get to Carnegie Hall? The answers to these kinds of questions can be found in Experimental Design for Biologists. Written in an engaging manner, the book provides compelling lessons in framing an experimental question, establishing a validated system to answer the question, and deriving verifiable models from experimental data. Experimental Design for Biologists is an essential source of theory and practical guidance in designing a research plan.
Written in November 2017: I read six chapters of it, but I prefer to read another book instead of this one and then come back and complete it. Reading this book needs both concentration and a higher level of English proficiency for me. I hope to enjoy and learn more from this book the next time I start reading it again.
UPDATE in February 2021: Finally, I read the book "Experimental Design for Biologists," written by David. J. Glass. Reading this book was a huge challenge for me due to its relation to the philosophy of science.
In 2017, I decided to approach the book, but I left it after reading six chapters. In 2020, I came back to read it with more confidence due to being more proficient at English. This time, I could understand many of its chapters better than before; however, I cannot claim that I have understood all of the book's parts. I should review the book later.
After reading this book and some other ones and finishing online courses such as "Statistics in Medicine" and "Writing in the Sciences" offered by Stanford Online, I feel more confident to move toward getting a PhD and becoming a scientist as soon as possible.
I should thank my professor, Dr. Yousof Gheisari, for introducing the book to me.
In lucid prose, Glass carefully takes the reader through the steps a scientist should take to carry out a successful and reproducible project. I especially enjoyed Glass's argument for the question over hypothesis to frame a project and his numerous examples regarding validating a system. I wish, however, more examples were given for the different types of controls. Overall, an excellent book for the experimental biologist.
Caveat: I have heard Glass talk about some of these topics before, but this book is a fine substitute for those who don't have the same opportunity.
the author addresses a very complicated issue that falls in the realm of philosophy, as simply as possible. he tries to demonstrate that "critical rationalists' approach has its pitfalls, and argues that "inductive reasoning" should be used in its stead. anyway this books is more about the philosophy of science than mere statistical design, and the argument certainly doesn't serve the biologists only, as per the title .
Warning: my review is based on the 2nd edition of this book. This is noteworthy, since as mentioned in the introduction to the latter, the content of the book has been majorly rearranged and expanded from the first to the second edition.
What I liked most of this book is that it doesn't deal with statistics at all. Normally, in experimental design books this would be a huge lack, and definitely not an upside. Nonetheless, in this case this is an advantage: 1) because it avoids repetition, in case you have a foundation in stats (and was it not the case there are plenty of options out there for you to catch up) and 2) because instead, it deals with topics that I haven't seen in other experimental design books, or at least it does in a simple and interesting way. If you want to learn about the general features of an experimental project, this is the book for you. Some of the main points treated here are: framing a research project,system validation (what type of data are we to measure? How? How do we ensure that the system is going to work?), Controls (positive, negative, controls for assumptions and so on) and model building and validation. Noticeably, a considerable part of the book is devoted to a discussion of two different experimental frameworks: the hypothesis-falsification one and the open-question one. Overall I do suggest this book but make sure to complement this one with another that goes into more detail on topics such as how to establish the sample size and how to carry out statistical analysis. Last, a small criticism to this book maybe be the way replication is treated. After reading "Experimental design for laboratory biologists" by E. Lazsic (similar title, I know) I found the treatment of this matter inappropriate in this one by Glass. The distinction between "biological replicates" and "technical replicates" is in my opinion a bit too superficial, so I would suggest to complement this part of the book with further reading as well.
Overall a very good book, addressing problems often overlooked in most of uni biology courses. However, despite the title, this book does not suit all biologists. It is tailored for molecular/cell biology, not for ecology or conservation biology.
I would recommend it to everyone who may forget the proper way to do biological research. Experimental design for biologists. I borrowed this book from the library (second edition). It helps me frame my research and determine to what extent I can interpret the results.