Journal of Youth and Adolescence, Vol. 18, No. 4, 1989

The Social Networks, Family Involvement, and Proand Antisocial Behavior of Adolescent Males in Norway M o n c r i e f f Cochran I and Inge B 6 2 Received June 24, 1987; accepted April 10, 1989

The focus of this study is on the ecology of pro- and antisocial behavior. The study was conducted in Stavanger, Norway, with a representative sample of ninety-two 16-year-oM boys. Data collected included socioeconomic background, neighborhood risk level, amount of time spent with parents and peers, maps of social network relations, self-reports o f alcohol use and criminal activity, and school reports of academic performance, truancy, school motivation, and social behavior. Analysis o f results produced two models linking background and process with outcome variables: (A) higher neighborhood risk and less time spent by the boys with their parents were linked with greater propensity for self-reported alcohol use and illegal activity, and (B) more educated parents and larger numbers of nonkin adults in the boy's network were related to better school performance, less absenteeism, and more positively evaluated social behavior. Discussion of these findings centers on the neighborhood and family processes involved in social control, and on adult network members in their roles,as positive models, norm reinforcers, and sources o f information for adolescent boys. INTRODUCTION

The focus of this study is on the ecology of pro- and antisocial behavior in adolescence. The study has been conducted partly in response to ~Associate Professor, Department of Human Development and Family Studies, College of Human Ecology, MVR Hall, Cornell University, Ithaca, New York 14853, USA. He is trained in psychology and sociology. His current interests are in development in context with particular reference to personal social networks, child care, and family policy. 2Associate Professor,:Hfgskolesenteret in Rogaland, 4001 Stavanger, Norway. He is trained in youth development. His current interests are in youth culture, including the influences of the social network, schooling and clubs and organizations. Reprint requests should be addressed to the first author. 377 0047-2891/89/0800-0377506.00/09 1989Plenum Publishing Corporation

378

Cochran and B6

a growing concern among both social scientists and youth work practitioners with two changes related to adolescents. The first involves their social worlds, which have become increasingly segregated by age, consisting more and more of peers and less and less of adults and children (Bronfenbrenner, 1973; Coleman, 1971; Condry and Siman, 1974; Coser, 1982; Nordland, 1980; NOU, 1977). The second change is in antisocial behavior and impaired mental health among teenagers, both of which have grown in incidence and in severity during the past 20 years (Cederblad and H66k, 1985; Advisory Committee on Child Development, 1976; Lavik, 1976; NOU, 1985; Otnes, 1983). Despite increased recognition of the possible causal relationship between these phenomena, a review of the literature on adolescence and youth uncovers surprisingly few studies addressing this issue. The impact of what is known is reduced both by conceptual gaps in existing studies and by the limitations imposed by restricted samples. Three shortcomings in these existing literature are of particular concern. First, conceptualization of the social environment having significance for adolescents has been fragmented. One reason for this fragmentation has been a preoccupation with research on friendship and the impacts of the peer group on adolescent development. As Ladd (1984) points out, there seems to have been a tendency to juxtapose peer group with adult and child influences, rather than viewing them each as different components in a more encompassing social system. Two recent and interrelated developments are serving to counterbalance this preoccupation with peer group influences. The first is a growing conviction, both in Scandinavia and in the United States, that portrayal of adolescence as governed primarily by a "youth culture" and a peer orientation is a serious oversimplification (Andersson, 1982; BO and Boyesen, 1984; Cooper and Ayers-Lopez, 1985; Henricson, 1973; Listhaug, 1983). The second development has been adoption of the social network approach to describing the range of "significant others" involved in social relations with adolescents (Blyth et al., 1982; Blyth and Traeger, 1988; Gardsjord, 1972). A finding common to these studies has been that core family membersespecially parents-are surprisingly important to young people in this age group. While respondents in the studies, as they grew older, gave increasing salience to peer friends, at no point did peers overtake parents as the most significant people in the lives of these teenagers. Our second concern is directed at the very studies we believe are at the forefront of research relating adolescent social relations with behavior and development--those using a social networks approach. This small number of studies is characterized by models of person-environment interaction that are incomplete. Missing is any recognition of the possible role of social structure in explaining outcomes in the adolescent. This oversight can probably

Social Networks, Family Involvement, and Behavior

379

be attributed to an interest by the researchers in psychological constructs rather than behavioral outcomes. Blyth and Traeger (1988) and Coates (1985) were interested in adolescent self-esteem as it might relate to social relations with members of the young person's social network. Vondra and Garbarino (1988) have examined such relations with respect to adolescent psychological adjustment. None o f these researchers use models that take the social structure of the larger society into account. The significance of this neglect can be illustrated by the fact that the single strongest predictor of the network characteristics of both children and young adults is the educational background of the child's parents or of the young adults themselves (Cochran and Riley, 1988; Fischer, 1982; Lewis et al., 1984). The third shortcoming of the existing body of knowledge relating the personal networks of adolescents to developmental outcomes is simply the scarcity of studies, a shortage that becomes still more acute when the focus is narrowed to pro- and antisocial behavior. Only Blyth and his colleagues have examined the possible links between network characteristics and such outcomes (Blyth et al., 1985). They found that drug users had less intimate relationships with their parents and were much more oriented to older youth than were those adolescents not involved with drugs.

The Stavanger Study The aim of this study was to examine the influences of parents, peers, and other network members on the pro- and antisocial behavior of antisocial boys. The conceptual model embracing the study, shown below in Fig. 1, includes dimensions representing environmental forces operating at the exo-, meso-, and microsystem levels of the young person's social surround, as well as aspects of behavior in the real world (Bronfenbrenner, 1979; Cochran and Brassard, 1979; Cochran, et al., in press). The influences of social position, as manifested through educational background and occupational prestige, are thought of as embracing the network and family systems, and so exerting their influence on the adolescent both through those systems and directly. Included as a separate background dimension is the social ecology of the living area containing each subject's residence. Personal

Socioeconomicand~ CommunityFactors ~.~,

FamilyP r o c e s s e s /~ NetworkProcessesFig. 1. Conceptual framework.

Behavioral Outcomes

380

Cochran and B6

network and immediate family, viewed as meso- and microlevel phenomena, respectively, are distinguished from one another because we view them as systems with differing socioemotional properties, exerting different kinds of social influences. Both of these social systems can be thought of as influencing adolescent behavior, but little is known about similarities and differences in the processes of influence involved in the two systems, or for which types of behavior each process is most likely to be manifested. At the righthand side of the model a rather broad range of possible outcomes is included, in order to ensure that the research be able both to address the concern about antisocial outcomes and to provide some insight into the ecological processes associated with more constructive adolescent behavior.

METHODS Setting The study was conducted in the City of Stavanger, on the southwest coast of Norway, after a pilot study in a "rurban" village on a nearby fjord. Stavanger has traditionally served an economic base built on shipping, farming, and fishing, but has doubled in size in the past 20 years, to 100,000 inhabitants, due to its emergence as the "oil capital" of Norway. Combined with a neighboring city of about the same size, the resulting metropolitan area has a population of about 200,000, and is the fourth largest urban area in Norway. The recent spurt in population growth, while bringing new opportunities and cultural activities, has been accompanied by an increase in negative youth behaviors-including criminal activities and alcohol and drug use.

The two junior high schools included in the study were selected because they could provide a socioeconomic cross section o f Stavanger youth. The two schools serve 18 typical Stavanger neighborhoods, each of which contains a mix of lower middle-, middle-, and upper middle-class families.

The Sample The sample consisted of ninety-two 16-year-old boys attending the 9thgrade classes in the two schools. (The sample was limited to boys for several reasons. First, pilot studies had shown very little variation in the antisocial behavior reported by girls or their teachers; they all clustered on the positive end of the scales for self-reported crime, alcohol and drug use, and school absenteeism. Second, financial limitations restricted the overall size of the sample.) Of the 97 boys in those classes, 5 refused to participate in the study,

Social Networks, Family Involvement, and Behavior

381

either on their own initiative or at their parents' request. (Data provided by school personnel about the 5 boys who refused to participate indicates that they came from neighborhoods with risk scores at the mean of the achieved sample, were slight underachievers in school, and displayed school behavior typical of the final sample.)

Measures and Variable Descriptions The four main instruments used in the study are described briefly below. Descriptions of the variables derived from these measures are provided in Table I.

The Network Interview Form (NIF) in this interview the boys were asked to identify people (I) who had significance for them, (2) whose name they knew, (3) who knew their name, (4) who they did something with, and (5) with whom they were in regular contact. Examples given included friends of the family, relatives, friends from school and other settings, neighbors, teachers, and club leaders. Probes also mentioned people of different ages, the possibility of telephone and letter contact, and the fact that the person did not have to be liked to "have significance" for the boy. A number of kinds of information were gathered for each network member, including sex, age, length of time known, frequency of contact, perceived importance of the relationship, the role(s) occupied by the person (friend, neighbor, etc.), and activities engaged in together. The variables derived from data gathered with the NIF are described in Table I (Nos. 3-7).

The Background, Attitudes and Behaviors Questionnaire (BAB) The BAB contained a number of different measures. In it the boys provided the information about parents' years of schooling and occupations needed for the socioeconomic status variable (Table I, Variable 1). The measure of time spent in direct contact with parents and with peers was adapted from Frederiksen and Sorensen (1977). The boy was asked to indicate on a 5-point scale (never to > 4 hours/day) how much time he spent with parents (mother and father separately), peers, or alone on a typical weekday and weekend day (Variables 8 and 9 in Table I). A measure of attitudes toward and use of alcohol and drugs, developed for this study, was made up of ten 4-7-point Likert scale items (Variable 11 in Table I). Self-reported criminal behavior was also measured using Likert-oriented items, elaborated from

382

Cochran and Bt~

0

~o

% c-

.":

-~

,.-

~

~

9~

~

.-

..

o~,O

o

oo~_ ~

,.4

-

:~

~,

"a

o

~. ~

~ ~

~

.~

,.,.'." o ~

s~

~o

~

~'~ ':x"c,

~,-.~:

_.> o

o :~

o

o~

~

r~ .c

: ~ ~ ~.~.~

~ ~

> .. = =

o~~

m

~

~ ~

~ ~=~.'~ 9--

O . C O

~ oo' ~: ~

The social networks, family involvement, and pro- and antisocial behavior of adolescent males in Norway.

The focus of this study is on the ecology of pro- and antisocial behavior. The study was conducted in Stavanger, Norway, with a representative sample ...
1MB Sizes 0 Downloads 0 Views