Statistically Off-Topic

B.A.: I was just wondering how many people here have read Darrell Huff’s How to Lie With Statistics. I always thought of it as somewhat obscure, yet every time I mention it among people with eclectic interests and knowledge, they always say “Oh, I read that too.”

Where better than here to check the hypothesis?


  1. Is there any way of making the query statistically meaningful? Bill, do you know how many unique people read a given post?

    But of course, yes, I have read it.

  2. I hadn’t heard of it, but I read the sample on Amazon, and what’s there is just what you’d get in an introductory stats course.

  3. Years ago I had a summer temp job in a college bookstore warehouse. During breaks I pulled books from the shelf to read, including that one. I barely remember it, though.

    I wonder how many of the people who say they read it are mistaken*. The book title has become a very common phrase, people may have heard the title so many times that it’s drilled into their heads that they read the book.

    * Maybe including me? I know I read The Peter Principle that year (a book that proposes people in a hierarchy are promoted until they reach their level of incompetence). Maybe reading that combined with repetition of the title convince me I read it. But I remember reading it.

  4. I’ve heard the title, but never read it. I’ve read a lot of other stats books, though. (I teach statistics.) And Usual John@1, I think in this format, the answer is no.

  5. Chak, the book’s not Statistics 101 (it was written by an English professor, I think), but rather about how statistics are used to influence, manipulate and mislead us, and how we can learn to analyze what what is and isn’t being said.

  6. I bought it used decades ago. Read and enjoyed it. Many years later found out it was considered a classic.

    At some time I bought his sequel.Not only is it not as good, but he definitely gets some things wrong. I would not recommend it. But I would *very much* recommend How To Lie With Statistics. It’s both informative and a pleasure to read.

  7. Read it decades ago, and still remember bits of it. I’m surpirsed it’s considered obscure now; I thought it was still a standard work.

  8. The title isn’t familiar to me. The idea is, of course, notably “lies, damned lies, and statistics.”

  9. I’ve either read “How to lie with statistics” or “How to take a chance” but it was about 50 years ago and I can’t remember which. Definitely read something by Darrell Huff though.

  10. Brian, maybe “lies, damned lies, and statistics” was his inspiration, since this book is an exploration on the axiom.

  11. There is also a “How to Lie with Maps” and the antidote: “The Visual Display of Quantitative Information” by Edward Tufte (among many other works).

  12. I read it around the time when I was in middle school, either late 70’s or early 80’s. The version I got had a copyright date of like 1950 or so My dad gave it to me to read. ( he was a doctoral candidate before he stopped going to school to work full-time. I didn’t have a copy of my own, so my daughter got a recommendation but not an actual copy.
    Recognizing when people are trying to manipulate you, and how, is a very valuable trait, especially now that (censored to comply with “discussion of politics policy” hereabouts)

  13. I was about to say: Not sure, but I definitely did read “The Visual Display of Quantitative Information” by Edward Tufte. But thanks to D McKeon for already introducing that to this discussion!

    Last month when I was at the veterinary office, at the clerk’s counter paying for my cat’s exam, a guy came in carrying the Tufte book. I tried to start a conversation, but the guy cut me off and said he was carrying it because he had taken it to get signed at a reading / signing where Tufte was promoting his latest one. A little curtly, if you ask me. Well, that’s dog people for ya, hmmph!

    I don’t know if it’s either of these titles or just common in math classes, to alert people to the dangers of graphs with a non-zero bottom line.

  14. Yes, I read it. I’m a geezer. I particularly liked the illustrations. It’s full of little cartoon illustrations. Also I liked the chapter on IQ. One kid tested at an IQ of 101. Wow, a genius, sure to get into college. The other had an IQ of 99. Oh dear. Better get some remedial education.

  15. No. Heard of it. Might read it some time. But I have read stuff like that. Also some “statistics for journalists” stuff and such. During my journalism days, being able to unpack the weasel words in a press release that was touting numbers of any kind was very helpful. Also, being able to look at an annual report or financial statement and get understanding out of it as well. Lots of players (companies, governments, individuals) would hope nobody would parse the actual meaning behind the numbers. “Average” is definitely a weasel-word in that context as the mean, median, and mode can all be wildly different and they can all be called “average’.

  16. SBill, according to the official statistics counter, I have more Samoans than anybody could ever ask for.

  17. @UrbanVariable, why can’t you do statistics while reading comics? Lots of people read (web-) comics at work.

    @CIDUBill, why are you placing a limit on how many Samoans someone can ask for? Allow me to disprove your assertion: I request aleph-null Samoans.

  18. I have not read it. Have been a reviewer for many statistically based academic journal articles. Last week published an article referencing Edward Tufte and “The Visual Display of Quantitative information,” still an Amazon Best Seller. (Tufte’s book, not my article.) And when I read your responses, I mutter Samoans.

  19. I’ve read that, and also the Tufte. I am pretty sure that all of them are in the room with me right now, although not within arm’s reach of my chair.

  20. Never read it, not sure if I’ve heard of it, but I did read both of John Allen Paulos’ big books (from the library.)

  21. CIDU Bill, no matter how many visitors you have, you can always use Samoa.

    Andréa, I think we should work at attracting more Milanos.

  22. Are you spying on me?? I was having some Double Dark Chocolate Milanos for brekkie.

    When Publix has a two-fer-one sale, we stock up on Milanos, Brussels [NOT sprouts!], and the seasonal varieties from Pepp. farm. Except pumpkin pie spice Milanos – they were pretty bad; so bad, there’s still half a bag left in the cookie cupboard.

    Hah! Bet you thought I was gonna write ‘cookie jar’, didn’t you . . . we need an entire cupboard for cookies. I’m Dutch and was raised on cookies. With coffee. With tea. With . . . more cookies.

  23. I haven’t read it, though I have read Huff’s How to Take a Chance and greatly enjoyed it. My second college major was mathematics, with a concentration in combinatorics and decision theory, so I’m interested in the subject…but I admit that the reason I remembered the book was that at the same time I was (re)reading a series by Tanya Huff and was amused I was reading so many Huffs…
    I was aware of the name of How to Lie with Statistics, but hadn’t realized it was by Darrell Huff – I’ll have to find a copy and read it.

  24. Read Huff maybe 50 years ago (age ~20), remember nothing about it, though I’m sure many of the lessons stuck with me, reinforced by later reading. Read the Tufte maybe 20 years ago, don’t recall anything about it except Napoleon’s retreat from Moscow and Florence Nightingale’s statistical diagrams. May still have it somewhere.

  25. Never read it, but it know how to do it.

    Based on our reenactment unit one could extrapolate a statistic that men in a group are more likely than women in the same group to have duplicate first names. In the decades we have belonged to our unit we have multiple Roberts, Jims, Pauls, and Chrises. We just had our first duplicate women’s names – Pat and Pat – but then again our unit commander (male) is Patrick.

    This is supported by the fact that we have had a number of presidents named John and a number named George and several other names are duplicated (lost my paper where I keep the count). Among their wives there a number of names that 2 of the wives have had, but only the Elizabeth/Betty/Eliza combination come in at more than 2 and they are variations of the same name, not duplicate names.

    Of course in the general public this is not true as many women have the same names as many other women. Hey, there are even other Meryls in the world. I have met 3 of them – 2 were girls and one was a boy. (Did not meet Meryl Streep as she was filming a hockey TV movie when we saw the Broadway play she had been in – and Meryl is her middle name.)

  26. That should be -” but I know how to do it ”

    (Was called upstairs midpost to deal with an emergency canceling order to pick up something at one Walmart to pick it up at a different one as the original order now claims not to have the item.)

Add a Comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.