cultural capital and school success: the impact of status culture participation on the grades of...

8/12/2019 Cultural Capital and School Success: The Impact of Status Culture Participation on the Grades of U.S. High School

1/14

Cultural Capital and School Success: The Impact of Status Culture Participation on the Gradesof U.S. High School StudentsAuthor(s): Paul DiMaggioSource: American Sociological Review, Vol. 47, No. 2 (Apr., 1982), pp. 189-201Published by: American Sociological AssociationStable URL: http://www.jstor.org/stable/2094962.

Accessed: 03/06/2014 05:16

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

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

.

American Sociological Associationis collaborating with JSTOR to digitize, preserve and extend access to

American Sociological Review.

http://www.jstor.org

This content downloaded from 188.173.197.158 on Tue, 3 Jun 2014 05:16:04 AMAll use subject to JSTOR Terms and Conditions
http://www.jstor.org/action/showPublisher?publisherCode=asahttp://www.jstor.org/stable/2094962?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/2094962?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=asa


2/14

CULTURAL CAPITAL AND SCHOOL SUCCESS:THE IMPACT OF STATUS CULTURE PARTICIPATIONON THE GRADES OF U.S.HIGH SCHOOL STUDENTS*PAUL DIMAGGIO

Yale University

Ethnographers and other students of interaction have documented the impact of statusfactors on students' success in school. Yet survey research data consistently show theabsence of family background measure effects on high school grades. It is argued thatconventional measures of family background fail to capture those cultural elements ofstatus that make a difference in school interactions. Drawing on Weber's work on statusgroups and status cultures, and on Bourdieu's work on cultural capital, this paper reportsthe findings of an effort to assess the impact of one component of 'status cultureparticipation-cultural capital-on one aspect of life chances-students' high schoolgrades. A composite measure of cultural capital has a significant impact on grades,controlling for family background and measured ability. The pattern of relationships,however, differs strikingly by gender.

It takes more than measured ability to dowell in school. From Warner et al. (1944) andHollingshead (1949) to Coleman (1961) andCicourel and Kitsuse (1963), ethnographershave chronicled the impact of class on almostevery aspect of the experience of Americanhigh school students. More recently,ethnomethodologists and constituent ethnog-raphers have documented the impact of cul-tural styles on students' relationships withcounselors (Erickson, 1975), test scores(Mehan, 1974), and classroom instruction(McDermott, 1977). Similarly, recent work inthe status attainment tradition finds that mea-sured intelligence explains no more than 15 to30 percent of the variation in students' highschool grades (Crouse et al., 1979; Sewell andHauser, 1975).At the same time, however, measures offamily socioeconomic status have been foundto have a negligible impact on grades when* Direct all correspondenceto: Paul DiMaggio,Departmentof Sociology, Yale University, NewHaven CT 06520.This paperhas benefited romthe adviceand crit-icism of Ann Swidler, Christopherencks, HarrisonWhite, SusanBartlett,RonaldBreiger,RandallCol-lins, Steven Brint, Kevin Dougherty, JeromeKarabel,David Kamens, David Karen, Sally Kil-gore, KatherineMcClelland,Susan Olzak, BernicePescosolido, Richard A. Peterson, Frank Romo,David Swartz,MichaelUseem, BlairWheaton,VeraZolberg,andtwo anonymous eviewers.Those listed

are emphaticallynot responsible or any remaininginadequacies.Computer and institutional supportfrom the HarvardUniversityand Yale UniversitySociology Departmentsand from Yale University'sInstitutionor Socialand PolicyStudiesis gratefullyacknowledged.

measuredability is controlled (Crouseet al.,1979; Sewell and Hauser, 1975; Portes andWilson, 1976). If measuredability is not thesole predictorof high schoolgradesandif mea-sureddifferences n familybackground re noteither,thento what do we attributevariationnstudentgrades?And how may we squareoursurveyresearch indingswith the observationsof ethnographersthat schools are places inwhich status and culture matter and par-ticularisticeakages Erickson,1975)abound?The answer may be that aspects of culturalstyle only loosely associated with such mea-sures of family backgroundas father's educa-tion or head of households occupationmakean importantdifference.'MaxWeber'snotionof status culture (1968) may be useful in thisregard.Webernoted thatelite statusgroups-collectivities bound togetherby personaltiesand a commonsense of honorbaseduponandreinforcedby sharedconventions-generate orappropriateas their own specific distinctivecultural traits, tastes, and styles. This sharedstatuscultureaidsgroupeffortsto monopolizefor the group as a whole scarce social, eco-nomic, and culturalresourcesby providing o-herence to existing social networks andfacilitating he developmentof comembership,respect, and affection out of which new net-worksare constructed.The contentof a statuscultureis arbitrary; tatus honor maybe con-I Afterassessingthe impactof demographicamily

background characteristics, Jencks and his col-leagues (1979) found that unmeasured ackgroundcharacteristics hatvary among amilieswithsimilardemographic rofiles seem to accountfor significantamounts of variance in occupational status andearnings p. 81).American Sociological Review 1982, Vol. 47 (April:189-201) 189

http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp


3/14

190 AMERICAN SOCIOLOGICAL REVIEWnectedwith any quality sharedby a plurality(Weber, 1968:part 2, ch. 9).The impact of a student'sculturalresourceson hisor her success in schoolhas been treatedexplicitly by Bourdieu (1977; Bourdieu andPasseron, 1977), Collins (1975, 1979), andothers.According o Bourdieu,schoolsrewardstudents on the basis of their culturalcapital,defined as instruments or the appropriationof symbolic wealth socially designated asworthyof being sought andpossessed (Bour-dieu, 1977). Teachers, it is argued,communi-cate more easily with studentswho participatein elite status cultures, give them more atten-tion and specialassistance, and perceive themas more intelligentor giftedthan studentswholack culturalcapital.If, indeed, participationn prestigiousstatuscultures representsa kind of culturalcapital,we would expect to find the following:

Hypothesis 1: Measures of cultural capitalare related to one another in a manner thatsuggests the existence of a coherent statusculture of which they are elements.Hypothesis 2: Cultural capital is positivelyrelated to school success, in particular, tohigh school grades.In much of both the ethnographicand theWeberian radition,status culturesare seen asresources used to promote intergenerationalstatus persistence; cultural capital is passeddownfrom upper-and upper-middle arents otheirchildren. If this is the case, thenHypothesis 3a: Cultural capital mediates therelationship between family background andschool outcomes

Whatis more, if, as Bourdieuhas argued,cul-turalcapital s inculcatednearlychildhoodandthe response of others to cultural capital ispredicated n parton the social positionof itspossessor, thenHypothesis 4a: Returns to cultural capitalare highest for students from high statusfamilies and least to students from lowstatus families.Let us call this, following Bourdieu,the cul-tural reproduction model.

By contrast,considerthe possibilitythat, asWeber predicted, the rise of the markethasseverely corroded the status order. Whileideal-typical tatusgroupsare welldefinedandstrictlydemarcated, n modernsocieties statuscultures are more diffuse and more looselybounded. As the potential membershipof astatus group becomes less knownto any singlemember,the importanceof the shared statusculture-those culturalcues thatdefinea per-son as a memberto other members-becomes

greater.2Individualsmay have a repertoireofstatus cultures that they draw on selectively(see, e.g., GumperzandHymes, 1972,on codeswitching). In such societies, status cultureparticipationmay be deployed unconsciouslyat the level of daily interaction.For this reason, it may be more accuratetospeak of status culture participation than ofstatus group membership, and to think ofstatus as a cultural process rather than as anattributeof individuals.A person who is athome in a prestigiousstatus culturecan dis-play tastes, styles, or understandingsthatserve as cultural resources, making communi-cation easier and indicating tatusgroupmem-bership(see Goffman,1951;Collins, 1981).Insuch a fluid world childhood experience andfamily background may only partially andmodestlydeterminea person'sstockof culturalcapital. Active participation in prestigiousstatus culturesmay be a practicaland usefulstrategy for low status students who aspiretowards upward mobility. By contrast, bothhigh statusstudents(who, presumably, eceiveculturalresourcesin the home)andnonmobilelow statusstudentsmayprefer o participatenadversarial youth subcultureswhile in highschool (Coleman, 1961).If this is the case, we would expect the fol-lowing:Hypothesis 3b: Cultural capital's impact onschool success is largely net that offamilybackground.Hypothesis 4b: Returns to cultural capitalare highest for students who are least ad-vantaged.

I will refer to this as the cultural mobilitymodel.Inthe sections thatfollow, I developa mea-sure of culturalcapital and then assess bothmodels usinganalysesof the relative mpactofmeasured family background, measuredability,and culturalcapitalon the gradesof anational cross-section of white U.S. highschool students.DATA AND MEASURESThe analysesreportedbelow were undertakenwith data from a random sampleof white re-spondentsto PROJECTTALENT. TALENThas by far the richest variety of measuresofculturalattitudes, nformation, ndactivitiesofapproximately wenty data sets reviewed for

2 Thus Coleman (1971) found that residents ofKansas City relied on much more indirectcues toassess the status of individuals han did citizens ofthe small towns studied by earliercommunityre-searchers.



4/14

CULTURAL CAPITAL AND SCHOOL SUCCESS 191this investigation.Its disadvantages ncludeitsuse of self-reportedgrades and the relativelylow correlationsamongits demographicmea-sures (Porter,1974),both of which shouldtendto depressthe relationships eportedhere. Thesample includes 1427 men and 1479 womenwho were in the eleventh grade in public,parochial, and private high schools in 1960,when they were surveyed. The sample isweighted to reflect a cross-section of whiteAmericanhigh school students.FollowingBourdieu,I measure high schoolstudents' cultural capital using self-reportsofinvolvement in art, music, and literature.While it would be preferable o ground thesemeasures in observed cultures of dominantstatusgroups,in the absenceof such a rigorousdata base, high culturalmeasuresrepresent hebest alternative or several reasons. First, art,classical music, and literaturerepresent themost popularof the prestigiousart forms. Pat-ternsof artmuseumvisitation,concert attend-ance, and literature reading in the UnitedStates are similar o those found in Franceandother western countries, with attendanceandreadingconcentrated n the upper middle andupper classes (DiMaggioand Useem, 1978).Second,to the extent that there is a commonculturalcurrencyamongAmericanelites, it in-volves at least a modest familiaritywith thearts and literature. Such preoccupationsasracquetball,wine, or ancienthistory are likelyto characterizesmaller,more localized statusgroups. Minimal amiliaritywith high culture,by contrast, transcendscleavages of age orregion. Third,art and musichave received rel-atively superficialattentionin the curriculaofAmericanhigh schools (Rindskopf, 1979). If,as Bourdieucontends, culturalcapitalconsistsof familiaritywithpreciselythose subjectsthatschools do not teach butthatelites value, thenincludingart and music permitsus to tap di-mensionsof culturalcapitalthat are inculcatedoutsideof the school. Finally,highculture s anelement of elite culture that school teachersappear o regardas legitimate.While Americanteachers are recruited largelyfrom the lowermiddle class (Lortie, 1975), they are overrep-resented n arts audiences in proportiono theirshare of the labor force) more stronglythananyothergroup(DiMaggioand Useem, 1978).The first step in the analysiswas to build ascale of measures of culturalcapital. Threekinds of measuresfrom PROJECTTALENTwere employed. 1) Attitude measures askedstudentsto rate their interest n specifiedartis-tic activities and occupationson a scale fromone to five. Unlike aspirationsquestions, theoccupational-interestquestions simply askedthe studentto ratethe attractivenessof a widerangeof careers. In addition,four inventories

or composites generated by TALENT are in-cluded among the attitude measures. Three in-ventories combine questions tapping, re-spectively, artistic, musical, and literary inter-ests.3 A fourth composite, the cultivatedself-image scale, is based on ten self-evaluationquestions, such as I enjoy beautiful things, orI am a cultured person. 2) Activities mea-sures are based on questions about the extentto which students have created visual arts,performed publically, attended arts events, orread literature. Except for the arts-attendancequestions, which could include school trips,these questions explicitly exclude activitiesundertaken for school course work.4 3) Infor-mation measures are based on TALENT-administered tests of information about lit-erature, music, and art. All these tests tappedfamiliarity, appreciation, and historical knowl-edge, rather than technical skills of the sortdeveloped in practice. In the music informationtest, for example, students were asked aboutfamous composers rather than about thestructure of tonic or dominant chords (MarionShaycroft, personal communication, June 27,1979).Fuller descriptions, including means andstandard deviations, of the measures used inthese analyses are provided in the Appendix.Boys' reported cultural interests and activitieswere low; girls' were moderately higher. Inshort, high culture was a minority taste inAmerican high schools in the 1960s.ANALYSISBoth the cultural reproduction model and thecultural mobility model yield the predictionthat separate measures of high cultural in-volvements should be positively correlatedwith one another. This prediction inheres inthe definition of cultural capital as the masteryof elements of a prestigious status culture.There is no a priori reason that students whocare about any one art form-art, music, orliterature-should be concerned about any

I These questionscover a broaderrange (andareless strictlyhigh cultural) hanthe specific interestquestionsincluded n Table 1. The Artistic InterestInventory, or example, includes responsesto ques-tions about interestin becominga decoratoror de-signer.4 It is possible that studentsfrom uppermiddle-class familieswere temptedto exaggeratethese re-ports, although the low correlations between theself-reportsand parentaleducationmake this seemsomewhat unlikely. False reports,if skewed in thisway, would attenuate the relationships betweencultural capital measures and grades once back-groundwas controlled.



5/14

192 AMERICAN SOCIOLOGICAL REVIEWTable 1. Median Correlations and Ranges (in parentheses) of Correlations Between Sets of CulturalMeasures

Interests Activities InformationMedian Range N* Median Range N Median Range N

InterestsW .71 (.47-.80) 9MB .45 (.32-.60) 27W .73 (.48-.82) 9F B .47 (.35-.63) 27AactivitiesMW .31 (.22-.40) 9 NAB .20 (.11-.38) 27 .20 (.12-.29) 6F W .33 (.21-.44) 9 NAB .18 (.06-.34) 27 .17 (.13-.22) 6InformationMW .14 (.06-.24) 9 .11 (.04-.14) 3 NAB .09 (-.01-.20) 18 .08 (-.03-.26) 9 .67 (.61-.68) 3W .29 (.18-.37) 9 .17 (.15-.28) 3 NAFB .26 (.09-.37) 18 .18 (.01-.31) 9 .66 (.60-.68) 3

* N refers to the number of correlations in each category.W = Correlations between variables in same discipline (art, music, or literature).B = Correlations between variables in different disciplines.NA = Not applicable (only one measure per discipline in this category).other. Indeed, psychologicalresearch ndicatesthat thepracticeof differentartforms drawsonsubstantiallydifferent cognitive skills (Wolfand Gardiner,1979). If we do find that mea-sures of involvement n differentartisticdisci-plines are related, we must look beyond psy-chologicalexplanationsfor the answer.The notion of status culture leads to justsuch an explanation. To the extent that art,music, and literature are part of a coherentstatus culture, we would expect students in-terested in music to be interested in literatureand art, and vice versa. Milieus that inculcatean interest in any single artisticdisciplinewillalso be likely to inculcate an interest in anyother high culture form. This expectation isparticularly trong for the attitude and infor-mation measures. Participationakes time, sostudents who value the arts may tend to spe-cialize in practicingone form, while maintain-inginterest in andknowledgeabout others. Ofparticular interest are correlations betweenmeasures of involvement n differentforms. Ifhigh cultural nvolvementsconstitute elementsin a coherent status culture, these between-discipline correlations should be consistentlyand significantlypositive.The findingsare summarized n Table 1. Asexpected, relationshipsamong highculture at-titude measuresare stronglypositive. The me-dianbetween-discipine orrelationsare .45and.47 for boys and girls, respectively. Relation-ships among culturalactivitiesare less strong,but all significantat p S .001. Cultural nfor-mationtest scores in different culturaldisci-plines are also stronglyassociated, even whenone controls for ability test scores in other

areas.5 Finally, students who engage in onekind of cultural activity are more likely thanothers to be interested in any other high cul-tural activity.It may be objected, however, that the posi-tive correlations simply indicate that all ofthese measures tap some underlying personal-ity attribute like creativity. Fortunately, TAL-ENT also reports activity measures for severalmiddlebrow cultural passtimes-photography,crafts, woodworking, and needlework. If highcultural involvements really constitute part of acoherent status culture, we would expect tofind these measures less strongly correlatedwith the high culture measures than the latterare with one another. Again, this is the case.Two-thirds and one-half of all correlations be-tween cultural attitudes and cultural activities,for boys and girls respectively, are greater thanor equal to .20. None of the correlations be-tween cultural attributes and middlebrowactivities reaches this level. Similar findingsemerge when we compare correlations be-tween pairs of cultural activities with correla-tions between cultural and middlebrow activi-ties. (Tables available upon request.)These findings are consistent, then, with thefirst proposition of each model, that different5 Ability is measured by the student's compositevocabulary core. Thecommonlyemployedmeasureof ability-the compositeof thirty-seven nformationtestsadministered yTALENT-is contaminated ythe inclusion herein of the cultural nformation estscores. Thus the vocabularycomposite, which cor-relatesat .94withthe information omposite, is usedas a proxy.



6/14


7/14

194 AMERICAN SOCIOLOGICAL REVIEWsists of nonhighculture creative pursuits, ex-cluding, for each gender, those in which thefewest studentsreportedparticipation. t alsoincludes drawing, which, for both genders,clusteredwith the crafts rather than with thearts activities.Each of these fourfactors representsa kindof culturalresource,and each representsa co-herentset of interrelated raits. Factor4 shouldhave little, if any, positive impacton students'grades, unless, perhaps, it represents a mea-sure of creativity. Factor1 should have less ofan impacton grades than factors 2 or 3, be-cause it measures attitudes ratherthan actualbehavior or information. f status culturepar-ticipation influences grades because studentsdisplay their knowledgein a mannerthat im-presses teachers or boosts their performanceon tests, we would expect factor 2 to have amajor mpact.Ifwe believe that cultural apitalconsists of a set of interests, dispositions, be-haviors, and styles that are learned andenacted socially, then we would predict thatfactor 3, cultural capital, would have thegreatest impact. This is the case not just be-cause factor 3 includes measures of high cul-turalactivities, but because the factor is theonly one thatcrosscutsquestion types as well.While all of these factors are, of course, onlyindirect measures of cultural resources thatstudents bringto interactionswith significantothers, it is predictedthat factor 3, culturalcapital, will have the greatest impact ongrades.To test the hypothesis that cultural capitalsignificantly influences grades, separate re-gressions were executed for male and femaleeleventh graders. Independent variables in-clude cultural actors 1, 3, and4, the student'sreportof his or her father'seducationalattain-ment, andthe student'scomposite score in theTALENTvocabulary ests.7 Factor2, CulturalInformation,was excluded from the analysisbecause of its high collinearity over .800) withthe composite abilitymeasure.The strengthof

this correlationsuggests that students'culturalinformationtest scores were largely deter-mined by some underlyingset of aptitudes,skills, and motivations hatlead studentsto dowell or poorly on tests. (Partialcorrelations,not reportedhere, indicatedthat the relation-ship between grades and culturalinformationtest scores largelyevaporatedwhen measuredability was controlled.)Dependent variables in the analysis werestudents' self-reportedgrades in English, inHistoryand Social Studies, and in Mathemat-ics, and a TALENTcompositeof self-reportedgradesin all subjects. The use of self-reportedgrades, with restricted distributions,can beexpectedto depressR2s ntheseanalyses;butitdoes not affect the utility of the datafor com-parisons of the relativeeffects of independentvariables (Picou and Carter, 1976). English,History, and Social Studies are subjects inwhichculturalcapitalcan be expected to makea difference;standardsare diffuse andevalua-tion is likely to be relatively subjective. Bycontrast,Mathematics equires he acquisitionof specific skills in the classroomsetting, andstudentsareevaluatedprimarily n the basisoftheirsuccess in generatingcorrectanswers tosets of problems.Thus Welch et al. (1980)re-port thatMathematicsachievement est scoresare muchmorestrongly nfluencedby yearsofschool subject matter instruction than areachievement est scores in EnglishandCivics.The regressionresultsaredisplayed nTable3. They provide strikingconfirmationof thehypothesis that cultural capital is positivelyrelatedto highschoolgrades.Standardized e-gressioncoefficientsforculturalcapital(factor3)aresignificantat p - .001forbothmalesandfemales for grades in all subjects but Mathe-matics,where effects are smaller,but still sig-nificant. For English, History/SocialStudies,andAll Grades,the impactof culturalcapital sof the sameorder of magnitudeas the effect ofmeasuredability. Cultural nterests (factor 1)andmiddlebrowactivity (factor4) have no sig-nificant impact on grades. As expected, theimpactof father'seducationis minimal.These results support the expectation ofboththe culturalreproduction ndthe culturalmobility models that participation n presti-giousstatuscultureshas a significantlypositiveimpact on grades. (Factors such as self-reportedgradesthatdepressthe R2sshouldnotaffect the relative weights of ability and cul-turalcapital. The latter, rather than the totalvariance explained, is the focus of thisanalysis.)Indeed, the magnitudeof the effectsrelative to those of ability was unexpectedlygreat. Thefindingsalso tendto disconfirm wopossiblealternativeexplanationsof the associ-ationbetweengrades andculturalmeasures.If

statusculture;arts attendance s the singlemeasuremost commonly used as a proxy for culturalcapitalin other work; literature eading s also a proxy thathas been used in the past; symphonyaudiencesarethe most elite of all arts audiences;performance sprobably elated o formal raining; nd each of thesevariables s relatively stronglyrelated to abilityandto familystatus.7 Father's education was included as the solebackground measure because analysis of similarTALENT samples (e.g. Crouseet al., 1979)showedlittle benefit from inclusion of multiplebackgroundmeasures as predictorsof grades, and because theculturalmeasures were morestronglyrelatedto fa-ther'sthanto mother'seducationfor both genders.



8/14

CULTURAL CAPITAL AND SCHOOL SUCCESS 195Table 3. Results of Regression of Grades on Ability (1), Father's Education (2), Cultural Capital (3),Middlebrow Activity (4), and Cultural Interests (5) for Male and Female Eleventh Graders

Reduction inbeta ofIncrease in father'sDependent R2 with education withVariable 1 2 3 4 5 R2 Vars. 3-5 Vars. 3-5MALESGrades in B .5078 .1158 2.2330 .1488 -.1256 .1228 .0286 .0202All Subjects s.e. .0623 .0971 .4330 .4830 .3830beta .2791*** .0409 .1706*** .0106 -.0113Grades in B .0493 .0223 .2730 -.0025 .0412 .1034 .0310 .0211English s.e. .0077 .0120 .0540 .0603 .0476beta .2255*** .0660 .1716*** -.0015 .0309Grades in B .0715 .0012 .2646 -.0059 .0259 .1279 .0254 .0193History s.e. .0080 .0125 .0561 .0627 .0495beta .3096*** .0033 .1577*** -.0033 .0184Grades in B .0510 .0189 .1685 .0133 -.0387 .0723 .0102 .0124

Mathematics s.e. .0082 .0128 .0575 .0625 .0507beta .2223*** .0531 .1011** .0075 -.0277N = 809FEMALESGrades in B .5988 .0939 2.4314 .4223 -.1250 .1897 .0338 .0297All Subjects s.e. .0575 .0857 .3887 .3925 .3361beta .3374*** .0345 .1901*** .0346 -.0124Grades in B .0602 .0150 .3412 -.0062 .0127 .1683 .0463 .0300English s.e. .0072 .0106 .0482 .0485 .0412beta .2797*** .0457 .2211*** -.0042 .0106Grades in B .0710 .0111 .3354 .0352 .0443 .1713 .0382 .0294History s.e. .0079 .0116 .0531 .0534 .0453beta .2991*** .0305 .1970*** .0218 .0333Grades in B .0653 -.0066 .1302 .0750 -.0671 .0857 .0079 .0091Mathematics s.e. .0082 .0121 .0551 .0555 .0471beta .2781*** -.0185 .0774* .0469 -.0511N = 917* p S .05, two-tailed.

** p S .01, two-tailed.*** p S .001, two-tailed.

these measures appedsome generaldimensionof academic achievement motivation, wewould expect the impact on grades in Mathe-maticsto equal those on other subjects. In fact,it does not. If the scores reflected some un-derlying dimension of creativity, factor 4would have a significant impact on grades;again, it does not.The findings provide limited support or theexpectationsof either model about the extentto which culturalcapitalmediatesthe relation-ship between family backgroundand schoolsuccess. While the inclusion of the culturalcapital measures does reducethe betas for fa-ther's educationby 20 to 80 percent,the origi-nal betas are so low that these figures aresomewhat trivial. The extent to which thesemeasuresaffect grades independentof the im-pact of father'seducationsquareswith the pre-dictions of the cultural mobilitymodel.The third propositionof the culturalrepro-duction model holds that returns to culturalcapitalwill be greaterfor students from high

status homes than from low status back-grounds. By contrast, the cultural mobilitymodel posits that the impactof culturalcapitalwillbe greateron thegradesof less advantagedyouth,for whomthe acquisitionand displayofprestigiousculturalresources may be a vitalpartof upwardmobility.The maleand female sampleswere each di-videdinto threegroups on the basis of father'seducation:sons and daughters respectively)ofcollege graduates,sons and daughtersof highschool graduateswho did not graduatefromcollege, and sons and daughtersof men whodid not hold high school diplomas. Separateregressionswere runon each of these six sub-samples.Tables4 and 5 indicate divergentresults formenand women.Amongwomen, the impactofculturalcapitalon all four grademeasuresrisesmonotonicallywith father'seducation.As thecultural reproductionmodel predicts, returnsto culturalcapital are greatestto women fromhigh status families and least to women from



9/14

196 AMERICAN SOCIOLOGICAL REVIEWTable 4. Results of Regressionsof Grades on Ability (1), CulturalAttitudes (2), CulturalCapital(3), andMiddlebrowActivity (4) for Male EleventhGraderswith Non-HighSchool Graduate,High SchoolGraduate,and College GraduateFathers

IncreaseinR2 withDependent Variable 1 2 3 4 R2 Vars. 2-4Males with Non-High School Graduate Fathers N=494

Grades in All Subjects B .4384 -.6358 2.0638 .7770 .0865 .0297s.e. .0724 .4530 .5302 .5828beta .2515*** -.0614 .1616*** .0583Grades in English B .0559 -.0743 .3026 .1267 .0998 .0404s.e. .0093 .0586 .0693 .0764beta .2604*** -.0577 .1885*** .0752Grades in History B .0632 -.0195 .2549 .0061 .0957 .0223s.e. .0098 .0618 .0732 .0806beta .2796*** -.0144 .1508*** .0034Grades in Mathematics B .0345 -.0560 .0490 .0492 .0257 .0025s.e. .0098 .0622 .0736 .0811beta .1572*** -.0426 .0299 .0286Males with High School Graduate Fathers N=298Grades in All Subjects B .5397 .6665 2.0630 -.3403 .1216 .0289s.e. .0986 .6205 .7008 .7307beta .2903*** .0596 .1568** -.0258Grades in English B .0504 .0172 .2396 -.0888 .0951 .0387s.e. .0129 .0804 .0910 .0954beta .2184*** .1264* .1482** -.0549Grades in History B .0857 -.0374 .3433 -.0570 .1760 .0422s.e. .0128 .0799 .0904 .0948beta .3568*** - .0264 .2039*** - .0338Grades in Mathematics B .0608 .0230 .2385 - .0731 .0809 .0189s.e. .0142 .0881 .0997 .1046beta .2424*** .0156 .1357* -.0415Males with College Graduate Fathers N= 130Grades in All Subjects B .4274 .2767 1.5914 .1906 .0702 .0178s.e. .1735 1.0292 1.0230 1.2428beta .2088* .0237 .1315 .0135Grades in English B .0242 .1253 .1774 -.0555 .0337 .0175s.e. .0214 .0493 .1243 .1553beta .1010 .0371 .1276 -.0337Grades in History B .0651 .2500 .0723 -.0339 .0959 .0285s.e. .0233 .1361 .1350 .1687beta .2421** .1676 .0463 -.0183Grades in Mathematics B .0625 -.1276 .0963 .1436 .0670 .0123s.e. .0233 .1362 .1351 .1688beta .2362** - .0867 .0625 .0787* p


10/14

CULTURAL CAPITAL AND SCHOOL SUCCESS 197Table 5. Results of Regressions of Grades on Ability (1), Cultural Attitudes (2), Cultural Capital (3), andMiddlebrow Activity (4) for Female Eleventh Graders with Non-High School Graduate, HighSchool Graduate, and College Graduate Fathers

Increase inR2 withDependent Variable 1 2 3 4 R2 Vars. 2-4Females with Non-High School Graduate Fathers N=582Grades in All Subjects B .5006 .3136 2.2030 .7400 .1453 .0310s.e. .0660 .3966 .5010 .4966beta .2930*** .0331 .1636*** .0614Grades in English B .0440 .0906 .3312 -.0195 .1148 .0452s.e. .0086 .0504 .0651 .0633beta .2093*** .0788 .2036*** - .0133Grades in History B .0654 .0958 .2638 .0213 .1358 .0272s.e. .0093 .0546 .0705 .0685beta .2840*** .0761 .1479*** .0145Grades in Mathematics B .0450 -.0693 -.0005 .1535 .0442 .0075s.e. .0098 .0577 .0745 .0724

beta .1943*** -.0548 -.0003 .0951*Females with High School Graduate Fathers N=342Grades in All Subjects B .6216 -.8526 2.7058 .3466 .1776 .0494s.e. .0980 .5542 .6034 .6064beta .3266*** - .0844 .2238*** .0301Grades in English B .0629 -.0831 .3641 -.0114 .1755 .0659s.e. .0115 .0657 .0715 .0722beta .2906*** - .0720 .2608*** .0086Grades in History B .0634 -.0008 .4029 .0278 .1640 .0611s.e. .0132 .0749 .0816 .0824beta .2585*** - .0006 .2546*** .0185Grades in Mathematics B .0870 -.1431 .2245 .0647 .1384 .0230s.e. .0141 .0804 .0875 .0884beta .3356*** -.1034 .1343* .0406Females with College Graduate Fathers N=113Grades in All Subjects B .7317 -.6325 4.1952 -.2707 .2034 .0910s.e. .1738 1.0617 1.1567 1.2902beta .3714*** -.0530 .2968*** -.0174Grades in English B .0913 -.0148 .4244 -.0566 .2093 .0729s.e. .0214 .1264 .1386 .1561beta .3894*** -.0108 .2649** -.0317Grades in History B .0673 -.0764 .6174 .0944 .1644 .1160s.e. .0247 .1459 .1600 .1801beta .2557** -.0496 .3431*** .0470Grades in Mathematics B .0702 -.0924 .3971 .0661 .1216 .0578s.e. .0235 .1387 .1521 .1713beta .2877** - .0647 .2380* .0355* p


11/14

198 AMERICAN SOCIOLOGICAL REVIEWNo such differencesappeared or boys. (Threeway cross-tabulations,controllingfor father'seducation, were executed for each pair of thecultural nterestand culturalactivity questions.The bivariate relationships were strongestamong college educated men's daughters, butnot among their sons.)These findings suggestthat cultural nterestsand activities were culturally prescribedforteenage girls, while for adolescent boys theywere less strongly prescribed,perhaps evennegatively sanctionedby peers. High culturalinvolvementsmay have been partof an identitykit that academically successful, high statusgirls, but not similar boys, possessed.This interpretation s consistent, as well,with an additionalgenderdifference revealedin the correlation matrix. Reading literaturewas more strongly relatedto culturalattitudesforgirlsthan otheractivities,while arts attend-ance and performingwere more strongly re-lated to attitudes for boys. If having culturalinterestswere part of the good studentrole forgirls, then we would expect literaturereadingto be most strongly relatedto interestin othercultural ields;for in literaturereading he cul-ture inculcatedby the uppermiddle-classhomeand that inculcated by the school reinforceeach other.As may havebeenthe case for sonsof college graduates,to the extent that cultur-ally oriented students were not necessarilysuccessful or conformist students, culturalinterests would most likely be expressed pre-cisely throughthose activities that the schooldoes not teach-for example, performingorarts attendance.Coleman found in The Adolescent Society(1961:118-27) that leading-crowdgirls weremore likely to exhibit conformist charac-teristics and less likely to be as interestedinpopular music as their peers than leading-crowd boys. The high school years-particularlyn an era in which highschool en-rollmentswere risingrapidly-may have beena kind of cultural atency period duringwhichhigh status boys rebelled against parentalvalues. Boys with educatedfathers who wereculturally oriented may have been less in-volved in their peer groups and less likely topossess othertraits that lead to school successthan were other boys. Those highstatusboyswho were not academicallyorientedmayhaveused high culture as an alternative arenaforachievement.By contrast, boys from humblebackgroundswho were upwardlymobile mayhave begun to enter the status cultureof theupper middle class during high school. Up-wardlymobile boys may have been more in-clinedto express cultural nterests and to par-ticipate in cultural activities than were uppermiddle-classboys, who took such interestsfor

granted. By the same token, teachers may haverewarded students from lower status back-grounds who exhibited interests and behaviorexpected from higher status students.Conversely, cultural capital seems to havebeen part of an identity shared by academicallysuccessful high school girls. In a society inwhich men monopolize careers and control thematerial rewards they carry, it is all the moreimportant for women to distinguish themselvesthrough fundamentally cultural markets. Elder(1974: ch. 8) has observed that women leadcontingent careers ; to a greater extent thanmen they are evaluated and rewarded on thebasis of ascriptive and diffuse criteria. Womenwho wish to be recognized as eligible partnersfor men from high status backgrounds mayneed cultural capital to a greater extent thanmen who wish to achieve in the world of work.For boys from high status families, it may bemore important, in high school, to develop ataste for women who appreciate culture than todevelop a taste for high culture itself.CONCLUSIONSThe relatively low correlations between pa-rental education and cultural capital are nota-ble. An analysis of the responses of a cross-sectional sample of American adults to ques-tions that included a broader range of culturalattendance activities, but required a greaterspecificity as to the extent of the activity,found correlations of both occupational statusand educational attainment with culture con-sumption of approximately .40 (Gruenberg,1975:200). To the extent that the TALENTdata on arts attendance are comparable to theISR data employed by Gruenberg, they indi-cate that well-educated parents passed down30 percent of their cultural advantage to theirsons and 60 percent to their daughters.8 If, asBourdieu argues, early socialization is criticalto an adult's inclination and ability to consumehigh culture, the inheritance may lessen withage. It is also possible, however, that opportu-nities afforded by the school and peer grouponly temporarily attenuate the relationshipbetween family background and cultural in-volvement during high school; or that the ado-lescent rebellion described by Coleman (1961)and Stinchcombe (1964) during the early 1960stemporarily depressed the correlation amongthe boys in the TALENT sample. While firmerconclusions await analysis of the impact of anarray of background measures on students'

8 These figures are the ratios of the correlationsbetweenarts attendanceandfather's ducation o the.40 correlationGruenberg eports, orboys andgirlsrespectively.



12/14

CULTURAL CAPITAL AND SCHOOL SUCCESS 199cultural capital, these findings lend tentativesupport to Jencks and Riesman's assertion(1968) that the level of culturalmobility n theUnited States has been relatively high. Thefindingsalso suggest that culturalcapital s lessstrongly ied to parentalbackgroundraits hanBourdieu's theory or similar discussions ofclass and culture in the United States wouldpredict. Whether more direct measures thaneducationor parents'culturalcapitalwouldre-veal a stronger nheritance emains o be seen.It follows from these findings that educa-tional attainment s a very imperfectproxyforculturalcapital.A second, and related,lessonis that single measures of culturalcapital orparticipationn status cultures are inadequate.Abandoning he use of such variablesas edu-cational attainment or self-reportedarts at-tendance as single proxies for status cultureparticipation aises formidablemethodologicalpuzzles, but these must be confronted.An ideal data set for our purposes wouldcontain measuresof culturalcapital groundedin research on adult elites in a single commu-nity; objective measures of grades, stan-dardizedby school; data on teachers' evalua-tions of students'charactersandaptitudes;andobservationally grounded measures of stu-dents' interaction style, both linguistic andnonverbal.The TALENT data were not col-lected for our purposes and are not ideally

suited for them. The use of self-reportedgrades, the eliminationof high school dropoutsand minoritychildren from the data, and theself-reportsof culturalparticipation ll can beexpected to reduce the impactof culturalcap-ital on grades. What is more, the use of na-tionaldata limits this assessmentto the impactof participation n a nationalelite status cul-ture, and does not permit estimation of theeffects of local status culturevariations.Even with these limitations,the data showthat cultural capital has an impact on highschoolgrades hat is highlysignificantandthat,in nontechnicalsubjects, approaches he con-tributionof measuredability.Thisfindingcon-firmsratherdramatically he utility of the per-spective advancedhere. It remains, however,to assess the impactof culturalcapitalon suchoutcomes as educationalattainment, collegequality, marital selection, and occupationalattainment; o develop better measuresof cul-turalcapital; to assess the differingrole cul-turalcapital mayplayinthe mobilitystrategiesof differentclass segments;and to compare heinfluenceof culturalcapital n differentkinds ofeducational and occupationalsettings. In allthese arenas,conceiving of status as a culturalprocess which influences success by affectingthe outcomes of interactionsmay yield impor-tantgainsin ourability o understand he statusattainmentprocess as a whole.

APPENDIXTable A-1. Means and Standard Deviations of Talent MeasuresQuestion Mean Standard Deviation N

INTEREST (5 = high; 1= low)In being a musical composer 2.29 (boys) 1.33 13552.84 (girls) 1.46 1401In being a poet 1.97 1.19 13542.49 1.47 1380In being an artist 2.48 1.43 13643.11 1.53 1406In visiting art galleries 2.10 1.29 13102.73 1.48 1385In reading literature 2.51 1.30 12583.14 1.47 1353In attending symphony concerts 2.33 1.38 12512.86 1.52 1356Artistic Interest Inventory (TALENT scale) 16.64 9.16 138421.43 9.87 1426Musical Interest Inventory (TALENT scale) 13.87 10.85 138418.80 11.41 1425Literary Linguistic Interest Inventory (TALENT scale) 15.58 8.29 138521.20 8.91 1426Cultivated Self-Image (TALENT scale) 4.64 2.30 14005.95 2.21 1464ACTIVITIESDrawing, etc., in the past three years 2.24 (boys) 1.27 1398(5 = very often; 1 = never) 2.69 (girls) 1.34 1416

Acting, etc., in the past three years 1.97 1.23 1403(5 = very often; 1 = never) 2.62 1.40 1411



13/14

200 AMERICAN SOCIOLOGICAL REVIEWAppendix (Continued)

Table A-l. Means and Standard Deviations of Talent MeasuresQuestion Mean Standard Deviation NAttending concerts, etc., in the past three years 3.85 1.46 1372(1 = very often; 6 = never) 3.37 1.42 1420Books, plays, poetry, etc. read the past three months 1.15 1.75 1340(0= none; 5 = 5 or more) 1.80 1.99 1396

INFORMATIONLiterature information test (low = 0; high = 24) 13.28 4.60 140213.12 4.47 1463Music information test (low = 0; high = 12) 6.29 3.04 14016.88 3.00 1461Art information test (low = 0; high = 12) 6.38 2.51 13866.62 2.58 1423MIDDLEBROW ACTIVITY (3-year frequency,5 = very often; 1= never)Photography (excluding occasional snapshots) 1.74 (boys) 1.15 1446

1.41 (girls) 0.87 1474Making jewelry, pottery, etc. 1.62 0.95 14421.72 0.96 1472Cabinet making or woodworking 2.36 1.24 14351.15 0.46 1469Sewing, etc. 1.27 0.73 14323.65 1.19 1469Mark the one answer indicating the highest level of educationyour father reached: 11.27 (boys) 3.53 11664 = none, or some grade school 11.14 (girls) 3.60 12378 = completed grade school10 = some high school, but did not graduate12 = graduated from high school13 = vocational or business school after high school14 = some junior or regular college, but did not graduate16 = graduated from regular 4-year college17 = master's degree18 = some work toward doctorate or professional degree20 = completed doctorate or professional degreeMark the one answer indicating the highest level of educationyour mother reached: 11.52 (boys) 2.75 120411.28 (girls) 2.99 1277The following questions ask you to report your grades in coursesyou have taken in the ninth grade or later. Please consider onlysemester grades. If you have not taken any courses in thetopic, skip the item. In these questions, choose the one answer

that best describes your grades.6 = all A's or equivalent 3 = mostly B's and Cs or equivalent5 = mostly A's or equivalent 2 = mostly Cs and D's or equivalent4 = mostly A's and B's or equivalent 1 = mostly D's and below or equivalentMy grades in history and social studies have been 3.52 1.27 12963.61 1.25 1361My grades in English courses have been 3.36 1.21 12993.59 1.02 1351TALENT Composite of self-reported grades 23.34 9.71 137026.19 9.52 1424

REFERENCESBourdieu, Pierre1977 Cultural reproductionand social repro-duction. Pp. 487-511 in Jerome Karabeland A. H. Halsey (eds.), Power and Ideol-ogy in Education.New York: Oxford.1977 Reproduction n Education, Society, Cul-ture. Beverly Hills, Calif.: Sage.

Cicourel,Aaron and John Kitsuse1963 The Education Decision-Makers. In-dianapolis:Bobbs-Merrill.Coleman,James C.1961 The Adolescent Society. New York: FreePress.Coleman,Richard1971 Social Status in the City. San Francisco:Jossey-Bass.



14/14

CULTURAL CAPITAL AND SCHOOL SUCCESS 201Collins, Randall1975 Conflict Sociology. New York: Academic.1979 The CredentialSociety. New York: Aca-demic.1981 The micro-foundations of macro-sociology. AmericanJournalof Sociology

86:984-1014.Crouse, James, Peter Mueser and ChristopherJencks1979 Latent variable models of status attain-ment. Social Science Research 8:348-68.DiMaggio,Paul and Michael Useem1978 Cultural emocracy n a periodof culturalexpansion: the social composition of artsaudiences in the United States. SocialProblems26:180-97.Elder, Glen1974 Children f the GreatDepression.Chicago:Universityof ChicagoPress.Erickson, Fred1975 Gatekeepingand the melting pot. Har-vard EducationalReview 45:44-70.Flanagan,John C. et al.1964 The American High School Student.Pittsburgh: University of Pittsburgh,PROJECTTALENT.Goffman, Erving1951 Symbolsof class status. BritishJournal fSociology 2:298-312.Gruenberg,Barry1975 How Free is Free Time? Analysisof SomeDeterminants f Leisure Activity Patterns.

Manuscript.Middletown,CT.Gumperz,JohnJ. and Dell Hymes1972 Directions in Sociolinguistics.New York:Holt, Rinehart,and Winston.Hollingshead,August B.1949 Elmtown'sYouth. New York:Wiley.Jencks, Christopher t al.1979 Who Gets Ahead? The DeterminantsofEconomic Success in America.New York:Basic.Jencks, Christopher nd David Riesman1968 The Academic Revolution. New York:

Doubleday.Lortie, Dan1975 Schoolteacher. Chicago: University ofChicagoPress

McDermott,R. P.1977 Social relationsas contexts for learning nschool. Harvard Educational Review47:198-213.Mehan, Hugh1974 Accomplishingclassroom lessons. Pp.76-142 in Aaron Cicourel et al., LanguageUse and School Performance.New York:Academic.Picou, J. Steven and Michael Carter1976 Sipnificant-other nfluences and aspira-tions. Sociology of Education49:12-22.Porter, James N.1974 Race, socialization, and mobility in edu-cational and early occupational attain-ment. American Sociological Review39:303-16.Portes, Alejandroand Kenneth L. Wilson1976 Black-White differences in educationalattainment. AmericanSociologicalReview41:414-31.Rindskopf, David1979 Arts Education in Public SecondarySchools: Offerings, Enrollments,and theirDeterminants. Technical Memorandum#AH-46. St. Louis: CEMREL,Inc.Sewell, Williamand RobertM. Hauser1975 Education, Occupation, and Earnings:Achievement in the Early Career. NewYork: Academic.Stinchcombe,Arthur1964 Rebellion n a High School. Chicago:Quad-rangle.Warner,W. Lloyd, RobertJ. Havighurst nd MartinB. Loeb1944 Who Shall Be Educated? New York:Harperand Brothers.Weber, Max1968 Economy and Society. New York: Bed-minsterPress.Welch, WayneW., RonaldE. Andersonand LindaJ.Harris1980 Theeffects of schoolingon mathachieve-ment. Mimeo, Universityof Minnesota.Wolf, Dennis and Howard Gardiner1979 Style and sequence. Pp. 117-138 inMargery Franklin and Nancy R. Smith(eds.), Symbolic Functioningn Childhood.Englewood Cliffs, NJ: LawrenceErlbaumAssociates.

cultural capital and school success: the impact of status culture participation on the grades of...

Documents