Difference between revisions of "How to Lie With Statistics"

From Conservapedia
Jump to navigation Jump to search
m (→‎top: HTTP --> HTTPS #3, replaced: http://www.amazon.com → https://www.amazon.com)
 
(3 intermediate revisions by 2 users not shown)
Line 1: Line 1:
−
'''How to Lie With Statistics''' is a breezy but accurate introduction to the use of statistics, with an emphasis on applying critical thinking to statistical presentations in order to asses their underlying validity.  The book, first published in 1954, has a light, humorous style that makes the author's points easy to understand.  While some of the examples are dated, ''How to Lie With Statistics'' has been remained bestseller in its field, with over a half-million copies of the English language version sold, and it has been translated into numerous foreign languages.<ref>http://www-stat.wharton.upenn.edu/~steele/Publications/PDF/TN148.pdf</ref>
+
'''''How to Lie With Statistics''''' is a breezy but accurate introduction to the use of statistics, with an emphasis on applying critical thinking to statistical presentations in order to asses their underlying validity.  The book, first published in 1954, has a light, humorous style that makes the author's points easy to understand.  As one might guess from the title, this is a "how not to do it book", like [[Daniel Boorstin]]'s ''[[The Image]].''
  
−
The author, [[Darrell Huff]], was not as interested in writing a textbook in statistical technique as he was in examining how statistics can be used to persuade and influence.  Statistics are rarely used to provide objective facts to a reader, he contends, but are instead used to persuade.  "The secret language of statistics, so appealing in a fact-minded culture, is employed to sensationalize, inflate, confuse, and oversimplify," warns Huff.<ref>http://www.amazon.com/How-Lie-Statistics-Darrell-Huff/dp/0393310728</ref>  His purpose in writing the book is to highlight the manipulative potential in presented statistics, and to equip his readers with a basic level of statistical literacy to allow them to interpret statistics instead of accepting them blindly.
+
While some of the examples are dated, ''How to Lie With Statistics'' has been remained bestseller in its field, with over a half-million copies of the English language version sold, and it has been translated into numerous foreign languages.<ref>http://www-stat.wharton.upenn.edu/~steele/Publications/PDF/TN148.pdf</ref>
 +
 
 +
The author, [[Darrell Huff]], was not as interested in writing a textbook in statistical technique as he was in examining how statistics can be used to persuade and influence.  Statistics are rarely used to provide objective facts to a reader, he contends, but are instead used to persuade.  "The secret language of statistics, so appealing in a fact-minded culture, is employed to sensationalize, inflate, confuse, and oversimplify," warns Huff.<ref>https://www.amazon.com/How-Lie-Statistics-Darrell-Huff/dp/0393310728</ref>  His purpose in writing the book is to highlight the manipulative potential in presented statistics, and to equip his readers with a basic level of statistical literacy to allow them to interpret statistics instead of accepting them blindly.
  
 
The value of this of the book for learning critical thinking can be summarized by the last chapter, "How to Talk Back to a Statistic", where Huff presents five questions anyone should ask when presented with a statistical "fact":
 
The value of this of the book for learning critical thinking can be summarized by the last chapter, "How to Talk Back to a Statistic", where Huff presents five questions anyone should ask when presented with a statistical "fact":
 
:*Who says so?  
 
:*Who says so?  
−
:*How does the source know?
+
:*How does he know?
 
:*What’s missing?
 
:*What’s missing?
 
:*Did someone change the subject?
 
:*Did someone change the subject?
 
:*Does it make sense?
 
:*Does it make sense?
−
 
−
For example, the book is currently rated about 4.5 stars on amazon.com. <ref>http://www.amazon.com/How-Lie-Statistics-Darrell-Huff/dp/0393310728</ref> Using the book's method, the five questions can be answered accordingly (in the same order):
 
−
:*Amazon.com claims the book is rated 4.5 "stars".
 
−
:*The source has gathered ratings from anonymous buyers of the book.
 
−
:*There is not sufficient evidence that all the ratings are posted legitimately.
 
−
:*The subject has not been changed, but assumptions have been made about the ratings' authors.
 
−
:*The ratings seem to vary in quality enough to be considered legitimate at a glance.
 
−
 
  
 
==References==
 
==References==

Latest revision as of 15:09, April 9, 2019

How to Lie With Statistics is a breezy but accurate introduction to the use of statistics, with an emphasis on applying critical thinking to statistical presentations in order to asses their underlying validity. The book, first published in 1954, has a light, humorous style that makes the author's points easy to understand. As one might guess from the title, this is a "how not to do it book", like Daniel Boorstin's The Image.

While some of the examples are dated, How to Lie With Statistics has been remained bestseller in its field, with over a half-million copies of the English language version sold, and it has been translated into numerous foreign languages.[1]

The author, Darrell Huff, was not as interested in writing a textbook in statistical technique as he was in examining how statistics can be used to persuade and influence. Statistics are rarely used to provide objective facts to a reader, he contends, but are instead used to persuade. "The secret language of statistics, so appealing in a fact-minded culture, is employed to sensationalize, inflate, confuse, and oversimplify," warns Huff.[2] His purpose in writing the book is to highlight the manipulative potential in presented statistics, and to equip his readers with a basic level of statistical literacy to allow them to interpret statistics instead of accepting them blindly.

The value of this of the book for learning critical thinking can be summarized by the last chapter, "How to Talk Back to a Statistic", where Huff presents five questions anyone should ask when presented with a statistical "fact":

  • Who says so?
  • How does he know?
  • What’s missing?
  • Did someone change the subject?
  • Does it make sense?

References