Understanding Psychology as a Science: An Introduction to Scientific and Statistical Inference

by Zoltan Dienes

An accessible and illuminating exploration of the conceptual basisof scientific and statistical inference and the practical impact this has on conducting psychological research. The book encourages a critical discussion of the different approaches and looks at some of the most important thinkers and their influence.

Genres: Psychology, Science, Nonfiction, Philosophy, Textbooks, Academic

184 pages, Paperback

First published February 15, 2008

About the author

b. 1963
Grandson of Zoltan Paul Dienes

Reviews

[Mihai Zodian · Rating 5 out of 5]

Common wisdom tells us science is clear and certain, based on a unique method. Z. Dienes reminds us that neither knowledge nor data interpretations are absolutes. Understanding Psychology as a Science is a handbook, useful for everyone interested in how to think about research, including in social sciences. The text is easy to read, even if the topics require a certain degree of patience.

Understanding Psychology as a Science is mostly about methodology, and the logic of research than about psychology. I recommend the book to anyone interested in understanding how science works. The book contains two parts, the first about K. Popper, Th. Kuhn and I. Lakatos, the minimum readings for contemporary epistemology. The other is about three rival statistical schools, and Z. Dienes focus is on them.

Two things I liked the most. First, Z. Dienes view about K. Popper and falsifiability. In political science, for example, the latter`s ideas became dogma and were used to police new arguments. When people started to read Kuhn, the throne crumbled, but people went overboard with relativism. In Understanding Psychology as a Science, the author shows that there is still a good point under all this rubble and insists that arguments should trump intellectual fashion.

Second, Z. Dienes explains in layperson language why statistical methodology is a mixture or a battleground of three major interpretations. These are the mainstream Neyman-Pearson approach, the Bayesian one, and the likelihood theory, and their description and analysis make Understanding Psychology as a Science a rare book among its kind; the schools are often not compatible and these differences are often masked by the technical language, but they do matter. For example, in political science, Gary King`s methodological project is based on the likelihood perspective, while the book he co-authored Designing Social Inquiry: Scientific Inference in Qualitative Research relies on Bayesian epistemology.

Understanding Psychology as a Science manages to explain and make accessible many topics that seem too technical for the layperson. Z. Dienes offers many illustrations, exercises, and online materials. Its drawbacks are that it demands some basic acquaintance with statistics, mostly t-tests, and it lacks conclusions. Nevertheless, it is always good to see that things taken for granted are often more debatable than they appear.

[Annas Jiwa Pratama]

This reread teaches me again that I’m very bad at learning about stats unless I’m actually using it for something. While the chapter on Neyman & Pearson flew by, reading through Bayes and Likelihood testing feels like chipping through concrete. I think I understood Bayes more this time around, but Likelihood testing, bar its utmost basics, is rather hard to comprehend. The book inevitably goes pretty technical as well for both Bayes and Likelihood, which for a total amateur like me is pretty hard if I’m not doing tinkering on hands (it has some matlab examples, though, if you have access to that).

The more philosophical early chapters are very digestible introduction to Popper, Kuhn, and Lakatos. There’s also a section on each chapters, even unto the more technical ones, about how we should use the ideas presented to critically assess research (particularly psychology), which certainly makes implementing these ideas a more practical and is especially useful for the those philosophy of science chapters as it takes them from being ‘mere’ theories and translates them into conceptual tools.

[Konrad · Rating 4 out of 5]

Dienes provides an introduction to some foundational topics that are sorely missing from the typical Psychology degree programme -- the philosophy of science (from several different perspectives) and a conceptual comparison of three main methods of statistical inference: the frequentist, Bayesian, and likelihood approaches. The book could have benefited from a more meticulous editor, but Dienes' writing is, for the most part, clear and insightful, and he provides very useful suggestions for further reading. I would highly recommend this book (alongside the free Coursera course on "Improving your statistical inferences" by Daniel Lakens) to anyone interested in the logic behind empirical science. (And I do hope that most students of related fields would feel spoken to here.)

[Chris · Rating 4 out of 5]

This is a very good 'further reading' or extension book for anyone with knowledge of basic statistics, and it's definitely relevant outside of Psychology. The book gets increasingly technical as it goes on, beginning with two chapters on the philosophy of science covering Popper, Kuhn, and Lakatos. These chapters are lucid and easy to understand. The book then goes on to set out three particular paradigms in statistical inference: the Neyman-Pearson method (which is what you were probably taught if you studied Psychology undergrad); Bayesian statistics, and finally likelihood inference. I learnt lots here despite my prior stats training.

I don't usually add textbooks to my read list but I did actually read all of this! I came across this book in Bodo Winter's 'Statistics for Linguistics' R textbook as recommended reading.

[Ylvali · Rating 5 out of 5]

Very clear and surprisingly interesting introduction into philosophy of science and statistics, much more fun than any statistics lecture I have ever been in. Must read for every psychology student, or other social and natural sciences that use hypothesis testing.

[Arjun · Rating 5 out of 5]

An interesting and an essential read. More so than usual statistics textbooks. Something that was disappointed is that the author doesn't criticize the likelihood approach (who he seems to be a fan of). However, for any budding social scientist, this should be a must read.

[Melissa · Rating 5 out of 5]

My best introduction to the philosophy behind statistical inference. It gave me some excellent foundations for understanding advanced statistics later, because it provides more on the logic of WHY - if you are someone who doesn't understand stats because you are not given enough of WHY something means something else- this is excellent. 10/10. Explains concepts really well with good examples.

[Sarah · Rating 5 out of 5]

Thorough dissection of the philosophy behind schools of statistics in psychology.

[Malin · Rating 3 out of 5]

Some parts were really interesting, but sometimes my brain was elsewhere.

[nano · Rating 5 out of 5]

Criminally underrated book
Best statistics book for non-mathematicians i have ever read

[Aleksandr · Rating 5 out of 5]

Highly recommended as a supplementary read after or during your Statistics undergrad class.

[John J Milton · Rating 5 out of 5]

I would recommend this book to anyone remotely interested in any scientific branch