ED 035 961

AUTHORTITLE

INSTITUTION

SPONS AGENCY

REPORT NOBUREAU NOPUB DATECONTRACTNOTE

EDRS PRICEDESCRIPTORS

DOCUMENT RESUME

24 CG 005 025

HAGSTRCM, WARREN O.; GARDNER, LESLIE L. HUGHCHARACTERISTICS OF DISRUPTIVE HIGH SCHOOL STUDENTS.TECHNICAL REPORT NO. 96.WISCONSIN UNIV., MADISCN., RESEARCH AND DEVELCPMENTCENTER FOR COGNITIVE LEARNING.OFFICE OF EDUCATION (DHEW) , WASHINGTON, D.C. BUREAUOF RESEARCH.TR-96BR-5-T0216SEP 69OEC-5-10-15431P.

EDRS PRICE MF-$0.25 HC-$1.65*ANTI SOCIAL BEHAVIOR, *BEHAVIOR PROBLEMS,DISCIPLINE, DISCIPLINE PROBLEMS, GRADE 11, *HIGHSCHOOL STUDENTS, *INDIVIDUAL CHARACTERISTICS,*MALADJUSTMENT, SCHOOL POLICY, STUDENT PROBLEMS

ABSTRACTTHIS REPORT CONTRASTS THE CHARACTERISTICS OF HIGH

SCHOOL STUDENTS WITH DISORDERLY HISTORIES AND THOSE WITHOUT SUCHHISTORIES. THE SAMPLE CONSISTS OF 1,318 ELEVENTH GRADERS IN EIGHTWISCONSIN SCHCCL SYSTEMS., THE MAJOR DEPENDENT VARIABLES ARE STUDENTS'REPORTS OF BEING SENT FROM CLASSES FOE DISCIPLINARY REASONS ANDSKIPPING SCHOOL WITH A GANG OF KIDS; THE MERITS OF THESE INDICATORSARE DISCUSSED. QUESTIONNAIRES COMPLETED BY STUDENTS PROVIDED ALL THEDATA, EXCEPT IQ, WHICH WAS OBTAINED F1OM SCHOOL RECORDS.DISORDERLINESS, OE REBELLIOUSNESS, IS CONTRASTED WITH OTHER TYPES OFSTUDENT DEVIANCE; A TAXONCMY OF SUCH DEVIANCE IS PRESENTED ANDDISCUSSED.. THE REPORT CONCLUDES EY NOTING THE IMPLICATIONS FOR SCHOOLPOLICY AND FOE FURTHER RESEARCH CF THE EMPIRICAL FINDINGS. (AUTHOR)

I.

CHARACTERISTICS OF DISRUPTIVEHIGH SCHOOL STUDENTS,

ti

4.,

ft,

WISCONSIN RESEARCH AND DEVELOPMENT

TER FOR

NITIVE LEARNING

Alt

C.)

S.

U.S. DEPARTMENT OF HEALTH, EDUCATION & WELFARE

OFFICE OF EDUCATION

THIS DOCUMENT HAS BEEN REPRODUCED EXACTLY AS RECEIVED FROM THE

PERSON OR ORGANIZATION ORIGINATING IT. POINTS OF VIEW OR OPINIONS

STATED DO NOT NECESSARILY REPRESENT OFFICIAL OFFICE OF EDUCATION

POSITION OR POLICY.

Technical Report No. 96

;7 rZ

1947 7-"?oClee

CHARACTERISTICS OF DISRUPTIVE HIGH SCHOOL STUDENTS

By Warren 0. Hagstrom and Leslie L. Hugh Gardner

Report from the Models for Planned Educational Change Project

Max R. Goodson, Burton W. Kreitlow, and Warren 0. Hagstrom,Principal Investigators

Wisconsin Research and DevelopmentCenter for Cognitive LearningThe University of Wisconsin

Madison, Wisconsin

September 1969

Published by the Wisconsin Research and Development Center for Cognitive Learning, supportedin part as a research and development center by funds from the United States Office of Education,Department of Health, Education, and Welfare. The opinions expressed herein do not necessarilyreflect the position or policy of the Office of Education and no official endorsement by the Officeof Education should be inferred.

Center No. C-03 / Contract OE 5-10-154

NATIONAL EVALUATION COMMITTEE

Samuel BrownellProfessor of Urban EducationGraduate SchoolYale University

Launor F. CarterSenior Vice President on

Technology and DevelopmentSystem Development Corporation

Francis S. ChaseProfessorDepartment of EducationUniversity of Chicago

Henry ChaunceyPresidentEducational Testing Service

Martin DeutschDirector, Institute for

Developmental StudiesNew York Medical College

Jack MingDirector, Teaching Research

DivisionOregon State System of Higher

Education

Elizabeth KoontzPresidentNational Education Association

Roderick McPheePresidentPunahou School, Honolulu

G. Wesley SowardsDirector, Elementary EducationFlorida State University

Patrick SuppesProfessorDepartment of MathematicsStanford University

*Benton J. UnderwoodProfessorDepartment of PsychologyNorthwestern University

UNIVERSITY POLICY REVIEW BOARD

Leonard Berkowitz John Guy Fowlkes Herbert J. Klausmeier M. Crawford YoungChairman Director Director, R & D Center Associate Dean

Department of Psychology Wisconsin Improvement Program Professor of Educational The Graduate SchoolPsychology

Archie A. Buchmiller Robert E. Grinder Donald J. McCartyDeputy State SuperintendentDepartment of Public Instruction

*James W. ClearyVice Chancellor for Academic

Affairs

Leon D. EpsteinDeanCollege of Letters and Science

ChairmanDepartment of Educational

Psychology

H. Clifton HutchinsChairmanDepartment of Curriculum and

Instruction

Clauston JenkinsAssistant DirectorCoordinating Committee for

Higher Education

DeanSchool of Education

Ira SharkanskyAssociate Professor of Political

Science

Henry C. Wein lickExecutive SecretaryWisconsin Education Association

EXECUTIVE COMMITTEE

Edgar F. BorgattaBrittingham Professor of

Sociology

Max R. GoodsonProfessor of Educational Policy

Studies

Russell J. HosierProfessor of Curriculum and

Instruction and of Business

*Herbert J. KlausmeierDirector, R & D CenterProfessor of Educational

Psychology

Wayne OttoProfessor of Curriculum and

Instruction (Reading)

Robert G. PetzoldAssociate Dean of the School

of EducationProfessor of Curriculum and

Instruction and of Music

Richard L. VenezkyAssistant Professor of English

and of Computer Sciences

FACULTY OF PRINCIPAL INVESTIGATORS

Ronald R. AllenAssociate Professor of Speech

and of Curriculum andInstruction

Vernon L. AllenAssociate Professor of Psychology(On leave 1968.69)

Nathan S. BlountAssociate Professor of English

and of Curriculum andInstruction

Robert C. CalfeeAssociate Professor of Psychology

Robert E. DavidsonAssistant Professor of

Educational Psychology

Gary A. DavisAssociate Professor of

Educational Psychology

M. Vere De VaultProfessor of Curriculum and

Instruction (Mathematics)

Frank H. FarleyAssistant Professor of

Educational Psychology

John Guy Fowlkes (Advisor)Professor of Educational

AdministrationDirector of the Wisconsin

Improvement Program

Lester S. GolubLecturer in Curriculum and

Instruction and in English

Max R. GoodsonProfessor of Educational Policy

Studies

Warren 0. HagstromProfessor of Sociology

Johri G. HarveyAssociate Professor of

Mathematics and Curriculumand Instruction

Herbert J. KlausmeierDirector, R & D Center

Professor of EducationalPsychology

Burton W. KreitlowProfessor of Educational Policy

Studies and of Agriculturaland Extension Education

Richard G. MorrowAssistant Professor of

Educational Administration

Wayne OttoProfessor of Curriculum and

Instruction (Reading)

Milton 0. PellaProfessor of Curriculum and

Instruction .(Science)

Thomas A. RombergAssistant Professor of

Mathematics and ofCurriculum and Instruction

Richard L. VenezkyAssistant Professor of English

and of Computer Sciences

MANAGEMENT COUNCIL*Herbert J. Klausmeier

Director, R & D CenterActing Director, Program I

Mary R. QuillingDirectorTechnical Section

Thomas A. RombergDirectorPrograms 2 and 3

James E. WalterDirectorDissemination Section

Dan G. WoolpertDirectorOperations and Business

* COMMITTEE CHAIRMAN

STATEMENT OF FOCUS

The Wisconsin Research and Development Center for Cognitive Learningfocuses on contributing to a better understanding of cognitive learning bychildren and youth and to the improvement of related educational practices.The strategy for research and development is comprehensive. It includesbasic research to generate new knowledge about the conditions and procedsesof learning and about the processes of instruction, and the subsequent develop-ment of research-based instructional materials, many of which are designed foruse by teachers and others for use by students. These materials are tested andrefined in school settings. Throughout these operations behavioral scientists,curriculum experts, academic scholars, and school people interact, insuringthat the results of Center activities are based soundly on knowledge of subjectmatter and cognitive learning and that they are applied to the improvement ofeducational practice.

This Technical Report is from the Models for Effecting Planned EducationalChange Project in Program 3. General objectives of the Program are to developand test organizations that facilitate research and development activities in theschools and to develop and test the effectiveness of the means whereby schoolsselect, introduce, and utilize the results of research and development. -Con-tributing to these Program objectives, the main objective of the Planned ChangeProject is to develop and test system-wide mechanisms which local school sys-tems can employ in utilizing knowledge and innovations of the type generatedby the Center. Change-agent teams have been organized in area school sys-tems and their effectiveness is being evaluated.

iii

CONTENTS

Page

List of Tables vii

List of Figures viii

Abstract ix

I Introduction: Conformity and Deviance in School Systems 1

Correlates of Rebellious Behavior 3

II Methods 5Sample 5Indicators of Rebelliousness 5

III Results 8Socioeconomic Status and College Aspirations 8Intelligence 10Peer Relations and Rebellion 14

IV Summary and Implications 18Research Implications 18Policy Implications 19Is the Nature of Student Rebellion Changing? 20

References 23

LIST OF TABLES

Table Page

1 Relative Rates of Being Sent Out of Class, by Sex 6

2 Relative Rates of Skipping School, by Sex 6

3 Correlations Between Being Sent Out of Class and SkippingSchool with Selected Attitudes to School, by Sex 7

4 Rates of Being Sent Out of Class for Disciplinary Reasons,by Socioeconomic Status and Sex 9

5 Rates of Skipping School with a Gang of Kids by SocioeconomicStatus for Girls 9

6 Rates of Being Sent Out of Class for Disciplinary Reasons byCollege Expectations and Sex 10

7 Rates of Being Sent Out of Class for Disciplinary Reasons bySocioeconomic Status and College Expectations, for Boys 11

8 Rates of Skipping School with a Gang of Kids by SocioeconomicStatus and College Expectations, for Boys 11

9 Rates of Skipping School with a Gang of Kids by SocioeconomicStatus and College Expectations, for Girls 12

10 Rates of Being Sent Out of Class for Disciplinary Reasons bySocioeconomic Status and Intelligence, for Boys 12

11 Rates of Being Sent Out of Class for Disciplinary Reasons byCollege Expectations and Intelligence, for Boys 13

12 College Expectations by Intelligence and Rebelliousness, forBoys 14

13 Rates of Being Sent Out of Class for Disciplinary Reasons byPerceived Distance from the School's "Leading Crowd," forBoys 16

14 Rates of Being Sent from Class for Disciplinary Reasons byLeadership Status and College Expectations, for Boys 17

vii

LIST OF FIGURES

Figure Page

1 Types of Student Deviation 2

2 Rate of Being Sent Out for Disciplinary Reasons by Numberof Choices Received as Leader, for Boys 15

3 Rate of Being Sent Out for Disciplinary Reasons by Numberof Choices Received as "Like Most," for Boys 15

4 Rates of Being Sent Out of Class for Disciplinary Reasonsby Leadership Status and College Expectations, for Boys 17

viii

ABSTRACT

This report contrasts the characteristics of high school students with dis-orderly histories and those without such histories. The sample consists of1318 eleventh graders in eight Wisconsin school systems. The major depend-ent variables are students' reports of being sent from classes for disciplinaryreasons and skipping school with a gang of kids; the merits of these indica-tors are discussed. Questionnaires completed by students provided all thedata, except IQ, which was obtained from school records.

Disorderly students are far more likely to be boys than girls and they areespecially likely to be those boys who do not expect to attend college. Thereis no clear correlation between father's occupational status and disorderly be-havior, although SES and college expectations do interact in affecting dis-orderly behavior: boys and girls from high SES families who do not expect toattend college are most likely to become disciplinary problems. Students ofaverage IQ are more likely to be disorderly than those with either low or highIQ. Intelligence interacts with SES and college expectations to affect disor-derliness; those of high SES origins and low intelligence, and those of highintelligence and low expectations of attending college, are more likely to bedisorderly than other students. Boys of marginal leadership status, who havesome followers in the classroom `but are not among the top leaders, are morelikely to be behavior problems in high school than either non-leaders or thetop leaders, although this is true only for those not expecting to attend col-lege. The disorderly students are likely to perceive themselves either as inthe leading crowd of their schools or as remote from the leading crowd.

Disorderliness, or rebelliousness, is contrasted with other types ofstudent deviance; a taxonomy of such deviance is presented and discussed.The report concludes by noting the implications for school policy and forfurther research of the empirical findings.

ix

INTRODUCTION: CONFORMITY AND DEVIANCE IN SCHOOL SYSTEMS

The generalization that the schools havea despotic political structure seems tohold true for nearly all types of sphools.. . . It is a despotism in a state"of per-ilous equilibrium. It is a despotismthreatened from within and exposed toregulation and interference from without.It is a despotism capable of being over-turned in a moment, exposed to the in-stant Joss of its stability and its pres-tige. (Waller, 1932, Pp. 9f.)

Schools ordinarily have difficulty maintainingorder because students are bound by many andcomplex rules while school authorities havelimited rewards and punishments to offer forconformity. Students are expected to conformto explicit rules related to the central tasks ofthe school; for example, rules about prompt com-pletion of homework, independent work on ex-ams, or appropriate spelling on written assign-ments. They are expected to conform to explicitrules related to the coordination of the activi-ties of many persons in schools; for example,rules about raising one's hand to be recognizedin a classroom, or using only certain stairwellsbetween classes. Further, students are ex-pected to conform to explicit or implicit stand-ards of deference and demeanor (Coffman, 1956),'deference to authorities and peers, demeanorfor themselves as- students and young people.The subtlety of some of these expectations isnot inconsistent with their importance or theharshness of sanctions for their transgression.The explicit rules governing adolescents in alarge high school might fill a large volume,and the implicit rules might be uncountable.We shall not attempt to devise a taxonomy ofthese rules here, although this would be a de-sirable task.

Schools in modern America are severely limi-ted in the sanctions available to them. Extrin-sic incentivesmoney, power, or prestige inthe larger communityare seldom available,

at lcasi for the large majority of students.Physical coercion can be used only in extremecases and is quite properly considered incon-sistent with the central goals of schools asinstitutions. The intrinsic rewards of learninghow to solve problems and acquiring excitinginformation are of great importance, but theyare often scarce and often irrelevant to therules governing studentsthese rewards arenot usually supposed to be contingent uponmany aspects of "good behavior." Thus, schoolsmust operate primarily on the basis of symbolictoken rewardsgrades, praise, private blameand public ridicule. A school can reward stu-dents by conferring upon them identities aspersons of merit, and it can punish them byidentifying them as failures (Stinchcombe, 1964).These incentives have great long-run importance,and most students are usually influenced bythem. But we ought not be surprised if moststudents sometimes and some students usuallyfind these sanctions insufficient and deviatefrom expectations.

This report is devoted to one type of studentdeviation: rebellionactive alienation fromauthority. This is but one of many types ofdeviation, and we shall describe other typesto justify our concentration on this particularone. Talcott Parsons (1951, Ch. VII) suggestsa taxonomy of deviation based upon three vari-ables. First, deviants tend to be ambivalentabout their relations to others and to rules,and this ambivalence may be expressedeither in terms of alienation or in terms of over-conformity. Second, deviation may be activeor it may be passive. Third, the focus of thedeviation may be on social objects (persons,roles, groups) or on norms. Figure 1 appliesParsons' scheme to the student role. (Ob-viously the scheme could be applied to otheractors in schools, particularly teachers.)

The most common type of deviation and per-haps the most costly in terms of the long-rungoals of schools is probably passive evasion

1

ConformativeDominance

AlienativeDominance

7

Active Passive

Compulsive Performance Compulsive Acquiescence

Focus onSocial Objects

Focus onNorms

Focus onSocial Objects

Focus onNorms

Dominance Compulsive Submission Perfectionistic

"Manipulatorsof authorityfigures"

Enforcement

E.g., the"teacher's pet""overdependentstudents"

Observance(Probably anempty cellfor students)

"Ritualism"

General Rebelliousness General Withdrawal

Focus onsocial objects

Focus onnorms

Focus onsocial objects

Focus onnorms

Aggressivenesstoward socialobjects:particularlyschoolauthorities.The rebel"

(failures ofdeference)

Rejection of valuesbydemeanor:"fools," "clowns,""regression in termsof age standards."Rejection of normalmeans, not ends:"Cheaters."

Withdrawal fromsocial interac-tion "truants""dropouts"

Evasion ofrulesthe "lazy,apatheticstudent"

Adapted from Parsons, 1951, p. 259

Figure 1. Types of Student Deviation

of rules. Students may attend classes butneither pay attention nor try to learn. Over-dependency and ritualistic behaviorwherethe student follows rules and instructions re-gardless of the fact that he isn't learning any-thing or otherwise attaining the formal goalsset by the schoolare also common, even ifthey are seldom recognized as problems ofdeviance, and they too are costly in a societywhich prizes creatively and imagination and inwhich the schools must help children attain ahigh degree of independence.

Despite the importance of these other typesof deviation, rebellion, the active alienationfrom authority, is critically important to thesocial organization of schools in the short run.The rebel, unlike the cheat and unlike theapathetic lazy student, presents an open chal-lenge to school authorities to which they mustreact, and these reactions may themselvesthreaten the order of the school. (Phillips,1959) . The rebellious student is likely toclaim some sort of legitimacy for his behavioror at least challenge the legitimacy of theauthority of teacher and principal. Stinch-combe suggests that the claim to legitimacy

2

usually takes the form of claims to symbols ofadult statussmoking, cars, marriage, etc.(Stinchcombe, 1964, Ch. 5), although todaysome rebellious students claim legitimacy onpolitical bases. Furthermore, the rebelliousstudent is likely to generate further disorderby stimulating various forms of collective be-haviorcrowdlike or moblike behavior. Crowdsand mobs occur commonly in schools for a num-ber of reasons. Students are generally ambiv-alent toward authority and will express aliena-tion in crowds or mobs that they might controlin ordinary situations. Large classrooms andespecially large study halls give participantsa feeling of anonymity, and as social reinforc-ers of particular roles and statuses are weak-ened, the student can participate in the undif-ferentiated crowd or mob. And noise andphysical movement in classrooms can generatethe kind of "circular reinforcement" noted bystudents of collective behavior.1 The rebel-

1 Cf. the excellent discussion of collectivebehavior in schools by Waller (1932, Ch. XII).

lious student, by expressing the alienativetendencies of others, can sometimes stimulatethem too to become disorderly. (It seemsreasonable to suppose that this would be morelikely to occur as class size increases.)

Foolish, clowning behavior also tends togenerate collective behavior, but fools ordi-narily pose a smaller threat to the order of theschool than rebels because the former tend notto claim legitimacy. The fool can display dis-respect for the normative order of the school(by failing to demean himself properly) and getaway with it because he does not claim respectfor himself. Fools may even be functional forgroups by providing a form of tension releaseand a focus for group identification (Daniels& Daniels, 1964; Dentler & Erikson, 1959).

Teachers often overreact to rebellious stu-dent behavior, partly because teachers arethemselves ambivalent toward students. (Itis not easy to wholeheartedly like adolescents.)Thus, a student's act of disrespect or disobedi-ence may stimulate a disproportionately aggres-sive response from the teacher. In Americanschools today this teacher aggressiveness islikely to be verbal; the teacher can "pick on"the deviant in a manner that can arise quitenaturally out of normal classroom activities.Teachers may also attempt to enlist the coop-eration of students by their use of ridicule, byusing the deviant student as a scapegoat, etc.2These practices are often "effective" punish-ments. However, use of them may lead to aloss of respect by students for the legitimacyof the teacher's authority. This may lead toother acts of deviance, producing still moreteacher aggression, and so forth in an everwidening vicious circle of deviance.

Parsons suggests that effective social con-trol must operate to stem these vicious circlesand that one essential element is the refusalby the offended party to reciprocate the devi-ance of the offender (Parsons, 1951, Ch. VII).Parsons also suggests the importance of sup-port, emphasizing the group membership of theoffender rather than threatening it; permissive-ness within limits; and the manipulation of re-wards or punishments. The failure of normaltactics of teacher control leads to the rejectionof the student.3 Rejection may occur in a variety

2Cf. the discussion by Jules Henry (1963)

of these forms of teacher behavior.3This is one manifestation of the very gen-

eral principle that deviation leads to rejection.Cf. Stanley Schachter (1951) and the criticalreview of the literature on this topic in Levineand Allen (1968).

of formal or informal ways, but in Americanschools one common procedure is the exclu-sion of the student from the classroom: he issent to the principal, the dean, or some otherpunitive agent. The frequency of occurrenceof such rejection will be our major indicatorof deviance in this report; its validity will bediscussed in the next chapter.

Deviance may be a transient and episodicaspect of a person's life; most students willhave engaged in most of the types of deviancenoted in Figure 1 by the time they graduate fromhigh school. On the other hand, when authori-ties react to deviance and label the offender,and when he comes to accept this labelling anddefine himself as a particular type of deviant,his deviation will become a more stable aspectof his role and his personality.4 We are par-ticularly interested in this report in the latterform, although our indicators do not enable usto distinguish between primary and secondarydeviation with much confidence. Thus, thisreport will deal with the correlates and causesof students coming to play the role of rebel inschools.

In the next few pages we shall discuss thevariables to be used in our attempts to explainvariation among individuals in rebellious be-havior. In the next section we shall describethe methods of the study, the sample and themajor indicators. Then we shall present ourresults, and this report will conclude with adiscussion of the implications of these findings.In a future report we shall discuss the charac-teristics of student groups and teachers inmore or less disorderly classrooms.

CORRELATES OF REBELLIOUS BEHAVIOR

Those who are familiar with the lives of ado-lescents usually point out that boys and girlsare markedly different in the way they relateto authority, whether that authority is domes-tic, academic, or legal. The sources of thesedifferences are varied. The research of Douvanand Adelson (1966), for example, clearly indi-cates that male adolescence is centered aroundthe problem of autonomy, while, adolescentgirls, on the other hand, are concerned withtheir expressive relationships. If boys aremore concerned with personal autonomy, theycan be expected to be more consistently en-gaged in confrontations with authority figures

4For a discussion of what has come to becalled "primary" and "secondary" deviation,see Lemert (1951).

3

who expect them, for better or for worse, tobehave passively. Arthur Stinchcombe (1964),in developing a more general theory of rebel-lion in schools, suggests that boys, unlikegirls, are forced to be oriented to a universal-istic labor market, with the result that failurein school is critical to their self-conceptionsin a way not true for girls. Current theories,empirical studies, and common sense all sug-gest, then, that boys will have a higher rateof classroom misbehavior than girls.

Of almost equal obviousness is the hypothe-sis that lower-class students will have a higherrate of classroom misbehavior than upper-classstudents; or, if not obvious, at least this isa notion suggested by a long tradition of socio-logical literature on juvenile deviance. Con-sidering school misbehavior to be a type of"juvenile delinquency," this hypothesis wouldbe suggested by studies of "lower class sub-cultures," such as Albert Cohen's DelinquentBoys, (1955). The socioeconomic status (SES)hypothesis is more directly expressed with re-spect to high school in August B. Hollingshead'sfamous description of Elmtown's Youth (1949).Various reasons for the differences between so-cial classes have been suggested: that lower-class students are discriminated against in theschools, partly because of culture conflict withmiddle-class teachers and other authorities,that lower-class culture emphasizes short-runand concrete goals instead of the long-run ab-stract goals of the schools, that lower-classfamilies give less support to school authoritiesthan middle-class families, etc.5 Whether theissue is lower-class subcultures or lower-class dignity, this sort of reasoning leads tothe prediction that high school rebellion willvary inversely with social class.

The one empirical test thus far given theSES hypothesis in terms of high school mis-behavior, however, failed to find that social

5For critical reviews of these theories seeStinchcombe (1964), Phillips (1959), andWerthman (1963).

4

class has any direct bearing on classroom be-havior, at least in the one California schoolstudied. Stinchcombe's Rebellion in a High,School (1964) reports that middle-class stu-dents were just as likely to have skippedschool or have been sent out of class as werelower-class students. This result, refuting asit does a major assumption of most thinkers inthis area, requires replication, and this weshall do here. We shall further replicateStinchcombe's contrary hypothesis. Since heobserved that class background predicted highschool rebellion so poorly, he reasoned thatit is not a student's background that accountsfor rebellion, but instead his expectations ofhis fu_ ture status in education and the labormarket. "The key to high school rebellion isto be found in the status prospects of students,rather than in their status origins" (Stinch-combe, 1964, P. 69). Thus, we expect tofind that students who plan on going on tocollege will have lower rates of classroommisbehavior than students who do not expectto continue. We shall also discuss howthese aspirations interact with social classbackground and IQ.

The possible correlates or student rebel-liousness discussed so far are remote fromthe classroom situation, but the social situa-tion in a classroom certainly presents thestimuli that trigger specific rebellious actsand may lead to a persistent pattern of rebel-lious behavior. We shall here examine thestudent's social position among his peers.Fox, Lippitt, and ,Schmuck (1964, P. 102 f.)report that students rejected by their peerstend to have more negative attitudes towardschool. Another study (B. A . Schmidt, 1958)found that sociometric rejects in elementaryschools are often hostile and aggressive intheir school settings. Other studies (Short &Strodtbeck, 1963) suggest that gang boys mayengage in delinquent acts when they are com-peting for leadership in the gang; this suggeststhat the highest rates of misbehavior will befound among those with some claims to leader-ship but without a secure leadership status.

II

METHODS

pre-dominantly rut al populations. None of the

limi-tation of the sample is that it includes onlyrelatively small communities, not selected in

The sample includes 1318 high school studentsfrom 56 classrooms in ten high schools. The

any random fashion. The largest city in the

communities is a suburb, although a smallfraction of the workers in three of them areemployed in Milwaukee or Madison. As a re-sult

schools. There is one other cityof 35,000 with a substantial industrial base inthe sample, while the other communities aresmaller. The two smallest systems serve pre-

eight Wisconsin communities in March 1967.

schools, two junior high schools, and sixteen

sample had a population of about 46,000 in1960; the school system included two high

sample is not random. The most serious limi-

SAMPLE

The data reported here were gathered in

suit of the relatively small size of these com-munities, none of the schools involved is ashomogeneous in terms of student social classorigins as the high schools found in big cities.On the other hand, the schools are homogene-ous in terms of race; not a single identifiableNegro is in the high school sample. These twocharacteristics of the sample present the mostserious limitations on the generalizability ofthe findings reported here. The sample ofschools is also unrepresentative in lackingeither Catholic schools or private academies.

Within each school a random sample of 11thGrade English classes was chosen. (More than95% of the students were 11 Graders in Englishclasses; a small proportion were 12th Gradersor were sampled in social studies classes.)The choice of English classes made administra-tion convenient by avoiding problems of dupli-cation or omission, but it limits our ability togeneralize when the variables concern teacheror classroom characteristics.

Many of the same questions were asked ofFifth Grade students in the same communities.Results from this sample will be reported in alater publication. The questionnaires adminis-tered were developed in cooperation with theCooperative Project for Educational Develop-ment (COPED), and they have also been ad-ministered to thousands of other students inother states. We hope to be able to replicateand extend the present analysis with the moreextensive data from these other states. (Fora detailed description of COPED's program,see Mial, 1967).

Questionnaires were administered by paidassistants during the regular English classperiod. The teacher was asked to leave theroom and did so in almost all cases. No at-tempt was made to contact students absent onthe day the questionnaires were administered,and students were assured that their responseswould be treated confidentially.

INDICATORS OF REBELLIOUSNESS

Two questions taken from Stinchcombe (1964)are our primary indicators of rebellious behavior.The first, and the one that will be used most inthis analysis, is Have you ever been sent outof the class to the office by a teacher you didn'tget along with? --- Yes, more than once; Yes,once; or No." The distribution of responsesto this question are shown, separately for boysand girls, in Table 1. This indicator has facevalidity, and it conforms to the theoreticalnotions about deviation and rejection discussedabove. However, student responses may havebeen dishonest (the actual frequencies of hav-ing been sent out are probably higher than thosereported in Table 1), and the behavior reportedclearly depends not only on the student con-cerned but his teacher as well. The likelihoodthat a teacher will send a misbehaving student

5

Table 1

Relative Rates of Being Sent Out of Class,by Sex

Times Sent Out

Sex 2+ 1 0 Total

Boys

Girls

22%

3%

17%

7%

62%

90%

100%*

100%

(N = 653)

(N = 665)

*Totals do not equal 100% because of rounding.

out of class will depend not only upon theteacher's ability to maintain control in theclassroom but on the policies of the principaland other disciplinary authorities in the school.Thus, this indicator is relatively crude. It is,however, correlated with other questions indi-cating deviance in the way we expect it to be.Some correlates of responses to this questionare shown on page 7; see also the discussionin Stinchcombe (1964, Ch. 2). The data onpage 7 show that those who have been sentout of class are substantially more likely thanothers to report: that life in their presentclassroom is unpleasant; that they do notwork very hard at school tasks; that they arebored with the work in their present classroom;and that, everything considered, their schoolis not very good. Since it is unlikely thateither student misreporting or teacher charac-teristics are systematically associated withother variables used in this report, the crudityof this indicator will ordinarily tend to reducethe correlations we shall present.

The second indicator of rebelliousness usedhere is the student's response to the question,"Have you ever skipped school with a gang ofkids (whether or not you got caught)? Yes,more than once; Yes, once; or No." Table 2presents the distribution of responses to thisquestion by sex. The phrase "with a gang ofkids" in this question is added to sharpen itsrelation with flouting authority rather thanmerely evading it; the student who wishes to

6

Table 2

Relative Rates of Skipping School, by Sex

Times Skipped School

Sex 2+ 1 0 Total

Boys

Girls22%

12%

10%

11%

68%

79%

100%* (N = 653)

100% (N = 665)

*Totals do not equal 100% because of rounding.

rationally evade authority can do it best byskipping out by himself, while students whoskip in a group are likely to be showing con-tempt for the school's authority to themselvesand others. This indicator, like being sentout of class, may be misreported by studentsand will be affected by the situation and poli-cies of the school. Skipping may be easierto accomplish in some locations than others,and some schools may punish it more severelythan others. The data in Table 3 show thatresponses to, the skipping question are cor-related with attitudes toward school in themanner to be expected. The data in Tables1, 2, and 3 show that skipping has more vari-ance than being sent out, especially for girls,and, for girls, its correlations with attitudestoward other aspects of school are consider-ably larger than the correlations betweenthese attitudes and being sent out of class.Thus, for girls skipping is a better indicatorof rebelliousness than being sent out of class.Girls do not confront the authorities of theschool directly, as do boys; their deviance ismore likely to be in the nature of collectiveevasion of rules. For boys, on the other hand,being sent out of class appears to be a some-what better indicator of rebellious behavior.

Most of the other indicators used in thisreport are responses to questions in the samequestionnaire. The number of sociometricchoices received was computed by summingchoices given withiri a class. Intelligencetest scores were obtained from official schoolrecords.

Table 3

Correlations Between Being Sent Out of Class and SkippingSchool with Selected Attitudes to School, by Sex

Correlation (Gamma coefficient)*

Times sent out ofclass by teacher(Never, once, morethan once)

Boys Girls

Times skipped schoolwith a gang of kids(Never, once, morethan once)

Boys Girls

1. Times sent out of class

2. "Life in this class withyour regular teacher. . "

1. "Has all good things; "2. "Has mostly good things;"3. "Has more good things than bad; "4. "Has about as many good things as bad,"5. "Has more bad things than good,"

MI MI ON OM .69 .66

6. "Has mostly bad things." .31 .08 .25 .24

3. "How hard are you working these dayson learning what is being taught at school?"

1. "Very hard; "2. "Quite hard;"3. "Not very hard;4. "Not hard at all." .23 .19 .22 .35

4. "When Pm in this class I. . ."1. "Usually feel wide awake and very

interested; "2. "Pretty interested, bored part of the

time; "3. "Not very interested, bored a lot of

the time; "4. "Don't like it, usually feel bored."

5. "This school. "

1. "Is my idea of a good school; "2. "Is O.K. but it could be better; "3. "Isn't very good; "4. "Is pretty bad-I don't like it."

.34 .05 .31

.38 .33 .40

.19

.46

* Gamma is a measure of correlation for ordinal variables. Consideringall pairs of cases in a bivariate table, it measures the proportionate ex-cess of pairs where both variables are ordered in the same direction overpairs where one member is higher on one variable, the other member higheron the other variable, with ties on either variable not counted. Gamma iszero when the variables are statistically independent, plus one when forall untied pairs the same member is higher on both variables, and minusone when for all untied pairs the member higher on one variable is loweron the other. For a double dichotomy gamma reduces to the familiar Yule'sQ. On this measure see Goodman and Kruskal (1954).

7

HI

RESULTS

The data in Table 1 show that more than one-third of the 11th Grade boys in this samplehave been sent out of class at one time oranother for disciplinary reasons, and more thanone-fifth have been sent out of class at leasttwice for these reasons. Thus, a large frac-tion of these boys have misbehaved at sometime in such a way that their teachers havebeen unable to cope with them unassisted. Onthe other hand, only 10% of the girls admittedhaving ever been sent out of class at all, andonly 3% have been sent out more than once.The difference between the sexes is muchsmaller when misbehavior outside the class-room Ls considered. About 33% of the boysskipped school with a gang of kids at leastonce, 22% more than once, while 23% of thegirls skipped with a gang at least once and12% more than once. Girls are much less likelythan boys to confront directly the authority ofthe teacher, but they are only somewhat lesslikely than boys to avoid publicly the authorityof the school.

It is tempting to explain the difference be-tween the sexes in terms of typical personalitypatterns and normative expectations for thesexes. However, we have no evidence to sup-port this argument, and it should not be ac-cepted without evidence. For an ingeniousexplanation of the sex differences in terms ofcareer expectations see Stinchcombe (1964,pp. 64-81); we shall return to this point below.It is clear, however, and no surprise, that boyspresent the major threat to the authority systemof the school.

SOCIOECONOMIC STATUSAND COLLEGE ASPIRATIONS

The socioeconomic status of a child, as in-dicated by the occupational status or income ofhis father, is one of the best predictors of aca-demic success and attainment (Sewell & Shah,

8

1968). There have been many speculations andsome investigations to interpret this associa-tion; some of these speculations imply that thesame processes leading to low academic attain-ments of students from low socioeconomicstatlks (SES) origins will lead these studentsto be behavior problems in the schools. Whilethese arguments may have some validity in ex-plaining different levels of academic aspira-tion and success, there has been no good evi-dence to support their validity in explainingdifferent levels of student misbehavior. Thedata in the present study lend no support tothe idea that social class of origin is a majordeterminant of student misbehavior. The zero-order association between father's occupationand misbehavior is shown in Table 4. Thedata show the highest rates of misbehavior forthe middle SES groups, lower rates.for thehighest and lowest groups. Table 5 showssimilar data for the relation between skippingand SES for girls; here there is essentially nocorrelation.

(In originally coding father's occupation anattempt was made to produce categories thatwould be ordered according to level of living;hence, farmers were included in the other oc-cupational categories. In retrospect it wouldbe desirable to have data stated separately forchildren of farm origins, but recoding wouldbe too costly. We were also interested in dif-ferences among the 10 high schools in our sam-ple with respect to the association between SESand rebelliousness. In what types of socialsettings, if any, are children of lower SESorigins more rebellious? Unfortunately, mean-ingful conclusions cannot be reached becauseof the small numbers in some SES categories inmany of the schools.)

If social class of origin is not associatedwith rebelliousness, social class of destina-tion is. Our indicator of this latter variableis intention to go to college. Sewell and Shahhave shown (1968) that high school students'

Table

Rates of Being Sent Out of Clby Socioeconom

4

ass for Disciplinary Reasons,is Status and Sex

Father's Occupation

Professional, managerial,owner of large businessor large farm

Sales and clerical, owneror operator of medium-sizedfarm

Skilled worker or foreman,service, small business orfarm**

Semi-skilled or unskilledworker, unemployed

Total

Percentage sent out of class more thanonce for disciplinary reasons(Number of cases in base of percentage)*

Boys Girls

15% (104) 3% (96)

29% (84) 2% (94)

23% (303) 4% (3k7)

18% (154) 1% (164)

22% (645) 3% (661)

* Non-responshere and in

** A small nuhas been

es to questions about being sent out or skipping excludedfollowing tables.

mber of cases where father's occupation was not ascertainedincluded in this category.

Table 5

Rates of Skipping School with a Gang of Kids bySocioeconomic Status for Girls

Father's Occu pation

Professioowner o

Saleso per

nal, managerial,f large business or farm

and clerical, owner orfor of medium-sized farm

Skilled worker or foreman,service, small business or farm

Semi-skilled or unskilledworker, unemployed

Total

Percentage of girls skipping with agang of kids

More than once Once Never (N = 100%)

12% 11% 76 (96)

9% 9% 83 (93)

12% 13% 74 (306)

11% 8% 80 (166)

12% 11% 77 (661)

9

stated intentions to go to college are well cor-related with actual attendance, and attendanceat college is highly correlated with later SES(Blau & Duncan, 1967). The data in Table 6show a strong association between collegeexpectations and rebelliousness; those boysfor whom high school is not a means to theend of college admission are far more likelyto misbehave than the others.

There is a moderate positive correlation be-tween social class origins and plans to attendcollege (gamma = +.33 for boys and +.28 forgirls). This might be expected to produce anegative correlation between SES and rebel-liousness. We have shown that this is notthe case, and the data in Tables 7 and 8 showwhy: within categories of students equallylikely to attend college, SES is positively re-lated to rebelliousness. The data from thepresent study do not enable us to interpretthis relationship, but Stinchcombe (1964, Ch.6) presents data supporting the following argu-ment. High school student rebellion is a re-action to failure to succeedan overt rejectionof the identity given the student by the school.This reaction will be stronger if there is greaterpressure to succeed. Since parental pressureto go on to college is positively associatedwith social class (see Stinchcombe, 1964;Sewall & Shah, 1968), the student of highersocial status origins who is doing poorly inschool and unlikely to attend college will bemore likely than others to display rejection ofthe identity conferred upon him in school byrebelling against the school. This socialclass effect will be less important for girlsthan for boys, since girls are less likely tohave parental pressure or support to attendcollege (Sewell and Shah, 1968), and the data

in Table 9 do show in fact little correlationbetween SES and skipping for girls when col-lege expectations are controlled.

With cross-sectional data such as we havehere the hypothesis that failure causes rebel-lion cannot be positively supported. It isalso plausible to argue that misbehavior inschool leads to poor performance and littledesire to go on to college. Both processesmay occur, and an examination of the effectsof intelligence on rebellion can present someevidence in support of both lines of reasoning.

INTELLIGENCE

Intelligence is an excellent predictor ofsuccess in school examinations and grades.It is not a very good predictor of student mis-behavior in school. The overall results arecurvilinear: % of boys sent out ofIntelligence6 class at least twice for

misbehavior

less than 100 20% (155)100-106 29% (147)107-114 22% (109)

115 or higher 14% (90)

6Intelligence was measured when the studentwas in 10th or 11th Grade with the Henmon-Nelson Test of Mental Ability (Henmon-Nelson,1942). This was taken from school records forseven of the eight school systems in our sam-ple; only Lorge-Thorndike IQ scores wereavailable for students in the eighth system,and they have not been included in these tab-ulations.

Table 6

Rates of Being Sent Out of Class for DisciplinaryReasons by College Expectations and Sex

"How likely do you think itis that you will go on to acollege or university?"

Percentage sent out of class more thanonce for disciplinary reasons (N)

Boys Girls

Definitely will go on 12% (222) 2% (218)

Probably will go on 22% (115) 2% (116)

May go on but not sure 21% (115) 2% (125)

Probably will not 23% (117) 4% (107)

Definitely will not 47% (79) 5% (97)

10

Table 7

Rates of Being Sent Out of Class for DisciplinaryReasons by Socioeconomic Status and College

Expectations, for Boys

Father's Occupation

Percentage sent out of class morethan once for disciplinary reasons (N)

Expectations to attend collegeNot sure,

Definitely Probably will probably not,plan go on or definitely not

White collar, owner oroperator of large businessor large or medium-sizedfarm *

Skilled worker, foreman,service, small businessor farm

Semi-skilled or unskilledworker, unemployed

13% (111) 34% (35) 34% (41)

12% (75) 18% (51) 30% (177)

9% (34) 14% (29) 23% (91)

* Two highest SES groups in Table 4 are combined here to assure enough casesfor reliable percentages.

Table 8

Rates of Skipping School with a Gang of Kids bySocioeconomic Status and College Expectations, for Boys

Father's occupationDefinitely

plan to

Percentage skipping school with agang of kids more than once (N)

Expectations to attend college

Not sure,Probably will probably not,

go on or definitely not

White collar, large business,large or medium-sized farm 13% (112) 31% (35) 17% (41)

Skilled worker, foreman, service,small business or farm 11% (75) 22% (51) 35% (176)

Semi-skilled or unskilledworker, unemployed 9% (35) 11% (28) 24% (92)

11

Table 9

Rates of Skipping School with a Gang of Kids bySocioeconomic Status and College ExpectaLions, for Girls

Father's OccupaticinDefinitely

plan to

Percentage skipping school with agang of kids more than once (N)

Expectations to attend college

Not sure,Probably will probably not,

go on br definitely not

White collar, large businesslarge or medium-sized farm 8% (99) 9% (32) 16% (58)

Skilled worker, foreman, service,small business or small farm 5% (85) 14% (52) 16% (168)

Semi-skilled or unskilledworker, unemployed 6% (33) 0% (32) 17% (101)

The highest rate of misbehavior occurs for boysof average intelligence; rates are lowest forthe highest intelligence category and next low-est for the lowest intelligence category. Thissuggests that it is not low intelligence or re-sulting poor performance in school that givesrise to misbehavior but the meaning given topoor performance by the student and his sig-nificant others.

We have suggested above that parental ex-pectations of sons' academic performance are

positively correlated with social status. Giventhese differential parental pressures, the highsocial status parent of the low IQ son can beexpected to find his poor school performancemore distressing than the low IQ son of a lowstatus parent. This implies that SES and re-bellious behavior will be pop4.tively correlatedat low IQ levels, but not necessarily at highIQ levels; the data in Table 10 show this tobe the case. Among those with IQ's less than100, sons of white collar parents are more than

Table 10

Rates of Being Sent Out of Class for DisciplinaryReasons by Socioeconomic Status and Intelligence, for Boys

Father's Occupation

White collar, owner or operatorof large business or large ormedium-sized farm

Skilled worker, foreman, service,small business or small farm

Semi-skilled or unskilledworker, unemployed

Total

Percentage sent out of class more thanonce for disciplinary reasons (N)

Intelligence*Less than 100 100-106 107-114 115 or higher

43% (21) 18% (40) 20% (41) 15% (40)

16% '(70) 31% (71) 26% (38) 14% (28)

9% (47) 32% (25) 17% (24) 14% (22)

20% (155) 29% (147) 22% (109) 1.4% (90)

* Henmon-Nelson Test of Mental Ability

12

twice as likely than sons of manual workingparents to have been sent out of class morethan once for disciplinary reasons. On theother hand, for those with IQ's greater than114, there is no association between SES andthis indicator of rebelliousness. Stated theother way around, only for sons of white col-lar workers is IQ negatively correlated withrebelliousness. (These conclusions must beespecially tentative because of the small num-ber of cases in some of the cells in Table 10.Small cell sizes are produced by the usualpositive correlation between SES and IQ; thecorrelation can be seen by inspecting the dis-tribution of bases of percentages in Table 10,where, for example, sons of white collar workersaccount for 44% of those with IQ's over 114 butonly 16% of those with IQ's less than 100.Furthermore, unreliability in response to thequestion about being sent from class may in-crease as IQ decreases.)

Other aspects of Table 10 are not readilyinterpreted. The negative or curvilinear asso-ciation between SES and rebelliousness for boyswhose IQ's range between 100 and 114 does notappear in other subgroups in this table or else-where. Looking at the data the other way, itis difficult to explain the strong curvilinearrelation between IQ and rebelliousness for sonsof manual workers. Perhaps those with low in-telligence and low SES have very low levels ofaspiration and find failure unthreatening, whilethose of very high intelligence and low SES arenot frustratedit turns out that almost 80% ofthe boys in this category expect to go to col-lege. Thus it is the boys of average intelli-gence and low SES who may have high aspira-

tions that are frustrated by their school ex-periences and who rebel as a result of thisfrustration.

This consideration of the joint effects ofSES and IQ on rebellion supports the line ofreasoning that rebellion is a consequence offailure, not an antecedent. However, an ex-amination of the joint effects of IQ and col-ege expectations does not provide data thatseem consistent with this line of reasoning.If low intelligence leads to low expectationsto attend college [and it does], and if rebel-lion occurs when boys find school irrelevantto their futures and a threat to their identities,then one might expect to find that there is lit-tle association between IQ and rebelliousnesswithin categories of boys equally likely toattend college. Table 11 shows that this isnot the case. Among those not expecting toattend college there is a strong positive cor-relation between IQ and rebelliousness. Thedata seem more consistent with the line ofargument that some boys become alienatedfrom school (for reasons not known but not asa result of poor performance) and then decidenot to continue to more of the same in collegeand also express their alienation in ways thatlead to their expulsion from class. The datafrom Table 11 have been percentaged differentlyin Table 12 to provide support for this argument;rebels are less likely than non-rebels to planto go to college, especially among those withhigh intelligence.

In summary, intelligence is not associatedwith rebelliousness in any simple fashion.Consideration of the joint effects of intelli-gence and SES supports the idea that rebellious

Table 11

Rates of Being Sent Out of Class for DisciplinaryRe'asons by College Expectations and Intelligence, for Boys

"How likely do you think itis that you will go on to acollege or university?"

Percentage sent out of class more thanonce for disciplinary reasons (N)

IntelligenceLess than 100 100-106 107 or higher

Definitely will go on orprobably will go on 13% (30) 20% (64) 13% (150)

May go on but not sure 14% (36) 24% (37) 28% (18)

Probably will not orDefinitely will not 25% (89) 43% (46) 40% (30)

Total 20% (155) 29% (147) 19% (199)

13

Table 12

College Expectations by Intelligence and Rebelliousness, for Boys

Intelligence

Percentage definitely or probably planningto attend college or university (N)

Rebels, thosesent out of classmore than once fordisciplinary reasons

Non-rebels, thosenot sent out of classor sent out only oncefor disciplinary reasons

Less than 100

100 to 106

107 to 114

115 or higher

13% (31) 21% (124)

31% (42) 49% (105)

50% (24) 73% (85)

62% (13) 93% (73)

behavior is a consequence of failure, whileconsideration of the joint effects of intelli-gence and college expectations supports theidea that both rebellion and lack of interestin going to college are the consequence ofprior alienation. Both types of causal proces-ses probably occur, but the data from thisstudy do not enable us to evaluate their rela-tive importance.

PEER RELATIONS AND REBELLION

We have shown the relations between re-bellious behavior and a student's social classof origin and his likely social class of desti-nation. But the acts of rebellion we are ex-amining here always occur in the context of aschool classroom, and the position of a stu-dent in the prestige system of the classroomcan also be expected to affect his conformingor rebellious behavior. The rebel may be ag-gressive against his classmates as well ashis teacher, or, conversely, he may be rebel-ling against his teacher on behalf of his class.The rebel threatens the prestige and authorityof the teacher, and the teacher's reaction tothis may be contingent upon the student's so-cial position among his classmates.

We measured sociometric status in the class-room by asking each student the followingquestions:

Who are the 4 students in this class-room group who you think most oftenget other students in this class tofollow them?

The number of choices received by a stu-dent on this question (which may rangefrom zero to about 30, the number of stu-

14

dents in the average classroom) is takenhere as an index of his leadership status.

Which 4 students in this classroomgroup do you like the most?

The number of choices received by a stu-dent on this question, which also mayrange from zero to about 30, is taken hereas in friendships, or popularity.

Before presenting the results of our analysis,two limitations must be noted. First, our de-pendent variable of being sent from class re-fers to behavior in the past in either the class-room where the questionnaire was administeredor other classrooms in the same or precedingyears, while the leadership and friendship in-dexes apply only to the specific classroom. Inwhat follows we shall assume that sociometricstatus in a particular class is a satisfactory in-dex of sociometric status in the larger schoolsetting and that it is relatively stable over time.Our measurements will be invalid to the extentthat these assumptions are mistaken; it seemslikely that this will be especially problematicamong those students with few or no choices re-ceived in the particular class. Second, in thetables to follow we pool students from many dif-ferent classrooms without standardizing for sizeof classroom. The larger the classroom, thelarger the number of possible choices; in addi-tion, the most important leader in one classroomof 30 students may receive many fewer leadershipchoices than the most important leaderin anotherclassroom with the same number of students,since the distribution of choices received mayvary greatly among different classrooms.

Despite these measurement problems, thedata show interesting correlations between so-ciometric status and rebellious behavior. The

relationship is shown in Figures 2 and 3. It isclear that sociometric rejects are not more likelythan others to be rebellious. Neither are socio-metric stars: the boys most likely to be rebel-lious are those with eight or nine leadershipchoices or five to eight friendship choices;those in these categories are about twice aslikely to have been sent out of class more thanonce for disciplinary reasons than those withthe fewest or those with the most choices.7

These curvilinear associations can be in-terpreted in the following way. First, somepeer support is necessary before boys willhave the courage to challenge authority; socialisolates lack this support. Second, teachersare more apt to react to disobedience with ex-pulsion when the offender has a following inthe classroom than when the offender is anisolate, for the isolate's actions are lesslikely to be contagious. Third, boys withmany leadership choices are probably securein their leadership statuses and are probablyconfirmed in them by both peers and teachers;it is unlikely that rebellion will help enhancetheir statuses. But boys with a small follow-ing (say eight or nine who would choose themas leaders) may contest for leadership of class-rooms by challenging the authority of the teach-er. A dramatic illustration of how somethingthat may be similar to this occurs in violentgangs has been presented by Short and Strodt-beck (1963). In a study of street gangs inChicago they noted numerous instances wheregang leaders precipitated acts of aggressionagainst outsiders when their leadership stat-uses were threatened, and they suggest that"leaders resort to this action because of thelimited resources they have for internal con-trol of their groupparticularly when theirstatus is attacked."8 High school students

7These data are for boys. A similar curvi-linear relationship between friendship statusand rates of skipping can be observed for girls,with the exception of a slightly higher skip rateamong girls with no friends than among girlswith one or two friends in the classroom; iso-lation evidently leads to physical avoidance ofthe setting.

8A study of a women's ward in a mentalhospital also illustrates this differentiationof leadership, with some inmates popular andalso respected by the hospital staff, other in-mates popular with the patients because theyexpressed hostility to the staff. See Perrucci(1963).

45-

40-

i t 35-1"a'A

O . 25-

e.) 20-o

4-4au11U°°

30-

15

10-

5-

0

0 1

(Number of casesin be ofpercentage) (195) (100)

2-5 6-7 8-9 10-13 14 or more

Number of leadership choices received

(173) (41) (29) (53) (57)

Fig. 2, Rate of Being Sent Out for Discipli-nary Reasons by Number of ChoicesReceived as Leader, for Boys

Eo 40-2w 35..

V p

o ra

O.

6C

Iu'au

ulo «

10-a.

30-

25-

20-

15-

5-

0

(Number ofcases in baseof percentage)

0 1-2 3-4 5-8 9 or more

Number of friendship choices received

(58) (210) (174) (170) (33)

Fig. 3. Rate of Being Sent Out for Discipli-nary Reasons by Number of ChoicesReceived as "Like Most," for Boys

15

are ordinarily ambivalent toward authority, andhigh school classrooms often include differentleaders, some of whom express tendencies toconformity and others of whom express the stu-dents' alienation.

This picture can be made clearer if one con-siders the student's perception of how hestands in the social life of the school as wellas his objective leadership position. Weasked the following question about this.(Taken from Coleman, 1961; see that volumefor other interesting uses made of the re-sponses.)

The people in this school who are mostimportant and most looked up to couldbe called the leading crowd. Supposethe center circle below represented theleading crowd. Ho'w far out from thecenter are you? (Place a check in thecircle where you think you are.)

The response categories to this question wererepresented by five concentric circles. Table13 shows that those boys who perceive them-selves to be either in the leading crowd or farfrom the center are mcist likely to be rebellious.Those who perceive themselves as part of theleading crowd include some of those contestingfor leadership in the school, while those per-ceiving themselves far from the leading crowdinclude some who reject the status structure ofthe school and lead others against it.

The contest for leadership can be expectedto be most significant to those boys for whom

the school is not a means to an end such asentrance to college and higher occupationalstatus., For them especially school has valuein terms of the immediate status gratificationsthey can gain from their peers. The plausi-bility of this line of reasoning is shown bythe data in Table /4 and Figure 4. Amongthose definitely planning to go on to college,leadership status has little association withrebelliousness, while the association is verystrong among those who do not expect to io onto collegeand among them the curvilinearitynoted above almost disappears. Conversely,college expectations influence the rebelliousbehavior of leaders far more than that of non-leaders: the difference in percentage rebel-lious between those definitely planning oncollege and those not planning on collegeranges from 7 percentage points among thosewith no leadership choices up to 28 percentagepoints among those with 10 or more leadershipchoices. (This suggests a way in which theoriginal curvilinear relation may arise, al-though the small number of cases does notenable us to test this reasoning in detail.Suppose leadership status is uncorrelatedwith rebelliousness among those expecting togo to college and has a linear positive corre-lation with rebelliousness among those notexpecting to go to college, and that the num-ber of leadership choices received is stronglycorrelated with college expectations. Thefollowing hypothetical figures show how thiscould give rise to the curvilinearity observed:

Table 13

Rates of Being Sent Out of Class for Disciplinary Reasons byPerceived Distance from the School's "Leading Crowd," for Boys

Perceived distance from"leading crowd" More than once

Times sent out of class for disciplinary reasonsOnce Never Total (N)

1 Member of the leadingcrowd 36% 27 36 100 (75)

2 16% 19 65 100 (203)

3 19% 12 69 100 (251)

4 17% 16 67 100 (59)

5 As far as possible fromthe leading crowd 41% 11 48 100 (46)

'16 GPO e14 -475 -4

Table 14

Rates of Being Sent from Class for DiLeadership Status and College Ex

sciplinary Reasons bypectations, for Boys

How likely do you think itis that you will go on to acollege or university?

Definitely will go on

Probably will go on

May go but not sure,probably will not, ordefinitely will not

Total

Percentage sent out of class more thanonce for disciplinary reasons (N)

Number of leadership choices receivedNone 1 to 4 5 to 9 10 or more

11% (46) 13% (82) 12% (32) 10% (60)

10% (40) 26% (35) 36% (22) 22% (18)

18% (1 08) 31% (129) 42% (40) 38% (32)

14% (194) 24% (247) 31% (94) 20% (110)

ii

;4)

-a-1u

8g gg gCid .5

SO-

4540353025

2015105--

0

/

II

A%

0 1-4 5-9 10 or more

Number of leadership choices received

Top line: May go to college but not sure, probablywill not, or definitely will not

Middle line: Probably will go on to college ------ -Bottom line: Definitely will go on to college

Fig. 4. Rates of Being Sent Out of Classfor Disciplinary Reasons by Lead-ership Status and College Expec-tations, for Boys

Rebel-Rebel- lious-lious- nessness Among Rebel-Among Those % of row lious-

Leader- the Not who are nessship College College College WeightedStatus Bound Bound Bound Average

1 - low- .10est

2 .103 .104 .105 - high-

est .10

. 10 .50 .10

. 25

. 40

. 55

. 60

. 70

.80

.16

.19

.19

.70 .90 .16

To make this kind of model fit the data availableone would have to assume that those sayingthey aren't sure about their college plans area mixture of the other two types: one can con-ceive of this kind of "mixing existing withina single boy unsure of his plans and of hisfeelings about school.

The data we have presented here suggestthat peer support tends to be a necessary don-dition for student misbehavior, and the dataare consistent with the theory that the rebelliousstudent contests for leadership by expressingshared feelings of alienation from authority.These effects are important only for those boysnot expecting to go to college, because for themthe peer group is a relatively more importantaspect of life in the school.

17

IV

SUMMARY AND IMPLICATIONS

In this report we have described the loca-tion of rebellious students, those who openlychallenge the authority of the school, in thepopulation of high school students. Rebelsare far more likely to be boys than girls, andthey are especially likely to be those boyswho do not expect to attend college. There isno clear correlation between father's occupa-tion and rebellious behavior, although SES andcollege expectations do interact in affectingrebellion: boys and girls from high SES fami-lies who do not expect to attend college aremost likely to become disciplinary problems.Students of average IQ are more likely to berebellious than those with either low or highIQ. Intelligence interacts with SES and col-lege expectations to affect rebellion; those ofhigh SES origins and low intelligence, andthose of high intelligence and low expecta-tions of attending college, are more likely tobe rebellious than other students. Boys ofmarginal leadership status, who have somefollowers but are not among the top leaders,are more likely to be behavior problems inhigh school than either non-leaders or the topleaders, although this is true only for thosenot expecting to attend college. The rebelsare likely to perceive themselves either as inthe leading crowd of their school or as remotefrom the leading crowd.

These results are consistent with Stinch-combe's (1964) in all respects. They servedecisively to refute the idea that rebellion inhigh school comes primarily from working classyouth. The data support but do not proveStinchcombe's theory that student rebellion isa reaction to failure: that school, for the stu-dent not planning to attend college, is notonly irrelevant but such a threat to self esteemthat open expression of alienation is necessaryto preserve integrity of the self. Other resultspresented above support the contrary theorythat alienation from school is prior to both re-bellion and failure. The data further support

18

but do not prove the theory that rebellion inthe classroom requires peer support and isused to enhance leadership among fellow stu-dents.

RESEARCH IMPLICATIONS

Although socioeconomic' status is not cor-related with rebellion in the Wisconsin highschools studied here or in the Northern Cali-fornia community studied by Stinchcombe, itseems quite likely that if a national surveywere conducted it would show that schoolsserving communities of working class or poorpeople would have more problems in control-ling behavior. That is, while there may be nocorrelation between SES and rebellion withinparticular schools, there might be amongschools. The data here did not allow us totest this hypothesis, although we hope to beable to do so when data from the larger COPEDsurvey become available. (We might note thatin the Wisconsin community studied that hadtwo high schools, the one with the higher pro-portion working class had markedly higherrates of rebellion than the other.) Furtherstudies are needed, then, to ascertain theeffects of class segregation and race segrega-tion on rates of student rebellion.

Further studies of student disciplinary prob-lems should also include other types of devi-ance. It is easy to suggest hypotheses re-lating the independent variables studied hereto these other types of deviance, and these,hypotheses need testing to establish a goodpicture of deviance and assure that policychanges to control one type of deviance do notproduce other types. For example, it seemslikely that parental pressure to succeed iscorrelated with both SES and cheating,9 that

9See the study of college student cheatingreported in Bowers (1964).

non-leaders are more likely than leaders toengage in passive withdrawal in the classroomor to drop out of school, that girls are morelikely than boys to be overdependent upontheir teachers, and that high SES students andtop leaders are more likely than others to at-tempt to manipulate school authorities.

It seems likely that longitudinal studieswill be necessary to firmly establish theoriesof the genesis of deviance. The same studentsmust be observed at different points in time toascertain the temporal and causal priority offrustrated aspirations, failure, and alienationand to show how variables such as these arerelated to status in peer groups.

We also need additional studies to showhow the background characteristics of studentsenter school situations in such a way as togenerate stress. Since rebellion occurs ingroup settings, we need to relate character-istics of individual disruptive students to thecharacteristics of the groups in which theyact. In our next paper in this series we shallpresent some data on this point as well asshowing how other characteristics of class-rooms and teachers influence the amount ofdisorder in classrooms.

POLICY IMPLICATIONS

If frustrated aspirations lead to aggressivebehavior in schools, as we have suggested,then it might be thought that what is needed isa way of gently reducing the degree of frustra-tion before it produces aggression. That is,just as the confidence man "cools out" themark, his victim, after initially giving himhigh expectations of gain (Coffman, 1952), sothe school as confidence man must find waysof cooling out students after it has helped mo-tivate them to have high aspirations of success.Burton R. Clark (1960) has shown how many ofthe procedures used in junior colleges, suchas extensive testing, counseling, and coursesin occupational selection, can be interpretedas ways of gently reducing the aspirations ofstudents who want but will not be able to ob-tain 4-year college diplomas. High schoolguideance counselor, porbably do not servethis function very often; ordinarily theywork to raise aspirations by directing quali-fied students to appropriate colleges. Itseems unlikely that high school counselorscould be as successful as junior college coun-selors in cooling out high aspirations, sincethe former, unlike the latter, have few choicesto offer the student who will not succeed aca-demically.

Ralph Turner (1960), in comparing Britishand American education, suggested that theformer blunted the aspirations of students whowould not be permitted to enter colleges at amuch earlier age than the latter; this, he sug-gested, would produce stability in a systemof "sponsored mobility." The English proce-dures leading to lower levels of aspiration in-clude early determination (at age 11 plus) ofwho shall be permitted to enter college andan early segregation of the college bound fromother students. It seems unlikely to us thatsuch procedures serve to reduce classroomdisorder. In fact, Hargreaves (1967) attributesrebellion in the English school he studied inlarge part to just such segregation. In futurereports we hope to present more data relevantto this point.

If cooling-out procedures are not likely tobe effective in reducing school disorder in theAmerican context, then policy changes to beconsidered might include making school morerelevant and rewarding for those students mostlikely to become rebellious. School can bemade relevant to the student not bound forcollege by giving instruction that has clearvalue for the sorts of jobs he can in fact obtain.Stinchcombe (1964, Pp. 107-109) presents im-pressions of apprentice training programs forthe manual trades and suggests that studentsin them are committed and conforming. Theseprograms include training in skills that areclearly relevant to work being performed, theygive the student rights in a labor market thatare at least as valuable as the right to attendcollege, and they give students symbolic re-wards that are clearly important to them. Or-dinarily nothing like this exists for studentsin vocational training programs in publicschools. (However, we do need studies ofvocational schools, good and bad, to see whatcan be accomplished in such settings to getcommitted and orderly student bodies. MartinMayer (1963, Pp. 363-369) gives anecdotesindicating that at their best vocational schoolsfor lower class students can be very effective.)Vocational programs are often irrelevant andunrewarding, and it will require much additionalfinancing to change them.

If, for some students, school cannot be maderelevant as a means to aid in achieving futureends, perhaps they can be made rewarding inthe present as places in which every studentcan gain significant social responses for theexpression of whatever skills he does possess.American schools have long recognized thisand have emphasized the value of art, music,athletics, and social activities in the schoolprogram. These activities help ameliorate the

19

tendency of schools to become unbearablymeritocratic institutions (where "Merit = IQ +Effort") ,10 for they demonstrate that an indi-vidual's value is not only a function of nar-rowly defined academic talents. However,since there tend to be strong correlationsamong the various abilities mentioned, e.g.,book learning, artistic expression, athleticskill, social skills, etc., and since some ofthem often clearly lack significance, they donot compensate for the injuries inflicted bythe school upon those likely to become rebel-lious. Students alienated from school alsotend to be alienated from athletics, schoolsocial clubs, and other non-academic activi-ties associated with the school. (Cf. Stinch-combe, 1964, pp. 93-100, where it is shownthat rebellious students are more alienatedfrom and less likely to participate in schoolsocial activities and athletics than non-rebellious students.)

Still other ways of making schools morerewarding for those most likely to become re-bellious would be to grant adolescents certainadult privileges at an earlier age, privilegesof symbolic and hedonistic value such as drink-ing, smoking, sex, and car ownership (cf.Stinchcombe, 1964, -pp. 178-185 and passim).It is unlikely that this will be done. Thus,it seems unlikely that any of the possible pol-icy changes discussed here would be success-ful in removing the stresses that lead to stu-dent rebelliousness. If rebellion and disorderare to be minimized it is likely that it will bedone by controlling student behavior in alreadystressful situations. Teachers and other schoolauthorities vary greatly in their abilities toeffectively exercise such control, and in a fu-ture report we shall discuss the characteristicsassociated with such effectiveness and thepolicy implications of them.

IS THE NATURE OF STUDENTREBELLION CHANGING?

We have the uneasy feeling that the datawe collected in 1967, and the theories relat-

10 Cf. Young (1961). Our major objection tothe policies suggested by James Coleman inThe Adolescent Society (1961) are that theywould tend to make schools even more unbear-ably meritocratic. In a sense, the rebellionwe have depicted in this report is a rebellionagainst the meritocracy having some of thecharacteristics of the societal revolution des-cribed in Young's novel.

20

ing to them, are becoming obsolete. In theSixties it seemed that one could make a 'sharpdistinction between the rebellion typicallyoccurring in high schools and that providingserious threats to the order of college cam-puses. In universities the individual offenderas described here seemed to be absent; stu-dents are not constrained to go to college.Instead, the major threats to order have beenmore or less political: student movementshave claimed legitimacy, and they have beenled by superior students and students who havehigher educational expectations than the aver-age. [See, e.g., Somers (1965) on the 'Berkeleystudent revolt and Peterson (1968) for summariesof other studies.] While high school disruptorshave been rebels, neither claiming legitimacynor attempting to change the system, universitydisruptors appeared to be revolutionaries.Since the initial revolt at Berkeley there havebeen some signs of change. High school stu-dunts have been increasingly politicized. Theyhave come to question the basic legitimacy ofadult authorities who pursued what seemed tothem to be an immoral foreign war and partici-pated in what seemed to be immoral repressionof domestic minorities. This alienation fromadult authority in general has been applied toadult authorities in school settings. There isnow an ideology of "participatory democracy"that justifies student demands for more power,and some tenets of this ideology are widelyshared. Thus, a Louis Harris poll of U.S.students (1969) found that 58% wanted morestudent participation in school policy makingand that 54% of them thought that this was a"very important"matter.

As demands for "student power" have grownso also has a tendency to use disruptive tacticsto achieve this goal and other goals. A Na-tional Association of Secondary School Princi-pals poll found that "Three out of five princi-pals report some form of active protest in theirschools" in the 1968-69 school year (Herbers,1969). Alan F. Westin found 348 newspaperaccounts of high school disruption in the periodfrom November 1968 through February 1969;about 37% were racial protests, 22% political(including Viet Nam), 20% against dress regu-lations, 17% against discipline, and 17% foreducational reforms. If less active protestsare considered, they are even more widespread;thus the NASSP survey reported that 82% of theschools have some form of protest againstschool regulations (Herbers, 1969).

The change in the nature of student aliena-tion and the ways in which it is expressed hasperhaps been most extensive in Negro schools.We need studies of these new expressions ofalienation in all settings to determine if high

school student revolutionaries are, like theircollege counterparts, superior students withhigh aspirations, and we need to know whatrelations exist between the newer kinds ofrevolutionary behavior and the rebellious be-havior discussed in this report. Does therebel become enlisted into the revolutionarymovement and use its ideology to justify dis-obedience? Or do the different types of alien-ation from authority exist parallel to and per-haps in conflict with one another?

At the same time, the ideologies of univer-sity radicals increasingly include anti-meritocratic aspects, such as an oppositionto testing, grading, admission policies basedupon tests and grades, and the appointment ofprofessors on the basis of scholarly produc-tivity alone. In many respects university and

high school students have become more simi-lar. High school students have become moresophisticated (for which superior instructionmay be partly responsible), an increasingproportion of all students attend college, andstudents increasingly feel constrained to at-tend college in order to escape military serviceand to assure themselves of white collar jobs.

Disorder in schools is increasing, frequentlyto the point where armed police officers mustbe present in school buildings to control it.School authorities probably will have to be-come more adept in the techniques of repres-sion. Yet repression is inconsistent witheducation, and no amount of naked power canassure order in schools. The authority ofschool professionals depends on the willingassent of students.

21

REFERENCES

Blau, P. M. & Duncan, 0. D. The Americanoccupational structure. New York: Wiley,1967.

Bowers, W. J. Student dishonesty and itscontrol in college. New York: Bureau ofApplied Social Research, Columbia Uni-versity, 1964.

Clark, B. R. The "cooling-out" function inhigher education. American Journal ofSociology, 1960, 65, 569-577.

Cohen, A. K. Delinquent boys: The culture ofthe clang. Glencoe, Ill.: Free Press, 1955.

Coleman, J. S. The adolescent society.Glencoe, Ill.: Free Press, 1961.

Daniels, A. K. & Daniels, R. R. The socialfunction of the career fool. Psychiatry,1964, 27, 219-229.

Dent ler, R. A. & Erikson, K. T. The functionsof deviance in groups. Social Problems(Fall, 1959), 7, 98-107.

Douvan, E. & Adelson, J. The adolescentexperience. New York: Wiley, 1966.

Fox, R. S., Lippitt, R. 0., & Schmuck, R. A.Pupil-teacher adjustment and mutual adapta-tion in creating classroom learning environ-ments. Ann Arbor: University of MichiganInstitute for Social Research, Final Reporton Cooperative Research Project No. 1167to the U. S. Office of Education, 1964.

Goodman, L. A. & Kruskal, W. H. Measuresof association for crps classifications.Journal of American Statistical Association,1954, 49, 733-764.

Goffman, E. Cooling the mark out: Some as-pects of adaptation to failure. Psychiatry,1952, 15, 463-541.

Goffman, E. The nature of deference and de-meanor. American Anthropologist, 1956,58, 473-502.

Hargreaves, D. H. Social relations in a sec-ondary school. London: Rout ledge & KeganPaul, 1967.

Harris, L. What people think about their highschools. Life, 66, (May 16, 1969), 22-33.

Henmon, V. A. C. & Nelson, M. J. TheHenmon-Nelson test of mental ability.Boston: Houghton Mifflin, 1942.

Henry, J. Attitude organization in elementaryschool classrooms. In G. D. Spindler (Ed.),New York: Holt-Rinehart-Winston, 1963.Pp. 192-214.

Herbers, J. High school unrest rises, alarm-ing U. S. educators. New York Times,30 (May 9, 1969), p. 1.

Hollingshead, August B. Elmtown's youth.New York: Wiley, 1949.

Lemert, E. M. Social Pathology. New York:Mcgraw-Hill, 1951.

Levine, J. M. & Allen, V. L. Reactions toattitudinal deviancy. Theoretical PaperNo. 14, University of Wisconsin Researchand Development Center for Cognitive Learn-ing, 1968.

Mayer, M. The schools. Garden City, NewYork: Doubleday Anchor Paperback, 1963.

Mial, D. Introductory Note What is COPED.In Goodwin Watson (ed.) Change in schoolsystems, pp. v-ix. 1967 Washington D. C.:National Training Laboratories, NationalEducational Association.

Parsons, T. The Social system. Glencoe,Ill.: 'Free Press, 1951.

Perrucci, R. Social distance and stratifica-tion on a psychiatric ward. American Socio-logical Review, 1963, 23, 951-962.

Peterson, R. E. The student left in Americanhigher education. Daedalus, Winter 1968,97, 293-317.

Phillips, W. The influence of social class oneducation: Some institutional imperatives.Berkeley Journal of Sociology, Fall 1959,5, 63-91.

Schachter, S. Deviation, rejection and com-munication. Journal of Abnormal and SocialPsychology, 1951, 41*, 190-207.

Schmidt, B. A. The relationship between so-cial status and classroom behavior. (Un-published study, Washington University,

23

1958.) Cited by Glidewell, J. C., Kantor,M. B., Smith, L. M. & Stringer, L. A.,Socialization and social structure in theclassroom. Review of Child DevelopmentResearch, Z, p. 246. In L. W. Hoffman &M. L. Hoffman (Ed.). New York: RussellSage, 1966.

Sewell, W. H. & Shah, V. Parents' educationand children's educational aspirations andachievement. American Sociological Review,1968, 33, 191-209.

Short, J. F. & Strodtbeck, F. L. The responseof gang leaders to status threats: An obser-vation on group process and delinquent be-havior. American Journal of Sociology, 1963,68, 571-579.

Somers, R. H. The mainsprings of the rebellion:A survey of Berkeley students in November

24

1964. In S. M. Lipset & S. S. Wolin (Ed.),The Berkeley student revolt: Facts and inter-Pretations. Garden City, New York: Double-day Anchor paperback, 1965.

Stinchcombe, A. L. Rebellion in a high school.Chicago: Quadrangle Books, 1964.

Turner, R. H. Sponsored and contest mobilityand the school system. American Socio-logical Review, 1960, 25, 855-867.

Waller, W. The sociology of teaching. NewYork: Wiley, 1932.

Werthman, C. Delinquents in schools: A testfor the legitimacy of authority. BerkeleyJournal of Sociology, 1963, 8, 39-60.

Young, M. The rise of the meritocracy: 1870-2033. Baltimore: Penguin Books, 1961.

GPO 014-475-3