ving, and Discrimination Austin & $, Worchel (Eds, ‘ooks/Cole, Ed), Cognitive processes Scovering the social group station for groups and ina. reas and intelligence info. ‘meeting ofthe Midwestern >cial categorization. British social perception and inter. cognition and intergroup {of persons and! groups sad Paychology, 20, 445-465, the ingroup: The interplay 257-295, ‘A subjective essontalisic (Eds), The psychology of >ution: On translating sit in, 24, 1089-1103, 1 O The Unbearable Accuracy of Stereotypes Lee Jussim, Thomas R. Cain, Jarret T. Crawford Rutgers University-New Brunswick Kent Harber Rutgers University-Newark Florette Cohen Rutgers University-New Brunswick ‘sixty years of empirical research has taught us much about stereotypes. Stereotypes can arise from, ‘and sustain, intergroup hostility, They are sometimes linked to prejudices based on race, religion, ‘gender, sexual orientation, nationality, and just about any other social category. They can serve to ‘maintain and justify hegemonic and exploitative hierarchies of power and status. They can corrupt {interpersonal relations, warp public policy, and play a role in the worst social abuses, such as mass murder and genocide, For all these reasons, social scientists —and especially social psychologists— have understandably approached stereotypes as a kind of social toxin. Perhaps equally understandable, but scientifically untenable, is the corresponding belief that because stereotypes contribute to these many malignant outcomes, that they must also be—in the ‘main—inaccurate. The tacit equation is, if stereotypes are associated with social wrongs, they must be factually wrong. However, the accuracy of stereotypes is an empirical question, not an ideologi- ‘al one. For those of us who care deeply about stereotypes, prejudice, and social harmony, getting to the truth of these collective cognitions should guide inquiry about them. Unfortunately, this has not always been our experience. Because of his inquiries into stereotype accuracy, the first author has been accused by prominent social psychologists of purveying “non- sense," of living “in a world where stereotypes are all accurate and no one ever relies on them any- way.” of calling for research with titles like “Are Jews really cheap?” and “Are Blacks really lazy?” of disagreeing with civil rights laws, and of providing intellectual cover for bigots.’ ‘These reactions are understandable, if one remembers that social psychology has a long intel- lectual history of emphasizing the role of error and bias in social perception, and nowhere has this emphasis been stronger than in the area of stereotypes. To enter this zeitgeist and to argue for the need to take seriously the possibility that sometimes, some aspects of some stereotypes may have some degree of accuracy, therefore, is to risk making claims that are unbearable fo some social sci- entists. However, science is about validity, not “bearability” Itis about logic and evidence. In this chapter we review conceptual issues and empirical evidence regarding the accuracy of ‘Mereotypes. By doing so we hope to correct some long-held beliefs about stereotypes, and to thereby remove some of the obstacles tothe systematic investigation of stereotype accuracy and inaccuracy. The chapter has three main objectives: providing a logically coherent, defeasible, and practical definition of “stereotype”; reviewing empirical research on stereotype accuracy; and considering the role of stereotypes in increasing or reducing accuracy in person perception. 199200 Handbook of Prejudice, stereotyping, and Discrimination ‘ARE STEREOTYPES INACCURATE BY DEFINITION? Ginen the frequency with which stereotypes are assumed to be inaccurate both in popular culture nt the socal scientific literature, the fist order of busines is defining stereotype. The accuracy issue becomes “settled” if stereotypes are defined as inaccurate. Ia th ‘more agnostic approach is needed, ‘We begin by seriously considering the consequences of have so many researchers before us. When researchers define stereotypes ae inherently inaceurate, they Name thet stereotypes are inaccurate, there are only two logical posabiltes regarding what {Rey might mean: 0) All beliefs about groups are stereotypes and all wre inaceas ta or (b) mot at beliefs about groups are stereotypes, but stereotypes are the subset ofall belicns about groups that ‘ie naccurate, We next consider the implications of each ofthese possibilities his section we explain way g defining stereotypes as inaccurate, 2s ‘Aut Brutts Asour Grours CaNnor Possiaty BE INaccuRatt No social scientist has ever ex; ‘Thus, toppling the assertion that all stereotypes are inaccurate might appear to be refuting g Father than won, Unfortunately, however, this straw assertion, ven if itis merely an implicit rather than explicit assertion, appears to have an ardent sciertsfe following. For decades, ste Iman’ were Predominantly defined as inaccurate, with virtually no evidence demonstrating SE ITREY (¢8, Allport, 1954/1979; Aronson, 1999; Campbell, 1967, Schulte & Oskamp, 2000; ca, 1981; Brigham, 1971), Furthermore, among those who shine stereotypes a5 inaccurate, statements regarding what sort of behers about groups are 5), almost never appear (for concrete examples, , 1999; Campbell, 1967; Devine, 1995; Jones, 1986; Schulte & Oskamp, 2000; Allport, 1954/1979, remains a lone exception). an empty set. Furthermore, in their empirical studies, Accurate beliefs about groups, therefore, appeared to be the social sciences have considered people's beliefs about almost any attribute (personality, behavior, attitudes, criminality, competence, demograph- ics) regarding almost any type of group (in adltion to race, sex, class, ‘occupation, dorm residenc: POO ree ett, college major, and many more) to be a stereotype (ee ep, reviene by AP? 1991; Fiske & Neuberg, 1990; Kunda & Thagard, 1996; or the ‘meta-analyses reviewed in Jussim, Laer, Cravford, Cain, & Cohen, 2005). It seems, then, that forall praciel Purposes, the social spn aide any and all claims and beliefs about groups tobe stereotypes, Putting these poins together: Stereotypes are defined (by some) as ineceuate, A beliefs about aay case atentypes. Therefore, regacles of whether any escarcher has explicitly made thi claim, ino, Derspective assuming that all beliefs about groups are stereotypes, and defining stereotypes as inaccurate is logically compelled to reach the conclusion that all belie. and logical coherence is a minimum condition for consider. ‘ing a belief to be scienifc. On logical grounds, alone, implying that all beliefs about groups are inaccurate. Many researchers donot hold such an extreme view. Next, therefore, we consider the less extreme Pal pe ony logical alternative (if one defines stereotypes as inavcurst) to claiming that all ate is the following: Not all beliefs about groups are inaccurate, bt >eliefS about groups that are inaccurate, therefore, we can reject any claim stating or“The Unbearable Accuracy of Stereotypes 201 tp Srencorvees Ane THe Susser oF Bautrs Anour Grours THAT ‘Ave INaccugare, Tueke Is No “Sterrorvre” REstARcH then only inaccurate beliefs about groups can be con- s, then, must constitute a different phenomenon s those who subscribe to this view stick to their {fall stereotypes are inaccurate by definition, fidered stereotypes. Accurate beliefs about group together. This is nota logical problem as long definition and live with its implications, ip Srencorvees Ane Drrinen As INaccurate Baers Agour Grours THEN COniy Eapreicatiy Ivauubareo Beuicrs CoNstirute STEREOTYPES Accurate beliefs about groups are not stereotypes. Beliefs about sroups of unkown validity cannot te tnown tobe stereotypes. Tis perspective has a major drawback: It invalidates nearly all exist- ing research on “stereotypes” This i because so file ofthe sereotype research has assessed the sae ey ofthe beliefs under investigation, Without suc an assessment, peliefs cannot be known to ae Mereotypes, No research on “stereotypes” has ever been framed as follows tuthis bei abot that group a teretype? We are gong to igure out wheter his belie show et we a aercoype by asrssing wheter bat bebe s nacerae If his ele nace Wy will Se er ive weretype If his bei accurately described that group, we will contode that isnot a stereotype. “This, however, is precisely how research must be framed before one can Know one stadying a ste- eine f stereotypes are the subset of beliefs about groups thet are inaccurate Wf he sms ofa weocic belie being researched has nt been rst determined, ten tis impossivie toknow whether eelisier is a stereotype. The nature, causes, and effects of beliefs of unknown accuracy can8ot craribute to knowledge of stereotypes if only inaccurate beliefs are stereotypes olin social psychology to this restrictive definition woulkdmesn discarding decades ofresearch purportedly addressing stereotypes. Why? Because almost none of i Bas empirically established Purported as about groups being studied ae in fac erroneous. There would be nothing i A saat of the role of “stereotypes” in expectancy eects, self-fulfilling prophecies, pers Pott vend so on. Poof! We would have to throw out the baby, the bathwater, the tion, subtyping, memory, al house in which all our tub the bathroom, and indeed tear down the entite scientific and empiri ‘current understanding of “stereotypes” exists ra ere those researchers who define stereotypes as inaccurate, oF even emphasize CP inaccuracy, must provide clear answers t0 each of the following definitional aves Do they aacerer cl beliefs about groups to be stereotypes? Do they define all beliefs about gross AF arate? Or do they define stereotypes asthe subset of beliefs about groups thet are inaccurate? If the Iaten hove do they distinguish between accurate beliefs about groups that are not stercoRYPeS and inaccurate beliefs about groups that are stereotypes? A Neutrat DeviNITion oF STEREOTYPE Fortunately, many modern definitions of stereotypes donot define stereotypes as ini ntly inac- cote and are tustead agnostic in terms of stereotype accuracy. One of the simplest of these defini- Cae ae ge ue throughout this chapter, was provided by Ashmore and Del Boos (S)) on one eeotype is set of beliefs about the personal atribates ofa social group” 3) Starsonypes, as defined by Ashmore and Del Boca, may or may not be accursle an rational, Wwdely shared, conscious, rigid, exaggerations of group differences, positive os nest or based Wiel ea te eal rationales. Stereotypes may or may not be the eause or the eect of ‘rejadice, or the cause of biases and self-fulfilling prophecies.202 Handbook of Prejudice, Stereotyping, and Discrimination The Unt Itis good that Ashmore and Del Boce's definition does not specify these things—that leaves Some P these aspects of stereotypes open tothe kinds of empirical investigation this topic deserves. One ot the great values of trly believing in the neutral definition is that it does not presume that anytime We next person holds or uses a stereotype, something inherently bad (or good) is happening, Instead, it openg ae the door for understanding when stereotypes wreak damage, when they simply reflect social realiy, undoutt and, possibly, when they actually perform a social good. unjust Our rejection of defining stereotypes as inaccurate is not equivalent to defining them as accurate, ae Accuracy is an empirical issue, which naturally raises a question: How Gn)accurate are peoples oe beliefs about groups? eves ha secure yan, 2 ‘ARE STEREOTYPES EMPIRICALLY INACCURATE? ‘Tit This section reviews empirical investigations of stereotype accuracy It includes a discussion of sifler ie an important level of analysis issue with respect to understanding stereotype (i)accuracy, bret eomat review of common methods for assessing stereotype accuracy (and thei limitations), anda discus- cole sion of the complexities and richness involved in assessing accuracy. After presenting an overview a of those conceptual issues, ths section then reviews the research that has assessed the accuracy of ieee people’ stereotypes sities Srertorvre ACCURACY AND LevEts OF ANALYSIS bia ‘The following statement summarizes a class of criticisms of stereotype accuracy that has period Tress « cally appeared in the social psychological literature (eg, APA, 1991; Fiske, 1998; Nelsan, 2002; Stereo Schneider, 2004; Stangor, 1995): mae Event canbe successfilly shown that percevers accurately judge two groups to differ on some at ei bute: (@ Perceivers should not assusse tht thir stereotypes of the group sutomtically it all members Be of the group (b) perceivers cannot apply their beliefs about the group when judging Individuals and oad (@ifpecchery do apnly Gel bale abn the reve whenfugay tions ey a ay be Res ‘wrong much of the time because few members perfectly fit the stereotype. ash According to this type of analysis, all stereotypes are already known to be largely inaccurate so cone there is no need to assess their accuracy, cate p ‘There is merit to these points. Few, if any, members ofa group are fully defined by stereotypes Dis Assessments of any individual based solely on stereotypes will generally be lacking, However, tis and pc logic implies nothing abou stereotype accuracy, Instead itis aclaim about the accuracy of ering them stereotypes of groups to specific group members. them retype accuracy issues occu, therefore, at two different levels of analysis, each captured by vome a different question, First, how accurate are people's beliefs abant groups? Jast as a person might about not accurately remember how many games Roger Clemens won in 2000 Gnacouracy in person perception) and still remember that the Yankees won the World Series that year (accurate bli WH about Clemens's group), inappropriate application ofa stereotype doesnot mean thatthe streotyPe ae is itself inaccurate. A person may correctly know that, on average, women earn about TO% of what men car, but have no accurate knowledge whatsoever about how much Nancy earns. Thee Second, does people's use or disuse of stereotypes in judging individuals increase or reduce Me | aby accuracy with which they perceive differences Between small groups of individuals with whom | et) they have personally come into contact? This isthe accuracy version ofthe “stereotypes and persot bet perception” question. Do, for example, general stereotypes of male superiority in athletics lead th tet Coach of a soccer team to erroneously view the particular boys on the team as beter than the Pat Xo ticalar girls on the team, when they rally have equal skill? : fois‘the Unbearable Accuracy of Stereotypes 203 Some PReuMunary CAVEATS We next briefly summarize some key points made in previous research on accuracy that we need to draw on here, although space considerations precinde an extended discussion. First, stereotypes ‘tndoubtedly sometimes lead to errors, biases, self-fulfilling prophecies, and a variety of unfair and ‘unjustified outcomes. The research on these topics, however, typically has provided litle informa tion about their accuracy (Funder, 1987, 1995; Jussim, 1991, 2005). ‘Second, methodological difficulties once plagued accuracy research. Those difficulties, how- ever, have been resolved by statistical, methodological, and conceptual advances within the field of faccuracy research over the last 20 years (e.,, Funder, 1987, 1995; Jussim, 1991, 2005; Kemny, 1994; Ryan, 2002), Accuracy is now a thriving area of research within social psychology. ‘Third, this chapter does not address the issue of how group differences originate. Why groups differs 2 fundamentally different scientific question than whether people perceive those differences accurately. Whether group differences result from genetics, childhood environment, socialization, coltute, o roles is beyond the scope of this chapter. Fourth, the genesis of stereotypes is also irrelevant with respect to evaluating their accuracy. A beliefs accuracy must be assessed on its merits, not on its sources. Assessing accuracy of beliefs is ‘different endeavor than assessing processes leading to those beliefs (Jussim, 2005). DIFFERENT ASPECTS OF STEREOTYPE (IN)ACCURACY “Twres oF SteRcoryPe ACCURACY ‘stereotype accuracy has been commonly assessed in either of two ways in the scientific literature Discrepancy scores assess how close to perfection people's beliefs come. The stereotype belief (eg, “pow tall [rich, aggressive, etc] is the average woman in the United States") is compared to criteria (egs the average height [wealth, aggressiveness, etc] of the average woman). The difference indi cates how far people are from perfection. Smaller discrepancies equal greater accuracy. Research on stereotype accuracy has also frequently used correlations to assess how well peo~ ple’s beliefs about groups correspond to what those groups are like. Stereotype beliefs are correlated with criteria (eg, people's ratings of women’s average height, wealth, and aggressiveness, could be correlated with criteria for women's height, wealth, and aggressiveness). Higher correlations indi- cate greater correspondence of the stereotype with criteria—that is, higher accuracy. Discrepancy scores and correlations have been used to assess two types of stereotypes: cultural ‘and personal stereotypes. Cultural stereotypes refer to the extent to which a stereotype is shared by the members of a culture, or a particular sample, and are usually assessed by sample means (@.g, the mean belief about women’s height in a sample is the best estimate of the cultural stereotype for ‘women’s height for the group sampled). Personal stereotypes are simply any individual's beliefs about a group, regardless of whether that belief is shared by others, WHAT IS A REASONABLE STANDARD FOR CHARACTERIZING ‘A STEREOTYPIC BELIEF AS “ACCURATE”? ‘There is no objective gold standard with which to answer this question. Perfect or near perfect accu- racy i reserved for a very select set of endeavors (¢.. moon landings, measuring atomic weights, tc). Even when social scientists generate hypotheses that predict differences on some outcome between groups (whether experimental or demographic), they do not require a correlation of 1.0 between group and outcome to consider their hypotheses confirmed. Indeed, social psychologists are often quite satisfied with correlations of .2 oF less (Richard, Bond, & Stokes-Zoota, 2003). So, the issue is, what is a reasonable standard for lay accuracy in stereotypes? Because there are (Wopat Handbook of Prejudice, Stereotyping, and Discrimination, ‘broad types of accuracy, discrepancy from perfection and correspondence with real differences, there must be two separate standards, Bach is discussed next. Discrerancies The Bull’s-Eye iAtul-eyeis as god at gots target practice, Bulls-yes are not microscope geometric pon ‘They asvally have perceptibie width, which means one can legitimately hit a bul'-eye withog! being Robin Hood, who could hit the target dead center, then split his own arrow on the next sho, ur standard is that, for the type of social perceptual phenomena usually stadied by social pry: hologists, a bull's-eye is within 10% of dead centr, There is nothing magic about 10%. Reasonable people may disagree about this standard, and i might no be always appropriate, but when jdgiag 4 proportions and probabilities, as is common in the study of stereotype accuracy, geting within 10%, is doing pretty well : ‘Some studies, however, do not report their results as percentages. Most that do not, howevey report thei results as effet sizes or can be readily translated into effet sizes—real and persed aTiferences between groups in standard deviation units, which can be translated into percentage If Kay perceives Group A as 25 SD higher on some atribote than Group B, this means tht Key perceives the sverage person in Group A (score higher on that variable thn 605 of the pope Grovp B. Bingo! That isa 10% difference (we assume a normal distribution, so the average pein | fn Group B scores athe SOth percentile 0 the difference is 10%), Therefore, for sedis astsing stereotype accuracy using effect sizes, we characterize a perceived difference as accurate i ti within 25 SD ofthe real difference. 4 (Our atandards often donot correspond to those used by the original authors, MeCauley’sresear (gee Tables 10.1 and 102) often used “less than 10% off” as his eiteria for accracy: we deb a single percent, because we characterize 10% off as accurate, Others used statistical significant «thes Mander’ (ete the perceived diference statistically exceeded or underestimated the ral difference, they concluded it was not accurate), Although these standards have their own advanlGe) fand disadvantages, discussing them is beyond the scope of this chapter. Near Misses : nccuacy is matter of degree—it is at allo none Jusim, 2005). Therefor it des no eg aca a coeterze a belief hat is 10% off as “accurate” and one thats 10.1% of a “I seareee So how dhould we characterize near misses? Ax near misses. near mss i ot 200% tar tis not too far oft Continuing with the archery metephor, one ean stil rack up some Point it an hits the target, even if one does not hit che bul'seye; not as many points as when one HS the bults-eye, but more than if one misses the target completely. Wick than, is veasonable standard fr a near miss? We use more than 10% of st ne than 20% off, Within 20% is certainly not a bull’s-eye, but it is not completely out of touch “a reality either tis certainly far more accurate, sa, than being 40% off ar more, Here (0% poo can disagree about what reasonably constitutes a near miss Following the same rationale as for accuracy, when results are only reported insta tions, we ase “more than. 25 SD disorepancy (between belief and criterion) bat no more te Ufeenopany” a our standard for ner misses. I To's bli is that Group As mean 8S ight than Grong B's mean, when there really sno difference, he erroneoasly believes the (PS oan Gao A roup exceeds the sors of about 70% of the members of Group B, when in 82+ ial ‘exceeds the scores of 50% of Group B; again, this is a 20% difference? nara devi 5080) unbearable “ypes of Disre she literature bi ant sere or sale diffe ey? THESE 1 fanhasized io * ‘These ciserePe ale atibuts, (ore intelier ‘on social esi ne another tha "There is 2 § aac” but Independent © height, some some ait and 66. in in, respective thn eects s inbeight. Consesrono! How much « Nonetheless, ‘sentists ho! 1.Cohen 1 ofthe effect: suggested th ‘correlation correlations) they represe ‘This sta sizes found ‘analyses that mean al, 2003). ( forthe pher obiains eff A strong st correlatic means the, ‘uot for « ‘Conresro. Moderate and inace sharacter tecurate,unbearable Accuracy of Stereotypes es of Discrepancies perature has focused on 1o broad types of discrepancies, By fa, the most interesting and ant discrepancy involves perceiving differences between groups. Do people pereive «larger impor ciffrence between groups than realy exists, or do they pereive the difference accu- spt These types of discrepancies directly test the exaggeration hypothesis that has been 80 long epasizod in the scolar literature on stereotypes, I is also important for practical reasons, ere scepancis, when they show that people exaggerste real diferences on socially desir Tre rsibuts, indicate whether people ujusiably perceive one group as “better” than another re intelligent, more athletic, etc). When they show that people underestimate real differences cially desirable atcibtes they indicate that people unjustiialy see groups as more similar to ‘one another than they really are. "There is a second type of discrepancy reported in the literature that is still relevant as “inac~ uray” but has considerably less theoretical or practical importance with respect to stereotypes. fhdependent of perceiving how two (or more) groaps mutually differ on a given attribute righ, sometimes people have a general tendency to overestimate or underestimate the level of tome atzibute for all groups. For example, let's say men and women inthe United States average 72 nd 66 in. in height, respectively, Fred, however, believes that men and women average 74 and 68 in, respectively. He consistently overestimates height by 2 in. (his isa fairly meaningless “eleva- tion” effect; see, Jad & Park, 1993; Jussim, 2005), but he does not exaggerate sex differences Inheight CConresronpence Witt REAL Dirrenences: Hick Accuracy How moch cocespondeace shuld be considered “acura”? Again, this is jogment cal Renstlrs ne alvcat holding peopl fo high stodsré—the sme standards 0 which social fens ol tenses Cohen 983) im is assests ease imploring sos to examine he size edi ets they ainda ther dis and not juste “tats significance” ofthe esl, tease that eet sizes above 8 cond be consteed “age” Soch anf siz sles ino {eatin of tou nthe fran ofthat effect snes are discaied excel os tert. Bystanders colton of and higher could be considered aerate becase they represent a “large” correspondence between stereotype and reality. This tana has been sported by evo recent tes hat have examined the ypc effect sizes found in clinical and social psychological research. One recent review of more than 300 meta- ‘analyses—which included more than 25,000 studies and over 8 million human participants—found tht mea and ean effect size ln social pyeblogcal research were hth boat 2 (Richard et 41 200) Only 2% of social psyhological ets exceed 3. A smile ater hs bee found fre penomera tute by sia pelos (Hemphil, 2003), Payne esearch ey bain ee sizes exceeding corrals of. ict aes of and bigher, therefore, conse {seg nar fo accra) Lass aecording fo Roseuta’s (99) Binomial fet size displ Aeon ft aniston t 70 fei Ts eas te ate ih mare tha easton ase ae wrong, That seams Hike a appropiate Sof forconsdeiyastreatypresonbly accuses a Cortesronpence Wirt Ret DirrereNces: Moperare ACCURACY Motes conesgotzcs of une is hn igh corapondence Telets a mix asiey id imcoy Flloving th sae tarts as scene, Coon, 188 Rich tl, 209, we shreds btn pope ais and relly ranging rm 25 4 ar rey se Sich eorclatons dont eet pret acute, nord hy eet compete natTABLE 10.1 Are Rat Study and Stereotype I/Ethnic Stereotypes Inaccurate? Perceivers and Criteria Personal Stereotype Accuracy Discrepancies Consensual Stereotype Accuracy Discrepancies Personal Stereotype Accuracy Correlations Consensual Stereotype Accuracy Correlations yan (1995 bli aboat Sieences inthe peso ‘patel alse Teligene. ec) of fen ‘Amencan an Wate Univesiy i Ceerdo sadets Perceive: Random samples fs Atom Arerian Sed 30 White Univer of loro sens ‘Citra: Seep rom the same samples ‘Notable ‘Aficn Anerens tli abont ‘am Ameren 55.7 ‘Whe Ameren: 5, 5,7 Diteeness 5.5.7 (Grom eed exaggerate rel dieeces) ‘ite American” bef boat: ‘Afncan American 55,7 ‘Wine American 5,2, Dison 9.4 4 (hema ptecnof exaggeration or White preevers: 36° ‘ica Asian resvere 73.53,7 ‘eee: 76 ‘NeCaay end SH (197 line aot demographic {itereses (ih schon dee, nvelfee 2) etree Aican Amiens an ‘tis Aresiens ‘acven: Fv baghacard ‘apes (bach eo, ‘ho meer, Sader, enna oe Citi: US. Cans dat "Waite Ameria ages 17,13,5 ‘tiara ages 17, 144 Dares between eso: Aercans nd ‘er Aaa 27, 8.0, (tt discrepancies understated the el ‘terenes). ‘Net alable ‘iranian: £0 ‘Armin 99 Dilereces ‘etweon Aen ‘Ansan and oer ‘Ameri: 58 | | “asics ad ses (0990); belts out be mcioveent of ne Caan fhe groups Posies 94 aves of Citi: Toone Boat of ‘Accra: 36 of 4 ‘Beaggeraton 33 of 94 “Underestinstor: Wo valable @ Nowavable “Wot, Pr, a an “ye (200) is on diflrenes pee ‘non Aeris ana White bts (work eis incigene, cml te) a Wie Unb of ‘Color undegnates ei: US. Cones data Sn te repens Teco ding toon abuts ‘de ses adeng meister rues. Perceive rerestmated eountertemtypieal ‘iaotes ore thm serene Sts, whieh aus ey destined how seeoeypieal each fonp was, Tey also underestimated race ‘terences, boty ies than 10%, —— ‘Not aniable ‘Not alle 90z uopeujwy9s1q pue BuidArooraig 'aorpniaid Jo yooaPUEH Tics Bois wambors represent eure jdemens, Hal percentage oF within 25 Sheree eal fferences, o were inte completly aca wre eenorted. See text for explanation of personal and consensual stereotype accuracy & 'SD. Fer dserepancy scores lead bcs Terese nace, and gue epetace morabers epee eee ements nda the numberof eros tat exaggerated rel dflerencss, ot reported and that we were waabe to compat i from is refer to her sterotycaliy veromg dection (versal) Not aveiable means both that twa Gscrepancies and cmelatons Ryan's (1996) ea a near aes. Accuracy sans Within 10% of thera qeseaqun autope "Noe Bald ambos repesen accurate Jodgmen "alcand mame Terese nacewracies, and Tague Wpetice numbers repreent oar mises, ASEOaEY mens within 15 of the sel ance or mn 23 SD For dncrepacy scores that were ot mses an nacuracies,prettia ssteea inate te number of eos tha exaggerate el dieses, aera sel diferenes, or were ia te completely wrong drestion (ever). Not avaiable means Doth tha twas not reported end that we were unable to compute it from ts tat were reported. Se ext for explanation of personal and consensal scope accaecy discrepancies and cmelaions. Ryans (1996) rests refer wher tereotyplelty resus, not ee dispersion real. Fee simple, ifthe stadyzepned rare than one individ! level (average carlton, we sip averaged al cr comcaions roe o give ax overall ene of the degree of accuracy, ‘These coocations donot appear the orignal arte, bat are computable fom dta that were reported. Recah pup of pocelves the fet conan i the conespoodence betwen hi judgment andthe seeps of thir own gongs; he send conetaton ste corespondence femmes tt figment un the set ops ofthe etx group, adh thir colton isthe cmespndence between the pected dense between he groups andthe diferene ia the se-reports af the ro groups. recon aa Hass (1999) computed a personal scepuncy score for each perceiver and then pated ibe numberof percevers who were within 2 SD of teeter the number that xa seta ea ciftrcces aw acifeense rene han 2D leech thee ieenc) or unrest eal direc (a iferace mot thn 2D salle tan hea i eae) Ashon and Ese (1995) examined bells boa sine diferent Canadian ethic groups; and dcepany results ero the number of pacipans showing cach pater, The ro seer Ae cfer to ber Tule 2, wtch repos the mmbe of preherswiin.2 of he rea diferece They i not eport ress om which te namber of eeies wis. 25 ‘SD ofthe el tference could be dented. saddqoosoig yo Aovunooy ajqeieaqun SUL oz“TABLE 10. ‘Are Gender Stereotypes Inaccurate? Personal Stereotype Personal Stereotype Consensual Stereotype Accuracy Consensual Stereotype Study and Stereotype Perceiversand Criteria __Accuracy Discrepancies _ Accuracy Discrepancies __Correlations __Accuracy Correa [ecasie Thapsel, and Roun (198), [Psi Clee ers, gh [Nowa T6286 understate 5 (all [NatAvalale 5-98" | Mecaoley aod Tampeea (1991 betas | schon! stfet all commates drei sre ewes | tovtteont Staton ine dies | (= S21 were ferme) Sim COD let seat nex aitanamon [Peter Indic rama 1,7 Gandcesiatin, [Nevado Sty 78 Vetces [xr ‘eh 7 Seen? sey 279 citer: Meramaaes ex . Hawa a (05) eatearsee —— [Peeaves: ti neoacany | Novwralabe Feoaeyecshewe9.0,8(5 |NovAviible | Femle paca 70 ieacs noted bin relay eat seaeratons anderson Male perceives: 6b eres prea eave ae pce 11,603 ndretinaons, sagas, 2 renal 0 and Cane (tlle abotaee | ertvrs 708 Insts Rees w 7 ‘Sineocs oT chntries Poenbg sce tesa Mealy of ex "la ani Eagly 900), berate eceners ADT Reema Nac dominatt oxapatons——|Novavlatie [57 |" tacn nn sere cxpasons poetnogy salen icone cic US, Conse Female domint cszaons Z Nears (gdeeston? I et Dinan Enya Kaa GHG) bles | Pemsven 617 elepratiens [Near neomeey a eae aa oe ‘ouike attics ot menaed woven’ | overdisee nies oa ergs, overs aide ty Fone tpic [Ronle gee 7 “vt opel septely Ys jain se | When ging sx eee: 80 esa Asma otc as | foreach) ‘athens ncetd athe Geer Seca ‘Sotvey, web ea reng 07 Luopeujunosiq pue SujdAvoasas ‘2>Ipnaid Jo yooqpueyy jer BH sir anti x ib oo eet jor and ean rd pot eag of in ed women se majors [Noe os oie sues theresa rie oak rege of ening stents Mae esebere 65 (Sentereinaons 2 ee (Uundrsination Pema perv: 7.4(1 cndrestion, | epg, reveals (J anderestnaen) Femle poser: 79 Peale pete: enlarges. 34 rete 22 Feasts | looy ajqezeaqun ULoui: Acca iat | fre i) aces 2 “indicated in the General Social me Sane ich in areing | . aes poe |e (807 bet aati ex ata [Pees vane aj a ‘nm difacal majors anda grade pit [25 cllege alent aie pecehere 6.5 Nae prev. ale peer: $0 sergeofmensnd woman in thom mor | (Guderesinations, eves, [Renae peeves [mle paces 9 cute: Caoge reece | rctderentnatn) 2 sens ateting sans Female peer: 7.4(1 ase point erg |Male ree 3S snreimaton, leeggenico,? |Nale ges 22 |Femae rps 34 foveal) (endeetinton) [Pome re ~08 emai eccive1.7@2 vel 4 (dentin 3 ‘Note: Bol pumber represent accurate judgments, jalicied pumbers represent inaccuracies, and regular typeface numbers epreseat near misses. Accuracy means within 10% of the eal et ‘centage or within 25 SD. For diacrepancy sores that Were neat musss and inaccuracies, pureelcal statements indicate the qumbe of eros thal cxaggraed real differences, nt tentimated real ferences, or were i the completely wrong direction (reversals). curacy dat ze Separated aconting o perceiver proupsorargt groups only wha these tase ‘ported inthe orginal studies. "Not availabe therefore, mean bd tha it was ao reported an that We were unable comput item the data hat Were feprted, See text for expla- halon of personal and consensual stereotype accuracy discrepancies and comelaons, + ‘hese sadies ae grouped togetierbacase thy Were So sala These corelations do ot appear in the eign arele, bat ae computable from data hat Were reported Swim (1994) sometimes reported more than one meta analysis at ucrtrion fora perceived dference. [a hat cate, we simply averged together the real ferences indicated by the mets tuulyseso have a singe crerionagalnst which o evaluate the accuracy of he peceved ference Cejka and Eagly (199) examined beliefs aboot 80 occupations, but dd not report esl separately for each occupation. Instead, they simply reprted overall discrepancies within male= ‘dominated ocopatons and sith female-domiated oocapatons. Thee was slight overall tendency to underestimate the sex diference i isbusion into ralo-Sominated occupations (6°39), whieh accurate by our suandads snd «somewhat stronger tendency to understinate the sex diferencia detibtin into female-documened occupations (17.9%), which goalies sa pear miss by our standards For Dickman et (2002) we have averaged several comeltons together to give an overall assesment of accaracy. Unfortunately, Dickman etal. id not report their ests seperately for ach ania, Instead, they reported results averaged overall atiudes. They de, however, report resus separately for men and women targets. Therefore, here Wee atl of six conseasual ereotype accuracy discrepancies reported studies by men and Women target Al six Were underestimates; tat i, tere Was a genrl tendency to underestimate how much people oppor the various postions, However, this tendency was nt very lage ll six consensual steresype isrepancy results underestimated suport by only 2% 10 8% For Beyer (1990), all sults rc reared separately for mea and wotnen perceiver, excep! the individual corelations fr grade point average (GPA), Becase thee was no significant sex of poteiverdiference ia these coelations, Beyer reported te results separately formal and female targets. GPA is ao given in percentages and Beyer did aot report stiadard deviations To ‘letomine whether result was accurate, Iheefoe, We eed the conservative clea of witia 10 grade pont for accutay, and within 20 grade pins for a ear miss. Thre was an overall tendency to inacuraely overestimate stents” GPA. This tbl, therefore only rept ess epading he accuracy ofthe pacavedaifrenes fa GPA (which most relevant wo sexo ing anyon). There are many reversals in Beyer's data because, although he stereotype is, appreny,that men perform beter in school han do women, women's GPAs were actually higher in every mor, ineiading the masculine ones (ath, chemist, et.) Voreujwnosig p sedkioora Jo Aovanoay ayqesvoqun 24 60z210 Handbook of Prejudice, Stereotyping, and Discriminaig ‘Using Rosenthal’ (1991) binomial effect size display, a correlation of 3 for example, means people are right almost cwo thirds ofthe time. Now, this also means they are wrong Title over third of the time, but two out of three isn't bas. Caveats AND CLARIFICATIONS systematic Errors T cocia seience research, “errors” ae usually random. Tn contrast, in stereotypes, even ithe rape bas considerable accuracy, a major source of concern isthe potentially nonrandom nae oe es Inthe preceding discussion on moderate accuracy, for example, a perceiver cul be Siam thirds ofthe time, Theres, ower, big difference ifthe one third worth of era ca eve tesos systematically biased against one gronp, Fortunately, discrepancy score analyse random Neconally designed to assess systematic errors (Judd & Park, 1993; Jussim, 2005), and we report such errors in our review. 4 ‘We Only Review Stereotype Accuracy Data ; “Many ofthe stadies reviewed herein dressed many issues other than accra), However, We iB ne gar discussion to those aspects that iavolve the accuracy of stereotypes. Other issues are eyo the scope ofthis chapter Differences in Terms None ofthe studies described in this chapter use our exact terminology of personal and consnsia stereotypes that can be evaluated using either discrepancies or correspondence with real differences eran often they simpy discuss “stcreotypes” Regardless, we do make that distinction and describe their results accordingly, regardless of whether they described their results this way. § Sie MGcarchers have distinguished between personal and consensual stereotypes, altos they generly use somewhat different terminology than we do, For example, consensual ee : tres ete sometimes disessed as “aggregated” rests of stereotypes (because they RETHNS arress all perotvers). Personal stereotypes are sometimes discussed as “individual serooyet and the Judd, Park, and Ryan group uses the phrase “within. subject sensitivity correlations” co refet to what we call personal stereotype correlations. ‘CarreRia FoR INCLUSION Tobe included here the empirical studies assessing the accuracy of sterentypes needed 0 Pet] roe ncteri, First, they had fo compare perceiver’ beliefs about one oF more tree Be er tes of what that group Wes actually like, Studies assessing social cognitive pronto with measores of wha at oe? idly presumed io be fawed an invalid, a not ni Fae pace snc tudes provide no diect information about accuracy (Funds, 1987 0) 2005). Trond, studies needed to use an appropriate target group. Sometimes, researchers hays ascesed | people's beliefs aboot a group, and used as citeria the characteristics of haphazar sample of Prembers of the target group (e.g. Alen, 1995; Martin, 1987). These studies have jmportant dif oer as tn stereotype they ae assessing andthe criteria they we. Consider fo: ee aaa pacelven provide thi bis regarding mea and woren in geet yi Petit bat haphazard sample of college stents (Aen 1988) 1 a geded prey wih ens women nee sare repo tthe characteristics ofthis erterion sample ifthe extern sample's ATE i toro cen and women in general. Consequen, such stades were not MEMES TI our review seunbearable ACCUraC accuracy OF ETH “publ 1.1 summarizes ‘hat met our eritet Pfrsensual stereotype d rents were either & Second, peopie’s cot renly more accarate th ips. For example, i fd regarding African thei judgments of dit 17 A similar pattern © ‘Third, these results Aifferences. Exagget®! ne exception. The ot of eorle wes accurat than to underestimate Fourth, the extent Consensual stereotyF correlations were sO° ACCURACY OF ¢ ‘eble 10.2 summatis son, Results are bs least a plurality of every study. Inaccu exaggerated real di generally leed peo} Anced exaggeration ‘Again, consens’ with most falling Once they were in female targets). Io (most corceations STRENGTHS A OF RACIAL, E° Several methodo alizabiity of the ‘graduates’ sterec McCauley & Th of accuracy ove: ‘ot represent se Nonetheless, ad ‘Second, the Education data ports, and so artifact resultDiscriminasy ‘ple, means alittle ovec cn | q even if the se vrandom natug ceiver could by score analyse 2005), and we.” ever, we cone res are beyend 16 consensual al differences istinetion end xis way es, although nisual stereo. ey aggregate stereotypes; fons” to refer ded to meet arget groups © processes, .ot included. 87; Jussim, awe assessed sample of portantdis- or example, and the cx this ease, they might acteristics included in she unbearable Accuracy of Stereotypes an ACCURACY OF ETHNIC AND RACIAL STEREOTYPES Je 101 summarizes the ut ofall studies asesing the accuracy of racial and te SEe> oat ou ertera for inclusion. We review the most noteworthy oftheir resuls hee, at gereoype discrepancies area mix of accurate and inaccurate belies. Nonetheless, most ces we io acct or ear mises, Only a minry ere mr than 20% off. iment peopl consensual teretype discrepancies for between group ¢ifeences ae cons vee acura than ate hr consensual steeotype dterepancis fr characterises WA Hy por example, i the Ryan (1956) study, Whites) consensual trots sen Whites som Ting Asean Americans each were accurate only 5 of 17 ines (10 of 34 oa, Hower and eno erences Deen Writes and ican Americans were cat 9 mes st of tr pattern ocurred inthe MeCavley and tit (1978) study ee Table 10.) a pene rests provide litle support fr the ia that stereotypes typically exaggere ot ‘curted, but it occurred no more often than did underestimation, with aierepton, The nly stay tases the accuracy of personal discrepancies found that lvally oe ten cc, ad that thre was alight greater tendency Yo exegparteren ferences re enleresimat real difernces (Ashton & Esse, 1999, summarized in Table 101) 2th te extent Which people's stereotypes corresponded with reality was strikingly High consensual stereotype accuracy correlations ranged from .53 to 93, Personal stereotype sesarat Constons were somewhat Lover, bt sil quite high by any standard, ranging from 36 wo 69 types hs cconsens differences. Exaggeration oct ACCURACY OF GENDER STEREOTYPES ‘able 102 summarizes the results of ll studies of gender stereotypes that met our criteria for incl Ton, Results are broadly consistent with those fo ethnic and racial stereotypes. In most cases, &8 Teast plurality of judgments was accurate, and accurate pls near miss judgments predomi 0 ery study. Inaccaracy constoted a minority of resol. Again, some results showed hat Pe Srenworted real differences, There was, however, no support forthe hypothesis char steroosyFeS generally lead people to exaggerate real differences. As with race, underestimstions ccounterbal- ‘anced exaggerations. "Again, consensual slereotype accuracy correlations were quite high, ranging from with most falling between .66 and .80. The results for personal stereotypes were more Onoe they were inaccurate, with a near-zro correlation with criteria (Beyer, 1999, perceptions of female targets). In general, though, they were at Teast moderately, and sometimes highly eccurste (most correlations ranged from 40-60; see Table 10.2) 4 10 98, variable. STRENGTHS AND WEAKNESSES OF RESEARCH ON THE ACCURACY OF RACIAL, ETHNIC, AND GENDER STEREOTYPES Several methodological aspects ofthese studies are worth noting because they bear on the gener- aliabiity ofthe results. Fest, although most of the studies only assessed the accuracy of under. itaitates’ stereotypes, several assessed the accuracy of samples of adults (McCauley & Stitt, 1978; MeCauiey & Thanguvely, 1991; McCauley, Thangavelu, & Rozin, 1988), Some ofthe highest levels of accuracy occurred with these adult samples, suggesting thatthe levels of accuracy obtained do rot represent some artifact resulting from the disproportionate study of undergraduate samples. Nonetheless, additional research onthe accuracy of noncollege samples is still need. Second, the studies used @ wide variety of criteria: US. Census data, self-reports, Board of Education dat, nationally representative surveys, locally representative surveys, U.S. government reports, and s0 on, The consistency of the results across studies, therefore, does not reflect some atifset resulting from use of any particular criteria.m2 Handbook of Prejudice, Stereotyping, and Discrimination "Third, the studies examined a wide range of stereotypes: beliefs about demographic character. istics (MeCauley & Stitt, 1978; Wolsko, Park, Judd, & Wittenbrink, 2000), academic achievement (Ashton & Esses, 1999; MeCanley & Stit, 1978; Wolsko et al, 2000), and personality and behavior ‘Ryan, 1996; Walsko eta, 2000). The consistency ofthe results across studies, therefore, does not reflect some artifact resulting from the study of a particular type of stereotype. Fourth, personal discrepancies were the least studied of the four types of accuracy. Thus, the studies do not provide much information about the extent to which individual people's stereotypes deviate from perfection. INACCURATE SreREorYPEs Despite the impressive and surprising evidence of the accuracy of stereotypes, there is some cons tent evidence of inaccuracy in stereotypes. In the United States, political stereotypes tend to have. Title accuracy (@g,, Judd & Park, 1993). Many people inthe United States seem to have little know! edge or understanding of the beliefs, attitudes, and policy positions of Democrats and Republicans ‘A recent large-scale study conducted in scores of countries found that there is also litle evidence ‘f accuracy in national stereotypes regarding personality (Terracciano et al, 2005). Its probably rot surprising that people on different continents have little accurate knowledge about one another's personality (¢., that Indonesians do not know mach about, say, Canadians, is not very surprising). However, somewhat more surprising is that people from cultures with a great deal of contact (vari- ‘ous Western European countries; Britain and the United States) also have highly inaccurate belies boat one another's personality characteristics. “Although the Terraciano etal. (2005) stady was impressive in scope and innovative in topic. it suffers from one of the limitations that excluded several studies from this review. Specifically, the criteria samples were haphazard samples of convenience, rather than random samples obisined from target populations. The extent to which this explains their low level of accuracy is unknown ‘until research is conducted oa the same topic that obtains criteria from random samples. in genera, ‘why some stereotypes have such high levels of accuracy and others such low levels is currently unclear and is an important area of future research, ‘THE ROLE OF STEREOTYPES IN ENHANCING OR REDUCING. ‘THE ACCURACY OF PERSON PERCEPTION Waar Stout Prorte Do to Be Accurare (On the Use of inaccurate Versus Accurate Stereotypes in Judging Individuals Relying on an inaccurate belief to judge an individual will not increase accuracy, as can easily Pe seen with a nonsocial example, If Armand believes that Anchorage, Alaska is warmer than Now York City, and he relies on that belie for making guesses about where itis going to be warms today, tomorrows; the next day, and so on, he will be wrong most of the time. Even though be m3¥ pp ap an oveasional ht on the rae days that Anchorage really s warmer thaa New York, he wil be wrong far more often than he i right Stereotypes are no different. If Celeste believes that professional (American) football playess 8° ‘unusually in, andif she relies on that stereotype to guess ther sizes she willbe very wrong, Rei" ‘on an inaccarate stereotype to judge an individual decreases the accuracy ofthat judgment ‘But what happens when people tely ona largely accurate stereotype to judge an indvieal Given that the prior section demonstrated moderate to high accuracy in many stereotyPe beliefs this becomes a pressing question. It turas out that there are some conditions under wich ing on an accurate stereotype can increase accuracy in judging an individual, and the ely: se are some ‘The Unbearable A. conditions under ¥ conditions are the Understanding | three different per possible will be di standing the role ¢ person's group me tion, in which the complicated by po Definitive Indivic ‘The first situation ticalar target. We ficient answer fo ° accomplishments, fora college appl salesperson; and \ and well-validatec quantity of inform If one discover in much of Alask: colder in Alaska i Professional bi 64" Its, therefor Once ina whit Spud Webb was « should one allow height is his heig pletely irrelevant. In situations v member of a g10 evant. One shoul Useful but Not Inmany other sit Presented in the { Small Amounts When ve meet a uly reveal se nes, wealth or Town Coun in which she ea Ambiguous Infe Some eforatic Payful hosing ‘Compliment flat tations uneDiscrimination © achievernent Y and behaviog fore, does not ‘acy. Thus, the 8 stereotypes is some consis 2s tend to have ve lite know 4 Republicans, > litle evidence 1 Ttis probably stone another's ry surprising), ‘Feontact (varie rccurate beliefs ative in topic, it cifically, the imples obtained vey is unknown oles. fe general, cls is currently as can easily be vrmec chan New 1 to be warmes, ‘though he may sw York, be will thal players are ‘wrong, Relying vdgment. + an individual? steotype belies, idee which rely There are some = eitions ae tbe pe unbearable Accuracy of Stereotypes 13 «under which relying even on an aocrate stereotype will not increase Secu Those jor focus ofthe next sections of this chapter. on andi the role of tereotypesin increasing oF reducing asoom6) involves understanding avon perception situations. How pereeivers should go about being a secant 2 peach ofthese three situations (because this discusion involves onder, perceivers know the target coition Unc © three different P sible wil be dif Me herle of stereorypes in person perception, in ll smatioes, v cmieship). I each eas, we first present an example iavolving nonsosil P=ISP- aases my, pethaps, be easier 10 understand, and which will certainly Pe less J correctness concerns, son's BCOUP tion, in which omplicated by politcal ividuating Information aving vividly clear and relevant individvating information abouts par ffinitive” beeause it provides a clear, valid, suf- Definitive Inc ‘The frst situation involves h ela arget, We refer co such individuating «8 weft answer to whatever question one bas aboot @ tArBe. Foe ‘example, when judging academic fef@hplishrments, we might have standardized test stores and class rank and grade point average ea college applicant, when judging sales socress, we might have 10 years of sales records for & tr perso: and when judging personality, we might have maliple SS jndges’ observations of, saree validate personality test cores for, a parGeslar individual: When We have this quality and Spun of infomation, bow much should we rely on stereo pes? Tove discovers from a credible source (ay, the Weather Channe}) hat 80 degrees today satu of Alaska, but only 60 in New Yor, what should one conchae? "The fact chat itis usually aie in Alaska is not relevant, Today, itis warmer in Alaska. . eee wal basketball payec tend to be tall—very tal tis very rere o ind one shorter than de tin therefore, reasonable to expect ll basketball players. be very el ‘onve in while, thovgh a short player makes it nto the National Basketball Association (NBA). po Web was a starting player in te 1990s, and he was about $7" Ones 4 ‘knows his height, sist one allow one's steretype to influence one's jodgment of his height? OF Sow E% His eight is hs beigh, and his membership in generally very tll group—NBA players—is com- pletely irelevant, ree where one has abundant, vividly cleat, relevant indvidaating infomation hou & ember of a group, the stereotype—its content, accuracy, and so on—becomes ‘completely irrel- ‘evant, One should rely entirely on the individeating information, Useful but Not Definitive Individuating Information Inman ober siations, people may have some useful information, butnot the definitive infornaton roesed inthe fist sean, Sometimes information i ambiguous or Kimited in quality or deBre= Small Amounts of information When we meet a person forthe ist time, we might have only physical appearance cues (which Wit taualyroveal sex, but which may or may not cue as in on race or ethnicity, attractiveness: NeOY pes, Wealth, consern with fashion, ete). Of, although we might nt be following the election for ‘Tors Council closely, we just happen to hear on the radio a 10-second sound bite from candidate in which she claims that property taxes are too high, Ambiguous Information Seo ihe nbignos—ts meaning ad inert are unclear. Isa shove Hoy torng sod or asa that ar friendly smile, ora superior sneer Is that extreme iment fattery ot areasm? In these cases people have information, bu its meaning or inter” Pretation is unclear. ae *214 Handbook of Prejuse,Seeotyping, nd Dict yest A Inferences Versus Observations = wat aboat jad Behavior can be observed directly. Most other aspects of psychology—belifs, attitudes, motia | ssertiveness, sock personality, intentions, and so on—are not directly observable; they mast be inferred on the bageae te and one only of behavior. Whereas itis possible to definitively know (most ofthe time) whether Davi si = gua bass for judg, without ots of other information, i is not so easy to figure out whether David is & “happy” avery basic applic ‘Whereas it is relatively easy to grade a student's test, without lots of other informs jon (Jussity ation, its gu hard to know whether that high test score reflects the students brilliance or the simplicity ofteagge'g ered fr vole ‘There isan inberent ambiguity in going from behavior to inferences about underlying attributes, dale behvion tnd rational if 02 Predicting the Future Versus Evaluating the Past 1 inthe absence of The fue is ners ambiguon, isnot pele 1 ow ently what wil pe il Tears fue (soy i itered with te iacutte preion of ic hot peat he eee gM wh incacsb the siperstod, Noneiles, we mat make pedo abou te teal ee Mesa ntsc we sls people fo asin to callepe, pada schol hte, we oes een i pba ore ttn hl heen pee ert pon yan, ar earceigae fe to sceeed reasonably wel Because the fut Isabel sakeoede hove oe Inova sou ‘never have enough information to render such predictions definitive. So, with ‘respect to mak pation pom cays redone he nell ebntee be oe teern weal What Should People Do With Useful but Not Definitive Individuating Information? Alaska Versus New York ‘Alaska and New You get one piece of information about each Tocetion. You learn that Jane, a lifelong resident IF you ase given Anchorage, considers it “cold” today and Jan, a lifelong resident of New York, considers it “cal “Anchorage and ? today. Note that the “information” that you have is identical regarding the two places. Should ya, place, you will therefore, predic that they have dential temperatures? EF time you are ask ‘That would be silly. It ignores the wealth of information you already bring (o beat onthe and igily resis tion: (a) tis usually much colder in Anchorage; (6) “cold” can mean lots of different things in regions sytem ferent contexts; and (©) people usually adapt to their conditions, o, if itis usually 40 degrees in ‘er informatio neighborhood, you would probably judge 20 degrees as cold; but if tis usually 60 degrees in you = tareis warmert neighborhood, 40 might be seen as quite cold. To ignore allthis would be footsh, and, most of ‘time, doing so will lead yon to an inaccurate conclusion about the weather in the two places, ‘Stereotypes anc In other words, in this situation, tothe exten that your beliefs about the general characteristics Ifyou are give ‘Alaska, Alaskans, New York, and New Yorkers are reasonably accurate, they should influence Bill, who is Ai interpretation of “cold” and your prediction regarding the weather in each place pout the ever No individuatin Predict that Ge Stereotypes and Person Perception A prediction ab ‘The logic here is identical. Consider stereotypes of peace activists and al Queda menber. YA 20 other infor hear the same thing abot an individual from each group: They have “tacked” the United Somehow irat Should you interpret this to mean tat they engaged in identical behaviors? Not kel. Toe “wal eogsie that erpeated by the peace activists most ikely a verbal “atk” on US war polices he led red te in stack is probably something far more lethal op cere te ‘The same principles hold regardless of whether the stereotypes involve groups for whom St types are deemed acceptable (¢.g, peace activists or al Qaeda) or groups for whom stereos deemed socially unaccepable (genders, nationalities, races, social classes, religions ei ties, etc). For example, if we learn both Bob and Barb are regarded as “tal.” should we covelide that they are exactly equal in height? Of course not. Undoubtedly, Bob is tall for a man, and B a is tall for « woman, and, because men are, on average, taller than women, tall means di Z objective heights for men and women (implicit acceptance of these “shifting standards” bas thoroughly demonstrated; e.g, Biernat, 1995). 'ypes when ja Many research1 and Discriminaigg ‘atid, Deine en rebon te 9 wate Davao weary tesco oe cniesig ate What will happen int in the reedy, the political, and IL the time, Whenever ©, essentially, making a 1, that he or she is ikely howe We cn ano ‘with respect to maki biguity, , 8 Information? we, a lifelong resident of ‘ork, considers it “sald” two places, Should you, ing to bear on the situa £ different things in dif. ually 40 degrees in yout sally 60 degrees in your ‘olish, and, most ofthe in the two places, eneral charecteristis of "y should influence your slace, { Qaeda members. You ed’ the United States. Not likely. The “attack” + policies; the al Qaeda roups for whom stereo whom stereotypes art sses, religions, ethnic |= 1." should we conclude all for a man, and Barb” 1, tall means diferent ng standards” has beet 7 re unbes rable Accuracy of Stereotypes 25 ot judgments about more socially charged attributes, such as intelligence, motivation, 1, hostility, and so on? The same principles apply. Ifthe stereotype is accu- 11 bit of ambiguous information about an individual, using the stereotype oto ve ti sl and one only hes a smal sy or judging the person will kel enhance accuracy. For the tatstcally inclined thsi sa fai application of Bayes theorem (eg, MeCauly, Sit, & Segal 1980) and principles of - 91). Let's assume for a moment that 30% of motorcycle gang members are sion (ussim, 19% vee for vclentbehtvior at some pont in ther lives, and 0.3% of ballerinas are arrested for arr evr at some pon nthe lives. People wo know this are being completely reasonable we tional ifn dark strets oat Jonely tai stations, they avoid the bikers more than ballerina, tripe absence of much othe individuating information about them. eof these cases, the stereotype “biases” the subsequent judgments, Atleast, tha is how such infuences have nearly always been interpreted in empirical social psychological research el eretypes (2, €8, Devine, 1995; Fiske & Neuberg, 1990; Gilbert, 1995; Jones, 1986). t ff probably mare appropriate, however, o characterize sch phenomena as stereotypes “infu- te rig” or “informing” judgments. Such effects mean that people are appropriately asing their (rowtedge about groups to reach as informed a judgment as possible under difficult and infor- tenon poor circummstances. If their knowledge is reasonably accurate, relying onthe stereotype Ill usualy increase, rather than decrease, the accuracy of those judgments (see also Jussim, 1991, 2005). ‘No individuating Information ‘Alaska and New York IT you are piven absolutely no information, and are asked to predict today’s high temperatore in “Anchorage and New York, what should you do? If you know anything ebout the climate in the two places, you will predict that it will be warmer in New York. Indeed, you should predict this every time you are asked to do so. Would this mean your beliefs about climate are somehow irrationally ‘and rigidly resistant to change? Of course not. All it means is that you recognize that, when two regions systematically differ and you are asked to predict the day's temperature, and are given no ‘other information, it will always be better to guess thatthe place with the higher average tempera- ture is warmer than the place with the lower average temperature. Stereotypes and Person Perception If you are given no information other than race, and you are asked to predict the income of Bill, who is African American, and George, who is White, what should you do? If you know about the average incomes of African Americans and Whites in the United States, you will Predict that George is richer. Indeed, you should predict this every time you are asked to make ‘prediction about the income of an African American and White target about whom you have 10 other information. Would this mean your beliefs about racial differences in income are 4omehow irrationally and rigidly resistant to change? Of course not, Al] it means is that you Tecognize that, when the average income of two racial groups differs and you are asked to Predict the income of an individual from those groups, and are given no other information, it ‘ill always be better to guess that the person from the group with the higher average income bas more income, Wiur Do Prorts Do Wien Tuey Jupce INpivipuats? Process People should primarily use individuating information, when itis available, rather. than stereo “Oi ‘when judging others. Do they? This area of research has been highly controversial, with MY researchers emphasizing the power af stereotypes to bias judgments (Devine, 1995; Fiske &26 Handbook of Prejudice, Stereotyping, and Discrimination evberg, 1990; Fiske & Teylor, 1991; Jones, 1986; Jost & Kruglanski, 2002) and others emphsning Fer ely modest insloence of stereotypes andthe relatively large role of indiviguating informay tion Gussisa, Eccles, & Madon, 1996; Kunda & Thagard, 1996). Trontenately literally hundreds of studies have now been performed that address this issue, and even more fortunately, muliple meta-analyses have been performed summarizing theit ress, ‘Table 10. presents the results from meta-analyses of stodies assessing stereotype bias #9 nny sera shows tha the effects of stereotypes on person judgments, eraged over undo of xperiments range from 0 to 25. The simple arithmetic mean ofthe effect s26s 10, which isan Xperia, because the meta-analyses with more studies yielded systematically lower eft sizes tees between effect size and numberof studies). The few natralise studies ofthe roe oh SE iriypes in biasing person perception have yielded similarly small effects (By Clarke & Campbell, 1955; Jussim et al., 1996; Madon et a, 1998). “Mw stnall isan effect of r= 107 Is small according to J. Cohen's (1988) heuristic eateesne ios oroftect sizes. tis among te smallest effects found in social psychology (Richard eta} 2900 ton esihatfect of 10 means that expectancies substantially infuence socal perceptions about S% a pe ame fs per Rosenthal’ (1991 binomial effect size display). This means that stereotypes do not influence perceptions 95% of the time. vn general therefore, based on more than 300 experimental stodies and « smaliet umber of nnatawstate studies, stereotypes have only very modest influences on person perception. OF couse TABLE TOS Meta-Analyses of the Role of Stereotypes in Person Perception i and Myers (1989), ‘women’s work? | See iia tepals tama? 7 5 a _| SeeaTengld | Des efint oi cgay Bs moe o"s “| oy oe L = : ll Nore Effect sas fa presented in terms of the correlation coefficient, bety sa wreotpe (or expectation) and asteome Ne ey esate Here ove econ on experimental starch, Invidasing infomation ao spe peronl carci, beheviors or accomplishment of india targets. The fT 2 ea ee fo ccf te etn represents th over verge ec se ins 8 aan fet cin afen varied for sbels of xpesimes cde inthe metals, Only means tf oxtcomes, pot of modertors ce mest, ae diapyed + negate ecfiet indicates Savoring en; apo coefcent inlets favoring Women. 1 Tysmeta analysis ncged tere becanse any ofthe studs involved stereotypes “The Unbearable Acct there is always the pc ‘way for powerful stet {for the existence of those emphasizing s. ‘The existence of generally rely beavil that they do. The ont jng information on [ "7h (Kunda & Thag: relying on individue ‘But what aboat ¢ least when the stere inaccuracy for two mentally created fi vyhich to assess ae as “bias” in the lit mean that people « ‘Therefore, the nex! jes that have ectua ‘Accuracy ‘Accuracy in Perc Madon etal. 099: performance, tale accuracy in the f ences by correlat social class. This ‘viduals from ont assessed actual | dents final grad self-reported mo was assessed by actual differenc ‘Madon etal perceived grout sex differences than boys, but t ‘was removed, t We are awa unjustifably fa Jussim etal 1 All three st texts—Tussim (1955) address pattern of acc’ ‘with whom o: ‘more broadly ‘This patter cceut for eth and judged te eeeind Discrimination others empha widuating inf ti tis in : ig the ean vope bas ie Gove heen si iomene iytweefeast sotttewleta Sit conga, seuristic categoriea. ichacd et al, 2003), exceptions about 5% + that stereotypes do \ smaller number of sreeption, Of course, pectin) and oteome Sunn information fe vidal ages. Te fe tne effec size obs it ysis. Only meoraaloet 27 = UF aways the possibly that researchers have aot searched > the right places or in the right I overt one bane Pr perouption. Atminimom, however, the burden of proof feceence of widespread, powerfol stereotype Bases in person perception) has shifted t0 Be cnkaseng sch posal ite se gitence of small stereotype elects, hovers does not necessarily mean that people do erally ly avy on ndiviNaE shformeation, However, the empirical evidence shows, in fact erp) do The one meta-analysis het has addressed this issue found that the effect of individuaty tration on person pereepion wis SNE tho largest effects found in social psychology, r= In ean & Thagad, 1996). lee Wort ‘peopl seem to be generally doing the right thing — 2 nid formation fa more han stereotypes. ye at aout hat 10 effet of tereoypes? Does hat demonstrate inaccuracy? Tt might, at yt when te stereotype sel is clearly inacouae However, it does not necessarily demonstrate gy fortwo reasons) Most of the studies xa these issues have examined experi ineallyereated fictitious targets who had no “eal” attributes, so that there was no eriteria with a assess accuracy; and (ba infuenee of am accurie “sereotype (typically characterized asim the itera) does not noesssariy ranlae 1o inaccuracy. Indeed, some “biases stat people are bein as accurate as possible under he tircumstances (Jussim, 1991, 2005). Ia ete ext soction eviews the very small handful of ereoo P= ‘and petson perception stad Frat have actually addressed the accuracy issue “ere is ‘ecuracy Accuracy in Perception of Small Group Differences Maton otal, (998) examined the accuracy of soventh-prade teachers pers oftheir tadents Formac, talent and effort math about 1 onthe School year, Madon etal. assessed Pe te fllowing manne. First they identified the weachery PoesTnee of group differ eeu Mrlating teachers’ perceptions of individual students with he students? race, sex, and Fa rs cena indicated the extent which faces sStemnaseey evaluated indi se Tone proup tore favorably than individuals from anther gow, We, Madon eta. a ema! grotp iffeences in performance talent, and efor by cores individual stu- cal pad the pio oar (before teachers know the student), standanized or and _ brepored motivation an effort with stodens ree sex and sos “The teachers’ accuracy seer cing te caches’ perceived differences etween groups With (he groups) actual differences, wal ileence sand tat eahers were mostly accurate, Tho coreation Between Washer en soup dfrences and actual poup differences was = 71 The ashes PES of eae ce went, wee highly inaourate—they believed is exerted more acs at hee was no soe ference in slFrepoted motivation an ff. Won outlier than boy ut ee ron between perceived and actual group dfferenesinreused 9 7 TE a te other studies that have adaressed wher people systematically Se FC ——-—T—T—™— Ihssim eat, 1996), Both yielded evidence of accuracy accompanied by smal its FA ae socio foaatting Madon otal, 1998) howeret, were condared ja educations om ie ei Creed teachers’ pereptons of students, and Clie and Canpte {0 sacs ao ereptions of one another. lemains an empiiel question whet 1 pater of scuracy and stl isin perceptions of demographic diffeeness Peo individoals Wilh yor one has extended contact is unique fo classrooms, or characterizes social PercePUOD more broadly _Tisinterot note igh scarey ercepionsof iferences between smal up 88 Beto tie wo reasons. Fister mi jettisoned ther stereotypes completely, judged targets primarily on the basis of relevant individnating information. ‘Second, perceivers18 Handbook of Prejudice, Stereotyping, and Discrimination right not have jettisoned their stereotypes. Ifthe stereotypes (e.g, “gies perform slighty higher {in math classes than do boys”) were accurate (girls actualy did perform slightly higher than boys teachers could also have reached accurate perceptions of differences between boys and girls by applying their stereotype, ‘The research described thus far does not distinguish between these explanations, Regardless of the explanation, however, this research does lead to one clear conclusion: In the few studies that have examined stereotypes and person perception under naturalistic conditions, there is no evidence of stereotypes powerfully and pervasively distorting perception. There was some evidence of bias and distortion, but the far stronger pattern has been accuracy. The next section, therefore, reviews studies that have not only assessed accuracy, but have also assessed the sources of both accuracy ‘and bias in person perception, Does Relying on a Stereotype Increase or Reduce Accuracy in Person Perce} ‘What does empirical research indicate about whether people's reliance on stereotypes increases or reduces the accuracy of ther judgments? Only a handful of studies provide data capable of addres: ing this issue, and they are discussed next Occupational Stereotypes: C. E. Cohen (1981) CE. Cohen (1981) examined whether people more easily remember behaviors and attibutes tht are consistent with a stereotype than those that are inconsistent with that stereotype. Pereeivert in her study viewed a videotape of a dinner conversation between a husband and wife (they were actually husband and wife, but they were also experimental confederates trained by Cohen), Half of the time, this conversation Jed perceivers to believe the woman was a waitress; hlf ofthe time, he conversation led perceivers to believe the woman Was a librarian, The remainder ofthe conversation conveyed an equal mix of librarian-like and waitress-like attributes and behaviors Perceivers were then given a series of choices regarding objective aspects ofthe woman in he videotape (e.g., wore glasses ... did not wear glasses). Their task was to select the corsect descrip tion, Perceivers consistently remembered 5% to 10% more behaviors or features that were consistent with the woman's supposed occupation than behaviors or features that were inconsistent with het sed occupation. For example, they were more likly to accurately remember thatthe “brat: ore glasses and liked classical music, whereas they were more likely to accurately remembet that the “waitress” had a beer and no artwork in her house (even though the tape was identi showing the woman wearing glasses, liking classical music, having a beer, and not having artwork in her apartment) This pattera occurred across two studies and regardless of whether the memory test occurred immediately after the videotape or up to 7 days later. Thus, it appeared that people selectively remembered stereotype-consistent information better than they remembered steretyPe inconsistent information. CE, Cohen (198!) also reported results regarding the accuracy of her petceivers’ memories, Across the (wo studies, accuracy levels were quite high—ranging from a low of 57% to high of £88¢ and averaging about 75% in the first study and about 66% in the second study. Overall, there fore, she found high (about 70%) accuracy and small (about 5%—10%) but real bis. ‘The results fromber second study were particularly relevant with respect to understanding whetber the stereotype increased or reduced accuracy In this study, haf ofthe perceives Teared ofthe WAY an's supposed occupation before viewing the tape (0 the stereotype was activated prior 0 view' haf learned of i after viewing the tape, In comparison to receiving the label after viewing the Pt ‘when people received the label firs, they more accurately remembered both stereotype-COmSET and stereotype-inconsistent information. On average they correctly remembered 70% ofthe a autibutes (Fegardless of their degre of stereotype consistency) when they received the Ibe] | they correctly remembered only about 63% ofthe targe’s attributes when they received the aber ‘The upshot here, therefore, is that, although the label biased memory in such @ mantet stereotype-consistent information, having the abel up front also increased overall accuse ‘the Unbearable Ac Why? Most likely sated their understa tent atributes. Store Residence Hall Ste ‘The wilty of an acc dents made predicti ‘one of two dormitor vative, wealttay, and Jeft wing with unco incither dorm) view then made predictio compared to the tar ‘When perceivers have peeppie attibt tions were correct ( sereotypes, and pr correct. Relying on tion predictions. Sex Stereotypes: Both Jussim et al. (Madon et al 1 ‘conducted in sixth ; differences betwee Jing for individuati ethnicity had litle social class and eth class and ethnic b ‘entirely on individ Sereotype increase Both studies, he Performance (stanc ‘effort, for Madon « Performing higher of models controll ‘encing teacher per Did these sex st ‘They did both, Inc eal performance « the 1996 and 1998 Producing the “ difference. In othe The small indep Mos ofthe sa teachers apparen sip iference Wel wih the sm tween boy an 1996,p. 348)1d Discrimination | 1m slightly higher higher than boy, boys and gis by ons. Regardless of, ve few studies that here is n0 evidence the Unbearable why? Most ent attributes Resi Accuracy of Stereotypes 19 euiely,thelbel provided some sort of organizing scheme for pam which facil- rcrpretation of oth streotype-consistent and steeotype-inconsis- Mrieir understanding a in increase accuracy. reotypes may “bias” perception and, simultaneous, idence Hal Seeotypes: Brock and Ross (1998) hrs — and Ross (1998). College sta- uly rotiions abot the behaviors and preferences of oe! college students who Tived in de iors, The siden inthe “preppi” dorm were widely #608 politically eonser- re evidence of bias one oft therefore, reviews on hy, and conentiona The stants nthe inp dorm were widely seen as politically 9 of both accuracy aie ith neonertonl practices and preference Pose (other students who did not live oer individual targets were informed of each target's Gorm, and ‘ception? ‘types increases of capable of address- s and atibues tha areotype. Perceivers and wife they were iby Cote) Halfot hal ofthe time, he «ofthe conversion ofthe woman inthe tthe correct desctp- sthat were consistent inconsistent with het ribo thatthe “libre accurately romembet @ tape was intel, 4 not having artwork whether the memery appeared that people nembered serotyPe serceivers’ memories. wy of 51% toa high of study, Overall, here> bias Understanding whetet ‘leaned ofthe wom ated prior to viewine aftr viewing the 0P8 7 stereotype consist 7 red 70% ofthe args 7 ‘eceived the label fi received the label ast fra manner a8 co fv0F rrall accureey, tether dorm) viewed photoes then made predictions abou compat ave preppie attributes or fora hip tions were correct — gereatypes, and preicted targets £0 Sorect Reiying on the preppie-hippie dor tion predictions. oth Jussi etal. (1996) and cond diferences between studk fing for individuating information (motivation, acievement, et) misty had litle or no effect on teacher expectations. Thos, ache sei case and ethnic stereotypes when judging differences between chiro from different social Gass and ethnic backgrounds. Although this finding is in many ways laudable ‘entirely on individuating information sHereotype increases or reduces accuracy. performance standardized regression coef effort, for Madon et al. and Jussim et al, respectively) performing higher and exerting more effort ‘of models controlling for individuating information, faving teacher perceptions—bias effects, in traditional social psychological pa © Trey did bot. Inthe case of performance, the sex stereotype effect increase ‘eal performance difference, as indicated by final grades the prior yeas, ws the 1996 and 1995 sais, especie, girls rosived sighs Nahe! grades) The regression mode! | roducing the “biasing” effect of stereotypes yielded a “bin © iiference. In other words: ; oe nget’s behaviors and attitudes, Pereeivers’ predictions were hen ae Pre earges’ self-reports on these same preferences and attitudes. co be consistent with their dorm (fora preppie dorm resident to we hippie attributes), 66% of their predic= they matched the targets” self-reports). When perceives jettisoned their dorm ‘be inconsistent with their dori, 43% of their predictions were stereotypes enhanced the accuracy of person percep ‘When perceivers predicted target se dorm resident to hay sex stereotypes: Jssim et al (1996) and Madon etal. (1998) ‘Madan etal. (1998) examined the accuracy of teacher expectations Joss 98, as described previously; Jssim etal, 1996, was sia sp that it was or faint grace rather than seventh grade, and it didnot examine Ne accuracy of perceived re ifferent demographic groups) Both found tat, when contro tudent social class and race or essentially jettisoned their teachers relying Toes not help address the question af whether relying on & Navever fund tha sex stereotypes based caches’ perceptions of Posy and gitls? cients of 09 and .10 for performance, and 16 and.19 for {in both stadies, teachers perceived girls as than boys. Because these effects occurred inthe contest ‘hey are best interpreted es stereotypes infla- ance. Die ee ac aterwotyping bas effects increase or reduce the accuracy of teachers! perceptions? sd teacher accuracy. The (08 and r= 10 for shat was virtually identical to the real ‘he smal independent eet of seat sex on teacher perceptions (of ptfomnanc®) secoune ff hes cation beeen se dhs PS (of pectormancs) This means thet Scapa oceatyped ssw pextorming Say higher than byt independent tbe we sities inperomaes Hower beeen vinien teachers di so coreesponded reasonably sawn elfen pronanee Ino Syotds, teaches” perceptions of differences icen Dy and ge were soraateDeceuse teacher flied onan accurate retype usin te 1996, p. 348)20 Handbook of Prejudice, Stereotyping, and Discrimination ‘The same conclusion, of course, also characterizes the results for the 1998 study. On the other hand, the results regarding effort provided evidence of bias that reduced accuracy, There was no evidence that girls exerted more effort than boys. Therefore, the influence of student sex on teacher perceptions of effort (.e, teachers’ reliance on a sex stereotype to arrive at judgments of effort) led teachers to perceive a difference where none existed. This is an empirical demonsira- tion of something that, logicaly, has to be (rue, Relying on an inaccurate stereotype when judging individuals ean only harm one’s accuracy. SUMMARY AND CRITICAL EVALUATION Our review has shown that itis logically incoherent to define stereotypes as inaccurate, that itis ‘unusual (but not unheard of) for stereotypes to be highly discrepant from reality, that the coctela- tions of stereotypes with criteria are among the largest effects in all of social psychology, that people rarely rely on stereotypes when judging individuals, and that, sometimes, even when they do rely on stereotypes, it increases rather than reduces their accuracy. Many scholars, scientists, and people of good will, we do not doubt, will find these conclusions unbearable. ‘Therefore, the next order of business is to identify important limitations and qualifications to these conclusions. We are going to (a) clearly state many of the things the stereotype literature does, not show; (b) state What it does show; and (¢) describe many of the limitations o existing research on stereotype accuracy, We hope that doing so reduces the extent to which some readers may mis- interpret our claims about what the stereotype research does show, and what lessons ean be learned from it. ‘Whur the Srereoryre Research Does Nor SHow 1. It does not show that al stereotypes are always perfectly 100% accurate. We know of no researcher who has ever made this claim. 2. It does not show that prejudice and discrimination do not exis, or are trivial and unimport ant. Prejudice and discrimination are terribly important, and can be terribly destructive ‘The research reviewed in this chapter has not addressed prejudice and discrimination. 3, It does not show that people correctly explain why group differences exist. Inasmuch social scientists do not agree as to why group differences exist is probably not possible to assess the accuracy of most lay explanations for group differences. 4, It does not show how people arrive at their stereotypes. There is very little research om ‘where stereotypes come from. Much speculative discussion emphasizes hearsay, fam ily socialization, and the media (eg, Allport, 1954/1979; Katz & Braly, 1933; Pickering, 2001), The extraordinary levels of accuracy shown in many of the studies reviewed in this chapter, however, do suggest another source is the primary basis of stereotypes—sociat reality 5. The amount of research that has addressed the accuracy of people's perceptions of diffe ences between small groups of individuals they know personally (stereotypes and person perception) is quite modest, and does not yet provide a sufficiently broad foundation of ‘which (o reach any general conclusions. It appears as if relying on accurate stereotyPes seems to mostly enhance accuracy, but that conclusion should be held tentatively, pending further stdies ‘Wiiar Tis Researc Does SHow 1. The claim that stereotypes, as beliefs about groups, are inherently inaccurate has been falsified “The Unbearable Accu! 2, A more modest that they are gen evidence provide most appropriate ate usually mod 3, This pattern of toany particula sroups, gender, 4. The pattern of m team or methot researchers; by ‘using noncomp through official target group. 5, This pattern of stereotype beli acteristics, ach 6, The strong for: tions or as cla data, Exagger quently than d ‘7. The exaggerat times does oc: ferences, more important que 8, In contrast to cies against aspect of ster. bout they were almost never Rosenthal’s ( meaningful types are abe ‘Table 10.4 com exceeding cortelat ing the extent to * Only 24% of soci 50.18 contrast, al Table 10.1 and Tat sonal sterearype & This is doubly in laypeople's ste Social sciences ar oF choose fo igno introductory cexts the mete exposun ‘weapons effect (¢ Speakers are mor ity for successes ‘much time and sjTABLE 104 ethnic and Gender Stereotypes Are More Valid Than Most Social Psychologica Hypotheses Proportion of Social Proportion of Consensual Proportion of Personal Foyeologcl Efects Stereotype Accuracy Stereotype Accuay Obtained in Research* Corrlations* Corelations Exceeding 30 288 roa (8/18) 818011) 5% 298 (1618) 268 (1) xceting 50 arbre rm Richt, Bona Zoo's (2003 sew fetes esa lates ofthe coreation ofa deanna Win ents, meroristbesankerof seco ser coon mi Fon ie ening 30 of 3 de Severe eal nb of tebe sanyo ect Tan 102, Bec Table 10mm he re fe es fs MC a eran te Ssh Apes cod fe in, Ter oie tty doin an Ne coe i gecesi es 10a 102013 corse Ons te ceeate average of several coneaions fond in the origin] ails incindingthoosands of sue, cople's stereotypes? Typically, none at all Fora field that aspires tob&) evidence of the accuracy of Ps ight even say unbearable, scientific, this isa (coubling state of affairs. Some mi InPoRTANT LIMITATIONS “There ae, of course, many important iittions tothe existing work on the accuracy of SeSoND Free ae cosy too tne oer major types of seretypes—teigon andsoial ss ea ce sp never been examined. Although we can think of no reason Wy pattern of WSN fa a ster thee type of groups, we will never know until the esearch s actually conse a ie pc eisng research has overwhcimingly examined the stereotypes hel by cle dene py pecase hoe samples ae cooveniet Is this important? Maybe, SuggesiTE V0 ae ess has ben he esearch By MoCanley ad colleagues 8 Tables 104 tie by Clabaugh and Moring (2004 showing that te accuracy of noncoe BS aie chat of college students, Nonetheless, more research with noncollege samples — ao or are many ifort types and aspects of accuracy, and few studies report addr al of them. dell, more research in the fotare wil provide more comprehensive rents ofthe various types of stereotype accuracy. ‘Fourth, most ofthe research on stereotype a6 ‘States and Canada. Peshaps stereotypes in other countries are suracy to date has been concieted in the Us Jess (or more) accurate, “Ane Srereoryees Ever HIGHLY INACCURATE? ‘The Evidence Reviewed in This Chapter Evidence of major inaccuracy is rare but it is not entire reviewed have shown that people are better at judging eee onder of atibutes within group, than they are at judging the exact eve of Caeser hina group. Tp other words, te analyses assessing correspondence, whieh peopl’ beliefs wi group atibutes or group differences, consistently found SO" evidel sevrracy, whereas the analyses assessing discrepancies provided a more mixed Dic incu ae pulls cjes,a fair amount of near misses, and fai aout of inaccuracy: PY i people do not exaggerate or underestimate rel differences, the evidence We reviewed showed ern, thy either consistently over-or underestimate the level of an atibute i 8 To¥P ly absent First, even the studies that we Mi tifferences between groups, and at jul groups “the Unbearable Acc Second, on averag vidual beliefs about; sonal stereotypes We sccuracy correlations these are simply mor ae systematically m candidates for indivi be intelligence (are + more accurate), exp 1954/1979, has long s erroneous stereotype ‘people lower in non’ see if this pattern re the intended henevol reotypes; Wolsko et prejudice and stereo deeply held prejudice Speculations on Ot Studies that examine haphazard samples ¢ 2005) consistently fi researchers who usec "Tables 101-10). Tk Aiticut to interpret. The existence of does justify the cone ‘moderate to high ac aecessarily justify cc just not yet looked ir For example, edu Camas, where mos fener, people are x Senocrces) than in nger inaccurate helps perpetuate ina. held; the Indian cas Hindus held by 2ist- Sereotypes might be Unfortunately, be lenges to their autty sccuracy in such cor Sereotype inaccura Serotype accuracy We Scennnc ax Stereotypes can be Ceaiting any secur Posie is more lik1g, and Discrimination iological Proportion of Personal ‘Stereotype Accuracy Correlations B18 e711) 366 @/11) udlog thousand fs, curacy conelation meting terotype accuracy cars five studies for McCavey, probably urderstinae the aoe, even ough hey ofa 1 field that aspices tobe veouracy of stereotypes, 4 social class-—have,as hy patterns of accuracy 1s actually conducted. pes held by college stu- ye. Suggesting it msy Tables 10.1 and 102), groups is nearly iden- samples is needed. ¥ studies report reslts comprehensive assess- onducted in the United accurate, the studies that we ave groups, and at judging xxact level of particular fence, which correlated and strong evidence of | xed picture, including 2 ‘inaccuracy. Even whet] > reviewed showed thal in a group. he Unbearable Accuracy of Stereotypes 23 second, on avecage, personal stereotypes corresponded well with groups’ attributes (ind isa etfs about groups correlated moderately to highly wih titer, Nonetheless, some pe wit derotypes Were highly inaccurate, Nesey all of the studies reporting personal serotype crreations found a least some people with very low—near zero—correlations. Whether ae ace simply more or Fess random fuctutions and measurement error, ot whether some people tbe gtematicaly more acorat than others, san important question for future researc. Possible a ines for individual differences that would predict systematic variations in accuracy would favineligence (re smarter people more accurate”), education (are more highly educated people ve accurate”), exposure 10 and experience with groups (the “contact hypothesis"; eg. Allport To54/1979, has long suggested that contact with a group reduces prejudice in pat, by dsconficming Pepeous stereotypes), nonverbal sensitivity (actually, Hall & Carter, 1999, already showed that seope lower in nonetbal sensvity hold less accurate sex stereotypes, bu it woud Be sel 10 reee tis patter replicates), and ideology/motivated egalitarianism/universalism (which, despite ‘ee intended benevolence of an egalitarian ideology, seems to lead people to hold less accurate ste- teatypes; Wolsko etal, 2000). Despite the existing evidence showing only weak relations between prejudice and stereotyping (Park & Judd, 2005) pethaps under the right (or wrong) conditions, {eeply held prejudices and hostilities ean sometimes lead to highly distorted stereotypes. accuracy 60 Speculations on Other Conditions of Inaccuracy Studies that examined people's beliefs about groups and then used as criteria the self-teports of taphazard samples of members of the target group (Allen, 1995; Martin, 1987; Terracciano etal 2005) consistently find more evidence of what those researchers interpret as “inaccuracy” than do researchers who used whole populations or random samples of targets the research summarized in ‘Tables 10.1-10.3). The disconnect between the stereotype and criteria, however, renders such results Aitfcue to imterpret. “The existence of so few clear and strong demonstrations of widespread stereotype inaccuracy oes justify the conclusion that research on the accuracy of stereotypes usually finds evidence of tmoverate € high accuracy, and only rarely finds evidence of low accuracy, It does not, however, tccessarily justify concluding that stereotypes are hacdly ever inaccurate. Perhaps researchers have just not yet looked in the right places or inthe right ways for stereotype inaccureey. For example, education and mass communication levels are so high in the United States and Canada, where most of the stereotype accuracy research has been conducted, that, pechaps, in general, people are more exposed to socal reality in these places (and, probably, in other Western democracies) than in many other places around the world. Perinaps poverty and ignorance help breed siconger inaccurate stereotypes. Perhaps the propaganda of demagogues in euthoritarian regimes belpspecpetuate inaccurate stereotypes. The Jim Crow American South; South Africa under apert- heid; the Indian caste system; the Nazis’ eacial beliefS; and beliefs about Christians, Jews, and indus held by 2ist-century Islamists area few examples where it seems plausible to speculate that stereotypes might be more inaccurate than found in the research reviewed here. Unfortunately, because the powers that be under such systems are not likely to be open to chal lenges to their authority, it will probably be very difficult to perform studies of stereotype (in) accuracy in such contexts. If is dificult to perform research in the contexts most likely to produce stereotype inaccuracy, the scientific literature will be skewed coward providing more evidence of stereotype accuracy than may be actually true of people in general, around the word Tie SciENTIFIC AND SOCIAL VALUE OF STEREOTYPE ACCURACY RESEARCH Stereotypes ean be accurate. Some scholars and lay people resist this concision, believing that crediting any accuracy to stereotypes is tantamount to endorsing prejudice, We argue that the ‘opposite is more likely true—that acknowledging the accuracy of some stereotypes provides theavearate stereotype assed i some dete ie was published in Distinguishing Accurate From Inaccurate Stereotypes oan Civil eights” is sO Notall stereotypes are accurate and those that are inaccurate may be the mast damaging. speci

Você também pode gostar

• What Determines Perceived Income Justice

  

  Documento33 páginas

  What Determines Perceived Income Justice

  Boagiu Răzvan

  Ainda não há avaliações
• Mesopotamia in The Early History of Alchemy

  

  Documento18 páginas

  Mesopotamia in The Early History of Alchemy

  Boagiu Răzvan

  Ainda não há avaliações
• Remembering The Past To Imagine The Future: The Prospective Brain

  

  Documento5 páginas

  Remembering The Past To Imagine The Future: The Prospective Brain

  Boagiu Răzvan

  Ainda não há avaliações
• • Revistas
  • Podcasts
  • Partituras
• The Path of Esoteric Hitlerism PDF

  

  Documento21 páginas

  The Path of Esoteric Hitlerism PDF

  Boagiu Răzvan

  100% (5)
• The Invocation of The Black Sun - Miguel Serrano

  

  Documento5 páginas

  The Invocation of The Black Sun - Miguel Serrano

  Sveinaldr Ingvarsson

  90% (10)
• The Apparition of The I PDF

  

  Documento5 páginas

  The Apparition of The I PDF

  Boagiu Răzvan

  Ainda não há avaliações
• Miguel Serrano On Integral Radical Traditionalists and Julius Evola

  

  Documento7 páginas

  Miguel Serrano On Integral Radical Traditionalists and Julius Evola

  Boagiu Răzvan

  Ainda não há avaliações
• Varieties of Moral Personality: Beyond The Banality of Heroism (2010)

  

  Documento36 páginas

  Varieties of Moral Personality: Beyond The Banality of Heroism (2010)

  Boagiu Răzvan

  Ainda não há avaliações
• Hitler and The Last Avatar - The 114th Anniversary of Adolf Hitler's Birth

  

  Documento1 página

  Hitler and The Last Avatar - The 114th Anniversary of Adolf Hitler's Birth

  Boagiu Răzvan

  Ainda não há avaliações
• Shoe Dog: A Memoir by the Creator of Nike

  

  No Everand

  Shoe Dog: A Memoir by the Creator of Nike

  Phil Knight

  Nota: 4.5 de 5 estrelas

  4.5/5 (537)
• The Subtle Art of Not Giving a F*ck: A Counterintuitive Approach to Living a Good Life

  

  No Everand

  The Subtle Art of Not Giving a F*ck: A Counterintuitive Approach to Living a Good Life

  Mark Manson

  Nota: 4 de 5 estrelas

  4/5 (5794)
• Hidden Figures: The American Dream and the Untold Story of the Black Women Mathematicians Who Helped Win the Space Race

  

  No Everand

  Hidden Figures: The American Dream and the Untold Story of the Black Women Mathematicians Who Helped Win the Space Race

  Margot Lee Shetterly

  Nota: 4 de 5 estrelas

  4/5 (895)
• The Yellow House: A Memoir (2019 National Book Award Winner)

  

  No Everand

  The Yellow House: A Memoir (2019 National Book Award Winner)

  Sarah M. Broom

  Nota: 4 de 5 estrelas

  4/5 (98)
• Grit: The Power of Passion and Perseverance

  

  No Everand

  Grit: The Power of Passion and Perseverance

  Angela Duckworth

  Nota: 4 de 5 estrelas

  4/5 (588)
• The Little Book of Hygge: Danish Secrets to Happy Living

  

  No Everand

  The Little Book of Hygge: Danish Secrets to Happy Living

  Meik Wiking

  Nota: 3.5 de 5 estrelas

  3.5/5 (400)
• The Emperor of All Maladies: A Biography of Cancer

  

  No Everand

  The Emperor of All Maladies: A Biography of Cancer

  Siddhartha Mukherjee

  Nota: 4.5 de 5 estrelas

  4.5/5 (271)
• Never Split the Difference: Negotiating As If Your Life Depended On It

  

  No Everand

  Never Split the Difference: Negotiating As If Your Life Depended On It

  Chris Voss

  Nota: 4.5 de 5 estrelas

  4.5/5 (838)
• The World Is Flat 3.0: A Brief History of the Twenty-first Century

  

  No Everand

  The World Is Flat 3.0: A Brief History of the Twenty-first Century

  Thomas L. Friedman

  Nota: 3.5 de 5 estrelas

  3.5/5 (2259)
• On Fire: The (Burning) Case for a Green New Deal

  

  No Everand

  On Fire: The (Burning) Case for a Green New Deal

  Naomi Klein

  Nota: 4 de 5 estrelas

  4/5 (74)
• Elon Musk: Tesla, SpaceX, and the Quest for a Fantastic Future

  

  No Everand

  Elon Musk: Tesla, SpaceX, and the Quest for a Fantastic Future

  Ashlee Vance

  Nota: 4.5 de 5 estrelas

  4.5/5 (474)
• A Heartbreaking Work Of Staggering Genius: A Memoir Based on a True Story

  

  No Everand

  A Heartbreaking Work Of Staggering Genius: A Memoir Based on a True Story

  Dave Eggers

  Nota: 3.5 de 5 estrelas

  3.5/5 (231)
• Team of Rivals: The Political Genius of Abraham Lincoln

  

  No Everand

  Team of Rivals: The Political Genius of Abraham Lincoln

  Doris Kearns Goodwin

  Nota: 4.5 de 5 estrelas

  4.5/5 (234)
• Devil in the Grove: Thurgood Marshall, the Groveland Boys, and the Dawn of a New America

  

  No Everand

  Devil in the Grove: Thurgood Marshall, the Groveland Boys, and the Dawn of a New America

  Gilbert King

  Nota: 4.5 de 5 estrelas

  4.5/5 (266)
• The Hard Thing About Hard Things: Building a Business When There Are No Easy Answers

  

  No Everand

  The Hard Thing About Hard Things: Building a Business When There Are No Easy Answers

  Ben Horowitz

  Nota: 4.5 de 5 estrelas

  4.5/5 (345)
• Yes Please

  

  No Everand

  Yes Please

  Amy Poehler

  Nota: 4 de 5 estrelas

  4/5 (1891)
• The Unwinding: An Inner History of the New America

  

  No Everand

  The Unwinding: An Inner History of the New America

  George Packer

  Nota: 4 de 5 estrelas

  4/5 (45)
• Rise of ISIS: A Threat We Can't Ignore

  

  No Everand

  Rise of ISIS: A Threat We Can't Ignore

  Jay Sekulow

  Nota: 3.5 de 5 estrelas

  3.5/5 (137)
• Principles: Life and Work

  

  No Everand

  Principles: Life and Work

  Ray Dalio

  Nota: 4 de 5 estrelas

  4/5 (599)
• Fear: Trump in the White House

  

  No Everand

  Fear: Trump in the White House

  Bob Woodward

  Nota: 3.5 de 5 estrelas

  3.5/5 (738)
• Angela's Ashes: A Memoir

  

  No Everand

  Angela's Ashes: A Memoir

  Frank McCourt

  Nota: 4.5 de 5 estrelas

  4.5/5 (440)
• Bad Feminist: Essays

  

  No Everand

  Bad Feminist: Essays

  Roxane Gay

  Nota: 4 de 5 estrelas

  4/5 (1016)
• Steve Jobs

  

  No Everand

  Steve Jobs

  Walter Isaacson

  Nota: 4.5 de 5 estrelas

  4.5/5 (806)
• The Glass Castle: A Memoir

  

  No Everand

  The Glass Castle: A Memoir

  Jeannette Walls

  Nota: 4.5 de 5 estrelas

  4.5/5 (1713)
• The Gifts of Imperfection: Let Go of Who You Think You're Supposed to Be and Embrace Who You Are

  

  No Everand

  The Gifts of Imperfection: Let Go of Who You Think You're Supposed to Be and Embrace Who You Are

  Brené Brown

  Nota: 4 de 5 estrelas

  4/5 (1090)
• John Adams

  

  No Everand

  John Adams

  David McCullough

  Nota: 4.5 de 5 estrelas

  4.5/5 (2409)
• The Outsider: A Novel

  

  No Everand

  The Outsider: A Novel

  Stephen King

  Nota: 4 de 5 estrelas

  4/5 (1839)
• The Light Between Oceans: A Novel

  

  No Everand

  The Light Between Oceans: A Novel

  M.L. Stedman

  Nota: 4.5 de 5 estrelas

  4.5/5 (789)
• Manhattan Beach: A Novel

  

  No Everand

  Manhattan Beach: A Novel

  Jennifer Egan

  Nota: 3.5 de 5 estrelas

  3.5/5 (792)
• The Sympathizer: A Novel (Pulitzer Prize for Fiction)

  

  No Everand

  The Sympathizer: A Novel (Pulitzer Prize for Fiction)

  Viet Thanh Nguyen

  Nota: 4.5 de 5 estrelas

  4.5/5 (121)
• The Woman in Cabin 10

  

  No Everand

  The Woman in Cabin 10

  Ruth Ware

  Nota: 3.5 de 5 estrelas

  3.5/5 (2322)
• Brooklyn: A Novel

  

  No Everand

  Brooklyn: A Novel

  Colm Tóibín

  Nota: 3.5 de 5 estrelas

  3.5/5 (1937)
• A Man Called Ove: A Novel

  

  No Everand

  A Man Called Ove: A Novel

  Fredrik Backman

  Nota: 4.5 de 5 estrelas

  4.5/5 (4610)
• The Perks of Being a Wallflower

  

  No Everand

  The Perks of Being a Wallflower

  Stephen Chbosky

  Nota: 4.5 de 5 estrelas

  4.5/5 (2104)
• Wolf Hall: A Novel

  

  No Everand

  Wolf Hall: A Novel

  Hilary Mantel

  Nota: 4 de 5 estrelas

  4/5 (3811)
• Little Women

  

  No Everand

  Little Women

  Louisa May Alcott

  Nota: 4 de 5 estrelas

  4/5 (104)
• Her Body and Other Parties: Stories

  

  No Everand

  Her Body and Other Parties: Stories

  Carmen Maria Machado

  Nota: 4 de 5 estrelas

  4/5 (821)
• The Art of Racing in the Rain: A Novel

  

  No Everand

  The Art of Racing in the Rain: A Novel

  Garth Stein

  Nota: 4 de 5 estrelas

  4/5 (4200)
• Sing, Unburied, Sing: A Novel

  

  No Everand

  Sing, Unburied, Sing: A Novel

  Jesmyn Ward

  Nota: 4 de 5 estrelas

  4/5 (1103)
• The Constant Gardener: A Novel

  

  No Everand

  The Constant Gardener: A Novel

  John le Carré

  Nota: 3.5 de 5 estrelas

  3.5/5 (104)
• A Tree Grows in Brooklyn

  

  No Everand

  A Tree Grows in Brooklyn

  Betty Smith

  Nota: 4.5 de 5 estrelas

  4.5/5 (1929)