Most of us at Atomic probably qualified as nerds in high school. If you agree with Paul Graham’s observations in his essay How to be Silicon Valley, this is no bad thing.
Today I happened to observe something strange about our nest of nerds: our last names fall disproportionately into the first half of the alphabet. In fact, 77% of our last names start with a letter in the range [A-M].
A quick search didn’t turn up what the alphabetical distribution of last names in the US actually is, but I guess it must be a lot closer to evenly distributed than our data indicates. We don’t keep a paper phone book anymore, so I couldn’t measure that way. The only data I had readily at hand was our customer database. In this much larger set of names, 64% fall in the first half of the alphabet.
Maybe nerds are more near-sighted, have to sit closer to the front of the room, and actually learn more in school?
7 Comments
I like to make a distinction between geeks and nerds (which is very geeky of me). Nerds are socially inept. Geeks are so into something (usually academic) that other people can’t relate.
Just sayin’.
It might make you feel better about the N word, Dave, if you read Paul Graham’s essay.
But I agree with you. Geek definitely has a better connotation.
A cursory search makes me think PG doesn’t share this view.
I prefer the term nerd over geek because originally a geek was someone who ate live animals. I hate dork because it is male privates.
For what its worth, this site (http://names.mongabay.com/most_common_surnames.htm) lists the 1000 most common names in the United States. Taking this list and alphabetizing it reveals that the median name is Kim/King and A-M accounts for 65.9% of the list.
I think that it is Scott who has fully earned his nerd/geek stripes in this match. ;)
Keep in mind that the second half of the alphabet has some infrequent letters, such as Q, X and Z, whereas the front half doesn’t really have any that standout.
Using the file http://www.census.gov/genealogy/names/dist.all.last and putting it through a Python script yields the following for names in the list comprising at least 0.001% of the population.
Percent including this letter
A 3.353436
B 12.561880
C 20.459857
D 24.839804
E 26.699334
F 30.126900
G 35.437869
H 43.200151
I 43.518030
J 47.129036
K 49.782636
L 54.432718
M 64.177661
N 65.875110
O 67.239603
P 72.140972
Q 72.330695
R 78.150521
S 87.434351
T 91.007664
U 91.171001
V 92.423671
W 99.151904
X 99.164468
Y 99.734891
Z 100.000000
These are pretty close to the above-cited results.