American Association for Public Opinion Research

Poverty as We Know It: Media Portrayals of the Poor

Author(s): Rosalee A. Clawson and Rakuya Trice
Source: The Public Opinion Quarterly, Vol. 64, No. 1 (Spring, 2000), pp. 53-64
Published by: Oxford University Press on behalf of the American Association for Public Opinion
Research
Stable URL: http://www.jstor.org/stable/3078840 .
Accessed: 15/06/2014 01:06

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

American Association for Public Opinion Research and Oxford University Press are collaborating with JSTOR
to digitize, preserve and extend access to The Public Opinion Quarterly.

http://www.jstor.org

This content downloaded from 128.83.63.20 on Sun, 15 Jun 2014 01:06:01 AM

All use subject to JSTOR Terms and Conditions
POVERTY AS WE KNOW IT
MEDIAPORTRAYALS
OF THE POOR

ROSALEEA. CLAWSON
PurdueUniversity
RAKUYA TRICE
Indiona
University

Introduction
On the campaigntrailduringthe 1992 presidential election,Bill Clinton's
stumpspeechincludeda pledgeto "endwelfareas we knowit"to thedelight
of mostaudiences.Two yearslaterduringthe 1994 congressional election,
one of themostpopularplanksof the Republicans'ContractwithAmerica
was the"PersonalResponsibility Act,"whichcalledfora majoroverhaulof
thewelfaresystem.The electionofthisRepublicanCongressinitiated a great
dealoflegislative
activityandpresidential
maneuvering ontheissueofwelfare
reform. The culmination of thoseeffortsoccurredin Augustof 1996 when
PresidentClintonsignedintolaw sweepingwelfarereformlegislation.By
endingthefederalguaranteeof supportforthepoor and turning controlof
welfareprogramsover to the states,thislegislationreversed6 decades of
socialpolicyandbegota new era of welfarepolitics.Throughout thisperiod
ofintensepoliticalactivity,
themediafocuseda significantamountofattention
on povertyand welfarereform.
In thisresearch, we analyzemediaportrayals of thepoorduringthistime
whenwelfarereform was highon thenation'sagenda.Weinvestigate whether
the mediaperpetuate inaccurateand stereotypical
imagesof thepoor.Spe-
we examinethephotographs
cifically, thataccompanystorieson povertyin
fiveU.S. newsmagazinesbetweenJanuary 1, 1993,andDecember31, 1998.

We would like to thankthe PurdueUniversityMARC/AIMSummerResearchProgramfor

providingsummersupportforRakuyaTrice,thePurdueResearchFoundationforproviding a
SummerFacultyGrantforRosalee Clawson,and thePurdueUniversity LibraryScholarsGrant
Program.We greatlyappreciatetheefforts
of JillClawsonand ChrisSalisbury,who werein-
strumentalin obtaininginformation
fromthe CensusBureauweb site.We wouldalso like to
thanktheeditorand anonymous reviewersfortheirhelpfulcomments.
PublicOpinionQuarterly Volume64:53-64 ? 2000 by theAmericanAssociationforPublicOpinionResearch
All rightsreserved.0033-362X/2000/6401-0002$02.50

This content downloaded from 128.83.63.20 on Sun, 15 Jun 2014 01:06:01 AM

All use subject to JSTOR Terms and Conditions
54 Rosalee A. Clawson and Rakuya Trice

Portrayals of the Poor

In a studyof news magazinesbetween1988 and 1992,Gilens(1996a) in-

vestigated theaccuracyof themediain theirportrayals of thepoor.Gilens
(1996a) foundthatpovertywas disproportionately portrayed as a "black"
problem.Blacksmakeup less thanone-third ofthepoor,butthemediawould
lead citizensto believethattwo out of everythreepoor people are black.
Moreover, Gilens(1996a) foundthatthe"deserving" poor,especiallytheblack
deservingpoor,wereunderrepresented in newsmagazines.For example,the
blackelderlypoorandblackworking poorwererarelyportrayed. In addition,
Gilensexaminedmediadepictionsof thepoorbetween1950 and 1992 and
foundthatblackswere"comparatively absentfrommediacoverageofpoverty
duringtimesof heightened sympathy forthepoor" (1999, p. 132). In this
research, we pickup whereGilensleftoffby analyzingmediaportrayals of
thedemographics of poverty between1993 and 1998.
In addition,we extendGilens's workby investigating whethercommon
stereotypicaltraitsor behaviors associatedwith the poor are portrayed in the
media. In our society,citizens believe poor people have many undesirable
qualitiesthatviolatemainstream American ideals.Forexample,manycitizens
say people are poor due to their own "lack of effort" and "loose moralsand
drunkenness" (Kluegel and Smith1986, p. 79). A majorityof Americans
believethat"mostpeoplewho receivewelfarebenefits aretakingadvantage
of thesystem"(Ladd 1993,p. 86). Another piece of conventional wisdomis
thatpoormothers onpublicassistancehaveadditional babiestoreceivegreater
welfarebenefits. People also believethatpoorfamiliesare muchlargerthan
middle-class families(Sidel 1996).
Severalmediastudieshavefoundsuchstereotypical representations ofpov-
erty(Goldingand Middleton1982; Martindale1996). The mediaoftende-
scribetheunderclassin behavioraltermsas criminals, alcoholics,and drug
addicts,andtheunderclass is linkedwithpathological behaviorinurbanareas
(Gans 1995). Parisi's (1998) in-depthanalysisof a Washington Post series
on poverty demonstrated thatthemediaperpetuate stereotypes ofthepooras
lazy,sexuallyirresponsible, andcriminally deviant.Coughlin(1989) discussed
themedia'semphasison "welfarequeens"-a phrasethatinvokesimagesof
poorwomenlivingthehighlifeby defrauding and takingadvantageof the
welfaresystem.Thesestudiesfocusedon howthepoorweredescribedin the
textof newsstories;in thisstudy,we analyzewhether stereotypical traitsof
thepoorare presented in magazinephotographs.
Whyis it important to studythevisual imagessurrounding theissue of
poverty? The visualrepresentation of a politicalissue is an integralpartof
thedefinition of thatissue.' Visualimages(alongwithmetaphors, exemplars,

linkedto otherissuessuch
1. See Entman(1995) fora discussionof how povertyis implicitly
as crime,drugs,and gangsthroughvisualimageson televisionnews.

This content downloaded from 128.83.63.20 on Sun, 15 Jun 2014 01:06:01 AM

All use subject to JSTOR Terms and Conditions
Poverty as We Know It 55
and catchphrases)defineand illustrate
particularissueframes(Gamsonand
Lasch 1983).Forexample,NelsonandKinder(1996) demonstrate thatvisual
frameshavea significantimpacton publicattitudestowardaffirmativeaction.
People and eventsthatappearin photographs accompanying newsstoriesare
notsimplyindicative ofisolatedindividuals
andoccurrences; rather,thepho-
tographsaresymbolic of"thewholemosaic"(Epstein1973,p. 5). Thepictures
providetexture,drama,and detail,and theyillustratetheimplicit,
thelatent,
the"takenforgranted," andthe"goeswithout saying."Furthermore,scholars
shouldpayattention to visualimagesbecausejournalistsandeditorsperceive
themtobe a centralpartofa newsstory.In hisclassicstudyofhowjournalists
selectstories,Gans arguesthatmagazine"editorsconsiderstillpicturesas
importantas text"(1979, p. 159).

Research Design
In thisresearch,we testthehypothesis thatthemediaportray poorpeople
inaccuratelyand stereotypically.
The datawerecollectedbyexamining every
storyon thetopicsofpoverty, welfare, andthepoorbetweenJanuary 1, 1993,
and December31, 1998,in fivenewsmagazines:BusinessWeek,Newsweek,
New YorkTimesMagazine,Time,and U.S. News& WorldReport.2 We used
theReader'sGuidetoPeriodicalLiterature tolocatethestoriesandtoidentify
othercross-referenced topics(e.g., incomeinequality).Seventy-four stories
wereidentified as relevantfora totalof 149 picturesof 357 poorpeople.3
See table 1 forthedistribution and peopleby magazine.
of stories,pictures,
The photographs were analyzedin two ways. First,we scrutinized each
pictureas a whole.For thosepicturesthatincludeda motherwithchildren,
we notedthesize and race of thefamily.Second,we examinedthedemo-
graphiccharacteristics
ofeachpoorindividual inthepictures.
Forcodingrace,
we departedfromGilens's codingprocedure.Gilens (1996a, 1999) coded
whether thepoorpersonwas black,nonblack, orundeterminable. In contrast,
we used a moredetailedclassification schemeand coded whether thepoor
personwas white,black,Hispanic,AsianAmerican, or undeterminable.
We coded each person'sgender(male or female),age (young:under18;
middle-aged:18-64; or old: 65 and over),residence(urbanor rural),and

2. Takenas a whole,thesefivemagazineshavea circulation of over 12 million:BusinessWeek

reaches1,000,000people;theNew YorkTimesMagazinehas a circulation of 1,650,179;New-
sweekhas an audienceof 3,100,000;U.S. News& WorldReporthas a distribution of2,351,313;
and Timehas thelargestreadership with4,083,105subscribers.
3. Therewereseveralstorieson poverty we didnotincludein oursamplebecause:(1) thestory
did notincludeanypictures;(2) thestorywas an editorialor opinioncolumnthatonlyincluded
a pictureof theauthor;(3) thestorywas foundto be irrelevantto ourresearchtopic-e.g., one
storywas cross-listedas incomeinequalityand poor,butactuallyfocusedon Democraticand
Republicanpartyefforts to winworking-classvotes;(4) thepicturesin thestorydid notpertain
to contemporary povertyin theUnitedStates;(5) thestoryitselfwas missingfromits bound
volume(n = 6); or (6) thestorywas in a magazinethatwas at thebinders(n = 1).

This content downloaded from 128.83.63.20 on Sun, 15 Jun 2014 01:06:01 AM

All use subject to JSTOR Terms and Conditions
56 Rosalee A. Clawson and Rakuya Trice
of Povertyby Magazine,1993-98
Table 1. Representations

U.S.
New York News
Business Times & World
Week Newsweek Magazine Time Report Total

Numberof
stories 18 13 8 13 22 74
Numberof
pIctures 21 24 18 30 56 149
Numberof
poorpeo-
ple 40 64 35 78 140 357

workstatus(working/jobtraining We also analyzedwhether

ornotworking).4
eachindividualwas depictedin stereotypical
ways,suchas pregnant,
engaging
takingor sellingdrugs,drinking
in criminalbehavior, alcohol,smokingcig-
or wearingexpensiveclothing
arettes, Formanyofourvariables,
orjewelry.5
we wereable to comparetheportrayal of povertyin newsmagazinesto the
as measuredby theCurrent
realityof poverty PopulationSurvey(CPS) con-
ductedby the U.S. Census Bureau or as reportedby the U.S. House of
RepresentativesCommittee on Waysand Means.6

Research Findings
Many citizensgreatlyoverestimate thenumberof black people amongthe
poor(Gilens1996a).Do newsmagazinesperpetuate andreinforcethatbelief?
Accordingto the 1996 CPS, AfricanAmericansmakeup 27 percentof the
poor,butthesefivemagazineswouldlead citizensto believethatblacksare
49 percentof thepoor(p < .001; see table2). Whites,on theotherhand,are
depictedas 33 percentof thepoor,whentheyreallymakeup 45 percentof
thosein poverty(p < .001). Therewere no magazineportrayals of Asian
Americansin poverty,and Hispanicswereunderrepresented by 5 percent.
of poorHispanicsand AsianAmericansmaybe
This underrepresentation

4. A "Do Not Know"categorywas includedforthesevariables.

5. To ensuretheintegrity of our data,we conducteda testof intercoder A second
reliability.
person,whowas unawareofthehypotheses, codeda subsetofoursampleofphotographs. Across
therewas an averageintercoder
thevariablesof interest, of .90.
reliability
6. Althoughwe are analyzingmediaportrayals of povertybetween1993 and 1998,forease of
presentationwe use CPS datafromMarch1996 or WaysandMeansCommittee datafrom1996
of thepoor.The 1996 datarepresent
to establishthetruecharacteristics a reasonablemidpoint.
Moreover,the relevantnumbersdo not varymuchacrossthe timeperiodof interest; in no
instancewouldtheminorfluctuations changethesubstantiveor statistical of our
interpretation
results.

This content downloaded from 128.83.63.20 on Sun, 15 Jun 2014 01:06:01 AM

All use subject to JSTOR Terms and Conditions
Poverty as We Know It 57
Table 2. The Percentof TruePoor and thePercentof MagazinePoorby
Race, 1993-98

African Asian
Whites Americans Hispanics Americans

Truepoor 45 27 24 4
Magazinepoor 33*** 49*** 19* 0**
Poor in Newsweek,Time,and
U.S. News & WorldReport 33*** 45*** 22 0**
SOURCE.-"MarchCurrent PopulationSurvey"(U.S. Bureauof theCensus 1996).
NOTE.-We conducteddifference of proportion testsin whichtheproportion observedin the
magazineis comparedto thetrueproportion as reported by theCurrent PopulationSurveyfor
each racialcategory(Blalock 1979). A statistically
significant
resultindicatesthatthemagazine
portrayalof a particular
racialgroupis notrepresentative of thetruepoor.Due to rounding, the
percentages maysumto morethan100 percent.The samplesize is 347 fortheanalysisbased
on all fivemagazines.The samplesize is 272 fortheanalysisbased on Newsweek,Time,and
U.S. News & WorldReport.
*p .05.
p<.01.
***p <.001.

partof a largerphenomenon in whichthesegroupsareignoredbythemedia

in general.For example,Hispanicsand AsianAmericansarerarelyfoundin
massmediaadvertising (BowenandSchmid1997;WilkesandValencia1989).
Similarly,Dixon (1998) documentedthe invisibility of Hispanicsin local
news; however,thereis evidencethatin particular regionsHispanicsare
represented in accordancewiththeirproportion in thepopulation (Greenberg
and Brand1998; Turket al. 1989). Unlikeblacks,Asian Americansare as-
sociatedwithintelligence, notwelfaredependency (GilbertandHixon1991;
Gilens 1999). Thus, theirabsencemay reflecta positivestereotype, but a
stereotypenonetheless. Clearlythesecomments regarding HispanicsandAsian
Americans arespeculative.Furtherresearchis neededonmediarepresentations
of thesetwogroups.
Focusingon just the threemagazinesGilensincludedin his study(i.e.,
Newsweek, Time,and U.S. News& WorldReport),whitesmakeup 33 percent,
blacksmakeup 45 percent,and Hispanicsare 22 percentof themagazine
poor (see table2). In comparison, Gilens(1996a) foundthat62 percentof
thepoorwereAfricanAmericanin thesemagazinesbetween1988 and 1992.
Althoughat firstglanceour statistics maysuggestthatthemagazineshave
becomeless likelyto puta black face on poverty, we hesitateto drawthat
conclusiongiventhecodingdifference mentioned earlier.Recall thatGilens
codedwhether thepoorpersonwas black,nonblack, orundeterminable. Since
Gilens(1996a) reports a higherpercentage ofpoorpeopleforwhichracewas
not identified(12 percentcomparedto our 4 percent),it seemslikelythat
manyof thepoorpeoplewe coded as Hispanic,Gilenswouldhavecodedas
undeterminable. If we treatHispanicsin thatfashionand therefore exclude

This content downloaded from 128.83.63.20 on Sun, 15 Jun 2014 01:06:01 AM

All use subject to JSTOR Terms and Conditions
58 Rosalee A. Clawson and Rakuya Trice
Table 3. The Percentof AFDC Parentsand thePercentof Magazine
AdultPoorby Race, 1993-98

NativeAmericans,
African AsianAmericans,
Whites AmericansHispanics andOther

AFDC parents 36 37 21 7
Magazineadultpoor 34 48** 18 0
SOURCE.-Overviewof Entitlement
Programs(U.S. House of Representatives,
Com-
mitteeon Waysand Means 1998).
NOTE.-N = 159.
**p<.01.
p< .001.

themfromour analysis,blacksmakeup 58 percentof thepoorand whites

makeup42 percent-figures thatmirrorGilens'sdataquiteclosely.Regardless
oftheexactproportion, itis clearthesenewsmagazinescontinue toracecode
theissue of poverty.7
Since we are examiningportrayals of thepoorduringa periodof intense
debateoverwelfarereform, perhapstheracialcharacteristicsofthemagazine
poormirror welfarerecipients morecloselythantheyrepresent poorpeople
in general.The House WaysandMeansCommittee providestheracialbreak-
downforparents onAid toFamilieswithDependent Children (AFDC). There-
fore,in table3, we comparetheracialcomposition of AFDC parentsto the
magazineportrayal of pooradults.Indeed,theportrayal of poorwhitesand
Hispanicsmatchesmorecloselythetrueracialcharacteristics of welfarere-
cipients;however,blacksarestillheavilyoverrepresented (48 percent)among
themagazinepoor.Moreover, blacksmakeup 52 percentof thepooradults
whoareportrayed in storiesthatfocusspecifically on welfare(rather thanon
poverty in general).
Gilens(1996a, 1999)foundthatblackswereevenmoreprominent in stories
on poverty topicsthatwerenotverypopularwiththepublic.Between1993
and 1998, therewere severalstorieson unpopularissues,such as welfare
reform and pregnancy, publichousing,and welfareand thecycleof depen-
dency.8 We examinedtheproportion ofblacksamongthepoorin thesestories
and foundthatit jumpedto 63 percent,whereaswhitesmade up only 19
percentand Hispanicswere 18 percent.In contrast, blackswereassociated
less oftenwithsympathetic topics.In storieson welfarereform andchildren,
welfarerecipientsanddaycare,andjob training, 46 percentofthepoorwere
black,while32 percentwerewhiteand 22 percentwereHispanic.9We also

7. Unfortunately,we faceda trade-off

betweenproviding a moredetailedanalysisof theracial
portrayalof thepoorand makingexactcomparisons withGilens'sresearch.
8. These storieson unpopularissuesincluded75 poorindividuals.
9. These storieson sympathetictopicsincluded100 poorindividuals.

This content downloaded from 128.83.63.20 on Sun, 15 Jun 2014 01:06:01 AM

All use subject to JSTOR Terms and Conditions
Poverty as We Know It 59
Table 4. The PercentofTruePoorandthePer-
centof MagazinePoor by Age, 1993-98

Under18 18-64 65 andOver

Truepoor 40 51 9
Magazinepoor 53*** 43** 4**
SOURCE. -"March Current
Population
Survey"
(U.S. Bu-
reauof theCensus 1996).
NOTE.-N = 347.
p<..01.
***p < .001.

analyzedtwo storiesthatfocusedon various"myths"surrounding welfare

reform. 16 ofthe22 poorpeopledepictedin thesetwostorieswere
Ironically,
black.
The newsmagazinesexaggerated thefeminizationof poverty by about14
percent.Accordingto the CPS, 62 percentof the adultpoor are women,
whereas76 percentof the magazinepoor are women(N = 161).1oAgain,
though,sincemostof thesestoriesdiscusspoverty in thecontext
specifically
of welfarereform,itis importantto comparethemagazinepoorto peopleon
welfare.The vastmajority of adultAFDC recipients are female,so thepre-
dominanceof womenamongthepooris fairlyaccurate(U.S. House ofRep-
resentatives,
Committee on Waysand Means 1998).
In termsof theage of thepoor people,we foundthatchildrenwereov-
errepresentedamongthemagazinepoor(see table4)." Childrenare usually
thoughtofas a fairly
deserving groupofpoorpeople(Cook andBarrett 1992);
however,the largeproportion of black childrenamongthemagazinepoor
may undermine thatbelief.In Iyengar's(1990) experimental researchon
attributions
ofresponsibility
forpoverty,subjectsindicatedthatblackchildren
shouldtakeresponsibilityfortheirown plight,whereaswhitechildrenwere
notexpectedto solve theirown problems.
In contrast,
theelderly,whoarethemostsympathetic groupofpoorpeople,
wererarely Most
portrayed. peoplebelievetheelderly reallyneedtheir benefits
andthattheyuse themwisely(Cook andBarrett1992).Iyengar(1990) found
thatpeople thoughtsocietyshouldaid (bothblack and white)poor elderly
widows.This sympathetic groupmakesup 9 percentof thetruepoor,but
only4 percentof themagazinepoor (see table4).
We also examinedwhether poorpeople wereportrayed in urbanor rural
The magazinedepictionsimpliedthatpoverty
settings. is almostcompletely

10. Therewas no race by genderinteraction.

11. Therewas no race by age interaction.

This content downloaded from 128.83.63.20 on Sun, 15 Jun 2014 01:06:01 AM

All use subject to JSTOR Terms and Conditions
60 Rosalee A. Clawson and Rakuya Trice
an urbanproblem. Ninety-six percentofthepoorwereshowninurbanareas.'2
According totheCPS, mostpoorpeople(77 percent) do resideinmetropolitan
areas;however, themagazineportrayals greatly exaggeratethetrueproportion
(p < .001).13Accordingto Gans (1995), theurbanunderclassis oftenlinked
withvariouspathologiesand antisocialbehavior.Thus,thisemphasison the
urbanpoordoes notpromotea positiveimageof thosein poverty.
The medialeave theimpression thatmostpoorpeopledo notwork:only
30 percentofpooradultswereshownworking injob training
orparticipating
programs (N = 198). In reality,50 percentof thepoorworkin full-or part-
timejobs, accordingto theCPS (p < .001).'4 Whenwe focussolelyon those
storiesthatspecifically discusswelfare,35 percentof thepoor are shown
eitherworkingor in job training. Accordingto theHouse Waysand Means
Committee, 23 percent ofAFDC recipients workedorparticipated ineducation
orjob training programs in 1995. Thesephotographs reflecttheemphasisof
manycontemporary welfarereformers, liberaland conservative,on "work-
fare"rather thanwelfare.Since manycitizenssupport workrequirements for
welfarerecipients (Weaver,Shapiro,andJacobs1995),theseimagesarepos-
itiveones. Not surprisingly, whitesweremorelikelyto be shownin these
picturesthanblacks.
Next,we analyzedtheextentto whichthenewsmagazinesreliedon ster-
eotypicaltraitsintheirdepictions ofthepoor.We examinedwhether themedia
perpetuate thenotionthatwomenon welfarehave lotsof children.Whena
mother was portrayed withherchildren inthesemagazines, theaveragefamily
size was 2.80. This is virtually
identicalto thefigureof 2.78 reported by the
House Ways and Means Committeeforthe averageAFDC familysize in
1996.In themagazines,therepresentation of poorwomenand theirchildren
differedby race. The averagefamilysize forwhiteswas 2.44, whereasthe
averagesize forblacks was 3.05 and 2.92 forHispanics.Althoughthese
differences are notstatistically
significant,thedirectionsuggeststhatcitizens
receiveda less flatteringviewofpoorminority families.TheWaysandMeans
Committee does notreportthetruefigureby race;however, theU.S. Census
Bureau (1995) providesdata on the averagenumberof childrenever had
(ratherthantheaveragenumberof childrencurrently receivingbenefits)by
AFDC mothersby race.'5These data show thatblack AFDC womenhave
onlyslightly (and nonsignificantly) morechildren thanwhiteAFDC women.

12. Please notethesestatistics are based on a reducedsamplesize (N = 205), because many

(43 percent)of thepoorindividualswerecoded as "Don't Know"fortheirresidency. In many
cases it was difficult
to ascertainwhether thesettingwas a ruralor urbanone, so we decided
to erron theconservative side and code onlytheunambiguous settings.
13. Therewas no race by residenceinteraction.
14. The dataon theworking poorfromtheCurrent PopulationSurveyincludepeoplewho are
16 and over,whereasthedata on themagazineworkingpoor includepeople who are 13 and
over.
15. These data are fromthe Surveyof Incomeand ProgramParticipation conductedbetween
Juneand September of 1993 (U.S. Bureauof theCensus 1995).

This content downloaded from 128.83.63.20 on Sun, 15 Jun 2014 01:06:01 AM

All use subject to JSTOR Terms and Conditions
Poverty as We Know It 61
HispanicAFDC mothers, on theotherhand,do havemorechildren thannon-
HispanicAFDC women.
To our surprise,the media did not overlyemphasizeotherstereotypical
associatedwiththepoor.Of the357 peoplecoded,onlythree
characteristics
were shownengagingin criminalbehavior,and anotherthreewere shown
withdrugs.No alcoholicswerepresented, andonlyone personwas smoking
a cigarette.
However,of thosesevenstereotypical portrayals,
onlytheperson
smokingwas white-theotherswereeitherblackorHispanic.Onlyone poor
womanwas pregnant, so the media werenotprovidingimagessuggesting
thatpoorwomensimplyhavebabiesto obtainlargerwelfarechecks.Again,
though, thisstereotypical is ofa Hispanicwoman.We also examined
portrayal
whether themediapresented imagesconsistent withthe"welfarequeen"ster-
eotype.We feltthatpoorpeoplewhowereshownwearingexpensive jewelry
or clothingwouldfitthisstereotype.Thirty-nineindividualswereshownwith
flashyjewelryor fancyclothes;blacksand Hispanicsweresomewhatmore
likelyto be portrayedthisway thanwhites.
In sum,themagazinesoftenportrayed an inaccuratepictureof thedem-
ographiccharacteristicsofpoorpeople.Thesemagazinesoverrepresented the
black,urban,andnonworking poor.Blackswereespeciallyprominent in sto-
rieson unpopularpovertytopics,and blackwomenwereportrayed withthe
mostchildren.Otherstereotypical traitslinkedwithpoor people werenot
commonin themagazineportrayals. Nevertheless,in thoseinstanceswhen
themediadepictedpoorpeoplewithstereotypical theytended
characteristics,
to be black or Hispanic.The mostsympathetic groupof poor people,the
elderly,was underrepresented amongthe magazinepoor.The media were
most accuratein mirroring the predominance of women among welfare
recipients.

Discussion
These portrayals of povertyare important because theyhave an impacton
publicopinion.A variety ofexperimental researchdemonstrates thatnegative
imagesofblacksinfluence publicopinion(Gilliamet al. 1996;Iyengar1990;
Johnson et al. 1997; Mendelberg1997; Peffley, Shields,andWilliams1996).
Furthermore, whitecitizens'stereotypicalbeliefsaboutblacksdecreasetheir
supportforwelfare(Gilens 1995, 1996b).
In turn,publicopinionhas an impacton publicpolicy(Page and Shapiro
1983). Thus,if attitudes on poverty-related issues are drivenby inaccurate
andstereotypical portrayalsofthepoor,thenthepoliciesfavoredbythepublic
(andpoliticalelites)maynotadequatelyaddressthetrueproblemsofpoverty.
Furthermore, theseinaccurateportrayals of theracial characteristicsof the
poor may prime the white public to favor politicalcandidateswho make
raciallycodedarguments a linchpinof theircampaignstrategies. Whenthese

This content downloaded from 128.83.63.20 on Sun, 15 Jun 2014 01:06:01 AM

All use subject to JSTOR Terms and Conditions
62 Rosalee A. Clawson and Rakuya Trice
candidatesare elected,theyfavorwelfare(and other)policiesthatare in
keepingwiththeirracializedrhetoric.
Itis possiblethatthetextofthesestoriesonpoverty contains datadescribing
thetruedemographic characteristics
of thepoor.It is unclearwhatimpacta
storythatdispels stereotypes in its textbut perpetuates stereotypes in its
photographs wouldhave on publicopinion.Graber'sresearchon television
suggeststhataudiovisualthemesaremorememorable thanverbalinformation
(Graber1990, 1991). Although newsmagazinesarea verydifferent medium
thantelevision, possiblethatmagazinephotoscapturetheau-
it is certainly
dience's attention inthesamewayas television visuals.Psychologicalresearch
suggeststhatvividimagesofparticular cases aremorememorable andinflu-
entialthandrystatistical data(Fischhoff andBar-Hillel1984).Indeed,Hamill,
Wilson,andNisbett's(1980) experimental researchshowsthata vivid,detailed
description of a poor womanon welfarehas a largerimpacton subjects'
opinionsaboutwelfarerecipients thanstatisticalinformationaboutwomenon
welfare.
Gilens(1996a, 1999) investigated severalexplanations forwhyblacksare
overrepresented amongthepoorand concludedthat,at leastin part,itis due
to journalists'stereotypes. Gilens'sresearchreceivedconsiderable attention
frommedia elites,includingbeingthe lead topicof discussionon CNN's
ReliableSourceson August24, 1997. Unfortunately, ourdataillustratethat
journalistsand editorshave continuedthepracticeof race codingtheissue
of povertyevenafterit was brought to theirattention.
We mustalso pointoutthatthisracecodingofpoverty in newsmagazines
is notan isolatedincident; rather,
theracialbias reported hereis a widespread
phenomenon. Forexample,Clawsonand Kegler(in press)conducteda com-
parableanalysison the portrayal of povertyin introductory textbookson
Americangovernment and foundthatblacksweredisproportionately repre-
sented.In addition,severalscholarshavedocumented thenegativeimagesof
blacksin newscoverageofcrime(Delgado 1994;Dixon 1998;Entman1990,
1992,1994;Johnson 1987).Anditdoes notendthere:whether itis children's
programs, "reality-based" programs, sitcoms,or advertising,blacksareoften
portrayed in a stereotypical fashion(Graves1996; Humphrey and Schuman
1984;Oliver1994;Poindexter andStroman 1981).Theseimagesarepervasive
in oursociety.

Conclusion
In conclusion,
blacksweredisproportionately portrayedamongmagazinepor-
trayalsof thepoorbetween1993 and 1998. Blacks wereespeciallyoverre-
presentedin negativestorieson povertyandin thoseinstances whenthepoor
were presentedwithstereotypical traits.In addition,the "deserving"poor
wereunderrepresented in themagazines.Overall,thephotographic imagesof

This content downloaded from 128.83.63.20 on Sun, 15 Jun 2014 01:06:01 AM

All use subject to JSTOR Terms and Conditions
Poverty as We Know It 63
poorpeopleinthesefivenewsmagazinesdo notcapturethereality ofpoverty;
instead,theyprovidea stereotypicaland inaccurate
pictureofpovertywhich
resultsin negativebeliefsaboutthepoor,antipathy
towardblacks,anda lack
of supportforwelfareprograms.

References
Blalock,HubertM., Jr.1979. Social Statistics. 2d ed. New York:McGraw-Hill.
Bowen, Lawrence,and JillSchmid. 1997. "MinorityPresenceand Portrayalin Mainstream
MagazineAdvertising: An Update."Journalism and Mass Communication Quarterly74(1):
134-46.
Clawson,Rosalee A., and ElizabethR. Kegler.In press."The 'Race Coding' of Povertyin
AmericanGovernment Textbooks."HowardJournalof Communications.
Cook,Fay Lomax,and EdithBarrett.1992.Support fortheAmericanWelfareState.New York:
ColumbiaUniversity Press.
Coughlin,RichardM. 1989. "WelfareMythsand Stereotypes." In Reforming Welfare:Lessons,
Limits,and Choices,ed. RichardM. Coughlin.Albuquerque:University of New Mexico.
Delgado,Richard.1994."Rodrigo'sEighthChronicle:BlackCrime,WhiteFears-on theSocial
Construction of Threat."VirginiaLaw Review80:503-48.
Dixon, TravisL. 1998. "Race and Crimeon Local TelevisionNews." Paperpresentedat the
annualmeetingof theNationalCommunication New York.
Association,
Entman, RobertM. 1990."ModemRacismandtheImagesofBlacksin Local TelevisionNews."
CriticalStudiesin Mass Communication 7:332-45.
. 1992. "Blacks in the News: Television,Modem Racism and CulturalChange."
Journalism Quarterly 69(2):341-61.
. 1994. "Representation and Realityin thePortrayal of Blacks on NetworkTelevision
News." Journalism Quarterly 71(3):509-20.
. 1995."Television,Democratic TheoryandtheVisualConstruction ofPoverty."Research
in PoliticalSociology7:139-59.
Epstein,EdwardJay.1973.NewsfromNowhere.New York:RandomHouse.
Fischhoff, BaruchB., and Maya Bar-Hillel.1984. "Diagnosticity and the Base Rate Effect."
Memoryand Cognition12:402-10.
Gamson,WilliamA., and KathrynE. Lasch. 1983. "The PoliticalCultureof Social Welfare
Policy."In EvaluatingtheWelfareState,ed. ShimonE. Spiroand EphraimYuchtman-Yaar.
New York:AcademicPress.
Gans,Herbert J. 1979. DecidingWhat'sNews.New York:PantheonBooks.
. 1995. The WaragainstthePoor. New York:Basic Books.
Gilbert,Daniel T., and J. GregoryHixon. 1991. "The Troubleof Thinking:Activationand
Applicationof Stereotypic Beliefs."Journalof Personalityand Social Psychology60(4):
509-17.
Gilens,Martin.1995. "Racial Attitudes and Oppositionsto Welfare."JournalofPolitics57(4):
994-1014.
1996a. "Race and Povertyin America."Public OpinionQuarterly 60(4):515-41.
1996b."'Race Coding' and WhiteOppositionto Welfare." AmericanPoliticalScience
Review90(3):593-604.
. 1999. WhyAmericansHate Welfare.Chicago:University of ChicagoPress.
Gilliam,FranklinD., Jr.,ShantoIyengar,Adam Simon,and OliverWright.1996. "Crimein
Black and White."HarvardInternational JournalofPress/Politics 1(3):6-23.
Golding,Peter,and Sue Middleton.1982. Imagesof Welfare.Oxford:MartinRobertson.
Graber,Doris A. 1990. "Seeing Is Remembering: How Visuals Contribute to Learningfrom
TelevisionNews."Journalof Communication 40:134-55.
. 1991. "WhatYou See Is WhatYou Get."Paperpresented at theannualmeetingof the
AmericanPoliticalScienceAssociation, Washington, DC.
Graves,SherrylBrowne.1996. "Diversityon Television."In TuningIn to YoungViewers,ed.
TannisM. MacBeth.ThousandOaks, CA: Sage Publications.
Greenberg, BradleyS., andJeffrey E. Brand.1998."U.S. Minorities andtheNews."In Cultural

This content downloaded from 128.83.63.20 on Sun, 15 Jun 2014 01:06:01 AM

All use subject to JSTOR Terms and Conditions
64 Rosalee A. Clawson and Rakuya Trice
Diversityand theU.S. Media,ed. YahyaR. Kamalipour andTheresCarilli.Albany,NY: SUNY
Press.
Hamill,Ruth,Timothy DeCamp Wilson,and RichardE. Nisbett.1980."Insensitivity to Sample
Bias: Generalizing fromAtypicalCases." Journalof Personality and Social Psychology 39:
578-89.
Humphrey, Ronald, and Howard Schuman. 1984. "The Portrayalof Blacks in Magazine
Advertisements: 1950-1982."Public OpinionQuarterly 48:551-63.
Iyengar,Shanto. 1990. "FramingResponsibility for PoliticalIssues: The Case of Poverty."
PoliticalBehavior12(1):19-40.
Johnson, JamesD., Mike S. Adams,WilliamHall, and Leslie Ashburn.1997. "Race, Media,
and Violence:Differential Racial Effectsof Exposureto ViolentNews Stories."Basic and
AppliedSocial Psychology19(1):81-90.
Johnson, KirkA. 1987. "Black and Whitein Boston."ColumbiaJournalism Review26 (May/
June):50-52.
Kluegel,JamesR., and Eliot R. Smith.1986. Beliefsabout Inequality.New York:Aldinede
Gruyter.
Ladd, EverettCarll,ed. 1993. "PublicOpinionand Demographic Report:Reforming Welfare."
PublicPerspective 4(6):86-87.
Martindale, Carolyn.1996. "NewspaperStereotypes of AfricanAmericans."In Images That
Injure,ed. Paul MartinLester.Westport, CT: Praeger.
Mendelberg, Tali. 1997. "ExecutingHortons."Public OpinionQuarterly 61(1): 134-57.
Nelson,ThomasE., andDonaldR. Kinder.1996."IssueFramesandGroup-Centrism inAmerican
PublicOpinion."JournalofPolitics58(4):1055-78.
Oliver,MaryBeth.1994. "Portrayals of Crime,Race, and Aggression in 'Reality-Based'Police
Shows:A ContentAnalysis."JournalofBroadcastingand ElectronicMedia 38(2):179-92.
Page,BenjaminI., andRobertY. Shapiro.1983."Effects ofPublicOpinionon Policy."American
PoliticalScienceReview77:175-90.
Parisi,Peter.1998. "A Sortof Compassion:The Washington Post Explainsthe'Crisisin Urban
America."'HowardJournalof Communications 9:187-203.
Peffley,Mark,Todd Shields,and BruceWilliams.1996. "The Intersection of Race and Crime
in TelevisionNews Stories:An Experimental Study."PoliticalCommunication 13:309-27.
Poindexter,Paula M., and CarolynA. Stroman.1981."Blacksand Television:A Reviewof the
ResearchLiterature." JournalofBroadcasting25:103-22.
Sidel,Ruth.1996. KeepingWomenand ChildrenLast. New York:PenguinBooks.
Turk,JudyVanSlyke,JimRichstad,RobertL. Bryson,Jr.,and SammyeM. Johnson.1989.
"HispanicAmericansin theNews in Two Southwestern Cities."Journalism Quarterly 66(1):
107-13.
U.S. Bureau of the Census. 1995. "StatisticalBrief: Mothers Who Receive AFDC
Payments-Fertility and SocioeconomicCharacteristics." CensusBureau WebPage. http://
www.census.gov.
. 1996. "March CurrentPopulationSurvey." Census Bureau Web Page. http://
www.census.gov/ftp/pub/income/histpov.
U.S. House of Representatives,Committee on Waysand Means. 1998. OverviewofEntitlement
Programs.Washington, DC: U.S. Government PrintingOffice.
Weaver,R. Kent,RobertY. Shapiro,and LawrenceR. Jacobs. 1995. "The Polls-Trends:
Welfare."Public OpinionQuarterly 59(4):606-27.
Wilkes, RobertE., and HumbertoValencia. 1989. "Hispanics and Blacks in Television
Commercials." JournalofAdvertising 18(1):19-25.

This content downloaded from 128.83.63.20 on Sun, 15 Jun 2014 01:06:01 AM

All use subject to JSTOR Terms and Conditions