One of the illuminating things that you discover when you study statistics and logic formally is that peoples' arguments are based on poor reasoning and/or dodgy math (particularly dodgy statistics). This is especially true of much of the arguments of so-called opinion-leaders that are communicated via the media. I suspect that, if you're reading this review, you already have a sense of this as an issue. The great thing about a book like this one is that it can give you the intellectual tools to unpack these arguments to identify their fatal flaws.
Part One of the book describes "some indispensable tools for critical thinking" which is really a discussion about how words themselves can be used to try to manipulate you into thinking one way or another about a subject without you realising it.
Part One then goes on to explain the basic construction of a logically valid argument and why some arguments are invalid purely by their construction (regardless of the merit of the issue being argued for). This is followed by an explanation of the common fallacies in argumentation and is great stuff because the author explains most of these fallacies very well.
After this, Part One deals with math (specifically probability, statistics and graphs, etc) and how it can be used to manipulate people. That said, let me rush to add that this isn't a math textbook so it's not heavy-duty. However, there is enough to give you a grasp of some of the basic issues that will help you to develop a healthy scepticism of the reported results of opinion polls and quasi-scientific research.
Part Two of the book is possibly more challenging. Not because the theory is difficult (actually, most of the theory is relatively straight-forward) but because it uses the latest research into human perception, memory and judgement to challenge what you think you see and hear. And, if it does, you'll certainly be more sceptical about what other people tell you that they saw and heard. In the end, you'll find that people just can't be trusted - Not because they're liars but because they're oh so human.
Do I think that there is anything wrong with the book? Well, yes - There are a few minor problems (but none that would stop me from buying the book). First, the author can be a little patronising at times (especially in Part One). I doubt that he feels superior necessarily. I just don't think he realises that his audience is reading the book because they're already clever enough to have a sense that something is wrong and they've come to book because they want to do something about it.
Second, I'm not sure whether it's the fault of the author or the translator but, unless you know the definition of words like coda, dissemble, utilitarian, etc, you'll need to keep a dictionary close by (though mostly for Part One of the book, Part Two is much better in this regard).
Third, although the author's examples are valid, some can be a little silly. For example, one that he uses throughout the book involves the New York Police Department and a rather silly brand of "billy club" (which I presume is a baton). Couldn't he think of a more inclusive example?!