Becoming a Successful Scientist: Strategic Thinking for Scientific Discovery

by Craig Loehle

Scientific research requires both innovation and attention to detail, clever breakthroughs and routine procedures. This indispensable guide gives students and researchers across all scientific disciplines practical advice on how to succeed. All types of scientific careers are discussed, from those in industry and academia to consulting, with emphasis on how scientists spend their time and the skills that are needed to be productive. Strategic thinking, creativity and problem-solving, the central keys to success in research, are all explored. The reader is shown how to enhance the creative process in science, how one goes about making discoveries, putting together the solution to a complex problem and then testing the solution obtained. The social dimension of science is also discussed from the development and execution of a scientific research program to publishing papers, as well as issues of ethics and science policy.

Genres: Science

258 pages, Paperback

First published December 1, 2009

Reviews

[Lendoxia · Rating 4 out of 5]

Some parts of the book are a bit boring than the others but all in all it's a really good and useful book. If you are or want to become a scientist it's a really good read, not to long either so it won't take too much of your precious time but it may give you perspective and interesting things to think about and maybe implement into your own work.

[Gavin · Rating 2 out of 5]

Pretty sensible but very long-winded and staid. I suppose it is actually quite vigorous and irreverent, coming from an organisation man:

study can be a substitute for productive work

Darwin considered himself to be a geologist, but the world remembers largely his biology. Should Goethe be in the literature, biology, physics, or philosophy department? He actually was most proud of his work on optics, though that work was largely flawed. Would Newton or Fisher find comfortable academic niches today?

All graduate students are taught that it is essential to become an expert. As a short-term goal this is, of course, valid. Academic search committees are also looking for experts. As a lifestyle, however, becoming an expert can inhibit creativity... As one becomes more of an expert, a larger and more complex network of facts and explanations accumulates and solidifies, making it difficult to entertain radical alternative ideas or to recognize new problems... An Aristotle or Freud may create a set of bars within which most people pace rigidly, never noticing clues from outside the cage

it is much more likely that one can work at 100% mental clarity for about four hours. If one keeps this in mind, then a distinction can be made between critical issues that need full clarity and intense effort, which become part of the four hours of work per day, and those parts of a project that are routine and become part of the rest of the day... returning calls, coding a clearly designed subroutine, ordering equipment, attending seminars, editing reports, etc

But these are the only interesting bits in 300 pages. This is true but the book doesn't help much:

scientists are largely uncoached and are rarely introspective. They spend a lot of
time studying their disciplinary subject matter, but almost no time learning strategies of problem solving

---

* He gives examples from many different domains (ecology, epidemiology, physics, hardware), but so I spotted some errors.
* A new record! Loehle cites a crashing falsehood on p.2, and several times elsewhere: Gardner's DOA theory of multiple intelligences. And he naively teases out the strategic implications. He harps on this theme repeatedly - e.g. this is also flatly false: above a certain minimal level, IQ and college grades are not predictive of productivity, success, or innovation.
* He tries to talk about software but is stuck in the 90s. He has no sense of open-sourcing software as the most successful strategy (witness XGB or Chollet).
* It tries to also appeal to business people, for some reason. Half the examples and advice are about corporate decisionmaking (The proper focus or perspective is essential when looking at business performance.) This is distracting and makes it feel generic, belies the title.
* His "new model of the scientific method" is vague and doesn't deserve the word.
* The chapter on social distortions (credit stealing, cherry-picking, trends and irrational effects of publication timing) is ok, just ahead of the Replication crisis curve.

---

Anyway: as a scientist you want problems. But not just any problem - something that both doesn't fit, & has important implications. Advice I read in this, or read into it:

* A theory can be inconsistent or incomplete: one generate contradiction, the other keeps explanations weak.
* Paradoxes are shortcuts to the frontier. Look for heated debates and find a synthesis / circumvention. When you have one: Find tacit assumptions; Make new distinctions; Operationalise!

---

(Alternative books: Medawar, Cajal, Polya, Hamming, anything by Feynman. Stenhardt's model is more rigorous than the rest put together but I don't know if it's helpful.)

[Zhenia Vasiliev · Rating 4 out of 5]

An enjoyable read, I quite liked the proposed central role of pattern recognition in scientific scholarship, as opposed to eureka view. And also the advice to walk more and work four hours a day!