Alphabetical Discrimination

Status
Not open for further replies.

Mel-o-rama

Sci-fi and Humor
Super Member
Registered
Joined
Apr 9, 2011
Messages
69
Reaction score
5
Website
meloramawrites.blogspot.com
Here's something on the lighter side. I wrote this blog post about Alphabetical Discrimination a while back. I halfway meant it as a joke, but then it's kind of festered inside of me, and I wonder if there's any truth in it.

The question is, do sci-fi authors having last names starting near the beginning of the alphabet have a better chance of selling books than do authors with names near the end of the alphabet?

To make matters worse, I had lunch with a publisher, and out of the blue, he told of a story where an author used an "M" pen name so as to position his book in the middle of the stack. The publisher turned to me and laughed. "You're a W, so you may want to think about it."

Is there any truth to this?

Then there's the NPR top 100 SFF Book List recently put out (a total popularity contest). I did a quick tally of all the representative last names (putting Ursula K La Guin under "G").

A 7
B 12
C 6
D 2
E 2
F 2
G 11
H 8
J 1
K 3
L 1
M 10
N 4
O 2
P 4
R 2
S 13
T 2
V 5
W 5
Z 2

This is slightly skewed toward the beginning of the alphabet, with S being the biggest outlier toward the end of the alphabet.

A-D make up 26% of the total.
A-H make up 48%.
J-N (middle) make up 18%.
S is the highest by itself at 13%.
T-Z make up 13%.

So, have a good laugh. And if I'm screwed, please tell me, and I'll just use the pen name "Melvyn Aardvark."
 

DeleyanLee

Writing Anarchist
Kind Benefactor
Super Member
Registered
Joined
Sep 6, 2007
Messages
31,667
Reaction score
11,425
Location
lost among the words
I don't know if it's true (LeGuin should be under "L", BTW), but that tally's making me laugh 'cause there's more for my real last name (end of the alphabet) than for my pen surname (L). LOL!

But I can see that, from a shopping POV. You get in the book aisle and most people will start from the beginning of the alphabet. Occasionally you'll get a rebel who starts from the end. And after a while, you just stop looking--which puts you about the middle of the alphabet. And "S" is simply too large a section to completely ignore if you're just walking by.

Completely unscientific, but I can see why it happens. With on-line shopping, though, I don't know since I don't "browse" for books on-line (yet).
 

Alessandra Kelley

Sophipygian
Super Moderator
Moderator
Super Member
Registered
Joined
Mar 27, 2011
Messages
17,769
Reaction score
7,752
Location
Near the gargoyles
Website
www.alessandrakelley.com
I wonder how the plotting of last names compares to the actual percentages of those initial letters in the general population. To be honest, I'm not seeing a lot of statistical persuasion here.

As for placement, well, my husband's last name is Garfinkle, which usually puts his books right next to Neil Gaiman's, which I suppose can't hurt. :)
 

Shadow_Ferret

Court Jester
Super Member
Registered
Joined
Apr 26, 2005
Messages
23,708
Reaction score
10,661
Location
In a world of my own making
Website
shadowferret.wordpress.com
I thought about this sort of thing when I was thinking of a pen name. Many businesses use this. That's why there are so many A businesses. Acme. Ace. And there are a ton of businesses called A&A This or AAA That.

Why then did I settle for a pen name that started with W? o_O
 

thothguard51

A Gentleman of a refined age...
Super Member
Registered
Joined
Oct 16, 2009
Messages
9,316
Reaction score
1,065
Age
74
Location
Out side the beltway...
The S stat does not surprise me as S is the most popular letter used in the Alphabet as well...
 

Allen R. Brady

Super Member
Registered
Joined
Aug 7, 2011
Messages
189
Reaction score
22
I wonder how the plotting of last names compares to the actual percentages of those initial letters in the general population. To be honest, I'm not seeing a lot of statistical persuasion here.

Yep. This data is meaningless without knowing the distribution of names by letter in general. Offhand, I would expect to see a lot of names starting with A, B, J, M and S. Four of those are in the front half of the alphabet. The latter half is also saddled with letters like Q, X and Z, which don't correspond to many names.
 

RichardFlea

Foo!
Super Member
Registered
Joined
Sep 20, 2011
Messages
509
Reaction score
78
Location
At home - Adelaide (if you know where that is!)
Agree with the distribution of letters based on last names. Still, S is a bit of a curiosity or is 'Smith' the most popular western surname, but for the life of me I cannot think of any Smith authors.

I wonder what would happen if China started using the western alphabet. The 'W' section would explode!
 

bettielee

I'm a sparkly fairy princess!
Super Member
Registered
Joined
Apr 7, 2009
Messages
24,466
Reaction score
12,761
Location
Enchanted Forest and/or editing cave
Website
bettielee.wordpress.com
I'll be next to Tolkien. :) Mighty fine comp'ny

Besides - it's not like they're all on one shelf. At my Borders (RIP) the C's were at the top of the 2nd shelf in the SF & F section.

I know Robert Jordan said he chose a pen name with a J for that reason.
 

Ian Isaro

New Member
Super Member
Registered
Joined
Feb 13, 2011
Messages
433
Reaction score
30
Location
Tanzania
No "I"s at all?

What actually surprises me is the "B"s. I didn't think of it as that popular of a letter.
 

lauralam

Moonshade
Super Member
Registered
Joined
Jul 22, 2011
Messages
896
Reaction score
84
Location
Alba
There are lies, damned lies, and statistics :)

Fortunately my (real) surname puts me right next to Scott Lynch - w00t!

I'd be sort of near you two, and right by Mercedes Lackey and her 348749 books.
 

Mel-o-rama

Sci-fi and Humor
Super Member
Registered
Joined
Apr 9, 2011
Messages
69
Reaction score
5
Website
meloramawrites.blogspot.com
I found these stats for US last names (2000 census):

A 3.42%
B 8.09%
C 7.52%
D 3.6%
E 1.41%
F 3.24%
G 5.71%
H 8.06%
I 0.14%
J 5.02%
K 2.13%
L 4.11%
M 10.48%
N 1.65%
O 1.39%
P 4.83%
Q 0.07%
R 6.9%
S 8.63%
T 3.41%
U 0.05%
V 1.14%
W 8.14%
X 0%
Y 0.67%
Z 0.21%

Read more: http://wiki.answers.com/Q/What_is_t...letters_in_last_names_in_the_US#ixzz1YmfGT8QV

The SFF list does have slightly more ABC combined, less W, and interestingly more S.

What would be more useful would be to see a compilation of data on books actually sold. Will search...
 

Deirdre

practical experience, FTW
Super Member
Registered
Joined
Aug 17, 2007
Messages
889
Reaction score
159
Location
Menlo Park, CA
Website
deirdre.net
My actual last name is Saoirse-nonbreakingspace-Moen, but I decided to go with D. S. Moen for my writing name based on a comment by Patrick Nielsen Hayden about my name being difficult to spell.

That puts me just to the right of L. E. Modesitt, which is okay by me.
 

Ardent Kat

Kill your television
Super Member
Registered
Joined
Aug 27, 2009
Messages
793
Reaction score
152
Location
Austin, TX
Website
www.katherineokelly.com
Ah, crap. I hope my name is next to someone famous. I'm stranded in the boonies with my last name. (At least the apostrophe at the beginning of "O'Kelly" means I'll probably be at the start of the 'O's)
 

Rachel Udin

Banned
Joined
Nov 19, 2010
Messages
1,514
Reaction score
133
Location
USA... sometimes.
Website
www.racheludin.com
I think I'd be the Only U on the entire shelf. But then, I've always been labeled an oddball.

(One of my ideas as a kid was to marry someone with a cool last name for the bookshelf positioning. Shallow, yes... but that definitely tells you something.)
 
Last edited:

jjdebenedictis

is watching you via her avatar
Super Member
Registered
Joined
Jun 25, 2010
Messages
7,063
Reaction score
1,643
Okay, so as a geek, I am honour-bound to protest that we need a larger sample set for the statistics to mean any dog-darned thing at all.

That said, if you normalize the NPR list's numbers against the US census numbers for last names, you get the following:

A 2.05
B 1.48
C 0.80
D 0.56
E 1.42
F 0.62
G 1.93
H 0.99
I 0.00
J 0.20
K 1.41
L 0.24
M 0.95
N 2.42
O 1.44
P 0.83
Q 0.00
R 0.29
S 1.51
T 0.59
U 0.00
V 4.39
W 0.61
X 0.00
Y 0.00
Z 9.52

The uncertainty is probably about +/- 0.75, assuming I'm doing my statistical analysis right--which is not a safe assumption. :) (I'm using [2*sigma/sqrt(N)]*sqrt(N-n/N-1), for anyone who knows stats. Feel free to correct my math!)

If you put these values in a histogram, Z and V are really monstrous, but I doubt that means anything real. It's more likely the fact that the NPR list happens to have a few of those letters, whereas they're rare in the general population.

Given most of the authors are American, British or Canadian, I doubt the fact we're using US census data skews things by much.

Looking at the histogram, the main thing I see is that--with the exception of V and Z--things look really flat (within uncertainty.)

S00per-scientific conclusion: I see no evidence of alphabetical discrimination.

Right. So now let's do a study to see if my really long-ass surname is going to hurt my popularity. :tongue
 
Last edited:

jjdebenedictis

is watching you via her avatar
Super Member
Registered
Joined
Jun 25, 2010
Messages
7,063
Reaction score
1,643
GEEK-UPDATE!

~~~~~~~
In fact, I just put the data into a graphing program, excluded V and Z, as well as the letters that weren't represented on the NPR list (i.e. the letters that had zero names), and the slope was:
-0.02 +/- 0.07

In other words, the graph is flat within uncertainty! There's no alphabetical discrimination.

~~~~~~~
With the zero-letters included, but V and Z still excluded, the slope is:
-0.04 +/- 0.06

~~~~~~~
With everything, including V and Z, the slope is:
+0.06 +/- 0.06

~~~~~~~
With V and Z included, but the zero-letters excluded, the slope is:
+0.12 +/- 0.06

Note: This implies alphabetical discrimination with the END of the alphabet being more favourable. However, as I said, I suspect V and Z's results are an artifact of the small sample size.

~~~~~~~

Translation for those uncomfortable with this stuff: If the second number (after the +/- symbols) is bigger than the first number, that means the slope of the graph is zero within uncertainty, which mean there is NO alphabetical discrimination.

A positive slope means there IS alphabetical discrimination, with the front of the alphabet being favoured. A negative slope means there IS alphabetical discrimination, with the end of the alphabet being favoured. However, you always want to take into account the size of the uncertainty (the second number) also.
 

MJNL

A Little Lost
Super Member
Registered
Joined
Sep 18, 2010
Messages
1,033
Reaction score
118
Website
lostetter.wordpress.com
I'm fairly sure we writers are the number one group most obsessed with obscure and useless statistics. At least it's fun :) .
 

Rachel Udin

Banned
Joined
Nov 19, 2010
Messages
1,514
Reaction score
133
Location
USA... sometimes.
Website
www.racheludin.com
I'm fairly sure we writers are the number one group most obsessed with obscure and useless statistics. At least it's fun :) .
Hear Hear! Like buttons are on the opposite side for women because women used to have dressers. (Maids who dressed them). Since it was in vogue during the industrialization period still, the buttons stayed on the same side.

My grandmother also gave me a manners book. So I have other useless facts. (Fantasy and Sci-fi tends to get me to pick up the most useless ones to today's day to day world.)

Still think I'm screwed with a U. Maybe I should pay 100 dollars US to change my last name.
 
Status
Not open for further replies.