Second paper
Laurence Steinberg, Ph.D. II' with B. Bradford Brown, Ph.D., and .Sanford M. Dornbusch, Ph.D. SIMON & SCHUSTER New York LondonToronto Sydney Tokyo Singapore SIMON &SCHUSTER Rockefeller Center 1230Avenue ofthe Americas New York, NY 10020 Copyright ©1996 by Laurence Steinberg Allrights reserved, including the right ofreproduction inwhole orin part inany form. SIMON &SCHUSTER and colophon areregistered trademarks ofSimon &SchusterInc.
Designed byJennifer Ann Daddio Manufactured inthe LJ:nited States ofAmerica 3 5 7 9 108 G 4 2 Library ofCongress Cataloging-in-Publication Data Steinberg, Laurence D., date. Beyondtheclassroom: why school reform hasfailed and what parents needtodo / Laurence Steinberg, with B. BradfordBrownandSanford M.Dornbusch. p. em. Includes bibliographical references and index. 1.Highschool students-United States-Social conditions-Longitudinal studies.
2. Academic achievement-United States-Longitudinal studies. 3. Home and school United States-Longitudinal studies.4.High school students-United States Attitudes-Longitudinal studies .. I.Brown, B. Btadford (BensonBradford), date.
II. Dornbusch, Sanford M. III.Title. LC205.574 1996 370.19'342'09794-.<1c20 96-3GG2 CIF ISBN 0-G84-80008-X EIGHT The Power of Peers Parents playa centralroleininfluencing theirchild's development andedu cation, but by the time children havereached thelater years of elementary school, friendshavetalcen on tremendous importance in theirschool life.In order to understand thefull complement of influences on school perfor mance and engagement, especiallyduringtheadolescent years-and in order to understand thecauses of America's achievement problem-we need to lookclosely atthe iroles played by peerJllIndeed, ourresearch indicates that peers shape student achievement patternsinprofound ways, and thatin many respects friendsaremore powerful 'influences thanfamily members are, Foralarge number of adolescents, peers-not parents-are the chief determinants of how intensely theyareinvested inschool and how much ef fort they devote totheir education,) THE SOCiAL WORLD OF ADOLESCENCE In our research, wedevoted countless hours to investigating anddescribing the social world of adolescence. Thisexpenditure of timeandenergy was necessary becausestudying peerinfluences onadolescent behaviorentails 138 THE POWER OF PEERS muchmorethanstudying theinfluence of theadolescent's closefriends. The closefriends ayoung person has,while significant influencesinayoung per son's life,areonly asmall part of thetotal complex of peerinfluence. The adolescent's socialworld canbedrawn as three concentric circles. In the innermostcirclearethe youngster's one or two ~est friene&-lThese arethe other children with whom thechild spends most of hisorher free time-at lunch, during freeperiods, and so on. If youarethe parent of achild be tween the ages of tenand sixteen, talce amoment and think about thechil dren whoareyour child's besrfriends. Thesearethe children whom your child greets first on arriving atschool, andthey arealso thechildren whom your child sees lastbefore leaving schoolatthe end of the day. When the telephone ringsforyour child on aweekday eveningorweekend afternoon, more often than not itisone of thesepals.
Best friends comprise theinner circle of thesocial map of adolescence. In thenext circle out arethe youngsters whoaremembers of the adolescent's rliquel These adolescents arealso thechild's friends, but theirfriendships are not nearly as intimate as arethe ones thechild haswith hisorher closest companions, andthese relationships mayfluctuate inimportance from week toweek. At anyone time,anadolescent's cliqueusually willhave somewhere betweensixand tenmembers. Cliquemembers willsitatthe same tables inthe school cafeteria, hangaround witheachother during re cess, andinteract witheachother as theytravel toand from class and school.
These arethe children whosenames typically popupinthe stories children share withtheir parents abourtheevents of theschool day. The third circle in themap of adolescent peerrelations iscomposed of the adolescent's~rowd.jrhe adolescent's crowd ismadeupoflike-minded in dividuals whoshare certain features incommon witheachother burwho are not necessarily eachother's friends. Indeed, it ispossible foradolescents ro bemembers of the same crowd without reallylmowing eachother wellat all. This isbecause adolescents aremembers of the same crowd byvirtue of their common interests,arritudes, andpreferred activities, not because of theirre lationships witheachother. What crowd members haveincommon is not intimacy, but shared identity. Buteven though crowdmembers do not share the intimacy of close friends, rhey influenceeachother inimportant ways. An adolescent may not talkmuch about themembers of hisorher crowd, 139 BEYOND THE CLASSROOM but, as youwill read, crowd members greatlyinfluence eachother neverthe less. THE IMPORTANCE OF PEER CROWDS When adults tendtothink of peer influences on adolescent behavior,they tend tothink mainly aboutinfluences withintheinnermost circle(best friends), secondarily abourinfluences inthe middle circle(theclique), and only marginally, if atall, abour influences inthe outermost circle(the crowd). This isquitereasonable, since we mightexpect thatindividuals will be most influenced bythe people withwhom theyareclosest, andleast af fected bypeople withwhom theyshare littleintimacy. Of course, it istrue that during adolescence, as during otherperiods of life,best friends influ ence each other'sbehavior agood deal.Burduring adolescence, peerinflu ence operates withincliques andcrowds inextremely important ways,and adults should probably paymore attention thantheydotothe ways in which adolescents areinfluenced bythese groups of friends. It isespecially easy tounderestimate thepower of theadolescent crowd, because itsinfluence istransmitted in less direct,andmore subtle, ways than istheinfluence of theadolescents' closefriends or clique. Closefriends and clique members influence eachother inadolescence much as theyinfluence each other inchildhood or adulthood: byproviding modelswhosebehavior can becopied (''I'mdressing thisway because Jamielooks so coolwhen she dresses thisway"); byrewarding orpunishing certainactions, therebyin creasing (ordecreasing) thelikelihood of our repeating them(''I'mnever wearing thisagain because Lukelaughed atme"); andbyexerting direct pressure tobehave oneway or another ("JessietoldmeIjust had towear this shirt tothe party"). Crowds, incontrast, are less common inadulthood, and their influence isless familiar toadults. IBecause crowdmembers donot necessarily knoweachother personally, they don't influence eachother directlyt-through modeling,reinforcement, or coercion. Crowdmembers influenceeachother indirectly, throughestab lishing normsandstandards thatthemembers feel they must adhere to.
Once theadolescent hasidentified withaparticular crowd,thecrowd's stan dards become internalized, incorporated intohisorher own sense of self. As 140 THE POWER OF PEERS aconsequence, adhering to thenorms and standards of the crowd does not feel to theadolescent likesuccumbing to peer pressure; it feels mo ce likean expression of his orher own identity.
Adults do not understand thisdistinction verywell. When wethink of "peer pressure," wetend toenvision anadolescent beingcoerced byfriends to engage inacertain behavior ("Come on, justtrythis cigarette"), inthe company of friends whomodel aspecific action and thenwaittobe imitated ("Everyone isdoing it"),orwho actively rewardorpunish theadolescent for behaving inagiven way("You're not really wearing that!"). To be sure,such instances of active peerpressure doindeed occurinthe daily lives of teenagers. Butmuch of thepeer pressure experienced byadolescents is not this active-nor is itnecessarily experienced as pressure-although itis no less powerfulinits own way than the more activeforms. PEER PRESSURE PEAKS IN EARLY ADOLESCENCE IThe less secure weare about ourown identity and our owndecision-malcing abilities, the more weare influenced byothers' opinions\ Burthesalience of the crowd as aninfluence on our behavior declines as we'become adults.Be cause adolescence is atime when individuals oftenhavequestions about their identity and theirability to function independently, it isinherentlya time of heightened vulnerability tothe influence of others. Whileadultsare not immune to pressure fromtheirpeers, theyaresignificantly less suscepti ble toitthan adolescents are.
In our research, wehave been able to chart changes inindividuals' sus ceptibility to peer pressure as theymove intoandthrough adolescence. In severaldifferent studies,wehave found thatvulnerability topeer pressure that is, how easily swayed aperson isby the demands of hisorher friends rises as children becometeenagers, peal
There is a specific periodindevelopment, then-roughly from agetwelve through sixteen-when individuals areeasily influenced bytheir peers. And it isduring thistime thatpeers begin ro playanenormouslYI importantrole in influencing achievement.
It isespeciallyinteresting to juxtapose thedevelopmental course of peer 141 BEYOND THE CLASSROOM pressure,withitspower peaking inearly adolescence, withwhat weknow about thedevelopmental course of parentalinvolvement inschooling. & noted in the last chapter, national studies of American familiesshowthat parental involvement drops off precipitously betweenelementary and sec ondary school-precisely at the time when youngsters' susceptibiliry topeer influence israpidly rising.Moreover, ourevidence indicates thatthissort of parental disengagement isnot limitedtoschool matters, but affectsawhole range of issues, including monitoring and regulating thechild's relationships with friends, thechild's useofleisure time, and thechild's choice of activi ties. Tothe extent thatdiminished involvement intheir children's livesren ders parents' influence less powerful, thedoor isopened forpeers tostep in and exert asignificant impact on eachother's behavior-including their be havior inschool. And this isprecisely whathappens between thesixth and renth grades. At thispoint, infact, children's achievement ismoreeasily in fluenced bytheir peers thanatany other time in theirschool career. SHOULD ADULTS PANIC? When adultsaretold thatyoung adolescents arehighly susceptible tothe in fluence of friends-and, in fact, thatfriends maybemore potent sources of influence than parents-their firstreaction isalmost alwaystingedwithanx iety and fear. The stereotype of the adolescent peergroup portrays it as an influence thatinevitably affectsteenagers forthe worse-that rempts them into trouble, steers them awayfrom theendeavors thatadults value, and co erces them toengage inrisky orillicit activities. In reality, thisview of theadolescent's socialworld isfartoo simplistic.
True, therearepeers whoencourage theirfriends tobe sexually activeorto experiment withdrugs, and there aresome whocajole theirclassmates into cutting class or skipping school.Butresearch tells us thatthere arealso teenagers who put pressure on theirfriends tostay away from drugs, remain committed toschool, and refrain from sex, and thatthese peers canbejust as powerful intheir influence on other teenagers as peerswhoaretrying to steer other students inthe wrong direction. Inother words, although peer pressure inearly adolescence isagiven, harmfUL peer pressure isnot.
Friends caninfluence eachother's schoolperformance positivelyorneg- 142 THE POWER OF PEERS atively. &; adultssuspect, anadolescent whoseftiends dispatage schoolsuc cess willbesteered awayfromscholastic achievement. All otherthings being eq ual,a Bstudent whosefriends areCstudents orwho arealienated from school, willusually see hisorher own grades decline overtime as aresult of associating withthese less academically orientedpeers.But,bythe same token, anadolescent whosefriends valuedoing wellinschool willbenefit by his orher contact withthese peers. That is, a Bstudent whosefriends areA students willimprove hisorher school performance overtime as aresult of these friendships. Althoughalladolescents willbeinfluenced bytheir friends, thespecific direction of influence dependsonwho one's friends are and whattheyvalue. Because of rhis, it is not enough todiscuss peerpres sure inthe abstract-as ifitwere amonolithic negativeinfluence on adoles cents'. behavior. Wemust alsoknow theclimate of anadolescent's peer crowd, thecharacteristics of hisorher close friends, and thevalues heldby the members of hisorher clique. IdentifYing anadolescent's closefriends andclique mares is simple enough; wecan just askteenagers totell us who their bestfriends are and whom theyhang around with.Butwhat about the more amorphous peer crowds?
How do we identifY the crowds thatinfluence youngsters' behavior? In ordertoanswer thisquestion in our research,weorganized, ineach school, aseries of small "focus groups" composed of students whorepre sented across section of thestudent body.Inthese focus groups, weasked students ro talk abour themain crowds intheir school andto identifY stu dents whowere exemplary members of each of thecrowds.
Although weconducted ourresearch in different parts of thecountry and in very different sorts of schools,wefound striking similarities inthe rypes of crowds thatwere identified ineach school. Let'slookfora moment atthis part of adolescent sociery. THE ADOLESCENT SOCIETY Allschools havecrowds thatemphasize socialstatus, socializing and popu larity, although inmost schools thisgroup of thesocially elite isdivided into two somewhat differentcrowds:"populars" (populariry-conscious students who have amoderately strongcommitment toacademic achievement but 143 BEYOND THE CLASSROOM reportmoderate involvement indelinquent behavior and illicitdrug use) and "jocks" (whoarequite similar to"populars," but less academically ori ented, and not as involved indrug use-except for alcohol, whichtheyoften use toexcess). Counterbalanced againsttheseelitecrowds areone ormore alienated crowds-which are referred toby students as "druggies," "burnouts," "greasers,"andthelike. Along withheavy involvement indrug use and delinquent activities,members of thesecrowds tendtobe inatten tive toschoolwork and often hostile toward teachers andother school per sonnel. Finally,nearlyallschools havealarge, amorphous crowd,consisting of "average," "norma!," or "in-between" students whodonot distinguish themselves in anyparticular area-including school performance. In addition tothese elire,alienated, and average crowds, schoolsalso typically haveatleasr onegroup of high achievers-sa-called "brains" or "intellectuals"-students who thriveonacademics, fotgecloserelations with school staff, and avoiddrugs anddeviant activities. Mostschools also feature avety small, socially inept crowd-"loners" or"nerds," as they ate often called-whose members aregenerally low in socialstatus and,conse quently, self-esteem.
Although thenames of these crowds mayvaty from school toschool, or region toregion (e.g.,"populars" mightbecalled "preppies," "stuck-ups," or "'''''d ." 'hb alld"c 1"" ") C SOCles; rugglesmigt e c e rrea {S or stoners ,as Lar as we can tell,their existence is ubiquitous, atleast within publicschools. In racially mixed schools, wealso find crowds that aredefined primatily byethnicity, and only secondarily byother attributes. Forexample, someschools have crowds that ate characterized bystudents only as beingcomposed of Black students, or Vietnamese students, or Mexicanstudents, andsoon. Once we had assembled alist of crowds foreach of the schools wewere studying, weattempted tolocate eachstudent in thectowd structure of his at herschool. In order todo this, weasked adolescents ineach school to classifY other students, ratherthanthemselves, in theschool's crowds.Wein terviewed severalpaits of students ineach school and asked them totell us which crowd each of theitclassmates wasapart of (we prompted them with class lists at yearbook pictures). By tepeatingthisexercise acrossnumerous pairs of tatets, wewete ableto identifY most students' crowdaffiliation. Al though mostteenagers say that theythemselves ateunclassifiable, adoles- 144 THE POWER OF PEERS centshavesurprisingly littledifficulty in identifYing whichgroup(or groups) theirclassmates belong to. THE PREVAILING NORM: GETTING BY With thissocial mapinmind, then,whatdidour study tell us about the peer norms and standards operativewithinthetypical American school?
Let's begin bylooking atthe most common crowdsfoundinAmerican schools andwhat theystand for. As youwillsee,there isn'tmuch of aplace in the typical American highschool forstudents whoseprimary concern is academic excellence. The popularity-conscious, sociallyelitecrowds, whoseconcerns tend to revolvearound socializing, dating, and maintaining socialstatus among friends, accOunt forapproximarely 20percent of students inatypical high school. Students in thesecrowds maydowell enough toget bywithout ger ting into academic trouble, but theyrarely striveforscholastic excellence most of rheir grades are B's. Another 20percent of students belong to oneor more of the alienated crowds,whereidentities arecentered arounddrugs, drinking, delinquency, ordefiance; thesestudents areopenly hostile to aca demics-on average, theyearn C's.About 30 percent of students describe themselves as "average"-not especially opposed to academicpursuits,but not exactlystriving forsuccess, either;likerhose inthe social crowds, their grades hoveraround straight B's. Andbetween 10 and 15 percent of stu dents belong toacrowd defined byethnicity, althoughthisfigure variescon siderably fromschool to school, depending on theschool's ethnic composition. The extent to which members of ethnically definedpeer crowds areinvested in academics dependslargely on theparticular ethnic group inquestion, as I'll explain laterinthis chapter. What about rheexplicitly academically oriented crowds-the "brains," rhe "intellectuals," and so on? Despite thefacr rhar these students are en rolledinmore difficult, moredemanding courses-many of them ralee hon ors and advanced-placement courses-they maintain an A- averagein school grades. Burwhereas some70percent of students belong to one of the solid-B, popularity-conscious elites,one of thelow-achieving, alienared crowds, or to thelarge mass of "average" students, less than 5percent of all 145 BEYOND THE CLASSROOM students are members of ahigh-achieving crowd that defines itself mainly on the basis of academic excellence. Not only is there little room in most schools for the academically ori ented, there issubstantial peerpressure on students tounderachieve. Adults might think that virtuallyallteenagers would rather do well in school than do poorly, but our studies suggest that this isnot necessarily the case.Tobe sure, the prevailing expectation among American teenagers isthat one ought to avoidfailing in school and do what ittal,es to graduate. But our surveys indicate that among American teenagers,there iswidespread peerpressure not to do too well. Forexample: •Although mostadolescents saythat their friends believe thatit isim portant tograduate fromhighschool (73%)and ro go tocollege (46%), fewer (32%) say thattheir friends thinkit isimportant to get goodgrades or ro go ro "one ofrhe best colleges inthe U.S." (20%). Nearly as many (16%) say that their friends thinkir isimportanr thatthey be "willing to parry." •One out of every sixstudents deliberately hideshis or herintelligence and interest indoing wellwhile inclass because they are "worried what theirfriends mightthink." Onein five studenrs say their friends makefun of people whotry to do well inschool.
• More thanone-half of all studenrs saythey almosr neverdiscuss their schoolwork withtheir friends. Morethanone-quarter saythey have never studied withtheir friends. Onlyonein five has studied withhisorher friends morethanfivetimes during thepast school year. •We asked theadolescents inour survey to tell us which crowdtheir friends belonged to andwhich crowd they'dmostliketo be apart of. When askedwhich crowd theywould mostlike to belong to, five times as many students saythe "populars" or "jocks" as saythe "brains." Three times as many saythey would rather be "parryers" or "druggies" than "brains." Of all of thecrowds, the"brains" were least happy withwhothey are-nearly halfwished they were inadifferent crowd. PEER INFLUENCES ON ACHIEVEMENT Although theprevailing norm in most high schools is,evidently, to "get by, without showing 0[£" there arepockets within each school in which aca 146 THE POWER OF PEERS demicachievement is admired, andothers inwhich ir is actively discour aged. These cliques andcrowds thatdefine ayoungster's socialworld aresig nificant influences on thechild's academic performance, becauseeachcrowd has itsparticulat set of normative standardsandexpectations forachieve ment and behavior inschool, andbecause adolescents attempttoconform to the norms andexpectations of their friends. As aconsequence, anindi vidual student's schoolperformance willdepend inlatge measure onwhich crowd thestudent belongsto,and what thatcrowd's expectations are fat be haviorinschool. Simply put,given several adolescents of equal scholastic ability, thosewhoaremembets of intellectual cliquesandcrowds will achieve moreinschool thanthose whoaremembers of the socially elite cliques andcrowds, andboth groups of adolescents willoutperform those who atemembers of alienated crowds.
This seems reasonable enough, of course-it is hardlysurprising that membets of academically orientedcrowdsdobest inschool andmembers of alienated crowdsdoworst. Perhaps it ismerelythatstudents who choose to associate withbrainy classmates arethemselves moreacademically inclined, whereas thosethatselect friends fromthealienated crowdsatethemselves less oriented towardschool. Afterall,it is not as if adolescents are placed withindiffetent peergroups. How canwebecettain thatfriendships really affect students'schoolperfotmance, ratherthansimply reflect it. Do friends really influence each other-is itteally acase of "the company they keep" or isitsimply that"birds of afeather flocktogether"? By tracking students overathree-year period, we wereabletosee how they were doing inschool atthe beginning of thetime period, whichfriends they were spending timewith, andwhether theirschool performance and behavior changedovertime as aresult. By comparing theacademic careers of studentswhobegan highschool withequivalent grades, but whohaddif ferent sorts of friends duringtheschool years, we wereabletosee whether the type of ftiends thatadolescents haveactually makesadifference intheir school performance. The answer is thatitmost certainly does,especially intwo areas: aca demic performance anddelinquency. Youngsterswhosefriends weremore academically oriented-that is, whose friends hadhigher grades, spent mote timeonhomework, hadhigher educational aspirations,andwho were more involved inextracurricular activities-did better overthecourse ofhigh 147 BEYOND THE CLASSROOM schoolthanstudents whobegan school withsimilat tecotds but whohad less academically orientedfriends.Similarly, studentswhosefriends were more delinquent-who used more drugs and alcohol and whohadmore conduct problems-developed more problems themselves overtime than did adolescents whobegan thestudy withthesame behavior profile but who had friends whowete less delinquent. , These findings tell us, then,thatparents havelegitimate reason ro be concerned aboutthequalities and values of their children's friends,espe cially during earlyadolescence, whensusceptibility topeer influence runs strong. There isalsoreason tobe concerned aboutthecharacteristics of the crowd towhich anadolescent belongs,sinceourstudy found thatthisin fluence matters, too.Allother things beingequal, adolescents whoaremem bers of more academically otientedcrowdsdobetter inschool thanorher students, whereasthosewhoaremembers of more alienared crowdsdo worse and ate more likely toget into trouble. How large adifference dofriends mal
Let's begin bylooking atthe first question: How dostudents endup in aspecific nichewithin theadolescent society?
HOW ADOLESCENTS SORT THEMSELVES INTO CROWDS What isthesorting ptocess through whichsomeadolescents becomepart of the "brain" crowdandothers become "jocks"? What makes some students become "partyers" andothers "druggies"? Why doadolescents end upwith the particular circle of friendsthey have? The results of our studypointtothree sets of forces thatdetermine in which crowdanadolescent willendup: (I) theadolescent's personalityand interests, which in partaredetermined bythe way thestudent hasbeen raised byhis orher parents; (2) thetypes of peer crowds available tothat stu dent inhis orher particular school;and(3)the tactics thatparents use to "manage" rheirchild's friendships. Indescribing howthese threesets of forces worktogether, Ihave found ithelpful touse asort of astronautical metaphor thathasthree main parts: thelaunch, theterritoty, andthenavi gational plan. THE LAUNCH The firsrset of factors-the child's personality andinterests-refers tothe general direction inwhich thechild is"launched" as parents prepare tosend the youngster onajourney through adolescence.
When thechild is sixor seven, adolescence seemsadistant destination, butparents are already "aim ing" thechild toward certain goals-even if theythemselves are not fully aware of what theyaredoing. As aresult of this goal-setting, duringtheearly elementary schoolyears,asort of "launch ttajectory" is established forthe child, especially withregard toschool matters. Launching thechild on acer tain trajectory does not guarantee thatheorshe will reach aparticular des tination, buritdoes point thechild inageneral direction. At oneextreme arechildren whoarelaunched on aroute that isheaded in the general direction of educational excellence,instilledwirhvalues that 149 BEYOND THE CLASSROOM stressscholastic success, and whoareexpected tomake school achievement a top priority. In thechild's upbringing, traitslikeperseverance, achieve ment motivation, and responsibility areemphasized, and parents put into place highstandards forachievement. At the other extreme arechildren whose launch trajectoty does not aimthechild toward schoolsuccess. The childmaybeaimed toward adifferent goalor,more likely, toward nospecific goal at all. Socialization inthese households maybeoverly permissive orin consistent, and parental expectations andperformance standardsareun clear. Between thesetwoextremes areother trajectories, whichvaryin the degree towhich theypoint thechild toward schoolsuccess and inthe strength and importance of schooling as an activity. THE TERRITORY Because peersplaysuch animportant roleininfluencing children'sday-to day behaviors oncetheyreach adolescence, theterritOlY intowhich achild islaunched-that is, the particular types of peers and peercrowds heorshe islikely to run into-is as important as thelaunch trajectory on whichthe child isinitially placed. Once achild becomes involvedwithacertain group of friends, thesepeers begin tohave an effect onhis orher behavior. Tocon tinue theastronautical metaphor,wecan think of peer crowds as sorts of "planets" towardwhichthechild islaunched. Once achild endsupinthe "orbit" of a given peergroup, thepower of that group keepsthechild within its orbit and encourages thechild toadopt acertain set of behaviors and out looks. The longer achild orbits around acertain group of friends, thetighter the rein thegroup has on thechild's behavior, and the more established that behavior patternbecomes.
We saw this quite clearly whenwelooked atadolescent drug use. The mostimportant determinant of anadolescent's initialexperimentation with drugs-primarily alcohol and marijuana-is thehome environment. Spe cifically, adolescents aremore likely tobegin drinking and experiment with marijuana whentheycome fromhouseholds thatareexceptionally permis sive orinwhich theparents aredisengaged, andthey are less likely toexper iment withthese substances whentheirparents areauthoritative. This isnot very surprising. But it is thepeer group, and not thehome environment, that determines whetheranadolescent willprogress fromexperimentation '50 THE POWER OF PEERS withdrugs toregular use.Adolescent "experimenters" whohaddrug-using friends werefarmore likely tobecome regularusersthanwere "experi menters" whosefriends were not usingalcohol orother drugs. Inother words, parental permissiveness ordisengagement maylaunch anadolescent in the direction of drug-using peergroups, but whether drugusebecomes a part of the adolescent's regularpattern of behavior dependslargelyonthe peer group that he orshe joins. An adolescent fromapermissive familywho does not connect withadrug-using peergroup isunlikely toget into trou ble with drugs, despite thepermissive homeenvironment.
We can apply thesame logictounderstanding thedual roles of parents and peers inschool achievement. Parentsmaylaunch theirchild on anaca demic trajectory, but ifthere isnoacademically orientedcrowdforthat stu dent toconnect upwith, thelaunching willhave little effect. On theother hand, ifthere are only academically orientedpeercrowds in agiven setting, what parents doathome, in terms of the trajectory theylaunch theirchild on, will mal
To acertain extent, then,theimpact of thehome environment on the adolescent's behaviorwilldepend toalarge measure onthe peer groups that inhabit theadolescent's socialworld. Knowing thishelps us understand why the impact of parents on their children's achievement, whilesignificant, is not all-powerful. Parentsmaysocialize achild inacertain direction, but whether thatsocialization willbe successful-that is, whether theadolescent will actually reachthedesired goal-will also depend onthe peer influences he orshe enCOunters duringthejourney. This,inturn, willdepend onwhich peers areavailable forthe child toassociate with, and howtheadolescent navigates amongthedifferent circles of classmates withinhisorher school. NAVIGATING THROUGH THE ADOLESCENT SOCIETY Although it istruethatparents have less of adirect effectontheir children during adolescence thanduring childhood, ourstudies showthatparents can have apowerful indirect effect bysteering thechild toward somepeer groups and away from others. Through suchpiloting, parentscanexert some control overthetypes of peers theirchild spends timewith and,con sequently, overthepeer influences to which theirchild isexposed. 151 BEYOND THE CLASSROOM Therearetwo chief ways inwhich parents dorhis. One way is byar tempring to exert some control overthechild's choice of friends andout-of school activiries. This, of course, is difficultoncethechild hasentered adolescence, but itisnot impossible. Indeed,incontrast to the widely held view thatthere is little parents candowhen itcomes to influencing their child's choice of friends, wefind inour research thatfamilies varyagreat deal in the extent towhich parents monitor andregulate theirchild's friend ships. Moreimportant, parentswhoexercise greatercontrol overwhich peers theirchildren spendtimewith havechildren whodobetter in school and who are less likely to getinto trouble.
A second, andpotentially morepowerful, wayinwhich parents influ encetheir child's choice of friends isbyselecting thesettings inwhich their child willspend time-by living inone neighborhood as opposed to an other, bychoosing oneschool overanother, andbyinvolving thechild in certain rypes of after-school andweekend activities. This isreallyamatter of playing thepercentages, ttyingtomaximize thenumber of "good" peersa child comes intocontact withandminimizing thenumber of "bad" peerin fluences in rhe child's environment. When parents maximize thenumber of good peers intheir child's environment, rheyare less likely to need toexert control overtheir child's choice of specific friends, sincetheodds aregood that bychance alonethechild willcome intocontact withpeers whoare likely to be positive influences onhis orher development. Inessence, al though parents can'tchoose theirchildren's friends,theycaninfluence rheir child's choices bydefining theavailable pool of possiblepeers. One waythat parents candothis is by malring sure thattheir child's world is adequately populated withother children whothemselves havebeen raised inauthori tative families-families that, as Ihave explained, tend to producethemost well-adjusted children. WHY NEIGHBORHOODS MATTER Aclear example of thisphenomenon wasrevealed when we lookedathow neighborhoods affectchildren's behaviorandperformance inschool. Be cause our sample was so large, we wereable to compare adolescents who went to the same school but lived in different neighborhoods withinthe 152 THE POWER OF PEERS school district.lWhat wefound was that adolescents who live inneighbor hoods inwhich alarge proportion of families areauthoritative performbet ter inschool andare less likely toget into trouble thanadolescents who come fromidentical home environments-and who go tothe same school-but who live inneighborhoods inwhich thepopulation of author itative families ismuchlower. \ Why wouldgrowing upinaparticular neighborhood matrer,aboveand beyond theinfluence of the home andschool environment? Becausewhere a family lives affects thepool of peers theirchild comes intocontact with, and this, inturn, influences thechild's behavior. If youareagood parent and you live inaneighborhood withother goodparents, chancesarethat the lessons youhave tried so hardtoteach yourchild athome willberein forced whenyourchild comes intocontact withother children, andother adults, inthe community.
I want tostress herethatchoosing a"good" neighborhood inwhich to settle andraise afamily is not thesame as choosing anaffiuent neighbor hood. Although, as ageneral rule,theprevalence of authoritative parenting rises, andthat of disengaged parenting falls, as onemoves upthe socioeco nomic ladder, parenting styleandfamily income arebyno means perfectly correlated. Withinanyparticular social class range,therefore, there is con siderable variability inhow children areraised, andit ispossible bothfora well-to-do familytoend upinaterrible neighborhood (so far as thequality of parenting isconcerned) andforafamily of more modest meanstoend up in aneighborhood thatprovides awonderful socialenvironment forchil dren. What specific factorsincrease thelikelihood thatagiven neighborhood will provide agood social environment forthe child? Basedonour research, parents shouldlookforahigh level of parental involvement inthe local schools, ahigh level of parental participation inorganized activitiesserving children (sportsprograms, artsprograms, etc.),andahigh level of parental monitoring andsupervision of children. Our research showsthatchildren who grow upinsuch neighborhoods farebetter. Even iftheir own parents are not especially involvedinschool, activeintheir child's life,orvigilant su pervisors of their child's activities, thechildren benefitfromcontact with peers whose parents havethese characteristics. Andforparents who are in- 153 BEYOND THE CLASSROOM valved,active, and vigilant, livinginacommunity inwhich there isahigh proportion oflike-minded parents gives anadded boost to thebeneficial ef fects of an authoritative homeenvironment. SCHOOL CHOiCE: CHOOSING SCHOOLS OR CHOOSING PEER GROUPS? Our findings on theimportance of peers as influences onadolescent achievement and behavior areinteresting inlight of current debates about school choice. Mostdebates aboutproposals toincrease parents'choice of schools-for example, tuitiontaxcredits, givingparents vouchers to usefor private schooltuition, orpermitting parents to chooseamong severalpublic schools withintheir area-have focused on theimpact of these policies on schools' practices. Proponents of schoolchoice have argued thatpermitting parents to choose among schools-either among privateandpublic schools, or among onlypublic schools-will enhance schoolquality because itwill force schools tocompete witheachother. Opponents of school choicepro grams contend thatproviding parentswithvouchers touse forprivate schools willundermine thequality of public schools (bysiphoning resources out of schools'coffers and directlyintoparents' hands). In addition, oppo' nentsargue thatencouraging competitionamongpublicschools willulti mately widenthegap between goodschools and badones, sincethegood ones, overtime, willbecome moreselective and attract better and better students, whilethebad ones willultimately havetoserve alarger and larger proportion of ill-prepared students. An importantpart of thecase made byschool choiceproponents isthe observation thatstudents attending privateschools outperform thosein public schools. One of themost important elements of thisargument isthat the observed achievement differencebetweenprivate and publicschool stu dents persists evenafter talcing intoaccount thedifferent familyback grounds of these twogroups of students (as onewould expect, privateschool students, onaverage, comefrommore affluent families). The usualinter pretation of the achievement differentialbetweenstudents fromprivate schools and those frompublic ones isnot,then, thatthestudents attending the two lcinds of schools areinherently differentfromeachother, but that 154 THE POWER OF PEERS privateschools havehigher standards, morerigorous requirements, and morestrenuous disciplinary practices. AB aconsequence, it is argued,stu dentsattending privateschools takemore demanding courses,workharder, behave themselves better,and,ultimately, learnmore inschool andperform better on achievement resrs. An equallyplausible alternative, though, isthattheachievement differ ential between publicandprivate highschool students isnot due ro differ ences between theitfamilies or between theirschools, but ro diffetences between theirpeergroups. Compatisons that talce familybackground into account donot control forthe mote intangible facrorsthatdistinguish stu dents whoatesent ro private schoolfromthose whoattend publicschool, such as motivation, self-reliance, andtheknowledge thatone's parents have made afinancial sacrificeforone's education. Inall likelihood, studentswho attend private andparochial schoolsareexposed ro ahigher proportion of peerswithhigheducational aspirationsandgood study habits, andthis ex posure positively affectstheirownbehavior, entirelyindependent of thein structional climate of theschool. Although our studydid not include students fromprivate schools, our findings on theimportance of peers, as well as awealth of research onthe minimal effects of schooldifferences on student achievement, are consistent withthisinterpretation. When parents arechoosing aschool, theyare not only choosing aprin cipal, aschool facility, andafaculty. They are also choosing classmates-and potential friends-for their child. Our study suggests thatitmay bethis as pect of school choice-the choice of apeer group-that may bethe most important, andthat parents shouldkeepthisin mind when selecting a school fortheir child. ETHNIC DIFFERENCES IN ACHIEVEMENT: THE IMPORTANCE OF PEERS Our findings on theimportance of peers inadolescent achievement also bear directly on thequestion of ethnic differences in school performance.
Remember, one of thepuzzles weencountered whenwelooked atthe role of thefamily inschool achievement wasrhat Asian parents didnot, on the surface, appear ro bedoing anything particularly specialthatwould account 155 BEYOND THE CLASSROOM fortheir children's remarkable success,norwere Black parents doing anye thingnoteworthy thatwould explain theirchildren's relativelyweakerper formance. Overall,Asianstudents in ourstudy wereperforming betterthan we would expectonthe basis of their parents' practices, andBlack students were performing worse.Something inAsian students' livesprotects them, even ifthey areexposed to less than perfect parenting, whilesomething in Black students' livesundermines thepositive effects of parental involvement and authoritativeness.
According to our study, this"something" is thepeer group. One clear reason forAsian students' success is thatAsian students arefarmore likely than others tohave friends whoplace agreat deal of emphasis on academic achievement. Asian-American studentsare,ingeneral, significantly more likely to say that their friends believeit isimportant to dowell inschool, and significantly less likely than other students tosay that their friends placea premium on having an active sociallife. Not surprisingly, Asianstudents are the most likely to saythat they work hardinschool tokeep upwith their friends.
Asian students' descriptions of theirfriends as hardworking andaca demically orientedarecorroborated byinformation we gathered indepen dently from thefriends themselves. You mayrecall thatone of theunique features of our study was our ability tomatch information providedbyado lescents withinformation provideddirectlybytheir friends. Thisprovided us withamore accurate assessment of eachadolescent's socialnetwork than would havebeen possible had we beenforced todepend onadolescents' per ceptions of their friends' behavior, sincesuchperceptions canbeerroneous (like adults, adolescents tend to overstate thedegree of similarity thatexists between theirfriends andthemselves). When welook atfriends' activitypatterns foradolescents fromdifferent ethnic groups, we see quiteclearly thatthefriends with whom Asianstu dents socialize placearelatively greateremphasis onacademics thanother students do,whereas theopposite istrue forBlack andHispanic teenagers.
Specifically, Asianstudents' friendshavehigher performance standards(that is, theyhold tougher standards forwhat grades areacceptable), spendmore time onhomework, aremore committed toeducation, andearn consider ably higher grades in school. BlackandHispanic students'friendsearnlower 156 THE POWER OF PEERS grades,spend less timeontheir studies, and havesubstantially lowerperfor mance standards. Whitestudents' friends fall somewhere betweenthesetwo extremes on these various indicators. When Ifirst sawthese findings, my presumption wasthat they were due entirely toracial segregation inadolescent peergroups. In other words, if Asian students areperforming better in school'than otherstudents, and Black and Hispanic studentsworse, and if peergroups areconstituted mainly alongethnic lines,itnecessarily followsthatAsian students willhave friends whoaredoing betterinschool, and Black and Hispanic studentswill have friends whoaredoing morepoorly. Itturns out thatthesegregation argument isonlypartly true.While it is certainly thecase thatadolescent peergroups arecharacterized byahigh de gree of ethnic segregation-about 80 percent of White and Black students, and more thanhalf of Asian and Hispanic studentshavebestfriends from the same ethnic group-there aresufficient numbers of cross-racial friend ships in any school toask whether thepattern described aboveholdsforstu dents whotravel inintegrated circles. The answer isthatitdoes, atleast for the most part.Even ifwelook solely atyoungsters whosebestfriends are from adifferent ethnicbackground, westill find thatAsian students' friends place agreater emphasis on doingwellinschool, and Black andHispanic students' friends,relatively less. Onceagain, White students fall somewhere in between.
Peer pressure amongAsianstudents and theirfriends todo well in school isso strong thatanydeficiencies in thehome environment-for ex ample, parenting that iseithertooauthoritarian oremotionally distant-are rendered almostunimportant. It is, of course, truethatAsian students from authoritative homesperform better in schoolthanthose fromdisengaged ones. ButanAsian student whocomes fromaless-than-optimal homeenvi ronment is likely to be "saved" fromacademic failurebyfalling inwith friends whovalue academic excellence and providethenecessary supportfor achievement. Why is it so likelythatanAsian student will fall intoanacademically oriented peercrowd and benefit fromitsinfluence? Ironically,Asianstudent success isat least partly aby-product of the fact that adolescents do not have equal access to different peergroups inAmerican highschools. Asianstu- 157 BEYOND THE CLASSROOM dentsare"permitted" ro join intellectual crowds,likethe"brains," but the more socially oriented crowds-the "populars," "jocks," and "parryers"-are far less open ro them.Forexample, whereas 37 percent of the White stu dents in our sample weremembers of one of these threesocially oriented crowds, only14percent of the Asian students were-even though more than 20percent of the Asian students saidthey wished they could bemem bers of these crowds (slightly less thanone-third of the White students as pired tomembership inone of these crowds). In essence, atleast some Asian students whowould like ro bemembers of nonacademically oriented crowds aredenied membership in them.
A similar argument hasbeen advanced byseveral Asiansocialscientists in explaining theextraordinary success of Asian-American students.They have noted thatacademic success isone of thefew routes tosocial mobiliry open toAsians inAmerican culture-think for a moment of the relative ab sence of Asian-American entertainers,athletes,politicians, and so on.For Asian youngsters, whoseemost nonacademic pathwaystosuccess blocked off, they have "nochoice" but ro apply themselves in school. This is why Asian students are so much morelikely than other youngsters ro subscribe ro thebelief thatacademic failurewillbring terrible consequences. When individuals believethatthere arefew opportunities ro success through routes other than education, doingwellinschool becomes thatmuch moreim portant. Because Asianstudents finditmore difficult than White students ro breal{ inrothemore socially oriented crowds,theydrift roward academically focused peergroups whosemembers value and encourage scholasticsuccess. The result of thisdrift isthat alarge number of Asian students, eventhose who are less academically talentedthantheir peers, end upincrowds that are highly oriented towardsuccess in theclassroom. Onceinthese crowds, Asian students benefittremendously fromthenetwork of academically ori ented peers. Indeed, one of thestriking features of Asianstudent friendships is howfrequently theyturn to each other foracademic assistance andcon sultation. The opposite is true forBlack andLatino srudents, whoarefarmore likely thanother students to findthemselves in peer groups thatactually de value academic accomplishment. Indeed,peerpressure amongBlack and 158 THE POWER OF PEERS Latinostudents not toexcel inschool isso strong inmany communities even among middle-class adolescents-that manypositive stepsthatBlack and Latino patents havetaken tofacilitate theitchildten's schoolsuccess are undermined. In essence, much of thegood work thatBlack and Latino par ents aredoing athome isbeing undone bycountervailing pressuresintheir youngsters' peergroups.
As aconsequence, parentaleffortsinthese ethnic groups do not havethepayoff thatwewould expect. This is true not onlyinracially integtated schools, but insegregated schools as well. In one well-known study of anall-black, inner-city high school, forexample, theresearchers foundthatstudents whotried todo well in school wereteased and openly ostracized bytheir peers for"acting White." Students werecriticized-accused of acting as iftheywere "better" than their peers-if rheyearned goodgrades, exerted effortinclass, orat tempted toplease theirteachers. Thosewhowished todo well academically wete forced tohide their success andtodevelop othermeans of maintaining theit popularity amongclassmates in ordertocompensate forbeing good students, such as clowning aroundin class orexcelling insome athletic ac tivity. Why would Black and Latino peergroups demean academic success?
In many minority peergroups, scholastic success is equatedwith"selling out" one's cultural identity) as somesort of surrender tothe control of White, middle-class America.
I found this so interesting thatIasked anextremely brighrAfrican American undergraduate inone of my seminars atTemple University, who was familiar with our research, tohelp mebetter undersrand rhisphenome non. The student saidthatthefinding rangtrueforher. Shehad been raised in dire poverty withininner-city Washington, D.C., and she was theonly one of her school friends tohave made itout of theghetto; as she explained, all of her former schoolmates wereeither on drugs, injail, onwelfare, or raising aninfant. Shewas torn about where shewould settleaftergraduat ing from college; thepull toreturn toher home community was verystrong, but shefelt that she'could nor face herformer friends. Whenever shere turned homeduring schoolvacations, she was tauntedforthinking too highly of herself and reased for not yethaving givenbirthtoachild. Shesaid that thepressure herfriends put onher over theyears todrop out of college and return toher roots was enormous. Infact, shesaid, herfriends inti- 159 BEYOND THE CLASSROOM matedthattheonly teason shehad gone off tocollege and avoided eatly pregnancy wasbecause shewas nOt physically attractiveenough to interesta man. Why is succeeding inschool equated insome circles with"acting "''hite'' or"selling out"? As Signithia Fordham and John Ogbu,twoAfrican American socialscientists whohave studied thisphenomenon explain:
[W]hite Ameticans traditionally refused to acknowledge rhatblack Amer icans are capable of intellectual achievement, and ... blackAmericans subsequently began to doubttheir own intellectual ability, began to define academic success as whitepeople's prerogative, andbegan to discourage their peers, perhaps unconsciously, fromemulating whitepeople inaca demic striving, i.e., from "actingwhite. One of my colleagues atthe University of Georgia, LayliPhillips, points out thatthis message-that academic success issomehow incompatible with a healthy Black identity-is perpetuated byamass media thatemphasizes and glorifies low-income African-American peerculture, malcingitattrac tive even tomiddle-class African-American youngsters.African-American parents whowant theirchildren to succeed inschool are not onlybattling the force of theBlack peerculture (whichinmany circles demeans academic success), but are fighting adifficult battleagainst thevery powerful images of anti-intellectual Blackyouth portrayed as normative in music, movies, and television.
Weheard variations onthe "acting White" theme many,manytimes overthecourse of our interviews withhighschool students. The sadtruth is that many students, and many Blackstudents inparticular, areforced to choose between doingwell in school and having friends. Although thereare crowds withineachhigh school inwhich academic success isvaluedandin which successful studentsarerespected, thesecrowds tend to bedominated by W'hite students, and peer groups in American highschools are so ethni cally segregated thatit isextremely difficultforBlack and Latino students to jointhese crowds. Thus,inmany schools, there isanear-complete absence of identifiablepeergroups thatrespect and encourage academic success and aregenuinely open to BlackandLatino students. As aconsequence, it isfar' 160 THE POWER OF' PEERS moredifficult foratalented Mrican-American studentthanit isforacom parably skilledAsian at White student tofind thenecessary peersupport for achievement. Among thehigh-achieving Blackstudents inour sample, forexample, only 2percent saidtheir friends weremembers of the"brain" crowd, as op posed to8percent of the White students and10percent of the Asian stu dents withthesame grades inschool. Interestingly, theproportion of high-achieving Blackstudents whosaidthey wished they were members of the"brain" crowd(6percent) wasabout thesame as itwas for' White stu dents (5 percent), Thus,whilejust as many Blackstudents asWhite students aspire toward membership inthe "brain" crowd,membership inthis group. ismore opento White thantoBlack students. It is important tounderstand thatthepressure againstacademic excel lence that ispervasive withinBlackandLatino peer gtoUpS isnotunique to these ethnic groups, Rather,whatwesee inthese peergroups isanextreme case of whar exists within most White peergtoupS as well. As noted earlier, the prevailing norminmost adolescent peergroups is one of "getting by without showing off"-doing whatittakes toavoid getting intotrouble in school, but at the same timeshunning academicexcellence. The chiefdif ference appears tobe not in the different ethnicgroups' avoidance of excel lence-this iscommon amongall but theAsian youngsters-but in how the different ethnicgroups defineacademic "trouble," We measured students'perception of this"trouble threshold" byasking them wharthelowest grade was thatthey could receive without theirparents getting angry, The students' answerstothis question confirmed oursuspi cion: Among BlackandLatino students, not untiltheirgrades dipped below a C- did these adolescents perceivethatthey would getinto trouble, Among White students, however,theaverage "trouble threshold" was oneentire let ter grade higher-somewhere between a Band a C. And among Asianstu dents, theaverage 'gradebelowwhich students expected rheirparents to become angrywasanastounding A-! One reasonforthe relatively poorer school performance of BlackandLatino students, then, is thatthese stu dents typically havedifferent definitions of "poor"grades, relative totheir White andAsian counterparts, And because peercrowds tendtobe ethni cally segregated, differentnormative standardsdevelopwithinBlackand 161 BEYOND THE CLASSROOM Latinopeergroups rhaninother crowds. Conversely, onereason forthe re markable success of Asian students isthatthey have amuch stricter, less for giving definition of academic failurethantheir Black, White, and Latino peers, and this definition shapespeernorms. Our findings suggest,then,thatalarge part of ethnic differences in high school achievement does not derivefromdifferences inthe ways in which parents fromdifferent ethnicgroups raisetheir children-that is, the "launch" theygetfrom thehome environment-but come instead fromdif ferences inthe peer environments-the "territory"-that youngsters from different backgrounds encounter. At atime in development whenchildren are especially susceptible tothe power of peer influence, thecircle of friends an adolescent canchoose frommay malce allthe difference betweenexcel lent andmediocre schoolperformance. 152 NOTES 119 MetLifesurvey.Metropolitan Life Survey of theAmerican Teacher, Violence inAmericas Public Schools: The Family Perspective (New York: MetLife, 1994). CHAPTER 7: THE HOME ENVIRONMENT OF ACADEMICALLY SUCCESSFUL STUDENTS PAGE 125 parental involvement wotks:Wendy Grolnick andMarcia Slowiaczek, "Par ents' Involvement inChildren's Schooling: AMultidimensional Concep tualization andMotivational Model," ChildDevelopment 64(1994):
237-52; David Stevenson andDavid Baker,"TheFamily-School Rela tion andtheChild's SchoolPerformance," ChildDevelopment 58 (1987):
1348-57.
126 parents of successful students: See alsoAnnette Lareau, Home Advantage: So cial Class and Parental Intervention in Elementary Education (New York: FalmerPress, 1989). 129American parents'involvement: SusanChira, "Parents, TakeLess of aRole as Pupils Age," The New York Times, September 5, 1994.
129 drop-off inparental involvement: StevensonandStigler, Learning Gap. CHAPTER S: THE POWER OF PEERS PAGE 139 The adolescent's socialworld: B. Bradford Brown,"PeerGroups," in S. Feld man andG.Elliott, eds., At the Threshold' The Developing Adolescent (Cambridge: HarvardUniversity Press, 1990),171-96. 141 changes in.individuals' susceptibility topeer pressure: Laurence Steinberg and Susan Silverberg, "TheVicissitudes of Autonomy inEarly Adoles cence," Child Development 57(1986): 841-51.
142 Friends can inllu~nce eachother's schoolperformance: Joyce L. Epstein, "The Inflllenc~ of Friends on Achievement andAffective Outcomes," in J. Epsteinand N. Karwei:, eds., Friends in School (New York: Academic Press, 1983),177-200.
143 All schools havecrowds: Btown,"PeerGroups." 149 three sets of forces: B. Bradford Btownetal., "Parenting PracticesandPeer GtoUP Affiliation inAdolescence," Child Development 64(1993):
467-82. 207 NOTES 152 how neighborhoods affecrchildren's behavior: Anne Fletcher et al., "The Company They Keep:Relation ofAdolescents' Adjustment and Behavior toTheir Friends' Perceptions ofAuthoritative Parentinginthe Social Net work," Developmental Psychology 31 (1995): 300-310; Nancy Darling and Laurence Steinberg, "Community Influences on AdolescentAchieve ment and Deviance," inG. Duncan, J.Brooks-Gunn, and L. Aber, eds., Community Influences on Childand Adolescent Development (New York:
Russell SageFoundation, inpress).
158 one of the few routes to social mobility: Sue and Okazaki,"Asian-American Educational Achievement." 159 students who tried to do well in school:Signithia Fordham and John Ogbu, "BlackStudents' SchoolSuccess: Coping with theBurden of 'Acting White,'" Urban Review 18 (1986): 176-206; see also Signithia Fordham, "Racelessness as aFactorinBlack Students' SchoolSuccess: Pragmatic Strategy or PyrrhicVictory?" Harvard Educational Review 58 (1988): 54-84. 160 "[W]hite Americans tradirionally refused":Fordham and Ogbu, "BlackStu dents' School Success," 177. CHAPTER 9: ALL WORK AND ALL PLAY MAKES JACK A DUMB BOY PAGE 165 The widespread employment of American high schoolstudents: SeeLau rence Steinberg and Elizabeth Cauffman, "The Impact of School-Year Employment on Adolescent Development," in R.Vasta, ed., Annals of Child Development, vol. 11 (London: JessicaKingsley Publishers, 1995), 131-66; EllenGreenberger and Laurence Steinberg, When Teenagers Work: The Psychological and Social Costs ojAdolescent Employment (New York: BasicBooks, 1986).
171 working teenagers get less rest. Mary Carskadon, Mancuso,J., and Rosekind, M.,"Impact of Part-time Employment on Adolescent SleepPatterns," Sleep Research 18(1989): 114.
172 The United States is the only country: Greenberger and Steinberg, When Teenagers Work. 174participation invarious extracurricular activities:Datafrom Laura Berk, "The Extracurriculum," in P.Jackson,ed., Handbook ofResearch on Cur riculum (NewYork:Macmillan, 1992). 20B
