Race effects in juvenile justice decision-making: findings of a statewide analysis.
Overrepresentation of minorities in the juvenile justice system is well-established. On a national level, minority youths are arrested in numbers greatly disproportionate to their numbers in the general population. While black youths comprise approximately 15% of the ten to seventeen year old population at risk for delinquency,(1) recent figures indicate that they constitute approximately 28% of youths arrested.(2) Further, according to the Office of Juvenile Justice and Delinquency Prevention's (OJJDP) "Children in Custody" census, minority overrepresentation increases dramatically as one moves beyond arrest to later stages in processing. For example, minorities constitute approximately 62% of youths held in short-term detention facilities, and approximately 60% of those committed to "deep end" long-term institutional programs.(3)
Quite apart from issues related either to the extent or causes of differential minority involvement in crime, a number of researchers have expressed concern about whether the juvenile justice system operates with a selection bias that differentially disadvantages minority youths. The research reported here is intended to add to the growing body of literature addressing that question.(4) However, this research differs in significant respects from past research because it focuses on differences between the processing of delinquency and status offense (dependency) cases, rather than simply the juvenile justice system in general. Additionally, we supplement our statistical analyses with qualitative data to aid in understanding sources of racial disparity.
Our discussion is divided into two parts. In Part I, we report the findings of quantitative analyses conducted using official records of cases processed through the juvenile justice system in Florida. In Part II, we supplement and provide a basis for a more detailed interpretation of the quantitative findings, drawing upon in-depth interviews with system insiders--juvenile judges, state's attorneys, public defenders, and social service personnel. Based on those interviews, we explore the social and organizational processes underlying the findings reported in Part I.
II. PART I
A CONSIDERATIONS GUIDING THE QUANTITATIVE ANALYSES
Because the juvenile justice system consists of multiple decision points, it is essential that researchers track cases from arrest to final disposition through as many stages as possible. This is desirable for at least two reasons. First, decisions made at different points reflect the actions of different decision-makers--such as social service workers at intake, prosecuting attorneys at case filing, judges at court disposition--whose professional philosophies, organizational subcultures, and discretionary authority differ in ways that may render either intentional discrimination or institutional discrimination(5) more or less likely to occur.(6) The identification of more and less problematic decision points may facilitate both an understanding of sources of racial disparity as well as the development of strategies to reduce it.
Second, if a researcher examines only a single decision point, such as judicial disposition, the researcher's analyses may underestimate or altogether miss the effect of race. If disparities occur at early decision points that are not examined, analyses of late-stage outcomes are likely to produce findings of no discrimination.(7)
Another consideration guiding the quantitative portions of the research is the importance of estimating multivariate models that include controls for legally-relevant factors that might explain or justify race differentials in processing outcomes. At a minimum, we wanted to include as precise a measure of offense severity as the data would permit, as well as a measure of offense history that would take into account both the frequency and severity of individuals' prior records.(8)
A final consideration guiding the quantitative analyses is the possibility that the effect of race might be conditioned by other variables. Frequently, those who have explored racial disparities injustice system processing have restricted their estimates of additive or main effects models, which can obscure substantial race differences in treatment. Suppose, for example, that nonwhites and whites charged with serious offenses receive similar dispositions, while nonwhites charged with minor offenses receive harsher dispositions than whites.(9) In this instance, an additive model might show little or no racial impact, while an interactive model would reveal a significant race effect contingent upon offense severity.(10)
B. THE DATA SET
Data for the quantitative portions of this study were obtained from the Client Information System maintained by Florida's Department of Health and Rehabilitative Services ("DHRS"). The data set includes the total population of youths referred for juvenile intake processing throughout the state between January 1, 1985 and December 31, 1987. Because Florida law requires that all juvenile complaint reports be processed through the intake division, the data set is quite comprehensive and includes records of all police contacts other than those resulting in informal field adjustments, as well as referrals from parents, school officials, and other non-police sources.'! The case records were organized to permit tracking of decisions made at multiple stages in processing, from initial intake through judicial disposition.
Because the Client Information System tracks referrals rather than individuals, we reorganized the data set around individuals so that multiple referrals of a youth to the juvenile justice system over the three year period could be chronicled and examined. We accomplished this by restricting our analyses to the last delinquency referral in 1987 for each individual, a procedure that allowed us to capture at least two full years of offense and processing history information for each youth. The total number of individuals at the point of initial intake is 161,369. This includes 137,028 youths referred for delinquent acts and 24,341 youths referred for status offenses. Because status offenses are treated as dependency cases in Florida, and because the actors/agencies and processing decisions involved in dependency cases differ from those involved in delinquency referrals, we analyze these cases separately.
In the analyses that follow, the juvenile justice system is viewed as a series of decision points, each of which is simplified to represent a dichotomous contrast. Four stages are involved in delinquency case processing, two in status offense processing.
1. Delinquency Case Processing
1. INTAKE SCREENING: DHRS officials review all referrals originating from police arrests or from complaints by non-police sources. In addition to reviewing the facts presented in each referral, they are expected to interview the juvenile and his/her parents or guardians. They then make nonbinding recommendations to the state's attorney regarding the preferred method of handling each case. Intake officers may recommend that a case be closed without action, that it be diverted from the juvenile justice system for informal handling, or that it be referred to court for formal processing. We classify intake outcomes to differentiate between those cases closed without action or handled informally (coded 0) and those cases recommended for formal processing (coded 1).
2. DETENTION STATUS: Decisions regarding detention status are made shortly after delinquency referrals are received. Detention decisions are made jointly by intake staff, law enforcement officials (when referrals are police-initiated), and state's attorneys. Juveniles held in detention for any period between initial referral and the ultimate disposition of their cases are coded 1; those released prior to disposition are coded 0.
3. PROSECUTORIAL REFERRAL: State's attorneys decide whether a delinquency case proceeds to court. We coded prosecutorial referral as 1 in cases in which a decision was made to file a formal petition of delinquency or to seek transfer to adult court. Cases in which a decision was made not to seek formal action (e.g., no petition was ever filed or, if filed, a petition was subsequently withdrawn) are coded 0.
4. JUDICIAL DISPOSITION: The final stage in the processing of delinquency cases modelled in these analyses is judicial disposition of cases. Although the court has a wide range of options, our analyses compare youths who were returned to the community (e.g., those ordered to do community work service, placed on informal probation, placed on formal probation) (coded 0), with those who were committed to residential facilities (e.g., youth camps, training schools) or transferred to adult criminal court (coded 1).(12)
2. Status Offense Processing
1. INTAKE REFERRAL: Status offenders also enter the juvenile justice system at initial intake. However, because status offenders are legally defined as dependents, their processing differs somewhat from youths charged with offenses that would be crimes if committed by adults. At the time these data were collected, specialized intake caseworkers were responsible for processing dependency actions. Because these are the same officials responsible for handling cases of child abuse and neglect, their orientations may differ from those of delinquency intake personnel. For example, dependency caseworkers may be more oriented toward responding to a youth's family situation, rather than to the actions of the adolescent. Consequently, they may be more likely than delinquency intake officers to choose formal intervention in cases where the family is seen as troubled or dysfunctional. Unlike delinquency intake decisions, dependency intake decisions are unaffected by the anticipated reactions of prosecutors; in status offense cases, caseworkers have sole decision-making authority to file a formal petition. The first stage in status offender processing, then, is intake referral and is coded to distinguish between cases closed without action or handled informally (coded 0) and those petitioned to juvenile court (coded 1).
2. JUDICIAL DISPOSITION: Status offenders referred to court face some of the same dispositional options available to delinquent offenders, although they are not eligible to receive the most severe of the dispositions applied to delinquents (e.g., they may not be placed in secure detention, committed to training school or youths camps, or transferred to criminal court). At the most severe end of the continuum, status offenders may be placed in non-secure residential facilities, such as runaway shelters, foster homes, or group homes. Judicial disposition is coded to distinguish between youths who were ordered into some sort of residential placement (coded 1) and those referred for counseling or some other community-based treatment (coded 0).
3. Independent Variables
SOCIODEMOGRAPHIC CHARACTERISTICS: The RACE categories in the Client Information System include "white," "black," "American Indian," "Asian or Pacific Islander," and "unknown."(13) Because the number of persons classified as "American Indian," "Asian or Pacific Islander," and "unknown" was very small (less than 1% of the cases), we restrict the analysis to blacks, whom we hereafter refer to as non-whites(14) (coded 0), and whites (coded 1). Other sociodemographic characteristics included in the analysis are GENDER (coded male = 1; female = 0) and AGE (coded in one year intervals from 7-18).
For the analysis of delinquency cases, we used the most serious offense cited in the arrest or complaint to characterize the CURRENT OFFENSE. We coded this variable using the following scoring scheme: felony offense against person = 6; felony property offense = 5; felony offense against public order = 4; misdemeanor offense against person = 3; misdemeanor property offense = 2; misdemeanor offense against public order = 1.(15)
We also included CONTEMPT STATUS as a variable to distinguish cases referred for contempt (coded 1) from all other delinquency cases (coded 0). At the time these data were collected, state law permitted judges to place juveniles found in contempt into secure detention facilities for up to five months and twenty-nine days, not in the pre-adjudicatory phase, but as a disposition of their cases. This authority was frequently used as a vehicle to place status offenders otherwise ineligible for secure placement into detention facilities. That is, judges employed their contempt powers to incarcerate status offenders who disobeyed court orders by refusing to attend school or running away from home. These violations of court directives were treated as delinquent acts even though the behavior in which the youth engaged constituted a repeat status offense.(16)
We operationalized PRIOR RECORD by measuring the severity of prior referrals to the juvenile justice system. This measurement allowed us to account for both the frequency and severity of prior offending, and we constructed it by adding the severity scores of all offenses in each prior referral (using the same severity values as described above for CURRENT OFFENSE),(17) then dividing by the number of prior referrals.
Where appropriate, we also included case processing outcomes as independent variables in the analyses. That is, we explored the effects of decisions made at earlier stages in processing on subsequent stage outcomes (e.g., the effect of being held in secure detention on judicial disposition). This procedure allowed us to identify and assess possible indirect effects of race on case outcomes.
C. ANALYSIS AND FINDINGS
Because we defined each of the processing outcomes in terms of a dichotomous contrast, we used logistic regression as the method of estimation. In addition to estimating the main effects of each predictor in additive models, we also estimated models for each processing outcome that included all two-way interactions involving race. These interaction models allowed us to determine whether the influence of race at each decision point is conditioned by values of other variables in the model. We report models containing interaction terms in the tables only where the interaction model produced a significant increment in fit over the additive model.
In Figure One we present a bar chart that depicts the juvenile justice system as a series of dichotomous decision points. The figure shows the relationship between race and each processing outcome, comparing the proportions of white and nonwhite youths receiving the most severe treatment at each stage. Before reviewing the bar chart, it is important to note that nonwhites comprise 21% of the population at risk (ages ten to seventeen) and 29% of the group referred to delinquency intake, but only 19% of the group referred to dependency intake.
Figure One indicates that, among those referred for delinquent acts, a greater proportion of nonwhites than whites received the more severe disposition at each successive stage in processing. Racial disparities are most pronounced at intake screening and judicial disposition. For example, 53% of nonwhite youths referred to intake are recommended for referral to court, compared to 42% of white youths. At judicial disposition, 31% of nonwhite youths are incarcerated or transferred, compared to 18% of white youths. The cumulative effect of these decisions is that the racial composition of the cohort becomes increasingly nonwhite as it moves through the system: while nonwhites make up 21% of the population at risk (ages ten to seventeen) and 29% of the cohort referred to delinquency intake, they make up 44% of the cohort incarcerated or transferred.
For status offenses, on the other hand, the picture is quite different. Figure One's last three comparisons show that there is evidence of racial disparity in processing which suggests a tendency to treat whites more harshly than nonwhites. At intake, decisions are made to refer to court slightly greater proportions of white status offenders than nonwhite status offenders. Although the proportions of whites and nonwhites receiving judicial dispositions of residential placement are approximately the same, there is a fairly marked tendency to incarcerate higher proportions of white repeat status offenders under the court's contempt power (36% of whites, compared to 26% of nonwhites).
Figure One is useful in two respects. First, it provides an indicator of potential discrimination at the bivariate level of analysis. It remains to be seen whether these race-patterned differences in outcomes can be explained by other variables in multivariate models. Second, the figure points to patterns of race differentials in processing that are quite different for delinquency and dependency cases. This is an important issue for further exploration in the analyses and discussion that follow.
Table One presents results of logistic regression analyses in which we have modelled the processing of delinquency cases. Panel One shows results for intake referral outcomes. Consistent with other studies, the findings indicate that the seriousness of the current offense weighs heavily in intake decision-making, and is the strongest predictor of outcomes at this stage. As might be expected, intake officers also consider youths' prior records of offending and are more likely to recommend formal processing for youths with lengthy and serious prior records. In addition, however, there is evidence that individual characteristics of youths influence intake referral decisions. Non-whites, older youths, and males are significantly more likely to be recommended for formal processing than are whites, younger adolescents, and females.
[TABULAR DATA OMITTED]
Because logistic regression coefficients do not have a clear, intuitive interpretation, it is helpful to discuss the effect of race on the probability that intake will recommend a case for formal processing. To do this, we illustrate with the case of a typical youth referral: we calculate the predicted probability of a recommendation for formal processing for white and nonwhite youths with values of other variables in the model set at their respective means. In these data, the typical youth referred to delinquency intake is a fifteen year old male referred for a misdemeanor against person (e.g., simple battery), with a prior record score consistent with having one prior referral for a misdemeanor against property (e.g., criminal mischief). The probability that a white youth with these characteristics will be recommended for formal processing is 47%. For a similar nonwhite youth, the probability of a recommendation for formal processing is 54%--a substantial difference of seven percentage points.
Panel Two presents logistic regression results for detention out-comes. The results indicate that detention decisions are influenced to a modest degree by race when other important variables are controlled. For the typical case, the probability of being held in secure detention is 12% for a white youth, compared with 16% for a non-white youth. The strongest predictors of detention status are the legal variables of current offense and prior record. Gender and age are also significant predictors, although, as is the case with race, their effects are modest.
Panel Three presents an interaction model that provides a significant improvement in fit over the additive model. The panel shows that the effect of race on detention status is conditioned by both gender and prior record. Nonwhite males and females are handled much more similarly than are white males and females: among whites, the probability of being detained for females is significantly lower than is the case for males. Nonwhite females, on the other hand, are detained at a rate that approximates that of nonwhite males. The effect of race is also conditioned by severity of a youth's prior record. When youths have no prior record, or their prior record is not serious, non-whites and whites are rarely detained, and there is little difference in their detention rates. When the prior record is indicative of serious or frequent offending, however, the risk of being detained is much higher for nonwhites than for whites.
Two illustrations may help to clarify the nature of these interaction effects. Consider, for example, a white male with a relatively high prior record score of eight. His probability of detention is 17%. A nonwhite male with the same prior record has a probability of detention of 23%, a difference of six percentage points. A nonwhite female with a similar prior record has a detention probability of 21% (higher than that of the white male), while a similar white female has a probability of detention under these circumstances of 16%. When the prior record score is low, however, these race and gender differences are almost nonexistent.
Panel Four presents model estimates for the prosecutorial referral stage. As was the case with the initial two processing stages, offense seriousness and prior record each have significant effects on prosecutorial decision-making, as do gender and age. The impact of race is very modest: the typical white youth has a 32% chance of being referred to court, compared to a 34% chance for the typical nonwhite youth. After controlling for other variables, being detained has the effect of increasing the likelihood of referral to court. Consequently, some of the influence of race on prosecutorial decision-making is subsumed by the effect of detention status. Nonwhites are more likely than whites to be detained, and those who are detained are more likely to be prosecuted. Thus, racial inequality at the prosecutorial referral stage is more pronounced than the race coefficient in this model would suggest.
Panels Five and Six present results for judicial disposition. The main effects model (Panel Five) indicates that severity of the current offense and prior record each have significant, though fairly modest, effects on dispositional outcomes. Juveniles who are detained are also more likely to receive dispositions of incarceration. Once again, race, operating through detention status, indirectly affects disposition. Those found in contempt also are significantly more likely to receive harsher judicial dispositions. At this final stage in processing, each of the sociodemographic characteristics has a significant effect on case outcomes, the effect of race being relatively strong. The typical white delinquent has a 9% probability of being committed or transferred, compared to a 16% probability for nonwhites.
The results for judicial disposition become considerably more complex when we examine Panel Six, which includes three significant interaction terms in a better fitting model. The effects of race on case outcomes at this stage are conditioned by gender, prior record, and contempt status. To summarize briefly, while nonwhite and white youths with more serious prior records are dealt with similarly, nonwhite offenders with nonserious prior records are more likely to be incarcerated or transferred than white offenders with nonserious prior records. The findings also indicate that the treatment of nonwhite females more closely approximates the treatment accorded nonwhite males than does the treatment of white females approximate the treatment of white males. Finally, being held in contempt increases the likelihood of a more severe outcome selectively among whites, but not among nonwhites. We will return to this finding later in our discussion of status offender processing.
Table Two presents results for status offenders. Panel One provides estimates for intake referral outcomes. The model indicates that those with prior records of offending are more likely to be referred for formal processing. Additionally--and in direct contrast to delinquency cases--whites, females, and younger youths are more likely to be referred to court. There is a small race effect. The typical non-white status offender has a probability of referral to court of 9%, compared to a probability of 11% for the typical white status offender.
Table 2 LOGISTIC REGRESSION MODELS FOR STATUS OFFENSE CASES
Court Referral Judicial Disposition 1 SE 2 SE Intercept -1.691(*) .070 Race .227(*) .05 .387 .140 Gender -.254(*) .04 .022 .118 Age -.023 .01 -.115(*) .031 Prior Record .061(*) .800 .067 .024 Y .13 .14 N 24,341 2,747 -2 Log Likelihood 18,768 2,181 Model [X.sup.2] 105,4 d.f. 34,4 d.f. (*) Significant at the .001 level SE = Standard Error
Judicial disposition decisions for status offenders are modelled in Panel Two. The variables in the model do a poor job of predicting status offense outcomes. Of the four predictors, only age reaches statistical significance. Younger youths referred to court are more likely than their older counterparts to be ordered into residential placements. Recall, however, that when repeat offenders are referred to delinquency court for contempt, white status offenders are significantly more likely to be .incarcerated than are nonwhite status offenders.
D. SUMMARY OF FINDINGS OF THE QUANTITATIVE ANALYSIS
Our analysis points to clear disadvantages for nonwhites at multiple stages in delinquency case processing. While the magnitude of the race effect varies from stage to stage, there is a consistent pattern of unequal treatment. Nonwhite youths referred for delinquent acts are more likely than comparable white youths to be recommended for petition to court, to be held in pre-adjudicatory detention, to be formally processed in juvenile court, and to receive the most formal or the most restrictive judicial dispositions. For status offense cases, a very different pattern emerges. Whites are slightly more likely to be referred for formal processing than nonwhites. Although there are no significant race differences in status offender outcomes at the judicial disposition stage, when repeat status offenders are referred to delinquency court for contempt, whites are significantly more likely than nonwhites to be incarcerated. This difference in the way race impacts juvenile justice processing for delinquents and status offenders is intriguing. It is an issue we explore in some detail in the following section.
III. PART II
A. INTERVIEW DATA
To supplement and provide a basis for interpreting our quantitative findings, we conducted telephone interviews ranging in length from one to four hours with a randomly selected sample of thirty-four juvenile justice officials. The sample includes intake supervisors, assistant state's attorneys assigned to juvenile divisions, public defenders assigned to juvenile divisions, and juvenile court judges from each judicial circuit. A primary reason for conducting the interviews was to examine more deeply and from different perspectives the race differences uncovered in the quantitative portion of our analyses. We wanted to determine what officials working in the system observed and believed with regard to race effects, as well as how they interpreted these effects. Although most of our respondents are seasoned insiders with years of experience in juvenile justice, the explanations they offer must be considered tentative, because the sample is small. We are confident, however, that their observations point to potentially fruitful avenues for further research.
Findings of race differentials in processing, while consistent with the notion of intentional race bias or discrimination, are subject to a number of other interpretations as well. It would be too simplistic to conclude that our findings provide evidence of widespread racial prejudice. As we shall see, the reasons offered by justice officials for the racial disparities that we have observed are multiple and complex.
A majority of the juvenile justice officials whom we interviewed were quick to indicate that our findings of racial disparities in processing were consistent with their experiences and observations. There was, however, variation by functional role. For example, all of the DHRS caseworkers and defense attorneys in the sample perceived race disparities in processing. A smaller proportion of prosecutors and judges perceived race effects. There was also variation in interviewees' perceptions of the kinds of racial bias present in the juvenile justice system. Some respondents, for example, believed the main problem was individuals who held and applied prejudicial attitudes. Many more saw the problem as endemic to the system, the consequence of well-intended policies and practices that impact differentially on whites and nonwhites.
This latter group of respondents suggested that racial disparities in delinquency case processing are in part a result of agency policies and practices that focus on family support and family cooperation as considerations for diversion, for detention, and for final disposition. They noted that, in some instances, these considerations are incorporated into formalized agency decision criteria.
For example, DHRS policy renders youths referred for delinquent acts ineligible for diversion programs if their parents or guardians (a) cannot be contacted, (b) are contacted but are unable to be present for an intake interview, or (c) exhibit attitudes and styles of behavior that are perceived as uncooperative to intake staff. It is important to note that availability of a telephone and access both to transportation to DHRS offices and child care for young children who must remain at home are all taken for granted in this diversion policy.
DHRS intake supervisors reported that minority parents often are single working mothers or single mothers on welfare with other young children at home. If employed, they are often employed in low-paying, low-status occupations; unlike those in managerial and professional positions, these parents often lack the flexibility to take time from work to be interviewed. In addition, many may be embarrassed to make such requests of their employers. Those who are unemployed and on welfare frequently lack access to child care for young children remaining at home. Many must depend on public transportation which may not operate near their homes or DHRS offices. Some do not have telephones and this makes it more difficult for DHRS officials to contact them. Intake officials also indicated that minority parents tend more often than white parents to be distrustful of the juvenile justice system. Intake staff tend to see these families as less cooperative with DHRS. Similar references to family support and cooperation were cited by prosecutors as key considerations in detention decisions. Generally, these considerations have a negative and differential impact on nonwhite delinquents. Typical is the view illustrated in the following comment by a delinquency intake supervisor:
Our manual told us to interview the child and the parent prior to making
a recommendation to the state's attorney. We are less able to reach
poor and minority clients. They are less responsive to attempts to reach
them. They don't show. They don't have transportation. Then they are
more likely to be recommended for formal processing. Without access
to a client's family, the less severe options are closed. Once it gets to
court, the case is likely to be adjudicated because it got there. It's a self
Thus far, we have noted that respondents identified criteria for diversion and detention that render nonwhite offenders more likely to be recommended for formal processing and held in secure detention. It is interesting to note the interface between these comments and the findings of our quantitative analyses. Note, for example, that race had no direct effect on prosecutorial filing decisions, and that both prosecutorial filing decisions and judicial dispositional decisions to incarcerate were influenced by detention status. The race effect on both the decision to formally prosecute and on the judicial decision to incarcerate appears to emanate in no small part from decisions made at earlier stages in the system -- decisions to recommend formal prosecution and secure detention -- that are tied to well-intended but inadvertently discriminatory front-end agency policies.
Many of the interviewees were aware that policies of Florida's juvenile justice system locked them into decisions that ultimately disadvantage nonwhites. Respondents from all levels of the system commented on the unfairness of a structure which renders nonwhite youths more vulnerable to formal processing because their families are unable to comply with agency policies.
Many respondents also reported that juvenile justice decisions in delinquency cases are affected by differentials in access to retained counsel and private treatment resources, differentials that impact negatively on low-income especially minority-clients. Especially in later stages of delinquency processing, respondents observed that the system emphasizes treatments (e.g., psychological counseling, drug treatment) that are often best obtained through private agencies. Youths from affluent families may take advantage of these treatment options and avoid formal processing. Minority youths who are less affluent can only obtain comparable services by being adjudicated delinquent and then committed to residential facilities.(18) As one of the judges in our sample observed:
Minorities and low income kids get more [juvenile justice system] resources. If parents can afford [an expensive private treatment facility], the child gets probation. If not, he gets committed. Income is significant in that a lot of early interventions are directed to middle income groups. If a child needs constructive activity, a middle class family can afford it. Maybe there is institutional bias.
As might be expected, some respondents were very critical of practices which resulted in minorities receiving harsher treatment by justice officials. Others argued that these practices were quite defensible. In their view, justice officials were merely trying to provide needed services to the disadvantaged that wealthy families could purchase on their own. To become eligible for these services, however, youths had to be formally processed--e.g., referred to court, adjudicated delinquent, and placed on formal probation or committed to residential programs. Only then could these services be provided at state expense. Moreover, this sort of policy negatively impacts on nonwhites anytime they come back to the system on a subsequent charge. A juvenile's prior record and prior disposition history are primary predictors of (and primary justifications for) formal processing and more severe sanctions.(19)
What may begin with good intentions at an earlier stage ultimately becomes a self-fulfilling prophecy. The influence of race is obscured as decisions to formally prosecute and detain in the past are used to justify more severe sanctions for youths returning to the system.
In addition, many respondents indicated that juvenile justice officials make decisions influenced in part by perceptions (or misperceptions) of youths' family backgrounds and circumstances. Respondents frequently reported that delinquent youths from single-parent families and those from families incapable of (or perceived to be incapable of) providing good parental supervision are more likely to be referred to court and placed under state control. In other words, when justice officials perceive that there is family strength and support, they are more likely to select less intrusive treatments and sanctions. For the most part, our respondents believed that these distinctions were fair and appropriate. They also indicated that, at least in delinquency cases, black family systems generally tend to be perceived in a more negative light, that pre-disposition reports give disproportionate attention to assessments of family situations, and that judges rely heavily on pre-disposition reports in reaching dispositional decisions.
Several comments made by state's attorneys and judges are instructive:
Judge: "Inadequate family correlates with race and ethnicity. It makes sense to put delinquent kids from these circumstances in residential facilities."
State's Attorney: "Detention decisions are decided on the basis of whether the home can control and supervise a child. So minorities don't go home because, unfortunately, their families are less able to control the kids . . . I think the way the system sets up programs shows some institutional bias. If family stability was not a prerequisite to admission to less severe program options, race differences would be less."
State's Attorney: "In black families who the dad is, is unknown, while in white families--even when divorced--dad is married or something else. The choices are limited because the black family is a multi-generational non-fathered family. You can't send the kid off to live with dad."
One of the key findings of our quantitative analyses was that nonwhites are disadvantaged in delinquency case processing while the processing of status offenders in dependency cases appears to disadvantage whites.(20) One of the more experienced judges in our sample, who had served on the bench in both delinquency court and dependency court, shed some potential light on this issue when he observed that there is a "mysterious irony" in the way black and white families are viewed in delinquency courts versus dependency courts. He reported that it is common for judges in delinquency courts to justify harsher sentences for black youths by noting that the black family structure is weak and incapable of dealing effectively with troubled youths. At the same time, he observed that black youths in dependency actions are less likely to be made wards of the court because, in this context, the black extended family network is perceived more positively. In fact, he noted that in the dependency system the black family in particular is seen as strong. Thus, in delinquency proceedings, the black family is stereotyped as generally inept and the white family is seen as generally capable, while in dependency proceedings, these stereotypes are reversed--whites need help and blacks can handle their own problems.
One plausible explanation for these divergent views regarding black and white family systems may be that different organizational cultures have evolved in delinquency and dependency processing agencies. Social workers in dependency cases, for example, are accustomed to seeking out extended family networks (grandparents, aunts, uncles, etc.) to assist in the care of abused and neglected children. In the traditional extended kin networks of black families they find alternative sources of support, so the black family may be viewed in a more positive light. Intake officials in delinquency cases, in contrast, tend to wear two hats: those of social workers and law enforcement officers. Their orientations are geared more toward ensuring public safety, and in their organizational cultures, the broken family is more likely to be seen as a harbinger of future trouble.
Another possible interpretation for these findings is suggested when we consider differences in the ideologies and orientations of delinquency versus dependency proceedings. In delinquency proceedings, the rhetoric of treatment and rehabilitation coexists uneasily with an orientation to punish those who have violated the law. Indeed, in recent years, much has been written about the shift to an increasingly retributive mentality in delinquency courts, a shift that elevates concerns about punishment and public safety over the historical concern for treatment.(21) Dependency proceedings, on the other hand, traditionally have been and remain more often couched in the language of care and protection.(22) Dependent children--who include abused and neglected children as well as status offenders--are more often viewed as victims (i.e., children from troubled families who are without proper care). Viewed from this perspective, court referral and incarceration in delinquency cases are means of providing sanctions for those whose behavior is most strongly condemned (older youths, males, nonwhites). In the dependency system, by contrast, court referral and even incarceration are regarded as treatments more in keeping with the traditional parens patriae goals of protective care and rehabilitation. In the dependency system, these responses are means of providing services to those for whom the system has the greatest compassion (younger youths, females, whites).
Following this line of argument, the different responses to nonwhites and whites in delinquency versus dependency proceedings may be understood as official manipulation of cultural stereotypes to fit justice system goals. From this vantage point, minorities are disadvantaged in both delinquency and dependency courts. When the system is oriented toward punishment (delinquency cases), nonwhites receive more of it. When the system is oriented toward beneficent care and protection, nonwhites receive less of it. The view of the family that is invoked may merely be a means by which race-biased organizational responses are rationalized.
Our findings show clear indications of race differentials in justice processing. The quantitative analyses demonstrate appreciable effects of race on delinquency case processing that disadvantage minority offenders. These findings are consistent with perceptions of juvenile justice officials at all levels of professional involvement. That minority offenders are disadvantaged is not surprising. A number of previous studies have reported similar findings. What is surprising is our finding that the effect of race on the processing of status offenders differs so markedly from that of delinquents. Here, differentials in processing by race are less pronounced and when they do appear, they indicate that whites, rather than minority youths, are more likely to penetrate further into the system and to receive dispositions involving incarceration.
Had we only the quantitative data, these contrasts would have caused us concern; we would have had no basis for offering an explanation. Interviews with justice officials conducted as a second phase of this research have offered an intriguing glimpse into the dynamics of justice work, into organizational policies, practices, and philosophies, and into the possibility of manipulative use of race realities and perceptions.
Our qualitative findings support several interpretations. Intentional race discrimination does not appear to play a major role in accounting for racial disparities in processing. Although some officials whom we interviewed believed that some justice officials were motivated by prejudicial attitudes, few recounted specific instances of racially motivated actions. Without question, there are some justice officials who hold and act upon racially prejudicial attitudes. As long as race bias exists in the general culture, it would be surprising indeed if it did not operate through individuals in the juvenile justice system as well. However, we are not inclined to conclude that the disparities we observed are largely attributable to intentional race discrimination.
Instead, we see much evidence of institutional racism. This is evident both in criteria for diversion and pre-trial release that focus on family support and cooperation, and in efforts to provide the economically disadvantaged with resources at state expense that the more affluent can purchase on their own. Obtaining these resources exacts a price in terms of adjudications of delinquency and sentences to confinement.
Finally, our qualitative findings support the need to take a closer look at the climate of organizations that do delinquency work versus dependency work. The differing orientations of these two systems, and the differing foci of the professionals who work within them, may support very different views of nonwhite families as facilitators or inhibitors of the achievement of organizational goals for children. Whether the differing orientations of delinquency processing versus dependency processing agencies account for the inverted race disparities we have observed across the two systems--or, alternatively, whether these inverted disparities reflect a manipulation of perceptions of race realities to consistently disadvantage minorities--is an important issue for future research.
In closing, we wish to emphasize the tentative nature of our conclusions. Though randomly selected, our interview sample was small. Moreover, the sample was not stratified to include representative numbers of intake officials and judges from both delinquency and dependency courts. Further research is needed to explore the alternative interpretations offered here for the quantitative findings. Interviews with large samples of individuals drawn from both the dependency and delinquency systems would be helpful. In addition, participant observation and other field studies may provide clearer insights into informal agency policies and practices, and organizational climates that provide the context within which race effects may be most fully understood. (1) U.S. Bureau of the Censuc, U.S. Dep't of Commerce, U.S. Population Estimates, by Age, Sex, Race, and Hispanic Origin: 1980 to 1991, at 2 (1993) (Table 1--Resident Population--Estimates by Age, Sex, Race, and Hispanic Origin). (2) Fed. Bureau of Investigation, U.S. Dep't of Just., Crime in the U.S. 1993, at 235 (1994). When national arrest data are examined by offense type, it appears that minority youths are vastly overrepresented among youths arrested for property offenses, drug offenses, and, especially, violent crimes, but only slightly overrepresented among status offenders. Id. The latter observation is based on arrests for curfew violations and runaways. The FBI does not report counts of other status offenses, such as truancy and "beyond control." (3) Office of Juvenile Justice and Delinquency Prevention, U.S. Dep't of Just., in Children in Custody 1989, at 6 (1991) (Table 5--Juveniles in Custody by Minority Status and Type of Public Facility: 1989). (4) Ear comprehensive reviews, see U.S. Dep't of Justice, Minorities and the Juvenile Justice System (1992); Carl E. Pope & William Feyerherm, Minority Status and Juvenile Justice Processing, Pt I, 22 Crim. Just. Abstracts 327 (1990); Carl E. Pope & William H. Feyerherm, Minority Status and Juvenile Justice Processing, Pt. 11, 22 Crim. Just. Abstracts 327, 527 (1990). (5) Following Joe R. Feagin, Racial And Ethnic Relations (3d ed. 1984), institutional discrimination refers to practices having a negative and differential impact on members of a subordinate race even though organizationally-prescribed norms or regulations guiding those actions have been established and carried out with no intent to harm. (6) For further discussion, see Barry Krisberg & James F. Austin, Reinventing Juvenile Justice (1993). (7) See Donna M. Bishop & Charles E. Frazier, The Influence of Race in Juvenile Justice Processing, 25 J. Res. Crime & Delinq. 242 (1988); Margaret A. Bortner & Wornie L. Reed, The Preeminence of Process: An Example of Refocused Justice Research, 66 Soc. Sci. Q. 413 (1985). (8) One might wish to include controls for social variables as well. For example, recent research suggests that juvenile detention decisions are affected by such factors as socioeconomic status, family structure, and school performance. See, e.g., Madeline Wordes et al., Locking Up Youth: The Impact of Race on Detention Decisions, 31 J. Res. Crime & Delinq. 149 (1994). Unfortunately, the data used in this research do not permit us to explore the impact of these variables. However, in a very real sense, this does not constitute an important limitation. Insofar as these variables are correlated with race, their inclusion might help specify the considerations that produce institutional discrimination, but they would in no way negate findings of racial disparity. Stated differently, that blacks and whites may be treated differently because blacks are more likely than whites to come from single-parent families would not alter a finding of differential treatment by race. It would merely specify the mechanism by which such differential treatment might arguably be justified. (9) This is not an unlikely scenario, given that justice officials exercise much greater discretion in decision-making with respect to minor offenses. For an extensive review and discussion, see Michael R. Gottfredson & Don M. Gottfredson, Decision Making in Criminal Justice: Toward the Rational Exercise of Discretion (2d ed. 1988). (10) See, e.g., Bishop & Frazier, supra note 7; Margaret Farnworth & Patrick M. Horan, Separate Justice: An Analysis of Race Differences in Court Processes, 9 Soc. Sci. Res. 381 (1980); Terance D. Miethe & Charles A. Moore, Racial Differences in Criminal Processing: The Consequences of Model Selection on Conclusions about Differential Treatment, 27 Soc. Q. 217 (1986); Marjorie S. Zatz, Race, Ethnicity and Determinate Sentencing: A New Dimension to an Old Controversy, 22 Criminology 147 (1984). (11) An important limitation of the data is that we were unable to examine the effect of race on police decision-making. The question remains whether race impacts police decisions to arrest youths and refer them to intake; there is evidence from much prior research that it does, especially with respect to minor offenses. See, e.g., Nathan Goldman, The Differential Selection Juvenile Offenders For Court Appearance (1963); Donald J. Black & Albert J. Reiss, Jr., Police Control of Juveniles, 35 Am. Soc. Rev. 63 (1970); Darlene J. Conley, Adding Color to a Black and White Picture: Using Qualitative Data to Explain Racial Disproportionality in the Juvenile Justice System, 31 J. Res. Crime & Delinq. 135 ( 1994); Dale Dannefer & Russell K Schutt, Race and Juvenile Justice Processing in Court and Police Agencies, 87 am. J. Soc. 1113 (1982);Jeffrey Fagan et al., Blind Justice? The Impact of Race on the Juvenile Justice Process, 33 Crime & Delinq. 224 (1987); Richard J. Lundman et al., Police Control of Juveniles: A Replication, 15 J. Res. Crime & Delinq. 74 (1978); Irving Piliavin & Scott Briar, Police Encounters with Juveniles, 70 AM. J. Soc. 20 (1964); James Q. Wilson, The Police and the Delinquent in Two Cities, in Controlling Delinquents 9 (Stanton Wheeler ea., 1968). Thus, there is a racial selection bias already built in that we are unable to estimate at the point in processing at which our data begins. (12) Transfer to adult court is the harshest sanction option for adolescent offenders. Once youths in Florida are transferred to criminal court, they may be fined, placed on adult probation, or sentenced to jail or prison. Moreover, their juvenile status is terminated and any subsequent offenses are automatically handled in the criminal courts. (13) Although Florida has fairly large Hispanic, Cuban, and Haitian populations, members of these ethnic groups cannot be identified with the data available to us. (14) The bulk of the persons coded as "black" are African-Americans. However, especially in South Florida, there are substantial numbers of persons of Hispanic origin who are frequently coded as "black" depending on their skin color. We use the term "nonwhite," then, to refer to a group made up predominancy, but not exclusively, of African-Americans. (15) Because many youths had multiple offenses recorded on a single day, we also constructed a measure of CURRENT OFFENSE chat summed the severity value of each of the allegations contained in the referral. The results using the summated score did not differ significantly from those obtained when we scored only the most serious offense. Consequently, we used the simpler measure. (16) While the Florida legislature subsequently abolished judicial contempt power in such cases, the new juvenile justice reform legislation reinstated the power. For further discussion of the uses and abuses of the law of contempt in juvenile proceedings, see Donna M. Bishop & Charles E. Frazier, Gender Bias in Juvenile Justice Processing: Implications of the JJDP Act, 82 J. Crim. L. & Criminology 1162 (1992); Randall R. Beger, Illinois Juvenile Justice : An Emerging Dual System, 40 Crime & Delinq. 54 (1994). (17) In cases where a youth's offense history included a status offense, the status offense was assigned a value of 1. (18) Even then, some respondents noted, there are too few spaces available for poor minority youth who need such treatment, and this results in their receiving the harshest dispositions, often for no real benefit. (19) Ronald A. Farrell & Victoria L. Swigert, Prior Offense Record as a Self-Fulfilling Prophecy, 12 L. & Soc'y Rev. 437, 450-51 (1978); John C. Henretta et al., The Effect of Prior Case Outcomes on Juvenile Justice Decision-Making, 65 Soc. Forces 554, 559-61 (1986). (20) Recall that in dependency case processing, whites are more likely to be petitioned to court by DHRS Intake and white repeat status offenders referred to delinquency court for contempt are more likely to be sent to secure facilities than their minority counterparts. (21) See, e.g., Barry Krisberg & James F. Austin, Reinventing Juvinile Justice (1993); Barry C. Feld, The Juvenile Court Meets the Principle of Offense: Punishment, Treatment and the Difference it Makes, 68 B.U. L. Rev. 821 ( 1988); Barry C Feld, The Punitive Juvenile Court and the Quality of Procedural Justice: Disjunctions Between Rhetoric and Reality, 36 Crime & Delinq. 443 (1990). (22) In many jurisdictions, the dispositions that status offenders may receive have been severely restricted, while in others, status offenses have been removed from juvenile court jurisdiction entirely.
DONNA M. BISHOP, Department of Criminal Justice and Legal Studies, University of Central Florida.
CHARLES E. FRAZIER, Department of Sociology, University of Florida.
|Printer friendly Cite/link Email Feedback|
|Author:||Bishop, Donna M.; Frazier, Charles E.|
|Publication:||Journal of Criminal Law and Criminology|
|Date:||Jan 1, 1996|
|Previous Article:||The social construction of a hate crime epidemic.|
|Next Article:||Preventative detention and the judicial prediction of dangerousness for juveniles: a natural experiment.|