“Ethics and Statistics” is descriptive but boring. It sounds like the textbook for a course which, unfortunately, nobody will take.
“Lies, Damn Lies, and Statistics” is too unoriginal.
“How to Lie, Cheat, and Steal With Statistics” is kind of ok, maybe?
“Statistical Dilemmas”: maybe a bit too boring as well.
“Knaves and Frauds of Statistics, and Some Guys Who’ve Skated a Bit Close to the Edge”: Hmmm….
Maybe we have to get “statistics” out of the title altogether?
“Knaves and Frauds of Data Science”?
“Date Science and Data Fraud”?
“10 Things You Really Really Really Shouldn’t Do With Numbers”?
And, if no better idea comes along, there’s always “Evilicious: Why We Evolved a Taste for Being Bad.” (Regular readers will know what I’m talking about here; the rest of you can google it.)
Or maybe just “The Wegman Report”?
It’s hard to come up with a good title. Even John Updike had difficulties in this regard.
If any of you can suggest a better title for my ethics and statistics book, please let me know in the comments. Thanks!
P.S. The starting point for the book will be the series of columns on ethics and statistics that I’ve been running in Chance magazine.
Dewey & Cheatum, Data Scientists at Large
One idea is to use the “dilemma” angle as a reference to ethics, to avoid sounding too preachy:
Dilemmas in data: ethical practice in statistical science
(Dilemmas with data?)
Ethical dilemmas in data science
Others:
Stretching the signal: [insert your favorite subtitle here]
Truth, noise and fraud: [insert your favorite subtitle here]
Probabilistic truth: [insert your favorite subtitle here]
None of these are good… but I’d avoid the word ethics
I rather like “Dilemmas in data”; not sure about the second part.
“Ctrl+C, Ctrl+V: Data Fraud in the Digital Age”?
“Ctrl+C, Ctrl+V: Data Fraud is Easier To Do But Harder To Hide Than Ever”?
Title:
Numerate knaves
Tagline:
We are numerate knaves, all. Believe some of us.
How about “Unethical numbers”?
I prefer “Guilty Numbers”.
“Dishonest Data”?
“Statistical Shams”?
“Knumerical Knavery”?
Lying down with statistical lies: A one night stand with the ethic of stats
Years ago, a friend of mine thought about writing a book called Procrustean Statistics (Procrustes). I don’t think he ever got around to writing it. It might fit your book.
Gregg: If it doesn’t fit Andrew’s book, all he needs to do is trim the book down until it does fit.
Could you share more about the contents? E.g., the outline?
I would suggest that you shorten one of them and use “Things You Shouldn’t Do With Numbers”.
“Mostly Harmless Statistics” would be really funny.
+1
There is already “Mostly Harmless Econometrics”, though.
That’s the point… ;)
Values and p-values
“Statistikill : Death By Numbers” – I’d read that. Is it medical ethics, drug companies etc?
You could do some sort of pun on “playing by the rules?”
It captures the idea of stats being a science of default rules or practices, and also points at the ethical side.
I would go with a title that makes clear that stating precise “numbers” is not equivalent to stating “solid facts”…
So perhaps:
The illusion of precise numbers
The numbers’ illusion
or
How (not) to cheat with numbers
Really : short slogans are not easy – that’s why advertisement agencies take so much money!
Statistical Virtue / Numerical Deontology / A Critique of Statistical Reason / …
+1 to “How to Lie, Cheat, and Steal with Statistics”
Add:
– “Statistics: You’re doin’ it Wrong”
– “Corrupt Statistics”
– “Priors, Principles, and Posteriors”
– “The Moral Statistician”
– “Lying with Data”
>– “Statistics: You’re doin’ it Wrong”
“Statistics – Wat r u doin – Statistics – Stahp!”
With Low Power Comes Great Responsibility
+1. Quite possibly the best I’ve heard in months!
YES
Tim Ferriss has an excellent quantitative method used to rate book titles:
“He took 6 prospective titles that everyone could live with: including ‘Broadband and White Sand’, ‘Millionaire Chameleon’ and ‘The 4-Hour Workweek’ and developed an Google Adwords campaign for each. He bid on keywords related to the book’s content including ’401k’ and ‘language learning’: when those keywords formed part of someone’s search on Google the prospective title popped up as a headline and the advertisement text would be the subtitle. Ferriss was interested to see which of the sponsored links would be clicked on most, knowing that he needed his title to compete with over 200,000 books published in the US each year. At the end of the week, for less than $200 he knew that “The 4-Hour Workweek” had the best click-through rate by far and he went with that title.
His experimentation didn’t stop there, he decided to test various covers by printing them on high quality paper and placing them on existing similar sized books in the new non-fiction rack at Borders, Palo Alto. He sat with a coffee and observed, learning which cover really was most appealing.”
Ref: http://boingboing.net/2010/10/25/howto-use-google-adw.html
Niall:
Anything Tim Ferriss does, I don’t want to do!
How about “Standing on the toes of giants”, or “xxxxing on the shoulders of giants” to be a little cruder
“Significant Dilemmas”
“The Immoral Statistician’s Guide to the Universe”
“Lying, Cheating, and Stealing: 101 Ways to Use Statistics”
Freakonumbics
For your consideration:
“The Morality of Numbers”
“The Ethics of Data”
“How to be a Mad *Data* Scientist”
“Evil Numbers, and How to Avoid Them”
“Damned Statistics”
“How to Lie With Data”
“Lies, and the Lying Statisticians that Tell them”
How about “Data-Driving Lying”
Riffing on others above, what about “Coloring by Numbers: Dilemmas in Creatively Applying Statistics”
And just a quick thank you to you and your commenters. As a lurker I’ve benefited greatly from all of your work through the years.
+1 to “Data Science and Data Fruad”, with Zach’s “With Low Power Comes Great Responsibility” as a chapter title….
Fraud, not Fruad
Don’t wrory aobut it. Msot poeple prabobly dind’t even ntoice.
There is always the old standby “Sex, Dope and Lies”. When I was in high school a friend and I use to go to downtown Philadelphia wearing raincoats and peddle the “Realist” (yes I know this dates me and if you don’t know the Realist google it) by hiding it under raincoats except for a little bit of the cover and yelling “sex, dope and lies”. It was a powerful sales message.
Or you could do a takeoff on the old, popular poster, “expose yourself to fraud”.
“how to lie with statistics” is a 1993 book from Darrell Huff and Irving Geis, so you might want to stray from a very similar title.
you might want to give a little bit more info on what exactly the book will be about and what your target audience is. i personally love “values and p-values” and “With Low Power Comes Great Responsibility”. but i should not be your target audience, for the simple reason that i will most likely be aware of some of the contents of your book already.
if you want a more commercially successful book, i would say stay away from statistics puns, any title conveying that it might be a textbook and any variant of the “freak-”, whether “onumbers” or “atistics”.
Looking at my copy of “How to Lie with Statistics,” the original copyright is 1954. 1993 is the “Norton paperback reissue” which is the edition that I have. I just wanted to point out that it has a much longer history.
Search this blog for Huff for much more on this!
Cheating with Data
Torturing the Data (and Other Crimes Against Statistics)
Bigger the data bigger the liar.
Frequently False: Avoiding the Pitfalls of Statistics in Practice
Statistically Significant: Psych!
As well as a +1 to Zach above.
I agree with the google adwords method as laid out in Supercrunchers, if you can afford it. Cost unknown. Good luck!
How about “I Stole This Title from another Statistics Book”?
The Data Jungle.
“Ethics of Statistics: (catchy subtitle)”
Subtitle can include keywords: criminal, p-values, offence, lie, justice…
Though it all sounds a bit too gloomy :) Maybe something more cheerful could be nice as well.
Another big +1 to “How to Lie, Cheat, and Steal with Statistics.” I also like Marcel’s “How (Not) to Cheat with Numbers.” You don’t want to sound too coy or too much like a journal article. Colons and subtitles are so 2005.
Swept Under the Rug: The Subtle Lies of Statistics
Statistical Areté /Areté of Statistical Practice ?
“Not do the Wegman”
“The Moral Statistician”
“We don’t have to lie with statistics”
“The Honest Statistician”
“Truth and Data”
“Statistics Gone Bad”
How about 2005-vintage Fox TV, “When Statistics Attack”?
Probable Temptations: Ethics and Statistics
The title refers to the fact that at some point most people are presented with the temptation to abuse statistics: discard that robustness test, grab headlines with a point estimate while ignoring the wide error bands, purposely neglect to show counter-arguments so the reader doesn’t even think of them.
I like the title because it suggests that the ethical problems aren’t necessarily just with “those other people”.
Type E Error: Ethics and The Statistician
I like this!
Presumably, you will have both positive and negative examples. So your title should have both. I favor a “Knights and Knaves” theme.
I also think “Data Science” is by far the better marketing term. You’ll probably double sales by simply substituting that for “Statistics” in the title. Now combine these in a variety of ways and test them in the market through a survey:
Knights and Knaves of Data Science
Knights and Knaves: The Chivalry of Data Science
Knight or Knave? The Data Scientist’s Dilemma
Data Scientist. Knight or Knave?
White Knight, Black Knight: Ethical Struggles of a Data Scientist
Personally, I favor the 2nd one.
Dodgy Data: ((subtitle))
That’s assuming “dodgy” is a common term in the U.S. Otherwise, for the British edition only.
“Gelman’s razor: a Bayesian approach to ethical statistical practice”
“Beyond significance: ethics in the age of data science”
“Suckers! Stats for and by the morally impaired” by Wegman and Morris
“Wegman shot first: Surviving data science’s wretched hive of scum and villainy.”
Not as punchy as I’d like, but I’m procrastinating, not optimizing.
The Uncertainty Principles
Estimating the Truth
How to be Right (Probably)
Lying With Confidence
or, if you want to make journalists giggle,
About My Posterior…
Many good ideas, but first step along lines Niall described, but quick:
For any title that makes the short list, Google it, both with and without quotes
The Odds of Getting Caught
(or: What Are the Odds of Getting Caught?)
Perhaps:
Odds and Ends
(“ends” as in teleology, i.e., purpose, i.e., values, i.e., what ethics are about)
Odds and Ends is brilliant! It’s short, like the titles on Phil’s best sellers list and is clever. But maybe insert some kind of break (comma, or change in font) between ‘Odds’ and ‘and Ends’ to indicate you’re using the phrase in an unusual way.
Most Probably Neither True Nor False
Andreas’ Fables
Do as I say, not as I do: Applied statistics in the “real” world
Don’t know whether it has been suggested. How do you think of this: What you Can and Cannot Do with Statistics? — ethnics and statistics
Let’s take a look at the nonfiction bestseller list and see what titles are selling. Of course presumably these books are selling mostly because of content, not style, but at least we know the titles aren’t hurting them too badly.
Here’s the NYT hardcover list:
Killing Kennedy
Thomas Jefferson
Killing Lincoln
No Easy Day
How to Create a Mind
The Signal and the Noise
The Joy of Hate
The Last Lion
Far from the Tree
The Outpost
Bruce
America Again
Unbroken
Roll Me Up and Smoke Me When I Die
The Untold HIstory of America
Here’s the first part of the Amazon list, just ’cause that’s what came up first.
The Devil in the White City
Half the Sky: Turning Oppression into Opportunity for Women Worldwide
The Hero with a Thousand Faces (The Collected Works of Joseph Campbell)
Teach Your Child to Read in 100 Easy Lessons
The Road to Serfdom: Text and Documents–The Definitive Edition (The Collected Works of F. A. Hayek, Volume 2)
The Omnivore’s Dilemma: A Natural History of Four Meals
I Thought It Was Just Me (but it isn’t): Making the Journey from “What Will People Think?” to “I Am Enough”
Moneyball and The Big Short aren’t on these lists anymore but I liked ‘em so I’m throwig them in here too.
Maybe it makes sense to throw out the books that are bestsellers because of the celebrity of their authors; that would get rid of both “Killing” books, and The Signal and the Noise, and America Again.
Overall it looks like you want a title that does not sound at all academic, and that is very short (or at least very short before the colon if there is one). And, oddly, these titles do _not_ tend to play up opposition or conflict (this surprises me). Only “The Signal and the Noise” and “I Thought It Was Just Me” even hint at a juxtaposition (and you have to to go number 32 on the NYT list to find another, “Thinking, Fast and Slow.”
I like some of the titles people have suggested above…but most of them are nothing like the bestseller titles. Just because I like them doesn’t mean they’ll get people to look at the book. There’s nothing on the current list that is anything like “Knaves and Frauds of Data Science.”
Tom suggested something like “Pooping on the Shoulders of Giants,” I love this one, but I fear it will turn people off.
Using all of the information I have carefully gathered to train my state-of-the-art neural network, I have generated the following list of three titles. Since the network has only very limited information about the content of your book, it can’t choose which of these is more likely to be appropriate. All are a good match to the sorts of titles that make the bestseller lists. My neural net doesn’t come up with subtitles so you’ll have to do that yourself.
1. The Liars
2. The Numbers Game
3. Trust Me On This
“Lying With Confidence” – I like this one.+
How about “Dirty Numbers”. Or “Filthy, Dirty Numbers”
I suggest:
Hoocoodanode
Misbehavior of Data Analysis
Fooled by statistical cheating and unethical behavior
Not an Ethic Police of Statistics, but…
When statistics didn’t meet Ethics
The Dark Side of Statistics
I’m a fan of the titular colon and, as such, recommend using a pithy title combined with one of your longer titles from above. For instance:
“Standard Deviations: How to Lie, Cheat, and Steal With Statistics”
Crimes Against Data
“Doing a Number on Numbers/Data/Statistics”, pick one. I also like “Lying with Confidence.”
10 things you don’t want to know about statisticians
There is already “A statistical guide for the Ethically Perplexed.” I’d suggest avoiding the word “ethics”. It brings up images of either nuns and priests going on about Aristotle and Moral Philosphy, or folks like PETA, pushing their brand.
Statethics
I think your goal should be simply to move as many copies of this book as possible. I give you the following: Sex, Bieber, Palin, Obama, Sex: A Statistical Guidepost
Cheap Shot
The ethics of working with numbers
or
Ethical number crunching
Getting it wrong
When the numbers just don’t add up
Come on the answer is obvious:
Zombie Statistics
+1
The best suggestion.
Lying by Numbers. Cheating by Numbers. Bad Data [definitely needs a subtitle]. Confidence Tricks.
These all assume that a lot of the book is about particular examples of unethical statistics. That would probably make for an interesting book but might also get you sued into oblivion, so maybe they’re titles for the wrong book. (I haven’t read your columns in *Chance*.)
I think “Confidence Tricks” is at approximately the outer limit of how much statistics you should put in the title. (That is: I largely agree with Jan.)
Numbers Don’t Lie, But You Just Might
Hiding in Plain Sight: Scientific Malfeasance and Statistics
…Not As I Do: Statistics, Science, and Ethics
Codebook: An Ethical Guide To Statistical Practice
Not sure you could do it, but I liked the play on Mostly Harmless Econometrics, and to take it one step further — Mostly Harmless Statistics: An Empiricist’s Ethical Companion
Some of my submissions:
-Science and Conscience
-My Data is Bigger than Your Data
-The Conduct of Science
-The Heuristics of Statistics
-The Ethics of Metrics
-Statistical Ethics
-Ethical Statistics
-I Caught a Number Thiiiiis Big
Justin
“Lying With Numbers”
Statistics, Damned Statistics, and Lies.
Borrowed from book titles by my friend Paul Wadell (a moral theologian: http://www.amazon.com/Paul-J.-Wadell/e/B001IR1D1O )
Happiness and the Statistical Moral Life
The Primacy of Numbers: Ethical Issues in Statistics
Morality: A Course in Statistical Living
Statistics and the Moral Life
The Moral of the Story: Reflections on Misbehaving Statisticians
All I need to know about using statistics I learned in kindergarten.
Right & wrong vs. p<0.5
Statisticians: Hey, at least we aren't politicians!
Statistically Correct
To sell at airport bookstores, pick a short and semi-mysterious title, plus a long subtitle that sounds like it could help the reader make money. E.g.:
Moneyball: The Art of Winning an Unfair Game
Freakonomics: A Rogue Economist Explores the Hidden Side of Everything
Blink: The Power of Thinking Without Thinking
Or maybe that would be unethical?
“Standard Deviations: How to Lie, Cheat, and Steal With Statistics”
That’s not bad, but the subtitle casts the reader as wanting to lie, cheat, and steal. Instead, use the subtitle to portray your reader as becoming able to avoid being victimized by unethical statistical ploys by reading your book.
I don’t think witty titles actually work. I think making up a stupid word like “Freakonomics” or “Moneyball” works best, as long as it seems to imply the book will make the reader smart and rich.
“Smart*Rich” might work. (Of course, if it works, it could work for any content whatsoever).
None of my ideas seem terribly ethical, however.
Unethical Data Scientist: the sexyest (and dirtiest) job of the 21st century.
XXX Data Scientist: sexy and dirty in the 21st century.
Causing to misunderstand: Notes on the management of effective misrepresentations of uncertainties.
How Liars Figure.
What about a combination of your first two hunches? “Ethics, Damned Ethics, and Statistics”
I’m a believer.
With pointers from Steve
Causing to understand: Notes on the deconstruction of harmful misrepresentations of uncertainties in data science.
Data Science Fiction
Using Numbers Ethically
Similar to one of yours, “Statistical Science, Statistical Fraud”. Then a subtitle (as all seem to have now): “The manipulation of data for nefarious means”
What they forgot to teach you in your stat class
Acting on Uncertainty
Risk and Rights
Numerate knaves, innumerate fools
Hmm…”Let The Numbers Do The Lying” or, just getting silly now, “FudgeNumbers”.
No apologies: Lying and cheating like a scientist.
May not please the marketing guys but captures something special about moral behavior in science.
“The Wegman Report” but put a picture of a couple weimaraners on the cover for plausible deniability.
This is an adaptation of a chinese proverb
“Fake numbers can give you flowers, but not fruits”
How about Statistical Unreasoning? Misreasoning? Maybe Misreasoning (with data and analysis)?
“Beyond Standard Deviation”
Buysian and Fraudquentist Statistics
“Beware the Statistician,” or something else following the famous misquote of St. Augustine about mathematicians: http://www.math.osu.edu/~easwaran.1/augustine.html
Even the title was somebody else’s work: a guide to plagiarism and other statistical sins
“Everything You Always Wanted to Know About Statistics (But Were Afraid to Ask)”
If you could work a Repo Man quote into it somehow, that would also be ideal.
P-ing on the data: Manipulation and ethics in statistical science.
That made me chuckle :).
“Odds As Ends: How to Detect When Data is Used to Deceive” (or maybe without the ‘how to detect’)
“Sure, The Numbers Add Up… (to Something)”
+1 to “with low power comes great responsibility” but it might be too geeky, depending on your target audience
“When the Means Don’t Justify the Conclusions”
The Long Spoon
No clever titles from me — just a note that I TOTALLY wish I could take this class.
Maybe someone already proposed this (if so don’t call the plagiarism police on me!) but Statistics as Unprincipled Argument (cf.–as sure you know–Aronson’s classic Statistics as Principled Argument) would be pretty cool
That’s Abelson.
http://www.amazon.com/Statistics-Principled-Argument-Robert-Abelson/dp/0805805281/
Doh!
Andrew, maybe you should choose your top 20 from all of these suggestions and take a poll.
Here Be Statisticians
I like “Standard Deviations: “
“Did Somebody Say Significance? Adventures in the Misuse of a Concept” (A paraphrase of Zizek’s “Did Somebody Say Totalitarianism?”)
“slodderwetenschap,”
“We’re also told that the Stapel mess spawned a Dutch neologism: “slodderwetenschap,” which means something like “sloppy science” and refers to ineffective peer review and a culture of science that allows fraud to go undetected for so long.
From Retraction Watch
I know what you did with statistics!
Factual Lying: Ethical Dilemmas in Statistics
The evil that stats do
‘Error Terms – Ethics in the ‘era’ of Big Data
My contribution:
“How not to lie with numbers”