UNRAVELING THE CONTEXTUAL EFFECTS ON STUDENT SUSPENSION AND JUVENILE ARREST: THE INDEPENDENT AND INTERDEPENDENT INFLUENCES OF SCHOOL,NEIGHBORHOOD, AND FAMILY SOCIAL CONTROLS*

Scholars of human development argue that a variety of social contexts affect youth development and that the interdependency of these contexts bears on the shape of human lives. However, few studies of contextual effects have attempted to model the effects of school, neighborhood, and family context at the same time, or to explore the relative and interdependent impact of these contexts on youth outcomes. This study provides an examination of the independent and interdependent influences of school, neighborhood, and familial contexts through an analysis of student suspension and juvenile arrest. Findings reveal that school‐based and family‐based informal social controls additively combine to reduce the likelihood of suspension and arrest. Moreover, for suspension, results support the hypothesis that an interdependent compensatory relation is present between the extent of collective efficacy in schools and in the surrounding neighborhood; school collective efficacy has a controlling influence on the likelihood of suspension that becomes even stronger in the absence of neighborhood collective efficacy. However, for arrest, an accentuating effect of school‐based social controls exists rather than a compensatory effect. A lack of neighborhood collective efficacy and a lack of school‐based social controls combine to exert a substantial increase in the likelihood of arrest.     

COMPARING DELINQUENCY CAREERS IN COURT RECORDS AND SELF‐REPORTS*

Most knowledge about delinquency careers is derived from official records. The main aim of this paper is to compare conclusions about delinquency careers derived from court referrals with conclusions derived from self‐reports. Data are analyzed from the Seattle Social Development Project, which is a prospective longitudinal survey of 808 youths. Annual court and self‐report data were available from age 11 to age 17 for eight offenses. The prevalence of offending increased with age, in both court referrals and self‐reports. There was a sharp increase in the prevalence of court referrals between ages 12 and 13, probably because of the reluctance of the juvenile justice system to deal with very young offenders. The individual offending frequency increased with age in self‐reports, but it stayed constant in court referrals, probably because of limitations on the annual number of referrals per offender. There was significant continuity in offending in both court referrals and self‐reports, but continuity was greater in court referrals. The concentration of offending (and the importance of chronic offenders) was greater in self‐reports. An early age of onset predicted a large number of offenses in both self‐reports and court referrals. However, an early onset predicted a high rate of offending in court referrals but not in self‐reports, possibly because very young offenders who were referred to court were an extreme group. About 37% of offenders and 3% of offenses led to a court referral. The more frequent offenders were less likely to be referred to court after each offense, but most of them were referred to court sooner or later. There was a sharp increase between ages 12 and 13 in the probability of an offender and an offense leading to a court referral. It is concluded that criminal career research based on self‐reports sometimes yields different conclusions compared with research based on official records.     

INTIMACY AND HOMICIDE: COMPENSATING FOR MISSING DATA IN THE SHR*

The Supplementary Homicide Report (SHR) of the FBI's Uniform Crime Reporting Program remains the most accessible and widely used database on lethal violence in the United States. However, researchers using this database must address the problem of missing data, which typically is the result of the failure to file, inconsistent filing of reports to the FBI by local police agencies, or incomplete records about the characteristics of specific incidents of homicide (particularly, missing information about perpetrators), even when reports are filed. Williams and Flewelling (1987) proposed methods of compensating for missing information, and this paper revisits their assessment by again determining the extent of the problem and the consequences of adjusting for it. Alternative methods are proposed and analyzed, with a focus on relationship‐specific rates (i.e., rates of family, intimate nonfamily, acquaintance, and stranger homicide). The implications of the results for further use of the SHR are discussed.