Thoroughly revised and updated, the second edition of this popular introductory psychology textbook introduces effective new teaching techniques as well as a range of new topics. Clear and engaging, the book provides a fundamental insight into how the mind works.
The book does an excellent job of explaining pyschological concepts, detailing the progress of psychological science, and relating material to useful, real-world examples. It even keeps up an interesting, sometimes sarcastic sense of humor. Perhaps its best aspect is its organization: rather than fall into the trap of historical progression (William James believed... but Freud said... and Maslow theorized...) that ensnares many intro psychology books, this one is organized thematically and works exceptionally well. The history is still there, but it doesn't overshadow our present knowledge of psychology. Overall, excellent textbook.
I really enjoyed the writing styles of the authors and found this book to be a very informative and entertaining read. That said, I was a little disappointed in the lack of rigor in the statistical methods section in Chapter 2. The authors use the word "odds" as a synonym for "probability" and this is not the case. If the odds in favor of an event are A to B, then the probability of the event is A/(A+B) (page 64). They use the word "correlation" when they should use "association." For example, on page 52 they mention that "being insulted" is correlated with "refusing to give the time of day." Technically, "correlation" is the measure of the strength of a LINEAR relationship between two QUANTITATIVE variables. Here the variables are categorical and the expression describing the relationship is one of "association" rather than "correlation." In their discussion of "statistical significance" on page 63, they are suggesting that the P-value (though they don't use this phrase) is the probability that the randomization "fails." The most important idea here is that the P-value is a CONDITIONAL PROBABILITY, conditional on the null hypothesis of "no treatment effect" being true. If there is no effect from the treatment, the P-value will measure the fraction of random assignments that will lead to a value for the test statistic at least as extreme as the one observed in the experiment. If the null hypothesis is false, then there is no probability model from which to calculate the P-value. It is only under the hypothesis of "no effect" that we can make a meaningful probability statement.
That said, I recognize that the authors are trying to avoid using technical jargon with social scientists, but I think that rigor in this area will help produce better social scientists in the future. I enjoyed the text immensely and, for that reason, would love to see this section improved.
