Error and the Growth of Experimental Knowledge

by Deborah G. Mayo

We may learn from our mistakes, but Deborah Mayo argues that, where experimental knowledge is concerned, we haven't begun to learn enough. "Error and the Growth of Experimental Knowledge" launches a vigorous critique of the subjective Bayesian view of statistical inference, and proposes Mayo's own error-statistical approach as a more robust framework for the epistemology of experiment. Mayo genuinely addresses the needs of researchers who work with statistical analysis, and simultaneously engages the basic philosophical problems of objectivity and rationality.
Mayo has long argued for an account of learning from error that goes far beyond detecting logical inconsistencies. In this book, she presents her complete program for how we learn about the world by being "shrewd inquisitors of error, white gloves off." Her tough, practical approach will be important to philosophers, historians, and sociologists of science, and will be welcomed by researchers in the physical, biological, and social sciences whose work depends upon statistical analysis.

Genres: Philosophy, Science, Nonfiction

Kindle Edition

First published July 15, 1996

Reviews

[Bolin Zhou · Rating 5 out of 5]

Growing up in two different worlds, (Spent 17 years educated in China and then over 8 years in UK & US), it is like reading a book from two different perspectives. Most of the world is not as simple as black and white, but when it is absolutely crucial to learn the world in the shortest, most efficient way. Simply put, learning from errors, not only is how any from blank kids learn to live but also how the programs like artificial intelligence and machine learning work. That's how scientists found out the logics behind cannot-be-explained phenomenon and how they revil the reveal the truth.

This book explicitly explained the idea from the Bayesian methods, Pearson Neyman Statistics, Type I Type II errors, Fisher's Frequentisms methods. It is no doubt that those are the methods developed in the last 100- 150 years and they are the ONLY methods we have to tackle the data world in the digital age.

So if you see Statistics and scientific methods as gibberish, there is no point starting this book, if you found mathematics too intricate and too boring, this is not the book for you. If you have the true passion for scientific methods, statistics and the ONLY way to tackle problems, this is your ride.

[Alexander · Rating 3 out of 5]

Long review in progress...