The Business Rusch: Lies, Damned Lies, and Statistics
Kristine Kathryn Rusch
The quote in my title comes from Mark Twain’s autobiography. Twain said:
“Figures often beguile me, particularly when I have the arranging of them myself; in which case the remark attributed to Disraeli would often apply with justice and force: “There are three kinds of lies: lies, damned lies and statistics.”
The problem with Twain’s attribution, however, is that no scholar can find anything in Disraeli’s papers that even resembles it. (Yes, scholars have that kind of time on their hands.) The website twainquotes.com cites an 1895 article by Leonard H. Courtney in which the quote first appeared—or so everyone thinks.
I find it hilarious that the source of this quote about statistics is almost impossible to track down. I also find it funny that Twain’s preface to the quote has gotten lost in the pithiness of the “lies, damned lies, and statistics.”
“Figures often beguile me,” he wrote, “particularly when I have the arranging of them myself.”
And thus, Mark Twain, who died in 1910, has poked at the heart of modern publishing. We all love statistics – or figures, as he calls them – but they prove nothing. In fact, this year, statistical analysis is harder than ever.
You’d think it would be easier. We have computers, after all. We have incredible processing speeds and more information at our fingertips than ever before. We can “crunch” the numbers quickly and easily.
The problem is in which numbers we crunch.
Let’s take, for example, the number of e-book sales versus the number of print book sales. We’re seeing a lot of statistics about the percentage of e-books in the marketplace. And those statistics come from reputable organizations.
I felt uncomfortable about those statistics at the end of 2011, and I feel even more uncomfortable about them now. These statistics purport to examine all books sold, and I know that’s not true. I also know that there are equations that supposedly take a statistical sample, and apply them over information not yet gathered (or information that’s impossible to gather). And even though I know the mathematical model is accepted, I’m still uncomfortable.
You see the mathematical model in polling all the time. Pollsters contact 1,000 or 10,000 or 100,000 sufficiently diverse people, poll them, and then use them as a statistical sample that supposedly represents the entire population. This same technique takes place in medical studies. Studies gather information from 50 to 500 to 5,000 people, gauge their reactions to, say, a medication over a period of time, and then use those as a basis for the result.
People who watch medical studies, for example, generally ignore the ones with less than 100 participants, and really believe the ones with tens of thousands of participants. And if those tens of thousands were studied over years, then the medical study is considered even more accurate than the one that follows someone’s reaction to a treatment or a medication over a few hours.
See why Mark Twain insisted that he liked figures if he arranged them himself? Or to put it in 2012 language: he liked statistics if he manipulated the information himself.
One of the first things I learned as a journalist, back in high school of all places, was how to look for statistical manipulation. “Four out of five dentists surveyed” might mean that five dentists were surveyed, and four of them (the ones who worked for the company) liked the product. Or it might mean that four out of five dentists in a survey that contacted 10,000 dentists (none of whom worked for the company) liked the product.
Both statements would be true. Four out of five dentists liked the product. But only one statement might be information that a consumer might benefit from.
As the past year has continued, it has become clear to me that e-book sales are rising. Anyone who watches numbers knows that. Every day there’s a new tablet hitting the market, or some new version of an e-reader. Just this week, Apple unveiled iPad 3. At the same time that Apple announced the New HD iPad (which is what they’re calling it), Google announced Google Play which it claims will rival iTunes. We’ll see.
I spent some time as I wrote this trying to find exact numbers of tablets and e-readers sold, and I can’t. Part of that is Amazon’s unwillingness to impart information on its sales in this area, and part of it is the rapidity of growth.
At the moment, there aren’t even enough statistics out there to manipulate. At the moment, even the folks in the know admit that they’re guessing.
The point here is that the ways to buy and read electronic books are multiplying daily. The low prices over the holiday season made it possible to give e-readers and tablets to people who wouldn’t buy the devices on their own. Most of those folks are using those devices.
We’re already seeing an impact on pricing. I will deal with that in a future post. But we’re also seeing the impact on e-book sales. Every indie publisher I’ve talked with has seen a serious spike in sales. Traditional publishers also discuss the increased sales.
But keep this in mind: As readers order a new e-book, they don’t notice who published it. In fact, over the past few months, Amazon has changed their algorithm, so you actively have to look for a publisher on many titles—including traditionally published books.
It’s a smart move on Amazon’s part, because it ensures that indie writers will sell based on the book itself rather than who published the book. But study after study after study for decades have shown that readers rarely pay attention to the publisher of a book. Readers pay attention to the author name instead.
Some of that is branding. Readers will buy an entire book line, like a Harlequin category novel, because the books provide a consistent and predictable reading experience. Not that the books are the same, mind you, but they’re of a type. Just like an author’s books are of a type. The “voice” of the book line is consistent. Daw books did that in its early years, and branded all of its books with a yellow spine. Daw published everything from fantasy to science fiction during those years, but the books had a consistent look, and they had an editorial quality control rare in traditional publishing.
To pull something like that off is hard. Most traditional publishers can’t do it, which makes them reliant on their authors (whom they’re afraid of losing right now—also a future blog post).
So the point is that if I publish a book myself, the readers don’t care as long as I make sure someone else edits the book, and a different someone else copyedits it. Readers only care about publishers if the publishers get in the way of the reading experience.
In the past, publishers have gotten in the way by no longer publishing an author or canceling an on-going series. Then readers might notice that the publisher has done this to a favorite author. But usually the reader writes to the author asking why she discontinued a series, and the author then points to her publisher. The reader might or might not write to the publisher to complain, but even if the reader does, the publisher doesn’t care. The publisher made a business decision long ago, and really doesn’t plan on revisiting that decision no matter how many readers say they will buy the books.
Keep this fact in mind: Readers don’t notice publishers. Readers only notice writers. Readers will buy a writer’s next book no matter who publishes it.
But most of the statistical measures of book sales track book sales by publisher.
Twenty years ago, when Dean Wesley Smith and I ran Pulphouse Publishing, a small press by New York standards, (although we sold as many sf/f books as many imprints of major publishers), no major statistical analysis of books sold counted our books. Those measurements rarely counted any regional book sales or any university press book sales, even if the university press published trade (commercial or non-educational) books.
When these analysts looked at books in print, they would be able to find our books because our books—and the regional press books and the university and small press books—had an ISBN number. The ISBN (which is short for the International Standard Book Number) gets sold around the world through different companies. (Not every book in print gets one, by the way, and never did. The ISBN has existed since 1966, and came into wide use in the late 1970s.) R.K. Bowker, which sells the ISBNs in the United States, keeps careful track of those numbers and can give books in print statistics to anyone willing to pay for them.
But books in print is a different measure than book sales. (Just because a book exists doesn’t mean it has sold a copy.) And the introduction of e-books has changed the system dramatically. Many (dare I say most?) e-books don’t have an ISBN. Amazon has its own e-book tracking system, and so do some of the other e-bookstores. That ISBN system has broken down as well.
Which means that right now, no one knows how many different titles exist in e-book form. We now have even less information about e-books and e-book sales than we did about paper books back in the day, which we only knew imperfectly anyway.
In the past, if someone really wanted to, that someone could make an educated guess about the number of books for sale in the United States, based on the books in print, the size of book stores, the number of readers, and a bunch of other factors. That guess had the possibility of being reasonably accurate.
That guess couldn’t take into account sales velocity (meaning how fast a book sold once it was on the stands), but it might be able to track certain kinds of sales, given the information available through publishing company sales departments, bookstores, advances given to authors, and the number of books going through web presses, etc. It could be calculated if—like those Disraeli scholars above—you really wanted to lose half your life to tracking down arcane information.
Book sales can’t be tracked now. Because e-books can be produced in someone’s bedroom and uploaded on a home computer (no web press), and sold through a website (no bookstore or sales department).
Even existing bookstores aren’t much help. Amazon doesn’t give out sales figures on anything. Apple keeps a lot of information about the iBookstore close to the chest. Barnes & Noble, which works on the old book publishing method, does release more information, but still not enough. (And with B&N thinking about spinning off its Nook/e-book business, that behavior might change.)
So add this as your second piece of information: No one can accurately track all e-book sales. There isn’t enough information for anyone to piece those sales together from disparate bits of data, like you could once do with print books.
Finally, add this to the equation: traditional publishing got into e-books late. They didn’t convert most of their backlist (and still haven’t), so there is no one-to-one measure of paper books in print to e-books in print. From traditional publishers only.
Many—dare I say most?—of the e-books published right now are through indie presses or individual writers. I can’t say that with complete certainty because, as I said, the statistics don’t exist.
But Amazon keeps track of the indie titles published through its KDP program, and B&N keeps track of the indie titles published through its PubIt program. Those book titles aren’t always the same, either. Some authors publish exclusively through Amazon. Others publish exclusively through B&N. Many authors avoid all the big publishers and go it alone, on their websites or developing their own stores.
So add this third piece of information: traditional publishers have not published their entire backlists as e-books. Meaning that all the books traditional publishers could have published have not been published.
And now, let’s look at last week’s news.
The American Association of Publishers reported that e-book sales rose 117% in 2011. The report stated that net book sales went down 3.5%. What that means is that even though e-book sales have risen, they haven’t risen enough to compensate for the decline in paper book sales. For example, mass market paperback sales were down 35.9% with adult mass markets dropping off 40.9%. E-books rose 72.1% over December of 2010.
Why are mass market sales down? Mostly because of the decline in slots to sell the books. Safeway, Albertsons, and other grocery stores reduced their book sections dramatically. Borders is gone. And traditional publishers have given up on the mass market form, trying to get readers to go to more profitable e-books, trade paperbacks or hardcovers. Even if you want a mass market, good luck finding it in a brick-and-mortar store. If the mass market version even exists.
That’s a sidebar to my real point, however. The AAP is the only organization that tracks book sales with any type of accuracy. When you see a statistic like this one—E-books are now 18.6% of the market—that statistic usually comes from the AAP.
The AAP is an organization with 300 members. Those members come from all aspects of publishing, not just trade publishing (which is the category most books you find in a brick-and-mortar store like B&N fall into). Only 77 publishers reported sales to the AAP for its 2011 report which is up by the way, from its 2010 report, which was based on 12-15 publishers.
Those publishers are commercial trade book publishers active here in the United States. They are self-selected and their numbers are not verified.
Got that? Self-selected and not verified.
Let’s assume, though, that the numbers are accurate. Let’s also assume that these numbers can be extrapolated to include all traditional trade publishers here in the United States.
What’s wrong with these statistics?
- Books sell because of the author. These statistics only show traditional publisher sales, not sales by author. More authors are publishing their own backlist than ever before. Barbara Freethy alone has sold a million copies of her backlist books as e-books. She published those books herself.
- No one can accurately track e-book sales. There are too many places to sell the books to get any accurate count. Also, no one knows how many e-book titles are actually in print, so we can’t even make an educated guess as to the sales. Plus the biggest bookstores selling e-books don’t give out sales information.
- Traditional publishers still have a small percentage of their available books in e-book form. So if you’re reading a backlist e-book title, chances are it was published by someone other than a traditional publisher.
What does this all mean? It means that e-book sales are most likely greater than the statistics show.
Traditional publishing saw the decline in book sales revenue in 2011 as a bad thing. Yet all of the evidence coming from studies done of people who have a new e-reader finds that readers increase their book-buying after they get an e-reader. (Because I searched for this fact, I could only find the 2010 Wall Street Journal article on one study. There have been more since.)
Traditional publishers assume that if their overall book sales went down, all book sales have gone down. But I think that extremely unlikely. Since readers only care about who wrote the book and not who published it, and since readers have a finite amount of dollars, there is a good chance that those readers bought books from their favorite authors somewhere else. The book sales that have disappeared from traditional publishers’ ledgers have actually shown up on the ledgers of indie-published writers. Plus some.
In other words, folks, it’s my personal opinion that the traditional publishing statistics on e-books are completely meaningless. Or to put it another way, the news is much better than these statistics lead us to believe.
At some point, we need to figure out a way to count the e-books being sold through all venues and from every kind of publisher, from traditional to indie. Then, perhaps, we’ll have an accurate picture of what’s really going on in the book business.
Right now, we get only a snapshot of what’s happening among a self-selected group of traditional publishers who give out unverified information. We make all kinds of proclamations based on statistics derived from that faulty information.
Good old Mark Twain. Who knew that he could peer more than a century into the future and see what would happen with publishing? He loved to arrange figures to suit his point. Traditional publishers are doing the same at the moment—intentionally or not.
Last week’s post brought a bunch of strangers to this blog to disagree with me viciously and vehemently. The point was (I think) to get me to change my mind about what I wrote. That didn’t work.
What that tidal wave of negativity did do, however, was make me even more appreciative of those of you who have come to the blog over the past three years. You haven’t always agreed with me, and that’s a good thing, because I learn from you. But you’ve been polite and respectful. You’ve sent me links and helped me find information. You’ve made great comments, and sent wonderful, informative e-mails.
You’ve also donated your hard-earned cash to keep this blog alive.
I want to thank you for all of that. You folks are the reason I write this blog every week, even though it takes time away from my fiction. You’re a spectacular group and it’s a pleasure to interact with you every week.
“The Business Rusch: “Lies, Damned Lies, and Statistics” copyright © 2012 by Kristine Kathryn Rusch.
Ah, I remember that very same quote being in the front of my statistics text book at Uni! I’ve never forgotten it.
As for sales that are not tracked, I know many children’s authors regularly sell books on school visits after buying them from their publisher. I understand these sales do not show up on any computer, so presumably do not “count” in the stats?
That’s right, Katherine. They don’t count in the short-term sales that show up as bestseller lists, velocity, and all the other measures that publishers use to see if an author sells a lot of copies. Plus no one really knows what’s in the warehouse for copies. With old-fashioned print runs, the number was plus or minus 10%. So your print run number is off by 10% even before a book goes on sale. Which direction it’s off–up or down–no one knows.
I am constantly seeing people drag out these statistics on the portion of the market that is eBook as though they were accurate and can’t tell you how much it frustrates me. Most of them at best were inaccurate and out-of-date when they were published. How can you possibly even begin to know the portion of the market without knowing Amazon’s sales?
Obviously, you can’t and the same people who drag out these stats often bemoan that Amazon has the lion’s share of the market. Thanks for pointing this out so cogently.
Thanks, JR. Exactly.
Yes, I thought so, too (that either use of the word fits). Nice when it happens that way.
Reflects better on Twain, though, that he is not boasting about manipulating numbers, but rather admitting that he is not good at interpreting them.
He lost a fortune in a publishing investment, as I recall. He invested in a new technology, an automatic typesetting machine for newspapers. I have always thought it strange that the business failed. I guess he did, too.
Donna, the whole Twain/business thing… From what I read, Twain made typical small business owner mistakes. He was a visionary, but being a visionary doesn’t mean you’re a businessman. He lost everything and had to start over financially rather late in life. I learned this after I lost a small fortune in publishing the first time. It’s never fun rebuilding. Hence–most likely–his quote. 🙂
Excellent post.
Kris,
In my day job, I do market research for mobile device management, which includes all versions of smartphones and tablet devices. The Pew Internet study (http://www.pewinternet.org/Commentary/2012/February/Pew-Internet-Mobile.aspx) might be a trusted source for device adoption.
Another link (http://www.pewinternet.org/Reports/2012/E-readers-and-tablets.aspx) shows the rapid rise of devices in 4Q2011, as folks received tablets and e-readers over the holidays.
I’ve been interviewing business owners and have discovered a fast growing market segment, the small business owner, now self-identifies as a READER, and talks about READING a book while traveling on planes. What kind of book? An e-book, of course. Never before have I had business folks SELF-REPORT reading as a main activity.
Another group, folks between 25-39, are READING e-books on their PHONES.
There’s a big movement afoot. As one research subject said, “People pay more attention to their smartphone than their three-year old. They take it everyone.”
My take-away? Anyone who is an avid reader with a smartphone or tablet takes their library with them. Everywhere. It’s easier to buy an e-book than a physical book. Of course traditional publishers, with their antiquated systems, are underreporting e-book sales. And Amazon is smiling all the way to the bank. As are indie-writers everywhere.
Wonderful links, C.L., and great evidence of a movement afoot. This is all excellent news for those of us who are going indie. And it pretty much confirms the anecdotal evidence I’m hearing as well. Thanks.
I think you have misunderstood the Twain quotation. He is using “beguile” in the sense of “deceive”–I don’t have his autobiography at hand, which I think the quotation from, but I would imagine he was thinking of his financial troubles and disastrous business ventures, and his talent for self-deception when going over the numbers.
Just to check myself, I went to the Twain concordance to see how he usually uses the word, and find it in The Gilded Age, A Connecticut Yankee, and a number of others, used to mean trick or deceive. I did find one use in the sense of “amuse.” But I think the context (“lies, damned lies,” etc) points to “deceive,” rather than saying he liked to arrange figures to suit his point.
It seems to me, Donna, that either use still makes my point.
Two comments on this article–At christmas time our grocery store offered an e-book reader for sale so this would skew the numbers of readers sold.
We have noticed a jump in sales for our after-market titles which would not show up anywhere. I am thinking that the demise of the big stores feeds into this. There is no longer a store which primarily sells new books in our good-sized city. There is Costco which is inconsistent as to what it offers for sale and then you have to go to a neighboring city/town to find a selection of new books. We also only have two good sized used book stores (Half-Price Books and an independent).
Thanks, Jane. For those of you who don’t know, Jane owns a bookstore, which is why she was mentioning after-market titles. We have the same issue in our region. Our small town lost its indie bookstore several years ago. Then the largest city near us lost Borders and all the indie new bookstores. So for the Oregon Coast and the central part of the Willamette Valley, book buyers have to drive one to three hours to get to a bookstore that carries new books exclusively. We have several bookstores in our town and nearby towns, but they’re mostly used or new/used, with the emphasis on the used or on special ordering. Our grocery store sold an e-reader at Christmas as well, as did the local discount stores like Bi-Mart and a bunch of the other office supply stores. So you’re right, Jane, for those of us who live in an area without a good exclusively new bookstore, we have online sales or e-readers only. And e-readers are more convenient.
Great article! I do, however, have one gripe.
Sales revenue cannot be used to gage popularity when you don’t have a consistent price structure. You CAN use gross sales revenue to figure out consumer interest when you are talking about items that cost the same like movie tickets. For example, if one movie has better gross ticket revenue you can assume more people bought tickets because the ticket price was the same (ignoring for a moment IMAX and 3D).
You CANNOT use gross sales revenue in the same way when the pricing of items is dissimilar. If a $5 ebook made $50 and a $10 paperbook made $50, could you say that both books had the same number of readers? Of course not, the ebook has twice times as many readers.
The AAP article from teleread is a great example.
Let’s take the following price structure and look at that teleread article: $7 for an ebook, $30 for a hardcover, $9 for a Paperback. In millions, there was $85 in eBooks sales, $130 in Hardcover and $112 in Paperbacks. After some quick math you would get 12.1 million ebooks sold, 4 million Hardbacks, and 12.4 million. Now, doesn’t that give a different picture?
I will admit that, of course, those numbers are probably wrong. But, wrong in which direction? Both Carina (ebook subsidiary of HQ) and Penguin report figures to the AAP. A Carina romance is usually priced at about $4 and there are many NYT bestsellers at $12.99. So, ebooks sold could be as low as 6 million sold or as high as 21 million, without a price break out we really can’t be sure.
That 20% figure every loves to throw out can be used to talk about the financials of a publishing company, but it is not indicative of reader behavior.
Whew, rant over. 🙂
Christian, excellent point, and not a rant at all. Wonderful.
I can see an upside to this: if no one can track e-book sales, how can Paypal, the government, or anyone else know what you’re selling? It’s harder to impose rules on something so chaotic.
OTOH, it’s probably going to make it even harder for us authors to track our book sales. How do I audit Amazon? (pause for hysteria)
As for iTunes deleting your whole music catalog, you do realize Amazon can do the same thing to stuff you’ve bought on your Kindle? Granted, in this case they were protecting copyright. But the problem is not the intent, but the capability. What happens if IPG and Amazon can’t come to an agreement over discounting? There are a lot of issues that will affect ebook sales, but how will we know what those impacts are unless we can accurately track sales? As the King says in “The King and I” — is a puzzlement.
Thanks for keeping this issue in front of us, Kris.
Actually, Sarah, you can audit Amazon. You’re a supplier. You have to think retail now. You have very interesting legal rights as a supplier that you didn’t have as a partner with a publisher. So if you feel Amazon isn’t reporting fairly, you work in retail channels now. Still not fun, but it can be done. The Passive Guy has some interesting insights into this on his blog today.
And Sarah, you’re right about Amazon. Which is why I still buy the books I care about in paper. I may read them in e-book form though… 🙂
Writers obsess far too much about ebook percentage figures anyway. At best, they tell you a LITTLE about the potential of the ebook market for your works. But not much.
Lets say ebooks are 20% of books sold.
Are all of the people buying those 20% the sort to be interested in your book? Almost certainly not. They’re a pool of readers with all sorts of interests.
Do the people buying that 20% reflect the total market for your book? No, because by the time your book comes to market (if it that’s tomorrow), more people will join the pool of ebook readers, a few will perhaps return to print, and if you’re lucky, some of your fans may start reading ebooks just to read YOUR books. So the number isn’t at all fixed.
And is that 20% fixed across all sorts of genres and books? No. There’s evidence, for example, that novel readers are migrating faster. Heavy readers seem more drawn to ebooks, and so genres with audiences that read heavily may have better numbers. Romance, for example, migrated to ebooks early, and I suspect that’s a factor. Other types of books, reference, or technical books with lots of charts and illustrations or complex layout, are slower to migrate just for technical reasons. Its still easier to use such a book in print form.
And of course, what you’re really concerned about is YOUR readership. If the 20% number reflects your own numbers, it probably has more to do with coincidence than science. Yours could be higher. Yours could be (though probably won’t be, as long as you have product to sell) zero percent.
So the number isn’t much of a number. It doesn’t tell you much about the books you write unless you write ALL kinds of books (and in the same distribution as general publishing), and it doesn’t tell you a thing about your own sales.
Looking to such industry numbers to chart your own sales is like looking at the weather report to figure out if you’re cold. Really, it may be more important that your furnace is in good repair and you own a wardrobe of sweaters.
So, the number isn’t really much of a number, and it doesn’t apply
Thanks, Steve. Good point.
I have to say that you are correct about ebook buying from my personal experience. After I got my kindle, my reading tripled that year. Being able to buy the next book in a series immediately after finishing the previous book was probably the single greatest factor. Buying friction was decreased because I no longer had to make a trip to the bookstore. Not only did my reading triple but, with Amazon’s “also bought…” suggestions my selection of authors expanded–especially for indies.
Josh, that’s my experience too. Amazing what having books at your fingertips does to your book-buying habits.
No one is counting my eBooks sales. Over 400,000 in 2011.
I’m tired of seeing “juking the stats” from my favorite TV series The Wire.
All those with a vested interest in print will continue to push their numbers until they hit the unemployment line.
Bob, between you and Barbara Freethy alone, that’s 1.4 million e-book sales not counted. I’m sure we could go on from there. Love the sentence about the unemployment line. 🙂
Along the same lines as yours and Mr. Twain’s opinion on statistics, I believe the AAP article skews numbers even more by refering to what looks like gross receipts. Wouldn’t looking at units sold be a more accurate reflector of the change in e-book/print book sales?
This is why, whenever someone quotes statistics to back their points, I have to gulp down the urge to cough—and why I avoid using statistics as backup, myself.
See, one of my classes in high school made us each write a survey to ask a bunch of random people. We had to create two versions: one biased, the other biased. The unbiased survey was a nightmare to write, mentally taxing on a level that I doubt most care to.
I’ve never trusted statistics since.
So when folks go off about “Oh, look! E-books have umpteen percent of the market—” blah blah blah blah…
Uh-huh. You go on spouting that and freaking out. I’ll be over here working and selling e-books.
Carradee, I love that. “You go on spouting and freaking out. I’ll be over here working and selling e-books.” Exactly.
It’s a real pleasure to read your well reasoned, clearly presented posts. Thank you.
You’re welcome, Mike. 🙂
Not to be conspiracy theorist or anything, but…
The MPAA and RIAA have been very strong in pushing for rather ruthless copyright law extension – extensions that work primarily in their favor as opposed to actually helping the artists involved.
Could the push to arrange the statistics to show a drop in readership be a prelude to push hard on book piracy, especially e-book piracy? I know many of the independent writers are releasing their books without DRM, but the Big Publishers have been down right nasty about it.
Tara, apples and monkeys in that analogy, sorry to say. This isn’t really organized, but instead, companies trying to find a way to measure the unmeasurable. And my theory on DRM is this: As readers find books easily, they’ll fight anyone who makes getting a book hard. Fight with their feet that is, and buy from the non-DRM publisher/writer. It’s what happened in the music industry with CDs and MP3s (and is still going on with iTunes. The first time iTunes wipes out your music library because you changed computers, that’s the day you become militant).
I used to be a statistician, working as a marketing research professional.
The real skill of the job entailed finding out what the client wanted the outcome of the survey to be. Most of them just wanted confirmation they could take to the board to back up their decision to take a certain product forward, or kill another, or do one advert over another.
You tell me the answer and I’ll construct the survey that will give you that answer.
Watch this, it explains it much better, and wittier, than I can:
http://www.youtube.com/watch?v=G0ZZJXw4MTA
Thank you, Lynne. A lot of people don’t realize that most of advertising (and a lot of corporation-sponsored “studies”) are done precisely the way you describe. I had a friend quit a science-based research job because her boss signed on with a megacorp and was told which bits of data to use and which to ignore in his study. The boss cooperated. My pure scientist friend did not.
Nielsen have been trying for a while to generate an ebook sales chart for the UK, but Amazon will not divulge their sales figures. Nielsen say there is no point in an ebook chart without Amazon because it dominates the market so much.
It was a fascinating discussion when I heard it on the radio a couple of weeks ago. More here http://www.bbc.co.uk/programmes/b01c7rqj#synopsis
Not sure if BBC iplayer works outside the UK (sorry if it doesn’t).
Thanks, Jane. The link does work here. And it’s interesting, y’all. Worth listening to.
Suggestion:
what about some stats from writer-publishers? Say B. Freethy, J. Barnes, yourself… Since it looks like the APA doesn’t get those numbers, and publishers querying for them would be suspect, I’d say SFWA, RWA…
Take care
Ferran, it would be nice if someone tried to organize writer-publishers, but I’ve herded enough cats in my life to avoid that job. If someone else wants it, that’s another matter. 🙂
Aaah, statistics. I’m a statistician. Well, I’m a mathematician with a Master in Applied Probability and Statistics. And am I glad to read such pieces of common sense !
I know a hundred ways to botch statistical studies, most of them actually implemented, with or without intent. From not defining the object of your study (“makes your skin look 35 % healthier” : how do you measure skin wholesomeness with a figure, anyways ? That’s qualitative, not quantitative !), to applying corrective coefficients (I’m French, this might not be idiomatic) to pool results because extreme right-wing voters tend to lie about it (which makes your figures kind of a guess, the guessing part being the coefficient), to eliminating “non-significant” data from the sample, to… Nah. Let’s stop there.
And as to statistics themselves, you can add murders and assaults (on the rise) to Internet scams and cars theft (more numerous, and diminishing) and claim that “criminality is down this month” – only it’s gotten worse. Or neglect crimes that go unreported. And so on.
I don’t mean to get political. In advertisement, administration, sales reports, wherever statistics are used, they are used for a purpose, and almost never without bias and for information’s sake.
Statistics are the science of turning a staggering amount of data into understandable figures ; making sense out of something too big for the mind to apprehend. That is the intent of the mathematical models, and in their way, they are very adequate – assuming two things : that the models apply (every mathematical model has requirements, which are often neglected), and that the person who uses them, uses them without bias. This latter is even more iffy.
Don’t find fault with statistics. Find fault with implementation. Question implementation.
So thank you, Mrs Rusch, for reminding your readers how much statistics need scrutiny and skepticism. Thank you very much.
Thanks, Rémi Billoir. Great analysis. It’s nice to have a mathematician back me up here. 🙂