Truth or Truthiness: Distinguishing Fact from Fiction by Learning to Think Like a Data Scientist is a new book by Howard Wainer that can serve as a manual for how to be a good skeptic.
Wainer is a statistician, formerly with the famous Educational Testing Service, and a professor at the Wharton School of the University of Pennsylvania. He is well known for his work in statistics and data presentation.
You know what "truthiness" is. It is a term coined by Stephen Colbert in 2005 to refer to assertions that are clearly true because of how they look, feel, smell, but that are in fact, not true. But they are truthy. You get the point.
Wainer's book is an exploration of cases that demonstrate the difference between truth and truthiness, with an eye towards training oneself to tell the difference, and in some cases, develop arguments about the true and truthy. Does Fracking really cause earthquakes? Are school children in the US over tested? Is tenure what it is claimed to be? For these and other questions, one needs to have evidence, and to know how to evaluate that evidence.
This is a good book, and it is fun. You can read many of the various chapters independently to follow your own interests. To give you an idea of what is included, here is the table of contents:
Part I. Thinking Like a Data Scientist:
Part II. Communicating Like a Data Scientist:
Part III. Applying the Tools of Data Science to Education:
Kind of pricey.
But I would be interested in reading chapter 6.
Maybe the cost at Amazon will come down for the Kindle version someday.
The lack of basic observational skills implies that Chapter 6 isn't going to help much, or at all.