Coupling Literature and Statistics


TITLE: 

AUTHORS: 

REPORT NO.: 

PUBLISHED 
IN: 

Coupling Literature and Statistics 

Robert Boruch and Judith Zawojewski 

A-415 

Teaching Statistics. 1988, 10(2), 37-42 
and 

D. Green (Ed.) (1994) Teaching Statistics at its Best. In press. 



NTL. LOUIS. UNIV TEL:1-708-853-0214 May 12'94 15:02 No.011 P.06 .. · 
.... · ~ ..... 

Coupling Literature and Statistics 

Robert Boruch and Judith azwojewski 
lllillois. USA 

INTRODU:COON 
Did you know that Alennder Damas wrote a short 

'StOiy eDtided Skllitdics!l ·Or·that .Jl.ohiMsn Cr:usoe. 

explore<l in the 1890's by both Itallan and American 
statisticians. 

Gravity's Rainbow. and innumerable other stories ILLUSTIL\TIONS IN LITERATURE 
often exploit simple ideas about probability and Statistical reasoning. coaect aud otberwiae. is not 
swlsdcs? Are you ~ware that lmmorlats, such as difficult to find in ~-20th century fiction. Por 
Mark Twain. fiction writers such as Len Deighton, instance, tbe gambler's fallacy is fouod in Edgar 
novelists such as Thomas Pynchon. and Allan Poe's TM Gold Bug, in fiction by Dorothy 
autobiograpbets such as Ben Franklin, present ideas Sayers, iD Artbur Conan-Doyle's SMrlock Holm.u 
(rigblly andothclwise) based on the gamblerS' fallacy stodes, amlin Kahn's The Code Breakers. From 
aud simple dlstrlbutloos? WDUam Kru.skal' s tboughtful1reafment of IIWDbers In 

Most people are not aware that a great deal. or . prose. we find Charles Dickens loathing CSreadful 
probabWty 8Dd SWistics can be found in prose. poetry statisti.~ and Anthony Tmllope loving tbem.· Rubin's 
and drama. Srneewn and Smeeron (1984, 198~~ remarkable series~ of papers for tbe American 
mustraled ways in wbich statistical ideas can be StiJii#ido1l ·leadB us through tbe stmlstlcal worlds of 
Introduced In many facets of English Studies, Herodotus, Thucydides, Darwilr and· Wtlliam · 
including· U~ 1beir studies have lndicated the Shakespeare. Some of the same spirit is r:dlectecs in 
reastbWty of bleudlni statistics and Euglish Studies. more a:udemporary tteatments of scientlfl~evideDCC 
It seems timely, tben,. to &hare .a collection· of' by WltltaUJ.. Ooldlng, author or Lord of lhe.FIJU, in 
Illustrations In literature and historY that have bc:cD N• reviewing tbe works ot 'HerodotU and H. G. WeDs. 
compiled by one of tbe authors ov:er a nUlllber of Other ~tw~cal Ideas can be. identified In 
years. They are based on reading; tliat are common coni:emporary'·work. Michael Cricilton· explores· 

·. In precollep 8Dd litJc:ral art9 college am:ialla. ~·. analysis. fox: l~e, .in !lis recem 

ILLUSTRATIONS IN HISTORY . 
'lbe 17th 8Dd 18th camules c.oDStltute a watershed 

period- for thlnldDg about probability and statistics. 
Tbe works by Gnmnt and others on civic and medical 
statistics ~ ridl examples. Poisson's work as a 
mlHtary .sratlstldail for Napoleon was· dis~nated 
ami adopted in 18th cenrury philosophical wrltlni-
That Jn tum was adopted durin.a tbe prc:colonial and 
colooial periods iD the United States. To judie from 
Garry W1lls' Inventing America. Jefferson's 
constlmtlooal writing and personal papers contain 
strong lDflUCDCM of numerical thinking c:haractmstic 
of ths earlier period. Madison's Federalist papers 
alSO Clddblt the tbeme8. 

The 19th century growth of applied statistics 
appears clearly In popular literature of the period. 
1bere were similarities in some respects to 
contempclla&y debates. As.o;aults on tbe validity of 
census data. for instance. worrled joumallsts, 
businessmen. . and poUdcinns, u well as census 
ot.ficlals. Coums were alllmportno.t in government 
representation, government subsidy or geographic 
explOJ;'ltion. business. and so on. Even the subtler 
matters appear early. The contllct between two 
cultures about wbich C.P. Snow has written was 

Congo. He alludes to it in Andronwda. Srfain to help 
delermine if his apes are really ,apes from -IaDpage. 
appeanm.ce ud behaviour. FraDds Crick's problem 
in extraCting ~-ray photOBfaphs t'mm ·Rasallnd 
Franklin in Double Helix has pfeCCdent in early 
sdemitfc Ideas about Pythago!ean privacy. of researob. 
(to-judge by de. Saati.lla' s biography of Gai.Deo ). Tbe 
problem appears In more modem fonn in popular 
treatments ot sdenttflc work. sw:h as Koestler's 
Mldwif~ Toml and Jere111y Bernstein's Scilmtific 
Exp~rience. John Cheever exploits prhnltlve nodons 
of coincidence and IUcellbood Jn tbe opeulng to Bulin 
Park.. ln Bombu. Len Deighton em in computiDg 
simple probabDittes for be:iq ShOt down In missions 
duriDg World Warn. 

In more popular writing, science tlcdon IDd spy 
novels an: a rich Naurce of cumplcs of the usc of 
stalist.ical ideas. Isaac Aslmov. c. Slmak. and other 
scieucc fidlon authors refer' repeuecny IDd explicitly 
to odds. num.elical prollabillty, distributiODS and otbt:r 
coDCeptS. In a recent science fiction television series 
and book. The Hitr:hhilcer's GuiM to th~ GakD:y, 
inlprobabillty ls a keyslooe of the stories. 



~TL. LOUIS. UNIV TEL:1-708-853-0214 

Teacblq Statistia throagb Eoglllb Classes 
Clearly. them am JDaDY JBustradoDs of probability· 

and stltiStics in prose, poeuy aud drama. The 
eumples given bere are the result of casual interest. 
or hobby. of collecting StiCh examples. If we are to 
serlously CODSider embeddiq statistics lessons into 
non-mathematics courses, we need to systemadcally 
assess the actual readings. assignments and 
discussions tbJu: occur as a part of regUlar 
coursework. We bavc ~tly begun sudl an attempt 
as part of an Natlonal Science FoUDdation fimded 
project on Statistical Li.Uracy: A Cross·Disciplin4 
Devdopmenl Project. Al present. we are compiling 
Dsts of "'typical" literature readings required of Dinrh 
grade students in the Olicago metropolitan area. 
lben. we are systematically reading those sources 
with the purpose of identifying any uses of 
probablUty. swlstlcs or statlstical reasoning. Based 

. .. 

2 

May 12'94 15:03 No.011 P.07 

on I!Je iiiDsttatlons already shared in tbls article. we 
baw-n=aaon-to-believe-tbat· we-wm-find-mmy-plaees-
m wbicb man tmroducf.ory 1essoas {or vtpeUeS) m 
statistics may be embedded. Coupled with tbe 
in10mWion from Smeeron and Smeeron (1985) that 
such lessons arc feasible. we bope we can begin to 
l!UeJrare IDstmctlon in statistical. tblnklng into all 
areas. 

Retereaces 
Smeeton,. N.C. and Smeeton. E.A. (1984). Statistical 
Ideas in English Stndies. Tt!DChing Statistics, 6(3). 
92-94. 
Smeeton. N.C. and Smeeton. ._EA (198S). The 
Interface Between EngliSh and Stal:i.SI:ics. Teaching 
Srazisrics, 7(2), 34-37. 

. . ·.