The scientific hypothesis is here to stay
- PDF / 432,513 Bytes
- 4 Pages / 439.37 x 666.142 pts Page_size
- 41 Downloads / 210 Views
The scientific hypothesis is here to stay Bradley E. Alger: Defense of the scientific hypothesis: from reproducibility crisis to big data. New York: Oxford University Press, xxvi + 416 pp, $45 HB Noah N. N. van Dongen1
© Springer Nature B.V. 2020
In a time of great methodological and philosophical upheaval, the publication of this book is auspicious and has the potential to be of great benefit to both junior and senior researchers. It ties together different perspectives and provides practical advice. However, the emphasis should be on potential, because this book also suffers from some serious defects that could have deleterious consequences if taken at face value. I will start this review with a summary focused on the laudable qualities. (a detailed evaluation of the individual chapters can be found at https://osf.io/9yp76 /). This is followed by highlights of the problematic elements of the book (a detailed list of all the errors I detected can be found at https://osf.io/5p2nx/). In conclusion, I attempt to embed the book into a broader context and, in light of this context, make some humble suggestions for improvement. Before I start, let me reiterate that I am impressed by this book. If my criticism seems overly harsh, it is because I want to see this book live up to its potential in the next edition. This book covers more than 400 pages over 15 chapters in 3 parts. The language is accessible, the chapters are clearly structured, and each chapter closes with a helpful summary. I especially liked the small jokes and the playful language [e.g., ‘… you are curious (OK, nosy) about …,’ 149], which made it easy to read a book on such a heavy subject. The first part, Chapters 1–9, lives up to its name ‘Fundamentals.’ Alger provides an accessible summary of the classical philosophy of science concerning the problems of induction, the limits of deduction, and that knowledge is always uncertain to some degree. From this philosophical basis, he argues for a fallibilist program of hypothesis testing, based on Popper (2005) and Platt (1964), where multiple potential explanations of phenomena, i.e., scientific hypotheses, are put to the test and only the survivors of such tests are retained. Alger makes a clear distinction between explanation, scientific hypothesis, and the predictions, statistical hypotheses, derived * Noah N. N. van Dongen [email protected] 1
Department of Philosophy and Educational Sciences, University of Turin, Turin, Italy
13
Vol.:(0123456789)
Metascience
from explanation. Two chapters are dedicated to this distinction and how the statistical school of Frequentism and Bayesianism relate to fallibilism and the testing of scientific hypotheses. Also, the lack of replicability of social sciences results, e.g., Open Science Collaboration (2015), is addressed in this part and mitigating arguments are provided against its severity. I especially liked the section on the cognitive advantages offered by the use of scientific hypotheses, such as how they guide thinking, clarify communicatio
Data Loading...
