189 教育研究集刊 第四十八輯第四期 2002 年 12 月 頁 189-220
美國中學生參與學校運動代表隊與
偏差行為發生的相關性之實證研究
董旭英 嬌妻 青少年參加學校運動代表隊是否能夠有效仿制偏差行為的發生，或者是增加青少年 偏差行為的發生率?這是一個非常爭議的問題。只可惜，很少經驗研究就此一議題 作詳細的探討。所以本研究嘗試回答下列兩個問題: (一)參與不同類型的學校運動 代表隊是否與青少年偏差行為存在著不同的關聯性?(二)參與某一類型的學校運動 代表隊是否與不同種類的偏差行為存在著不同的關聯性?本研究主要使用一九九 ζJ 美囡囡家教育長期追蹤資料庫，其中樣本包括 20，706 名中學生。本研究包涵三個依 變項﹒違反校就行為、暴力行為、吸食大麻;而自變項則是參與不同類型的學校運 動代表隊。在分析資料數據方面，主要採用巢式多元迴歸及邏輯迴歸統計分析技巧。 研究結呆顯示，參與不同類型的學校運動代表隊與青少年偏差行為有著不同程度的 關聯性。其次，參與某一類型的學校運動代表隊與不同種類的偏差行為也存在若不 同程度的關聯性。這些發況可能暗示若不同類型的運動項目有著不同的內容特性、 訓練形式及比賽規則，所以導致與青少年偏差行為的發生產生不同程度的關聯性。 在研究報告中也對未來相關研究提出具體建議。 關鍵字﹒學校運動代表隊、青少年偏差行為 本文作者為國立成功大學教育研究所助理教授 電子郵件為 yytung@mail.ncku.edu.tw 投稿日期 2002 年 3 月 13 日;採用日期 2002 年 6 月 10 臼Keywords: interscholasticspo吋s ， delinquency
marijuana. Participation in different kinds of interscholastic sports was the independent variable. Moreover, the method of neste廿 multiple and logistic regression was employed to analyze the collected data. The results displayed two meaningful findings. First,participation in dif的erent kinds of interscholastic sports had various relationships witha吐olescentdeviant behavior. Secondly,participation in the same sport had various associations with different types of deviant behavior committed. This could possibly be attributed to the different activities and processes (i.e., rules, construction, practice, etc.) associated with each sport Clearly
,
various relationships between participation in different kinds of interscholastic sports and adolescent delinquency were. Suggestions for future research are also presented in this study190
Bulletin of Educational Research
December,2002, Vol. 48No.4 pp. 189-220
The Relationship between Participation in
Interscholastic Sports and Delinquency in
American High Schools
Yuk-Ying Tung
Abstract
Does participation in interscholastic sports have positive or relationships with a益。lescent delinquency? Though an arguable question
,
empirical studies have focused on this topic. This study therefore, a社emptsanswer two questions. First，位adifferent kinds of interscholastic sports have relations with the tendancy toward of adolescent delinquency? Second
,
participation in one kind of interscholastic sport have variou effects on forms of adolescent delinquency? This study used data from the first survey of the National Education Longitudinal Study in 1990 which 20,706 high school students. This research included three dependent misbehavior at school,engagement in physical fighting at school,and theYuk-Ying Tung is the assistant professor at the Insti阻te of Education, National Kung University
E-mail address:yytung@mail.ncku.edu.tw
Manascript received ,Mar. 13,2002; Accepted ,Jun. 10,2002
together to determine how likely adolescents are to commit deviant behaviors. Attachment refers to connections with other persons,adults,or groups,such 品 parents
and peer groups,which serve as reference groups. When youths have close affectional ties to others and identifY with others, they care about others' expectations (Ake凹，
1997). Adolescents who participate in sport care about their coach's order and teammates' opinions. As a result,athletes are more likely to obey the rule of group and conform their peers. Commitment refers to the results of a cost-benefit analysis. Infa哎， commitment is the investments accumulated in terms of sake in conformity (Shoemaker,1990). In other words,adolescents who participate in sports activities are less likely to engage in deviant behaviors based on the cos的 and the benefits of conformity to rules of these activities. Youths understand that they will be asked to leave the team when they break the rule or commit delinquency. Involvement refers to the assumption that indi利益uals who spend considerable time and energy participating in conventional and legitimate activities have no time to engage in deviant behaviors Therefore, if adolescents spend more time and energy engaging in sports activities, they will have less time to commit delinquency. Final妙， belief refers to the acceptance of the conventional value system. Based on Hirschγs argume剖， youth who participate actively in sports activities are expected to be less likely to participate in deviant behaviors because these activities can establishan吐 strengthenthe social bondthrοugh
obeyiilg the values and rules of sport as a buffer against 'deviance. According to
Hirschγs argume帥， youth who a閃 involved in sports activitiesa把 less likely to participate iil deviant behavior because their participation in sports activities can strengthen the socialbOnd that acts as a social control
Thesetο吋 pe叩ectiveiscatharsis theory of aggression. The geneh\l premise here is that pecillle always face a lot 'of pressure from internal instinct and external environment,and individuals nee吐 tofind a means to relieve the degree of pressure in order to maintain a safe setting (Storr, 1985). Based on catharsis theory, human organism is seen as a steam boiler. To maintain optimalop叮ation，the boiler must have 192 Bulletin of Educational Research Vol.48 No.4
Introduction
Are there positive or negative relationships between participation in school an通 adolescent deviant behaviors? This is an arguable question. The results of studies indicate that participation in sports contributes to physical well-bein臣， social a西ljustment， and selιesteem for adolescents (Kane, 1988;Thirer & Wright,
Zarbatar哼" Hartmann, & R由J!dn， 1990). On the other hand,the results of other research shows that athletes are more likely to engage in deviant behaviors (Buhrmann Bratton, 1978; Coakley, 1981; Schneider & Eitzen, 1983). There may be differ influences of participation in di臼erent kinds of interscholastic sports on
well-being. Few empirical studies have been interested in the study of
between different kinds of school sports and adolescent behavio凹.If participation in different school sports are not controlled, the inconsistent results be reported in different empirical studies ofthe effects of participation in school on the development of adolescent behaviors. The purposes of this study have two The first is to find out whether there are significant relationships between
in school sports and adolescent 吐eviant behaviors. Second, the influences participation in different school sports on adolescent behaviors will be examinee tries to show whether different kinds of school sports have discrepant impac臼 development of adolescent well-being and deviant behaviors.
Positive Effects of Participation in Sports
There are two general theoretical perspectives that address the
part比ipation in activities as a link to adolescent behaviors/development. The Hirschi's (1969) social control theory in which he examined conventional organized activities as one type of social control.According to Hirschi,the social consists of four parts: attachment, commitment, involvement, and belief,
Based on these scholars' ar且umentsand research findings,there are functions of sports to positively influence the development of adolescent behaviors,such as school performance, interpersonal relat凹的h巾， personality,etc. Landers and Landers (1978) stated
structure,i止，coaches and teachers (Agnew& Pe記自en， 1989; Crain,1981)
Empirical researches also suggested that the function of athletics is to enhance the adolescent's psychological as well as physical development (i.e.,learning sports skills,positively influencing one's self-concept,and developing intimate relationships with peers). Sports exercise is good not only for the body but also for the ethical development of adolescents (Dozier, Lewis, Kersey, & Ch缸pin皂， 1978). Moreover,
Maton (1990) asserted that aspiration level,participation in meaningful instrumental
actlvltI間，i.e. sport,might affect both life satisfaction and perceptions of value in daily activity. Adolescents are able to satisfy their aspirations through the process of participation in sports activities as socially accepted channels
The results of Schafer's (1969) study from Midwestern high schools showed that young athletes were less likely to commit delinquent behaviors than young nonathletes Thefindin且s of Browne an西 Francis' (1993) empirical research indicated that students who frequently participated in sports activities were more likely to be a社ractlve，
popular,and socially skilled. Moreover,Koss and Gaines(1993) reported that drug use is lower among intercollegiate athletes than among nonathletes
For the school performance of athletes, Schafer (1969) tried to understand whether participation in high-school athletics influenced academic achievement. The samples of this study were collected from two Midwestern high schools. He found that athletes are more likely to earn higher grades than nonathletes in junior high school The results of Sabo and his colleagues' (1989) study supporte廿 Schafer's findings Students engaging in school sports were more likely to earn better grades,drop outIe品，
have higher aspirations, attend college more often, and attain more years of college education
Relationship between Participation in Interscholastic Sports 195 Yuk-Ying Tung
194 Bulletin of Educational Research Vol.48 No.4
a safe way to reduce excess force (Figler& Whitaker, 1995). According to this theory, 在 adolescents cannot find an adequate way to release their internal an凸 external pressure,they are more like to rely on violent reaction to face the st閃ss. Conversely, P缸ticipation in spo吐s activities may provide adolescents with an additional source satisfaction;therefo間， it can prevent adolescents from engaging in reactive aggression The evidence that sport provides a reliable avenue to discharge excess pressure because physical exercise can generate a sense of physical and psychological well-being (Benne缸， 1991) 以keFigler and Whitaker's (1995) statement:“sport provides a ready outlet for hostile aggressive tension of nations and is the best hope forhumani句" given thepre目前esof modemIi色，politics,war,and weaponry" (p.238).
According to the above theoretical viewpoints, pa吐icipation in spo吐s h由 m句or functions to prevent adolescents from engaging in delinquency. The function is the social integration. Adolescents who associate themselves with teams need to attach to their coach and teammates very well in order to have a performance. Young athletes always have 古eelingsof friendships and belonging a tearn through exercises and competitions in sports (Figler& Whitaker,1995). order to keep the status in aspo吐s te副凡 athletes need to follow the rules of sports face the supe鬥站的n of coaches. As a result干 young athletes are able to learn approved attitudes and values in sports activities. Like Hirscl吐's (1969) adolescents who closely attach to adults and peers 缸e less likely to delinquency. Moreover,adolescents participating in sports activities can improve social skills because they usually and closely interact with peersan吐 adults
The second is the socioemotional function. Sports activities can help release psychologicalten日on. Young athletes learn how to manage tension and through the process of practice and competition in sports. The process of exercises can give individuals a lot of enjoyment and rei缸ation. Participation activities can reduce adolescent deviant behaviors by allowing release of strengthening bon吐s between participants and members of the
sample was administered to 2,605 high school students throughout the province of Saskatchewan, Canada. The results showed that participation in sports is the major factor which causes male adolescents to be aggressive and abusive to female
祕olescents (Bernard, 2000). The adequate reason to explain this finding is that athletics-based subcultures significantly construct masculiniti自由1d femininities in
a益。 lescents in which male adolescents are more likely to aggress female adolescents
(Miedzi血， 1991). Moreover, according to various reports in the press from 1993 個
1994,between 70 and 100 athletes and coaches have been accused of assault against a woman each year (Lapchick,2000).
The second is the perspective of positive deviance. Hughes and Coakley 例如) mentioned IIpO日tivedeviance" as the over conformity to the norms and values of sports by athletes through transforming positive behaviors into negative behaviors, for example,the use of performance-enhancing drugs in sports. In other words,in order to win a game or have a goodperform叩間， athletes would use every stra扭皂y， even use illegal means. Following Hughes and Coakley's statement,Miracle and Rees (1994) 缸guedthat "the conformity of athletes to this culture helps to create very strong bonds that can form the b扭扭 ofnegative behaviors,especially among athletes who have low self-esteem and are vulnerable togroup 吐emands ，or among athletes who regard sports as their only opportunity for success in life"(p. 118). Athletes are more likely to use
社rugs than nonathletes. The reason is that st叫ent athletes deal with the maintenance of
athletic performance while responding to i吋 uries and st問ss by using illegal 的gs (Wechsler
&
Davenport,1997)Based on the above theoretical viewpoints,we may be able to understand why Segrave and Chu (1978)foun吐 ahigher rate of delinquency among young male athletes than among young male nonathletes. The results of Buhrmann and Bratton's (1978) empirical research also showed that female athletes are more likely to engage III delinquent behaviors,including smokin臣， drinking,breaking school rule,cheating on tests,and getting into trouble with the law,than female nonathletes
196 Bulletin of Educational Research Vol.48 No.4
some of the often claimed contributions of athletics have development of a competitive spirit,abilit于 tocooperate,sportsmanship, manner, courage, a greater capacity for delay of gratification, resistance to pain and fatig凹， and a release from tension and impulses (p. 299)
The light of these empirical researches indicated thatp訂ticipationin school have positive effects on the development of adolescent behaviors. In other students who participate in schoolspo吐sare more likely to be attractive,popular, socially killed,and so on
Negative
E証ects
of Sports
There are two major perspectives to argue that participation in sporth自 由sociation with the development of individual behaviors, for example violent behaviors and using drugs
The first is the perspective of transformed aggression. Butt (1987) argued sports behaviors possess a unique position in physical and psychological She define吐 aggressIon 缸 litheenergetic assault on animate or inanimate
pu中ose"(p.15). Sport activities cover different types of aggression (Bu扯，
aggression means that sports rewar往 persons who are high in biological Therefo間， athletes are more likely to be aggressive. A second type of sports is called socialized aggression in which athletes become more through training and practice. The experience ofp前ticipationin aggression in brings more aggression during the athletes' life. In other wor吐 s， aggression transformed from sports acti、lities to daily activities. According to aggression students who participate in school spo吐s are more likely to engage in physical Youths involved with sports learnto 吐efineviolence as a useful tool for
goals. Bernard (2000) used the data of Saskatchewan Youth Attitudes 1997 test the relationship between participation in spo口s and adolescent violence
Problem Statement
According to theoretical arguments and findings of empiricalstudi前，partICipatIon in interscholastic sports has influences on the development of adolescent behavIOrs. Participation in school sports is able to limit or 吐evelop adolescent delinquency. The explanation 品。 rthis puzzle may be that different kinds of interscholastic activity have positive or negative effects on the development of adolescent deviant behaviors Therefore, this study tries to answer the following two questions. First, do different kinds of interscholastic sports have varied relations with the occurrence of adolescent delinquency? In this study,different types of interscholastic sports will be classified in the analysis in order to understand what kinds of school sports have positive effects and what types of school sports have negative impacts on adolescent engag10g 10 由vtant
Relationship between ParticijJation in Interscholastic Sports 199
experiences that have positive implications for or時's psychological well-being and sense of environmental mastery"(p. 112). The positive or negative effects of participation in interscholastic sports on adol臼cent behaviors are dependent on those sports constructions,rules,qualities,and the like
Smit (1996) 由serted that athletes participatin皂 in di叮erent sports are related to different degrees of violence. Each interscholastic sport has the official rules of body contact perfonned within the competition. Smith (1996) also mentioned that “such contact is inherent in sports such as boxin皂， wrestlin臣， ice hockey, rugby, lacrosse,
football,and to lesser 吐egrees in soccer,basketb訓， water polo,team handball,and the like" (p.162). In other words,the practices of some sports are significantly related to violence,while other sports are not
Based on Zimmerman and Maton's statement and Smith's argument, different types of interscholastic sports have different competition rules, practice processes, cooperation levels,and aggression degrees. As a result,when researchers attempt to understand the relationships between athletes and deviant behaviors,di旺erenttypes of school sports should be controlled in their study.
Yuk-Ying Tung 198 Bulletin of Educational Research Vol.48 NO.4
The results of an empirical study showed that a number of athletes believe the use of drug supplements can increase testosterone, as performance enhancers
1998). Additional 紗.， Ray (2000) examined the attitudes of college athletes toward the use of painkilling drugs. A number of student athletes think that there is nothing with usingpainkille自由 anillegal drug for spo此sactivities. Bower and Martin(1999) developed a study by sampling 50 A企ican American basketball players,representing 11 southeastern colleges. These players ranged in age from 18 to 23 ye缸 s. The results displayed that of the 50 athletes in this study,72% reported having consumed
beverages,and 46% had engaged in binge drinking. Eight percent of selected athletes used alcohol and tobacco,and their use of alcohol and tobaccowoul廿 reduceduring the competitive season. In fact,there are a number of empirical studies found that athletes tend to use illegal drugs frequently (i.e., Anderson, Albrecht, McKea臣，
McGrew, 1991; Denham, 1997; Selby, Wechsler, Moeyke肘， Davenport,
Hansen,1995; Weinsten,Bird,1990)
Relationship between Different Kinds of
Interscholastic Activities and
Di在erent
Types of
Delinquency
The inconsistent results are fo由ld in di在erent studies for the impact participation in school sports on adolescent well-being. These studies do not
that 出e effects ofdi鉛hent types of school sports on the development of behaviors are varie吐 Therefore ， if our explanation of relationships between and deviant behaviors are based on one particular school sport or combination several school sports,the inconsistent results will be found
Zimmerman and Maton (1992) stated that only some kinds of activities deviant behaviors, and others may cause adolescent deviance. They argued meaningful instrumental activities may preclude adolescents engaging in behaviors. They define meaningful instrumental activities 品“task-or
Data
Research Method
Relationship between Participationil11nterscholastic Sports 20 I Yuk-Ying Tung
This study attempted to find out the complex relationship between participation in interscholastic sports an西 delinquency. Therefore,the analysis of this study included three types of 臼dolescent misbehave. The independent variables were mainly participation in different kinds of interscholastic sport. A set of control variables was also control in order to avoid the spurious 記lation.The following was to describe the measurement ofe血hvariable in detail.
Deoendent Variables Both theoretical viewpoints and empirical studies have identified that participation inin阻rscholastic sport had negative orposit凹 eimpacts on adolescent deviant behavior, such 由 misbehavior at school, drug use,and physical fight. Consequently, this study included three dependent variables: misbehavior at school,commitment to physicalfight 剖 school ，and using marijuana.
Students' misbehavior at school consisted ofsix items: (1) respondentw品 late f，前
school, (2) respon位 ent cut or skipped classes, (3) respondent got in trouble, (4)
respondent" 血 puton in-school suspension,(5) respondent was suspended or put on probation from school, (6) respondent was transferred to another school for disciplinary reasons. The students were asked“howma且ytimes did such things happen to you in the first half of the current school year?" Stud叩ts who had h也her scores were more like to engage inmisbehavior 副 schoo l.The reliability for these questions is 0.65. Commitment to physical fight at school was dummy recoded: O=never; l=one or more than one. To measure using drugs,studentswe阻 asked
“
on how many occasions have you used marijuana in your lifetime?" Drug use was also recoded: O=never; 1=one or more than one occasionIndeoendent\祖riables In this study, interscholastic activities were only considered in the measurement of school sports. Interscholastic sports meant one school team competed with other school teams. To measure schoolspo哎， students were Measurement
and analyzed. 4
Vol.4日 200 Bulletin of Educational
behaviors.Secon忌，does participation in one kind of interscholastic sports have effects on different types of adolescent delinquency?
The National Education Longitudinal Study of 1988 [NELS:88] h由 been
data for the study of the relation between school athletes andadolescent 廿eVlance，
as misbehavior at scho仗， drug use, and physical fights at school. The data developed and practiced by the National Opinion Research Center at
Chicago and sponsored by the U.S. Dep缸tment of Education. The sample two-stage stratified probability sampling design that was utilized to build nationally representative sample of schools and students. The first stage was to 815 public and 237 private schools from four areas in the Unites States,
Northeast, South, North Central, and West. The second stage developed a selection of 26
,4
35 students among the sampled schools. NELS:88 include吐school students' transition from 8th to 12nd grade as a nationallongitudinal study data series begins with the year 1988. The first follow-up was surveyed in 1990, the second follow-up was conducted in 1992. NELS:88 not only included about students,but also surveyed students' parents,teachers,and the school
(Horn & West, 1992). This study uses data from the first follow-up survey in which consisted of 20,706 students and 695 variables
These data include measures of many important individual backgrounds, and behaviors of American youth. In addition,the respondents were also asked they participated in different kinds of interscholastic activities, such as basketball, footba缸， soccer,and swimming team. Some stu社ents participated in than one interscholastic sport. Base吐 on these data, the effects of different interscholastic activities on the development of adolescent well-being can be
There were three dependent variables and a set ofin社ependentvariables in this Analysis
did get along with their parents. The response categories were: 1=not along with their parents,2=only along with either one,3=along with both. The measures of parents' helping and checking respondents' homework consisted of two questions: (I) how often p缸ents check respondents' homework,(2) how a立en parents help respon益en臼 with homework. The r自ponse categories were
",
coded: I=never, 2=rarely, 3=sometimes, 4=often. The reliability for these questions is 0.70. A set of items that W品 used to measurethe 社egreeof students' discussion with theirp缸entsconsisted of:(I)discussed school courses with parents,(2) discussed school activities with parents, (3) discussed things studied in cl品s with parents,(4) discusse吐 grades with paren包
The response categories were recoded: 1=never,2=sometimes,3=often.The 問liability
for these four questions was 0.77. Family socia-economic status (SES) was the combination of father's education level,mother's education level,father's occupation, mother's occupation,and family income. Family socia-economic status composite was standardized to a mean of 0 and a standard deviation of1. The higher the score the higher socia-economic status respondents' families have.
For school variables, the evaluation of school grades contained four subjects mathematics grades,English grades,history grades,and science grades. The reliability for these four grades was 0.82. The gra社esranged from I to 8. High scores indicated better school grades. School region was recode社品 a set of dummy variables (I=yes; O=no): Northeast,North Central,West,and the South as the reference category. School area was also dummy recoded (I=yes; O=no): urban areas,suburban areas,and rural
缸e晶晶 the reference category. Two items that measured time spent in homework consisted of: time spent on homework in school and time spent on homework out of school. The response included a set of categories: O=none, 1=1 hour or less,2=2-3 hours,3=4-6 hours,4=7-9 hours,5=10-12 hours,and 6=over 15 hours
Relationship between Participation in Interscholastic Sports 203 Yuk-Ying Tung
202 Bulletin of EducationalResearc狙 Vol.48 NO.4
asked
“
did you P盯ticipate in any interscholastic sport?" Moreover,these sports were recoded 品 a set of 社ummy variables (l=yes; O=no): baseball, b品ketball， football, soccer, swimmin且， pa叫 icipation in more than one spo吭， and nonparticipation in any interscholasticsport 品 thereference category. The impactsof 自fiend's a吐吐udes to附ardspo此 were also examined in this study. Respondents were asked whethertheir 世iends felt sports were important. The 間sponse choices consisted of: 1=not important,
2=some啊hatimportant,3=very important
Control Variables Phillips (1993) argued that when researchers study relationship between 剖hletic an吐 adolescent delinquency，如nily background personal characteristics，凹的品 social cl血s，包租ily s仕llct叮e， parental gender,race,etι ， should be co由trolle吐 Forexample,Miracle and Rees (1994) that high school students from lower socioeconomic background we閃 mo間
experience frustration in school. Sports activities may give those students The results of this study implied that students in di缸erent social classes had ef良心tsofsch悶。Isport participation on the development of adolescent delinquency
部 ad目nicachievement
For personal ch缸ac扭 ristics， gende雪 was dummy coded as 1=male or Race was recoded into three dummy variables: White, Hispanic,and other r，血e，
Black as thereferen臼 category. Individual self-esteem contained eight items:(I)
good about himlherself,(2) does nothav官 enoughcontrol ov前 li色，(3) feels he or is a person ofworth,(4) able to dothin侈品 wellas others,(5)s剖isfteshimlherself,
feels useless at times,(7) thinks he or she is no good at all,(8) does not have be proud of. The response choices for self-esteem w品 from strongly disagree strongly agree. Higher scores indicated respondents have higher selιesteem
reliability for these eight questions was 0.82.
A set of variables was used to control the imp血ts of family factors structure was dummy recoded 品 1=intact family or O=non-intact family. To the relationship between students with their parents,students were asked whe
physical fights at school,personal characteristics,family backgrounds,school factors, and participation in interscholastic spo付:s. The mean score of students engaging in misbehaviors at school was 0.58, with a standard deviantion 3.8. A higher score indicated greater commitment to school misbehaviors. With regard to drug use,of the respondents, twenty percent 間ported that they had used marijuana. Furthermore, seventeen percent had the experience of physical fights at school. The detailed distribution of students participating in副部erentkinds of interscholastic sportw品 also showed on Table 1
The model I in Table 2 indicated that students participating in baseball (B=-.094; p;;;三 .001)， basketball (B=-.099; p 三三 00I),and swimmin直 (B=-.079; P三三 05) were less likely to engage in school misbehaviors. However, adolescents who participated in football (B=.162; P 三至 .001) were more likely to commit school misbehaviors. Students whose friends felt school sports were important were less likely to engage in school misbehaviors (B=-.057; P三 001). Finally,participation in soccer and more than one sport was not significantly related to school misbehaviors
After including e宜的ts of controlled variables, the significant 悶1ationship between participation in basketball, swimmin皂， and friend's attitudes toward to sports and misbehaviors disappeared (See Model 2,Table 2). On the other hand,participation in baseball and football still had significant relation with misbehaviors. Interestingly,
the relationship between participation in more than one spo此s and misbehaviors became significant (B=.05 I; P 三乏。 1). The full model with twentY-fo!Jr independent variables accounted for only 21 percent ( R2=.212) of the variance in thedependent variable,engaging in school misbehaviors
204 Bulletin of Educational Research Vol.48 No.4
study. Multiple regression analysis w品 used to test the usefulness of predictors of students' misbehaviors at school. The results of regression analysis could describe different aspects of the relationships between students' misbehavior at school and participation in differentkin也 ofinterscholastic
Two of the th閃e dependent variables had very skewed distribution with many cases having a value of zero. There were lifetime using marijuana and engaging in physical fights at school.Lifetime marijuana use was recoded to zeroan位 onewith zero meaning no experience, and one meaning one or more instances of marijuana lise. Engaging in physical fights at school was also recoded to zero and one, with zero meaning no reported physicalfigh話，and one meaning one or more. Logistic
analysisw血 usedto test the usefulness of predictors of each of the dependent variables that were dichotomous variables. Logistic regression estimated parameters using maximum likelihood method.
Three reduced models,without effects of controlled variables,were first see participation in which school sports variables were useful for predicting each three dependent variables: stu丑聞脂， misbehaviors at school, marijuana use, physical fights at school.The fullmo位elswere then examined to find out whether
we自 stillsignificant effects of participation in sports on the three dependent while con甘olling e能cts of personal characteristics, family backgrounds, and factors. In addition,the light of full models could also indicate whether there discrepant impacts of part比ipation on different interscholastic sports. After restricted models to the data,the results could be used to make inferences validity of hypotheses. This analytical strategy was also to evaluate the between participation in interscholastic sports and adolescent delinquency, possible spurious relations.
Results and Discussion
Table 1 showed summ叮y statistics for school misbehavior,marijuana use,
Table 2. Ordinary Least Squares Estimation (OLS) for the Sports' Effect on Adolescent Misbehavior at School
坐坐且i 坐里坐i主
Variable B S.E. B S.E
(Adjusted) (Adjusted) 。 14** .012 。 14*** .046件* 。37 013 135 1.887 212 11198 -.094 。25*** -.068 .023" -.099 022*** -.005 .020 162 。26*** 。77 。21*** -.026 。31 。43 .028 -.079 。36' -.023 .032 。35 。 18 。51 .017料 -.057 008*** -.007 。07 。68 。 10材* 018 。 17 。09 。24 。86 。23*** -.027 OIl' -.094 010*** -.049 006*** -.122 。 12*** -.084 。 10*** 。28 007*** -.116 004*** -.020 004*** III 。 15*** 。73 。 13***
Relationship betweenP訂tlClP酷的nin Interscholastic Sports 207
659 .015'" 015
n 11198
*Significant 訕。 5L口e1，**Significant at .0.1 Level,***Significant at .001 Level
Note: In order to avoid overestimating the significance test whileweighting 血e s紅np峙， the original
stan吐 arderrors were multiplied by following formula: SQRT(N吐。/ SQRT (n-dt). N is the population estimate. n is the actuals也nplesize. dfis the degrees offreedom in the model Yuk-Ying Tung
坐坐~:
P甜 1口pationin School Sports (Nonparticipation in Any
Sports 品 Referen臼)
P缸ticipationin Baseball Tearn
P缸ticipation in Basketball Team
p叮叮C中 ationin Football Tearn
Participati 凹1in Soccer Team
Participatio叫 inSwimTeam
P甜1口p目的明 inMore Than One Sport
Frien血，Attitudes toward Sports PersonalCharacteris刮目: Gender(Du凹nmy I~Male)
Race(BI臨k 且 Reference)
White Other Races Hispanic
SelιEste副主1(Index)
包且i1y Fact盟:
Along with Parents
Parents Check or HelpStud扭扭， Homework
Discus訕。 nwith Parents Intact Family (DummyI~Yes) Socio-EωnomicStatus (Index)
坐坐泣E經些單:
School Grade
Time Spend onHome"" 叮k SchoolA自由 (Rural 目 Reference)
Urb阻
Suburban
School Region(Sou晶晶 Reference) No汀theast North Central West (Constant) R n 16191 648 971 971 18097 16661 3332 13329 18013 3062 14951 Table 1. Descriptive Statistics ofthe Variables in the Analysis
Variable Mean Percent
DeDendent Variables: Misbehavior at school(Index) Mariiuana use
Yes No
Enga四 ngin physical fight剖 school
Yes No
S且且E
Participation in School Sports
p 缸ticipationin Baseball Team Participation in Basketball Team
p叮ticipationin Football Team
P血ticipationin Soccer Team
Partido 酷的nin Swim Team
Participation in More Than One Sport
Nonpartici口前ioninAnv Sport (Reference) Friends' Attitudes toward Sports
Per盟里迢主垣且益.eristics: Gender Male Female (Reference) Race White Other Races Hisp扭扭 Black (reference) Self-Esteem(Index) 祖且i1y F醒鈕旦河 AlongwithP缸eots
Parents Check or HelpStude混阻，Homework Discussionw社hParents
Intact Family Yes
No
Socia-Economic Status(Index)
釷垣且 Fae鈕n:
School Grade
Time Spend on Homework SchoolAreas Urban Suburban Rural (Reference) SchoolRe第IOns Northeast North Central West 坐坐k悠些旦旦旦
997 981 954 扭過且主 S.E. Exp(B) (Adjusted) 166967.409 17 11174 。41 036 日 43 347牢牢* -.003 -,019 -.047 2,085 B -.004 。67 996 。30 。 67 1.030 -.195 。66 亭亭 823 -.060 067 942 447 。47*** 1.563 170 。51*** 1.186 117 。86 890 -.022 。90 979 -.329 118** 720 -.178 123 837 258 。42* 牢牢 1.295 193 。44*** 1.213 。39 。38 1.040 100 。42* 1.l05 606 。32*** 1.834 。 27 。50 974 -.007 。69 993 。 17 。66 1.017 -.300 。65*** 741 -.244 。53*** 783 -.010 一038 990 -.331 。70*** 718 -.059 。31 943 -.177 。42*** 838 -.279 。20 字本* 757 -.068 。24** 934 127 。47** 1.135 173 。41*** 1.188 Relationsbip between Participation in Interscholastic Sports 209
Table 3. Maximum Likelihood Estimates ofLogistic Models for the Sports' Effects on Adolescent Physical Fight at School
也必吐1 S,E, Exp(B) B 巴拉笠金 Yuk-YingTun車 Variables 包且II
Participation in Baseball Team Pa巾crp酷的nin Basketball Tearn
p叮ticlpat 的nin Football Team
p 叮叮 cipation in Soccer Team Participation in Swim Team
p 叮ticipation in More Than One Sport Friends' Attitudes toward Sports Pe 鬥on31 Characteri 甜的: Gender (Dummy I~Male) Race (Black asRefe白血 e)
White Other Races Hispanic Self-Estee叩 FamilyF單位s: Alongwi由 Parents
Parents Check of Help Students' Homework
Discussion with Parents Intact Family (DummyI~Yes) Socia-Economic Status
坐包且正堡包 2
School Grade
Time Spend on Homework SchoolAreas(Rural 血 Reference)
Urban Suburban
SchoolRegie 的 (South 品 Referen 自) Nor血e 品t North Central West Constant -1.781 .234 料* Model Chi-Square 25649.490 M 7 n I Il74
*Significant at .05 Level**Significant at .01 Leve!***Significant at .001 Level
Note: In order to的 Didoverestimating thesignificance 甘st while weighting the sample,the original standard errors were multiplied by following formula: SQRT(N-d 叮 I SQRT (n-dl). N is the populationestimat巳 nis the actual samplesize 西 fis the degrees offreedcm in the model This result implied that controlled variables had impacts on the relationship
between pa吋icipationin more than one sport an社 school misbehaviors. These results indicated that participation in different types of sports had different impacts of! engaging in schoolmisbehavio呵.Participation in football or more than one sports had positive relations with adolescent engaging in misbehaviors at schoo!. However,
there was a negative relationship betweenpa前icipationin baseball team and engaging in schoolmisbeha可VIors.
The results for the relationship between participation in sports and physical fights at school were showed in Table 3. A吐olescents who p缸ticipated in basketball
(B=弋 195;Exp(B)=.823) and swimming (B=-.326; Exp(B)=.720)we間 less engage in physical fights during than those who did not participate in interschol品ticsports. Converse紗" participation infootball 阻dmo閃 thanone a positive relationship with physical fights. Students who participated in (B=.447; Exp(B)=1.563) were about 56 percent more likely to engage In
fights than those whoparticipate吐 in no sports. Moreover,adolescents who joined than one sport(包=.582;Exp(B)=1.295)were about 30 percent more likely to physical fights at schoo!. No difference was found in the log odds of physical fights from adolescents participating in baseball and soccer.
After controlling a set ofpe自onalcharacteristics,family backgrounds, factors,the significant relationship between participation in football and more sports was not chan伊拉 On the other hand, participation in b由eball and became no significant 間lationswith physical fights at schoo!' One interesting W品 that friends' attitudes toward sports became significantly related to physical fights (B=.lOO; Exp(B)=1.l05). These findings in吐icatedthat the between participation inspo此sand physical fights
Table 4. Maximum Likelihood Estimates of Logistic Models for the Sports' Effects on Adolescent Using Marijuana
variables 坐坐蛙1 世盛益主
B S.E. Ex尹(B) B S.E. Exp(B)
(Adjusted) (Adjusted) .894 .944 1.140 112 .038** -.058 .033 131 .038*** 4.102 .317*** 179308.562 17 11029 。65 914 -.028 .067 973 。65*** 747 -.136 .067郁 873 051** 1.164 。64 .056 1.067 。74 1.012 177 。78 草 1.194 。90 901 。49 。94 1.050 。42*** 1.164 247 .046*** 1.280 。36*** 667 -.256 。38*** .774 。95 。28*** !.lOO 212 。48*** 1.236 214 .065*** 1.238 129 .064* 1.137 -.233 。59*** 792 -.224 .050*** 780 -.244 .034*** 784 -.223 。65*** 801 -.336 。27*** 。 715 -.051 。38 951 .326 .019*** 722 -.077 .022*** 926 也 182 .043*** 1.120 214 .038*** 1.239 -.404
Relationship between ParticipationinInterscholastic Sports 211
Parti口pation in School Sports
(Nonparticipation in Any Sports 自 Refer問問)
Participation in Baseball Team -.090 Participation in Basketball Team -.291 Participation in Football Team .152 Participation in Soccer Team .012 Participation in Swim Team -.104 Participation in M前e Than One .152 Sport
Fr記nds'Attitudes toward S恨orts Personal Characteristics: Gender (Dummy 1~Male) Race(BI前kasReferen 臼)
White Other Races Hispanic Self-Esteem FamilyF堡包益: Belong with Parents
Parents Check of Help Studen脂， Homework
Discussion with Parents Intact Family (Dummy I=Yes) Socio-EconomicS個tus School Factors: School Grade
Time Spend on Homework School Areas (Rural asRefer凹 ce)
Urban Suburban
School Regions (South as Reference) Northeast NorthC目前ral West Constant -.747 .205*** Model Chi-Square 31482.572 M 7 n 11029
*Significant at .05L口 d 抖 Significant at .0 IL的eI料 *Significantat .001 Level
Note: In order to avoid overestimating the significance 阻st while 叭 eighting the sample, the original st，"吋剖 errors we 問 multiplied by 自ollowing formula: SQRT 叭"的 I SQRT(恥的 N 的 the population estimate. n is the actual sample size. dfis the degrees offreedom in themode才 Yuk-Ying Tung
Based on these results, different types of sports have different effects on the occurrence of violent behaviors at school. An additional result concluded that a friend of the respondent who felt that the sports was important h祕 a significant, positive impact on the respondent engaging in physical fights at school 品or both th
a.
t participated in spo前s and those that did not. These results might implyinfluence played an important role in determining whether or not one who in sports would engage in physical fights at school. The findings of Table 3 displayed an interesting result. Participation in school sports could not engaging in physical fights at school.Neverthele品，joining some interscholastic might be related to violent behaviors.
Table 4 showed the relationship between participation in sports and marijuana. Without including the effects of control variables,the results
students who pa巾cipatedin basketball were less likely to use marijuana than who p的cipatedin no sports by about 25 percent (B=-.291; Exp(B)=.747,See Table 4). However,respondents who participated in football (B=.152; or more than onespo社s(B=.152; Exp(B)=1.164) were more likely to smoke than those who did not p缸ticipate in any interscholastic sports by about 16 Friends' attitudes towards sports had a negative relationship with using «B=、404;Exp(B)=.66). In the modell,however,the results showed that in baseball, soccer, and swimming had no significant relation with physical during school (See Table 4)
While controlling the effects of personal characteristi白，臼mily
school factors,the relationship betweenpa此icipationin basketball and more than 210 Bulletin of Educational Research Vol.48 No.4
understanding coach's attitudes toward winning a game and achieving a championship. For instance,Waser (2000) argued that when researchers study the situation of athletes using illegal drugs,they must consider coaches attitudes toward drug abuse. We also neede廿 to help young players to release stress from the practice and championship. If we could follow the above suggestions, the negative impact of p叮ticipation in interscholastic sports might be limited and the positive influence might be improved We hope that participation in sports became a positive process of socialization
In addition, inconsistent results were found when examining friends' attitudes towards sports and the three dependent variables. Friends' attitudes towards sports were 品Dun吐 to have a significant, negative relationship with marijuana use at school However,a positive relationship was found when examining physical fights at school Thus,peer's influence positively affected the
“
macho behavior syndrome" in physical fights at schooL In other words,friends might expect that those who participated inspo位s would more likely engage in obvious physical fights before,during, an吐 after
participation in these activities. However,they did not expect that their friends would engage in misbehaviors and marijuana use because of their participation in sports These actions could possibly ejecttheir 飢endsfrom participating inspo舟.
No direct and complete theoretical viewpoints coul吐 be used to explain why participation indi仔erentkinds of sports had various relationships with deviant behavior. According to the findings of this study, researcher tried to develop a perspective of
spo吐s construction to interpret why participation in some kinds of sports might be related to engaging in deviant behavior, while other kinds might not be. The perspective consisted of three major dimensions. The first was the construction of spo吐sviolence. Smith(1986) differentiated four levels of sports violence,and football,
hockey,and boxing were defined on higher level ofspo此saggression. Individuals who participated in spo吐s with higher level of aggression were more likely to engage in violent behavior. The results of this study supported this argument. For instance,
participation in football had relationship with involving in violence; however,
212 Bulletin of Educational Research VoL 48 No.4
sports and using m缸ijuana was still significant. Friends' attitudes toward sports were still a useful predictor of adolescent using marijuana. The model 2 in Table 4 also showed an interesting finding in which the relationship between participation in football and using marijuana disappeared. On the other hand, participation in soccer became significantly related to the use of marijuana. Adolescents who participated in soccer were more likely to use marijuana than those who did not participate in any interscholastic sports by 20 percent (B=.I77; Exp(B)=1.194). As a result,this implied that when researchers try to explain the relationship between participation in sports and using marijuana,they need to consider the effects of personal characteristics, backgrounds,and school factors. Based on the results ofthe full model,participation baseball,football,and swimming failed to pre副 ct adolescent using marijuana. results again showe吐血atthe e臼ect of different sports on respondents either or increased thelikelihoo吐 ofmarijuana usage
.
Thus,some sports may possibly a greater opportunity for the use of marijuana. Conversel弘 participationin some might limit the occurrence of using marijuana. Additionally,friends of the who feel that spo吐S 缸e important had a significant,negative effect on for the respondents.The results of this stu位y found out that participation in different interscholastic sports has various relations with the involvementof 吐evtant
including school misbehaviors, physical fights, and using marijuana participation in one school sport had varied influences on different kinds of behavior. These findings supported the previous argument that different interscholastic sports had varie廿 constructions ， rules, qualiti凹， and so on; they ha廿 different directions and levels of influence on the development of behavior (Zimmerman & Maton, 1992). Although this research found some sports had positive relationship with adolescent delinquency, it did not mean needed to limit or prohibit our kids to participate in those spo此s. In fact,we pay attention to the characteristics of rules and pract時間 of different
and delinquency could not be drawn. The suggestion w品 that future studies shoul社
include effects of personal characteristics to find out what kind of individual background varied the link between participation in sports and involving delinquency
Finally,the data of this study were sampled from America,and the development of school sports between America and Taiwan was not similar. As a result,we could not directly explain the relationship between pa的cipation in sports and Taiwanese delinquency based on the data from other countries, because different societies had various cultural backgrounds. In spite of that, Taiwanese researchers and educators could realize some paths from the above findings to study how young athletics behavior was related to delinquency in Taiwan.Fir泣，few empirical studies focuse吐 on the impact of participation in sports on the development of delinquency in Taiwan. Consequently,educators and researches should develop empirical studies based on data 仕om Taiwanese situation to draw up a more adequate and complete depiction to describe the influence of involvement in sports on the development of adolescent delinquency. Second,when we studied the relationship between participation in sports and delinquency,different types of sports should be categorize 吐 bythe level ofru峙， violence, and practice. This process might generate a more detailed illustration to indicate what kinds of sport were related to adolescent delinquency,while other kinds might not be. Third, Taiwanese 阻d American society had different cultural backgrounds. So, when Taiwanese researchers developed an empirical study to interp記t the association between participation in sports and adolescent delinquen句， they should add some variables into the analysis ofthe model.For instance,the impact of family life on the development of Taiwanese in副vidual behaviors was greater than that of American individual behaviors (Thornton and Lin, 1994). According妙" the factor of [;剖nily relationships could be considered 品 a significant role in the explanation of the relationship between participation in sports and delinquency. Finally, for the application, Taiwanese sports coaches and teachers should figure out the characteristics of sports rule and practice in order to suppress the negative impact and 214 Bulletin of Educational Research Vol.48 No.4
participation in swimming and baseball was not related to aggressive behavior. The reason might be that these sports were defined as a“noncontact sport" and the rules were intended to control player violence (Figler and Whitaker, 1995). The second direction is the construction of spa此s rules. Symbolic interactionists believed that adolescents could learn how to take a role from joiningspa此s or games (Mead,1977). In other words,youths were able to practice how to adequately playa role in a group
m廿 cooperate with others through participation in sports. Conversely, the rules of different sports ha西 various effects on the development of adolescent behavior. Baseball players,forexamp峙， had to understand very well not only of their own role requirements but also of the roles and likely responses of every other teammate. If baseball players 吐id not follow the rules,they could not take the role of other players, thus they could not ant比ipate how others would respond to their actions (Roberts凹，
1987). In this study, the measurement of school misbehavior included a set of behaviors in breaking school rules. This might explain why this study found that participation in baseball team was able to prevent the involvement of misbehaviors at school. On the other hand, the rule of football game not only directly encouraged player to practice aggressive behavior,but also indirectly implied the spirit of personal hero to break the rule (Figler and Whitaker, 1995). Consequently,the results of this study showed that there was a significant relationship between engaging in misbehaviors and participation in football. The third construction was the personal background of athletes. The findings indicated that participation in soccer became significantly related to 廿ruguse, while controlling personal background. This implied that there might be interaction e仔ects between participation in soccer and personal background in the explanation of the development of adolescent deviance. As a result, when researchers focused on the relationship between participation in different
sports and engaging in deviance,they also needed to consider the personal
of athletes. Unfortunately,based on the data analyses ofthe study,a detailed picture for the impact of personal characteristics on the association between participation in
216 Bulletin of Educational Research Vol.48 No.4 Yuk-Ying Tung Relationship between Participation in Interscholastic Sports 217
According to the findings of the study, we were able to conclude three major points. First, there were significant relationships between participation in interscholastic sports an吐 adolescent deviant behavior. Secondly, participation in different school sports had various relationships with the development ofa益。lescent devi由1t behavior. Thi的妙， participation in the same interscholastic sports might have
di臼erent 由 sociations with different adolescent deviant behaviors. The results might also be used to understand why there were inconsistent findings in the empirical studies of impacts of participation in school sports. Like Zimmerman and Maton's(1992) and Smith's (1992) argument,different types of school spo抗s and activities had di的rent
competition rules, quality levels, practice processes, cooperation levels, aggression degrees, and so on. As a result, these school sports and activ詢問 had different directions and degrees of influence on the development of adolescent behaviors.
Finally, there were several suggestions for future researchers studying the relationship between part比ipation in interscholastic sports and adolescent behaviors First, researchers who wanted to test the relationship between school sports and adolescent delinquency should classify the effects of participation in different school sports. Only some kinds of interscholastic sports might be helpful for preventing adolescent 街叫胡間， while other kinds might not be. Secondly, the cross-sectional survey was not adequate to examine the causal relationship between a dependent variable and independent variables. Although this study found that pa吐icipation in different interscholastic sports had different associations with the development of adolescent deviant behaviors,the reasons could not be explained by the analyses of this study. Miracle and Rees (1994) state廿 that the results of cross-sectional survey could not be used tointe中retthe causal relationship between schoolathlet比如巾cipationand adolescent delinquency. Panel study data could be useful for the causal relation between participation in sports and adolescent behaviors. Thirdly, coach's attitu吐es
toward deviant behaviors as a useful means to win a game need to be considered in the future studies for understanding the relationship between participation in sports and
+ +* +*** +司，*， at School
-*
+ +* + +*** -牢牢* + -*' +字* +***Table5.Summary for Relationships betweenDi能目前 typesof Sport and Deviant Behavi肘s，Controlling for PersonalCh由這cteristics，Family Factors,and School Factors
Misbehavior Physical Fight at School Using 恥1arijuana
立坐坐位旦旦
participation in baseball team pa前 icipation in basketball team participation in football team participatIOn10soccer team
participatiOnInsWlmmmg team participation in moreth叩 onesport
Conclusion
improve the positive effect of participation in sports on adolescent behaviors.
In summary,the results of the study supported the hypothesis that participation in different kinds of interscholastic sports had various relationships with adolescent deviant behavior. For instance,participation in basketball had a negative relation with marijuana use, but not with physical fi皂hts at school and school misbehaviors (See Table 5). On the other hand,those who participate in more than one sport are more likely to engage in these behaviors. Again,the effects of participation in different sports allowed for more or less chances to participate in deviant behaviors inside and outside the school boundaries. This could possibly be due to 吐時 different activities and processes (i.e., rules, construction, practice, etc.) associated with each sport. Thus,
these activities and processes might have different effects on athletes who engage in misbehaviors, physical figl郎， an吐 marijuanause. Secondly,these results confirm 出e
hypothesis that participation in the same sport had various associations with different types of deviant behavior committed (See Table 5). For example, participation in soccer team had a positive relationship with using marijuana, but no significant relationship with engaging in misbehaviors and physical fights at school
*Signific叩tat.05Level 料 Significant at .01 Level 抑郁 Slgm t1cantat .00 I Level “-"=negative relationship
Horn, L., & West, J. (1992). Notional education long的Idinal study of1988: A pr«丹Ie of
porents ofeighth graders.Washington,D.
c.:
U.S. Department of E由cationHughes,R.,& Coakley,1.(1991). Positive deviance among athletes: The implications of over
∞nfonni句 tothe sport ethic. Sociology of SportJourn叫 8，307-325
Johnson,K. (1998). As drugs in sports proliferate,so do ethical questions. New York Times,147, pel
Kane,M.1.(1988). The female athleticrole 描 astatus determinant within the social systems of high school adolescents. Adolescence,23,253-264
Koss,M. P.,& Gaines,J. A.(1993). The prediction ofsexual aggression by alcohol use,athletic
p甜icipationand fraternity. Journal of Interpersonal Violence,8,94-108
Landers,D. M.,& L阻 ders，D. M. (1978).Ociali血仇onvia interscholastic athletics: Its effects on
delin中ency.Sociology of Education,51,299-303
Lapchick,R. E. (2000). Crime and athletes: Newrae凶 stereotypes.Society,37,14-21 Mead,G.H.(1977). Social psychology. Chicago: University of Chicago Press
Miedzian,M. (1991). Boys will be boys: Breaking the link between masculin你 and violer.間
New York: Doubleday
Miracle,A. W. Jr.,& Rees,C.R. (1994). Lessons of the locker room: The myth ofschool sports New York:Prome也eusBooks.
Phillips,1.C.(1993). Oncologyof 司port.Boston: Allyn and Bacon
Ray,T. (2000). Painkilling drugs in collegiate athletics:Knowled侈，attitudes,and use of student athletes. Journal ofDn喀 Education，30,313-324.
R咄咄son，L. (1987). Sociology. New York: Worth Publishers
Sabo,D.,Melnick,M. M. & Vanfossen,B. (1989). Minorities in sport. New York: Women Sports Foundation
Schafer,W. E. (1969). Some social sources and cons問 uences of interscholastic athletics: The case ofparticipation and delinquency. International Review of Sport Sociology,4,63-79 Schneider,J.,& Eitzen,D. S. (1983). The structure of sport and participant violence. ARE!v也
Review， 咒 1-16
Segrave,J. 0., & Chu, D. (1978). Athletics andjuv叩 ile delin中間cy. Review ofSpo叫 and
Leisure,3, 1-24
Selby,R.,Weinsten,H. M.,& Bird,T. S. (1990). The health of university athletes: Attitudes,
behavior,and stressors. Journal ofAmerican CollegeHealth， 玖 11-18
218 Bulletin of Educational Research Vol.48 No.4
deviance. Finally,although this study consists of several kinds of interscholasticspo巾， some popular sports, i.e. tennis, track-and-field, gymnastics, an位 volleyball， are not included in this research. Future studies should consider containing more kinds of sports in the examination for the relationship between participation in sports and delinquency
References
Anderson,W. A.,Albrecht,R. R.,McKea臣，D. B.,Hough,D.O.,& McGrew,C. A. (1991). A national survey of alcohol and drug use by college athletes. The Physician and Sports
Medicine， 凹， 91-104
Bennett,J. C. (1991). The irrationality ofthe catharsis: Theory ofaggression asjustification for
educat。凹，support of interscholastic football.Perceptual and MotorSkills， 刀， 415-418
Bernard,S. (2000). Boys against girls. Violence Against Women,6(9),960-987.
Browne,B. A.,& Francis,S. K. (1993). Participants in school-sponsored and independent sports Perceptions ofself and family. Adolescence,28,383-391.
Buhrmann,H. G.,& Bratton,R. D. (197的 Athletic pa的cip甜on and deviant behavior of high school girls in Alberta. Reviewof 學artandLeisu月， 2， 281-289
Butt, D. S. (1987). The psychology of sport: The behav悶 motivation ， personality and
performance of athletes. New York: Van Nostrand Reinhold
Coakley, J. J. (1981). The sociological perspective: Alternate causation of violence in sport
ARENAR，的Jiew，5,44-56
Colem間， J. S. (1961). The annals of the American athletics in high schooL Academy
Political and Social Science,338,33-34
Denham,B.E. (1997). Sports illustrated,the“war on drugs,"and the anabolic steroid control
ofl99 日 Journalof Sport
&
Socia/Issues， 刃，260-274Dozier,J. E.,Lewis,S.,Kersey,A. 丘，& Charpi時，J. W.(1978). Sports group: An treatment modality for emotionally disturbed adolescents. Adolescence,51,483-48日
Figler, S. K., & Whitaker, G. (1995). Sport and ploy in American life. Chicago: Brown Benchmark
Hirsc划，T. (1969). Causes ofdelinquency. California: University of California,Berkeley.
220 Bulletin of Educational Research Vol.48 NO.4
Smi曲， M. D. (1986). Sports violence: a definition. In D. S. Lapchick (Eds),Fractured/acus
Sport as are..月'ectiona/society,221-227. MA: D.C.Heath.
Sm拙， M. D. (1996). Typology ofsports violence.ln D. S. Eitzen (Eds),Sport in contemporary
society: An anthology,161-172. New York: St.Martin Press
Sto汀，A. (1968).Human Aggression.New York:Atheneum
Thirer,J., & Wrigl哎， S. D.(1985). Sport and social status for adolescent males and females
Sociology
0/
Sport Journal,2, 164-171.Tho肘rn岫1
Universit可yof Cαhicago Press.
Was缸，D. (2000). Alcohol and recreational drug education program forathletes 句artsCoach,
23,24-26
Wechsler，日， & Davenport,A. E. (1997). Binge drinking,tabacco, and illicit drug use and involvement in college athletics: A survey of students at 140 American colleges.Journal
of American CollegeHealth， 衍， 195-211
Wechsler, H., Moeykens，丘， Davenport, A叮 Castillo， S., & Hansen,J. (1995). The adverse impact of heavy episodic drinkers on other collegestud凹ts可 Journal
0/
Student Alcohol,5丘 628-634
Zarbatany, L., Hartmann, D. P., & Rankin, D. B. (1990). The psychological functions of preadolescent peer activities.Child Development,61,1067-1080
Zimmerman, M. A. & Maton, K. I. (1992). Life-style and substance use among male african-american urban adolescents: A cluster analytic approach. American Journal of
Community Psychology,20,121-138 《教育研究集刊》徵稿辦法 221
《教育研究集時》徵稿辦法
90.08.06 編輯委員會會議通過 9 1.08 日 6 編輯委員會會議修正通過 9 1.11. 08 編輯委員會會議修正通過 一、本刊以論述教育理論、教育實際問題及教育研究成果為主，並已列入國科會 人文處「臺灣社會科學引文索iJ lJ 資料庫 (Taiwan Social Science Citation Index,TSSCI)正式名單中，歡迎踴躍賜稿。 二、本刊仝年收稿，隨到隨審，來稿將於收件後五個丹內回覆審查結果。 三、本刊一年出版四期，於每年三、六、九、十二月出刊，各期分別以教育學不 向領域為徵稿範圈。 四、各期徵稿範園、截稿日期、出 f'J 日期表列如下. 期 耳目 第期 第一期 第二期 第四期 徵稿範區 課程與教學 教育行政與政策 教育哲史 教育社會學 截稿日期 前年 10 月 31 日 1 月 31 日 4 月 30 日 7 月 31 日 出于自日期 3 月 31 日 6 月 30 日 9 月 30 日 12 月 31 日 五、撰稿原則 I.來稿請用電腦橫打，文長以一萬五千字為原則，至多請勿超過兩萬字(合中 英文摘要、吉主釋、參考文獻、附錄、麗表等)。 2 來稿正文與中英文摘要請自行印出一式三份，連同「投稿者基本資料表 J ' 寄交編輯委員會。於通知接受刊登，益修改完畢後，再繳交磁片(請用 word 文字檔儲存)。 3 投稿請附中、英文摘要，中文摘要請勿超過 350 字，英文摘要勿超過 200 字，並請列出中、英文關鍵宇各三個。 4 來稿之編排順序為中文摘要、英文摘要、正文(<.王解請採當頁在方式)、附錄、 參考書目。若有致謝詞，請於通知稿件接受刊登後再加上，並置於正文後，
