The book speaks mostly about how the scientific method works, applying it to psychology. It attempts to show the difference between psychology as a science and all the pseudosciences that pretend to be part of the field: "In short, psychology has a kind of Jekyll and Hyde personality. Extremely rigorous science exists right alongside pseudoscientific and antiscientific attitudes."
Chapter 1 - Psychology is alive and well
Psychology is a young science and, because it puts under scientific investigation what was before considered to be "common sense", some people regard it with hostility.
Chapter 2 - Falsifiability
Falsifiability is introduced through the example of a doctor in 1793 who applied bloodletting to patients during an epidemic of yellow fever. If the patient became better, the result was attributed to the treatment, otherwise the doctor considered that the treatment was performed too late. Therefore, the theory that the removal of blood would heal the fever was impossible to falsify.
A theory must issue specific predictions. It should state what should and what should not happen.
As an example, the very complex freudian theory explains things related to human behavior after the fact, it can explain everything (not specific), but is unable to make predictions, therefore not falsifiable.
The author develops a theory that explains what drives human behavior: two little green men in the head that control the electrochemical processes taking place in the head. The little green men hide as soon as one would try to look for them (with X-rays for instance). Extrasensory Perception works the same: as long as nobody intrude to look into it, it functions.
Falsifiable theories allow science to move forward: as soon as a prediction made by a theory is falsified by experiments, an update or a new theory is required.
Chapter 3 -Operationism and Essentialism
Operationism: how does it work ? Concepts of scientific theories need to be linked to observable events.
Essentialism: What is the real meaning of .. ? What does it ultimately mean to speak of .. ?
Preexisting bias problem - when it comes to psychology, people come with emotionally held beliefs. People seem not to accept the need of operationalism. One problem is the use by the scientists of words from day to day life as technical terms. After all, don't we all understand concepts like smart, aggressive, anxiety ? We do, but not in the same way a psychology scientist does: he needs to specify how another lab could measure these in the same way and reach the same results.
The public expects psychology to answer essentialist questions, and, as it does not, the popular belief that there are no advances in psychology.
Chapter 4 - Testimonials and Case Study Evidence
Testimonials and case studies are useless in the evaluation of psychological theories: any practitioner of any pseudoscience can bring together a group of people to testify how effective his procedures are. This is mostly because of the placebo effect.
Chapter 5 - Correlation and causation
Two major classes of ambiguity in a correlation: the directionality problem and the third variable problem.
Several examples of selection bias are given. More than average deaths caused by respiratory problem in a region with clean air, Arizona: because people with respiratory problems tend to go there for the better air. And they tend to die there.
Chapter 6 - Getting things under control
During an experiment, scientists need to manipulate experiment variables in order to rule out alternative explanations. A control group is required to decide if a treatment is really effective.
Chapter 7 - But it's not real life
Psychology experiments sometimes attract critics such as "this is not real life".
Distinction between two kind of research: applied research, where findings of the study are applied directly to a situation and basic research, which "tests theories of the underlying mechanism that influence behavior". For the later, artificial situations are built to isolate the critical variable for study.
Chapter 8 - Avoiding the Einstein Syndrome
Connectivity principle: a new theory "must not only explain new facts, but account for the old ones". Most scientific discoveries are gradual, they rely in scientific consensus and not on breakthrough.
Unlike many pseudosciences, in science nobody has privileged access to the truth.
A theory is usually backed by many converging studies. Even if each of them may have specific flaws, the flaws are not the same and what matters is that the results converge. Views of a single expert have to be regarded with skepticism if the go agains the scientific consensus.
Chapter 9 - The misguided Search for the Magic Bullet
There is not a single cause to look after when explaining a phenomenon.
Chapter 10 - The Achilles Heel of Human Cognition
People are very bad understanding statistics and let themselves convinced easier by a single story than by numbers. There will always be a few people that goes against the strongest trend.
Especially in psychology there is a tendency to overweight single cases over abstract probabilistic information.
Chapter 11 - The role of chance in psychology
Should not try to come up with complex theory there where chance alone is enough to explain a phenomenon.
Clinical prediction does not work: when a practitioner tries to predict the behavior (for example recidive of aggressive behavior) based on clinical evaluation of a subject, they do a lot worse than what statistics predict for the case. When going with the statistics, we do know the rate of error and we have to accept it. The book calls this "accepting error in order to reduce error".
Chapter 12 - The Rodney Dangerfield of the Sciences
The bad image the psychology has is due to the pseudo-scientific literature addressed to the large public as well as "anti scientific attitudes within parts of psychology itself". The latter is about practices in psychology with no scientific grounds, and practitioners reluctant to submit themselves to experiment on the grounds that what they do is "more art than science".
The medias are more interested in quacks and pseudo scientists as they make better ratings. This also contributes the the bad image psychology has.