Filetype PDF | Posted on 20 Sep 2022 | 4 years ago
Authentication
digital resources
265x Filetype PDF File size 0.25 MB Source: u.osu.edu
Personality and Social Psychology Bulletin
http://psp.sagepub.com
Consuming Images: How Television Commercials that Elicit Stereotype Threat Can Restrain Women
Academically and Professionally
Paul G. Davies, Steven J. Spencer, Diane M. Quinn and Rebecca Gerhardstein
Pers Soc Psychol Bull 2002; 28; 1615
DOI: 10.1177/014616702237644
The online version of this article can be found at:
http://psp.sagepub.com/cgi/content/abstract/28/12/1615
Published by:
http://www.sagepublications.com
On behalf of:
Society for Personality and Social Psychology
Additional services and information for Personality and Social Psychology Bulletin can be found at:
Email Alerts: http://psp.sagepub.com/cgi/alerts
Subscriptions: http://psp.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav
Citations http://psp.sagepub.com/cgi/content/refs/28/12/1615
Downloaded from http://psp.sagepub.com at UNIV OF WATERLOO on May 20, 2010
10.1177/014616702237644
PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
Davies et al. / CONSUMING IMAGES
Consuming Images: How Television Commercials
That Elicit Stereotype Threat Can Restrain
Women Academically and Professionally
Paul G. Davies
Stanford University
Steven J. Spencer
University of Waterloo
Diane M. Quinn
University of Connecticut
Rebecca Gerhardstein
Florida State University
Women in quantitative fields risk being personally reduced to here?”...There’s a perceived threat that feeds male
negative stereotypes that allege a sex-based math inability. This anger about bright women. People tell jokes that are
situational predicament, termed stereotype threat, can under- negativetowomen,whichjustfuelsit.Anditencourages
minewomen’sperformanceandaspirations in all quantitative the women to pull back into safer ground—to choose
domains. Gender-stereotypic television commercials were more traditional majors, or to adopt a more neutral
employed in three studies to elicit the female stereotype among style—asawayofprotectingthemselvesfromthatanger.
both men and women. Study 1 revealed that only women for —Female science majors
whomtheactivated stereotype was self-relevant underperformed (cited in Seymour & Hewitt, 1997)
-
on a subsequent math test. Exposure to the stereotypic commer
cialsledwomentakinganaptitudetestinStudy2toavoidmath Judgingfromtheabovequotes,stereotypesregarding
itemsinfavorofverbalitems.InStudy3,womenwhoviewedthe theattributes of men and womenhaveremainedalarm-
stereotypic commercials indicated less interest in educational/ ingly stable over the past 25 centuries, which partially
vocational options in which they were susceptible to stereotype explains the clear gender divide in achievement-related
threat (i.e., quantitative domains) and more interest in fields in -
choicesthatpersistsinourculture.Forexample,astrik
which they were immune to stereotype threat (i.e., verbal ing gender disparity is found in the selection of college
domains). majors.Women,includingthosewhoareequallyskilled
and experienced in high school mathematics as their
Doyouknowofanythingthatispracticedbyhumanbe- Authors’Note:ThisarticlewasbasedonaportionofPaulDavies’sdoc-
ings in which the class of men doesn’t excel that of toral dissertation, which was conducted at the University of Waterloo.
women?Orshallwedrawitoutatlengthbyspeakingof We are grateful to Ziva Kunda, Dov Cohen, Claude Steele, Duane
weaving and the care of baked and boiled dishes—just Wegener, Mikki Hebl, Joshua Correll, and Michael Dahlin for their
thoseactivitiesonwhichthereputationofthefemalesex comments on an earlier version of this article. The research was sup-
is based and where its defeat is most ridiculous of all? ported by grants from the Social Sciences and Humanities Research
CouncilofCanadaandbydoctoralscholarshipsfromtheOntarioGov-
—Socrates (cited in Bloom, 1968) ernment.Correspondenceconcerningthisarticlecanbeaddressedto
Paul G. Davies, Department of Psychology, Jordan Hall, Building 420,
Stanford University, Stanford, CA 94305-2130; e-mail: pgdavies@
They just don’t know how to act with women students. psych.stanford.edu.
-
They don’t know what to do with you. Their whole atti PSPB, Vol. 28 No. 12, December 2002 1615-1628
tude, and facial expressions and body language says, DOI:10.1177/014616702237644
“You belong in the kitchen. What are you doing © 2002 by the Society for Personality and Social Psychology, Inc.
1615
Downloaded from http://psp.sagepub.com at UNIV OF WATERLOO on May 20, 2010
1616 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
malecounterparts,stillavoidmajorsinvolvingmoderate gender stereotypic, mirroring those of children raised
to high levels of math (Lefevre, Kulak, & Heymans, on television (Kimball, 1986).
1992). Lefevre and her colleagues (1992) found that
amongtheseequallygiftedstudents,only19%ofwomen STEREOTYPE THREAT
reportedchoosingmajorsthatinvolvedevenamoderate Negative stereotypes are so pervasive in our mass-
level of mathematics (e.g., architecture, business, and media culture that those who are stigmatized have
economics), whereas 43% of men reported choosing detailed knowledge of the accusations targeted at their
majorsinthesedomains.Thisgenderdividewasmagni
- group.
fiedwhentheresearchersfocusedonmajorsconsidered
highinmathcontent(e.g.,computerscience,engineer- African Americans, for example, are likely to be well
ing, and mathematics), in which men are 4 1/2 times aware that stereotypes accuse them of being intellectu
morelikelythanwomentopursuemajors.Exacerbating -
ally inferior and aggressive; and women are well aware
this problem, the college drop-out rate for women in that stereotypes accuse them of being emotional, bad at
mathematics,engineering,andthephysicalsciencesis2 math, and lacking leadership aptitude. (Crocker et al.,
1/2timesthatofmen(Hewitt&Seymour,1991).Natu 1998, p. 518)
-
rally, these gender differences extend beyond the class
-
room to the workplace, where disproportionate num- Thisdetailedknowledgeofrelevantstereotypescanlead
bers of men and women enter careers in quantitative to a situational predicament for stigmatized individuals
fields (Eccles, 1994; Stangor & Sechrist, 1998). Even (Steele, 1997). Specifically, the risk of being personally
thoughtherehasbeenamajorincreaseintheparticipa- reduced to a negative stereotype can elicit a disruptive
tionofwomeninthegeneralworkforce,sometradition- state among stigmatized individuals that undermines
ally masculine domains (e.g., engineering and physics) performance and aspirations in any alleged stereotype-
have seen no significant increase in women’s participa- relevant domain—a situational predicament termed
tion over the past 50 years (Culotta & Gibbons, 1992). “stereotype threat” (Spencer, Steele, & Quinn, 1999;
These gender differences in career selection are most Steele, 1997; Steele & Aronson, 1995). When members
pronouncedinthefieldsofmath,engineering,andthe of a stigmatized group find themselves in a situation
physical sciences, in which women occupy only 10% of wherenegative stereotypes provide a plausible explana-
the jobs and earn just 75% of what their male counter- tionfortheirbehavior,stereotypethreatistheunsettling
parts earn (Crocker, Major, & Steele, 1998; Hewitt & awareness that they risk being judged by, or treated in
Seymour, 1991). terms of, those negative stereotypes (Steele, 1997).
Women in traditionally masculine fields must con- Whereas susceptibility to stereotype threat requires the
-
tend with the added burden that accompanies cultural stigmatized to have specific knowledge of negative ste
stereotypesallegingasex-basedinability.Culturalstereo- reotypesandthedomainstowhichtheyapply,thissitua-
typesareprevailingattitudesabouttheattributesofstig- tional predicament does not require the stigmatized to
matizedgroupsthatareconveyedthroughsocietal-level have any internal doubts about their ability, or their
messages and images, such as those broadcast by our group’s ability, in those domains. In fact, the effects of
mass media (Ashmore, Del Boca, & Wohlers, 1986). stereotype threat may be most acutely felt by those indi-
Despitewidespreadconcerninoursocietyoverthecon- viduals who are invested and skilled in the targeted do-
sequencesofbeingraisedonasteadydietofculturalste- main,orbythoseindividualswhoatleastcareaboutthe
reotypes, there has been remarkably little experimental social consequences of being judged incompetent in
researchconductedontheeffectsofexposuretogender- that domain(Aronson,Quinn,&Spencer,1998;Steele,
- 1997).
stereotypicmediaimages.Numerouscorrelationalstud If this situational predicament is the root of
ies, however, have documented the consequences of -
heavy television consumption. For example, Gerbner, underachievement in targeted domains, then eliminat
ing the situational factors that give rise to stereotype
-
Gross, Morgan, and Signorielli (1993) found that indi threatshouldenablestigmatizedindividualstoperform
vidualswhowatchagreatdealoftelevision,comparedto to their full potential. Steele and Aronson (1995) tested
light viewers, believe that women have limited abilities this notion by varying the degree of stereotype threat
andinterests compared to men. These researchers also that African American students experienced during a
found that heavy viewers expressed more stereotypic difficultverbaltest.Aspredicted,whenparticipantswere
views regarding what professions are best suited for told the test was nondiagnostic of intellectual ability,
-
women.Evenmoreincriminating,anaturalistic experi -
whichmakestheintellectual-inferiority stereotype irrel
ment found that after television was introduced to a evanttotheexam,BlackandWhitestudentsperformed
small community,thechildren’sattitudes became more equallywellonthetest.Whenparticipantsweretoldthe
Downloaded from http://psp.sagepub.com at UNIV OF WATERLOO on May 20, 2010
Davies et al. / CONSUMING IMAGES 1617
testwasdiagnosticofintellectualability,however,thetra of anxiety as a mediator. Relying on self-reported mea
- -
ditional gap in performance between Black and White sures of anxiety also may have contributed to the
students resurfaced. The researchers then examined inconsistencyofpreviousmediationalfindings—partici
-
whether priming African American students to think pants may not always be willing or able to accurately
about their race could expose them to the detrimental report their own level of anxiety. Blascovich, Spencer,
effects of stereotypethreatevenonanondiagnostictest. Quinn,andSteele(2001)testedthisnotionbycollecting
Steele and Aronson(1995)confirmedthatmakingrace bothself-report andphysiologicalindicatorsofanxiety
salient, which elicits the threat of being judged in terms (i.e., blood pressure) while Black and White partici
-
ofracialstereotypes,underminedtheAfricanAmerican pants took a difficult verbal test described as being
students’ performance even on the nondiagnostic ver- eitherculturallyunbiasedordiagnosticofability.Repli-
sion of the test. cating previous stereotype-threat findings, the African
Becauseculturalstereotypesquestionwomen’sability American participants only underperformed on the
to succeed in any traditionally masculine domain, diagnostic test. According to the self-report measures,
womenare susceptible to stereotype threat in all fields their under-performance was not accompanied by an
considered inherently masculine. Again, if this situa
- increaseinanxiety;however,thephysiologicalmeasures
tional predicament is the basis for women’s perfor
- confirmed that their underperformance was actually
mance deficits, then removing stereotype threat from accompaniedbyasignificantincreaseinbloodpressure.
traditionally masculine domains should eliminate Therefore, both the timing and sensitivity of measures
women’s underachievement. Spencer et al. (1999) will be critical in determining whether anxiety is a reli
-
testedthishypothesisbymanipulatingthelevelofstereo- able mediator of stereotype-threat effects.
type threat that their female participants experienced Thepresentresearchwilltestanotherpotentialmedi-
whiletakingadifficultmathtest.Halfoftheparticipants ator of stereotype-threat effects, one which Steele and
were told that the math test had revealed “no-gender- Aronson (1995) exposed when they discovered that
differences”inthepast,whichmakesthemath-inferiority AfricanAmericansfacingadiagnostictestactivatedself-
stereotype irrelevant to the exam, whereas participants relevantstereotypes.SteeleandAronsondidnotadmin-
inthecontrolconditionwereprovidedwithnoinforma- ister the abovetest; therefore,afullmediationalanalysis
tion about the relative past performance of men and was not possible. Nevertheless, we believe that suscepti-
women. Not surprisingly, the well-documented gender bility to stereotype threat involves confronting activated
gapinmathperformanceemergedamongparticipants self-relevant stereotypes and knowing that one risks
in the control condition. When stereotype threat was being personally reduced to those stereotypes in tar-
eliminatedfromthetestingsituationviatheno-gender-dif- geteddomains.Thus,activationlevelofself-relevantste-
ference instructions, however, women and men per- reotypesshouldreliablymediatetheeffectsofstereotype
formed equally well on the difficult math test. threat. Testing this particular mediational pathway also
allowsustoclarifythedifferencesbetweenthestereotype-
MEDIATORS OF threat paradigm and the ideomotor paradigm.
STEREOTYPE-THREAT EFFECTS Ideomotor theory suggests that primed stereotypes can
Stereotypethreatisamultifacetedsituationalpredic- automatically evoke stereotype-consistent behavior,
ament whose mediational path can be shaped by fea- unbeknownsttotheindividual displaying that behavior
tures of the person, the context, and their interaction (e.g., Bargh, Chen, & Burrows, 1996; Dijksterhuis et al.,
(seeSteele,Spencer,&Aronson,inpress).Despitethese 1998). For example, Bargh and his colleagues (1996)
complexities, considerable progress has been made showed that undergraduates subtly primed with an
investigating potential mediators of stereotype-threat elderlystereotypewalkedmoreslowlytoanelevatorafter
effects (e.g., effort, expectancies, biological processes, leaving the laboratory. According to the ideomotor per-
stereotype activation, evaluation apprehension, anxiety, spective, because behavior stems mechanically and
-
etc.). Of these potential mediators, anxiety has arguably unconsciouslyfromprimedstereotypes,whetherstereo
-
receivedthemostattention,butreliableevidenceforthe typesareself-relevanttothetargetisimmaterial—alltar
mediational role of anxiety has been elusive (e.g., gets cognizant of the stereotypes are equally susceptible
-
Leyens, Désert, Croizet, & Darcis, 2000; Osborne, 2001; toideomotoreffects(seeWheeler&Petty,2001).Incon
Spenceretal.,1999;Stangor,Carr,&Kiang,1998;Stone, trast, our stereotype-threat paradigm requires targets to
-
Lynch,Sjomeling,&Darley,1999).Ofcourse,anxietyis confront activated self-relevant stereotypes and to real
-
only a potential mediator of stereotype-threat effects izethattheycouldbepersonallyreducedtothosestereo
-
- types in a given situation. Therefore, stereotype rele
whenthetargetisactuallyexperiencing,oratleastcon
- vance should moderate whether level of stereotype
templating, a threatening domain. Researchers over
lookingthisfactmayhaveunderestimatedthereliability activation mediates our stereotype-threat effects.
Downloaded from http://psp.sagepub.com at UNIV OF WATERLOO on May 20, 2010
no reviews yet
Please Login to review.
