AT VARIOUS DECISION POINTS IN THE STATE OF MARYLAND

Transcription

1 THE DISPROPORTIONATE REPRESENTATION OF AFRICAN-AMERICAN YOUTH AT VARIOUS DECISION POINTS IN THE STATE OF MARYLAND AVERAGE OF FY DATA FULL REPORT xxxxxxx xxxxxxx xxxxx xxxxxx xx xxxxxx xxxxxx xxxx December 1995

2 TABLE OF CONTENTS MAJOR FINDINGS AND RECOMMENDATIONS CHAPTER ONE Introduction... 1 Purpose... 2 Scope of the Study: Statewide and Small Area Analysis... 2 Maryland Juvenile Justice System Processing... 3 Maryland Data for FY Maryland Data: FY National Studies: A brief review...11 CHAPTER TWO Objectives Discussion of the Objectives Research Methodology A Data Acquisition B Determination of Small Areas in the State C Computation of Race-Gender Indices and Rates Per 1,000 youth Population CHAPTER THREE Section I: Data Analysis Format Description Summary of Results: Statewide Data State of Maryland Field Services Area Map Summary of Results: Small Area Analysis Area I Total Baltimore City East Baltimore West Baltimore South Baltimore Area II Total Baltimore County The Arbutus area The Dundalk and Essex area The Towson area Carroll County Harford County... 35

3 TABLE OF CONTENTS (continued) Area III Allegany County Frederick County Garrett County Washington County Rockville & its surrounding areas Other Montgomery County areas Total Montgomery County Area IV Caroline County Cecil County Dorchester County Kent County Queen Anne s County Somerset County Talbot County Wicomico County Worcester County Area V Calvert County Charles County St. Mary s County Prince George s - North Prince George s South Total Prince George's County Washington County Area VI Annapolis area Glen Burnie area Total Anne Arundel County Howard County Regression and Chi Square Tests Analysis CHAPTER FOUR Major Findings, Action Plan and Conclusion... 67

4 TABLE OF CONTENTS (continued) APPENDIX A: APPENDIX B: APPENDIX C: APPENDIX D: APPENDIX E: APPENDIX F: The list of small areas Decision points included in the study Select offense Category Disproportionate Representation Detailed Tables Tables D-1 to D-42 Offenses by Offense type by small areas Tables E-1 to E-42 Regression analysis data

5 LIST OF TABLES Table Number Title Page 1 Percentage of Population and Representation of Youth in the State of Maryland by Race-Gender: FY Number and Percent of Intake Referrals: FY FY Number and Percent of Delinquent Youths in Detention Facilities: FY FY Number and Percent of Youth in Secure Commitment: FY FY The Disproportionate Representation of African-American youth in the Maryland Juvenile Justice System: State of Maryland - Average of FY Population Index Values: State of Maryland Number & Percent of Alleged Offenses by Race-Gender: State of Maryland: Average of FY Data Population Index Values: Area I Total Population Index Values: East Baltimore City Population Index Values: West Baltimore City Population Index Values: South Baltimore City Number & Percent of Alleged Offenses by Race-Gender: Average of FY Data for Area I Population Index Values: Area II Total Population Index Values: Total Baltimore County Population Index Values: Arbutus area Population Index Values: Dundalk & Essex areas Population Index values: Towson area... 34

6 LIST OF TABLES (continued) Table Number Title Page 18 Population Index Values: Carroll County Population Index Values: Harford County Number & Percent of Alleged Offenses by Race-Gender: Average of FY Data for Area II Population Index Values: Area III Total Population Index Values: Allegany County Population Index Values: Frederick County Population Index Values: Garrett County Population Index Values: Washington County Population Index Values: Rockville and its Surrounding areas Population Index Values: Other Montgomery areas Population Index Values: Total Montgomery County Number & Percent of Alleged Offenses by Race-Gender: Average of FY Data for Area III Population Index Values: Area IV Total Population Index Values: Caroline County Population Index Values: Cecil County Population Index Values: Dorchester County Population Index Values: Kent County Population Index Values: Queen Anne s County Population Index Values: Somerset County... 48

7 LIST OF TABLES (continued) Table Number Title Page 37 Population Index Values: Talbot County Population Index Values: Wicomico County Population Index Values: Worcester County Number & Percent of Alleged Offenses by Race-Gender: Average of FY Data for Area IV Population Index Values: Area V Total Population Index Values: Calvert County Population Index Values: Charles County Population Index Values: St. Mary s County Population Index Values: Prince George s County North Population Index Values: Prince George s County South Population Index Values: Total Prince George s County Number & Percent of Alleged Offenses by Race-Gender: Average of FY Data for Area V Population Index Values: Area VI Total Population Index Values: Annapolis area Population Index Values: Glen Burnie area Population Index Values: Total Anne Arundel County Population Index Values: Howard County Number & Percent of Alleged Offenses by Race-Gender: Average of FY Data for Area VI... 62

8 LIST OF FIGURES Table Number Title Page 1 Flow Chart of case referrals in the Juvenile Justice System Percentage of Youth Population and Representation of Youth at Various Decision Points by Race: FY Percentage of Population and Representation of African-American Males Versus White Males - State of Maryland: Average of FY Percentage of Population and Representation of African-American Females Versus White Females - State of Maryland: Average of FY Comparison of Index Values between African-American and White Males at Various Decision Points: State of Maryland Comparison of Index Values between African-American and White Females at Various Decision Points: State of Maryland Comparison of Index Values between African-American and White Males at Various Decision Points: Area I Comparison of Index Values between African-American and White Males at Various Decision Points: Area II Comparison of Index Values between African-American and White Males at Various Decision Points: Area III Comparison of Index Values between African-American and White Males at Various Decision Points: Area IV Comparison of Index Values between African-American and White Males at Various Decision Points: Area V Comparison of Index Values between African-American and White Males at Various Decision Points: Area VI Percentages of Jurisdiction Waivers by Offense History Scores African-American Males Versus White Males Percentages of Secure and Residential Placements by Offense History Scores: African-American Males Versus White Males Percentages of Probation Youth by Offense History Scores African-American Males Versus White Males... 66

9 Definitions of the Terms Used in the Study Population: General population of youth, ages 11 through 17 years. Minority Group: This refers only to the African-American population. The other race category refers to Hispanics, Native-Americans, Asians and Pacific Islanders. Small Areas: These are cities or counties or urban neighborhoods based on income or ethnic homogeneity. Counties are the geographic unit of analysis for the rural portions of the State. Urban areas are divided by grouping the adjoining ZIP codes based on relative homogenous racial and socioeconomic characteristics. (see Appendix A for details). Decision Point: The various steps through which the cases of youth are processed in the juvenile justice system. The decision points are intake, formal petitions, probation, detention, residential placement and secure commitment. Intake: Referral to DJJ intake through arrest by police officers or other sources such as schools, parents, friends, citizen or other State Departments. Formalization: This is a complaint that an intake officer has authorized the State's Attorney to file a petition. The State's Attorney has the right to review felony and handgun related cases and can file a petition in which the Department has proposed an alternative, i.e. informal adjustment or close. A police officer or a victim may appeal a proposed informal adjustment by an intake officer to the State's Attorney to file a petition. Probation: Probation is the court disposition for adjudicated delinquents and refers to delinquency referrals. Children in Need of Supervision (i.e., status offenders) may be placed on protective supervision, but not on probation. Detention: Detention is a temporary secure commitment for alleged delinquent behavior. Youth detained are awaiting a court hearing or placement. Residential Placement: These are placements in a non-institutional program providing 24-hour care and supervision. They include state and privately-operated programs. Secure Commitment: Placement in an institutional program which is staff and architecturally secure.

10 CHAPTER ONE INTRODUCTION There is a growing concern nationally about the disproportionate representation of minority youths in the juvenile justice system, especially in detention and secure juvenile facilities. During the 1988 Congressional Re-authorization of the Juvenile Justice and Delinquency Prevention Act (JJDPA), Section 223(a)(23) was amended to establish a new requirement that each state receiving funds from the Formula Grant Program address efforts to reduce the proportion of juveniles who are members of minority groups confined in secure detention facilities, secure juvenile correctional facilities, adult jails and adult lockups, if such proportion exceeds the proportion such groups represent in the general population. The 1992 amendments to the Act upgraded this issue to a mandate. The Maryland Department of Juvenile Justice (DJJ), being similarly concerned, has been pro-active in addressing the issue of disproportionate representation of minority youths. In 1990, the Juvenile Justice Advisory Council (JJAC) conducted an initial assessment by examining the following historical data: arrest, referrals to DJJ, waiver of jurisdiction, disposition, detention, and commitment to secure care. The preliminary results of this assessment indicated that minority youth, especially African- American males, in Maryland appear to be arrested, referred to DJJ, formally handled by DJJ, securely detained, and committed to juvenile correctional facilities at rates which exceed their representation in the general juvenile population. In 1993 and 1994, the Juvenile Justice Advisory Council sponsored two statewide conferences on Children of Color and the Juvenile Justice System involving key decision-makers in the juvenile justice system, including representatives from the Department of Juvenile Justice, law enforcement, prosecutors, defense counsel, juvenile courts and the community. The purpose of these conferences was to increase the awareness of juvenile justice professionals to the extent of the problem of disproportionate representation of minority youth and the need for cultural competency training. As a result of these conferences, DJJ undertook further a study of the issue of disproportionate representation. This report is the product of that work. Mr. Richard Friedman, Program Administrator of the Juvenile Justice and Advisory Council, Dr. David Altschuler, Associate Professor at the Institute for Policy Studies, Johns Hopkins University, and Dr. Charles Wellford and Dr. Denise Gottfredson, Criminal Justice Professors at University of Maryland at College Park were available in a technical assistance capacity to DJJ in the course of carrying out this study. This study identifies the disproportionate representation of African-American youth within Maryland jurisdictions for the average of three Fiscal Years (FY 1990 through FY 1992) at various decision points. The various decision points are referral, formalization, probation, detention, residential placement, and secure commitment. DJJ's trend data reveals that African-American youth are underrepresented at closed at intake, disapproved, and informal supervision decision points. Whites are overrepresented at these decision points. Therefore, these decision points are not included in the study. This study takes a step beyond previous work by investigating whether or not African-American youth are overrepresented in the juvenile justice system because they have committed more serious offenses. Although offense severity and prior offense history are important criteria in deciding case disposition, many other qualitative factors such as family composition, education progress, poverty and unemploy- 1

11 ment, mental health status, and willingness to accept responsibility for the delinquent act must be considered. This study did not control for these other decision criteria due to lack of data. More complete data collection and further research is required to determine the extent to which other essential classification criteria may further explain why African-American youth are overrepresented in more secure dispositions. PURPOSE The purpose of this study is twofold. One is to determine if there is any evidence of disproportionate representation after controlling for the offense history, race, age, and gender. The other is to identify in what small areas and at which decision point the disproportionate representation occurs. For the purpose of this study small areas are cities or counties or urban neighborhoods based on income or ethnic homogeneity (see Appendix A). SCOPE OF THE STUDY: STATEWIDE AND SMALL AREA ANALYSIS The scope of the study includes statewide and small area analysis of data covering three fiscal years from intake to secure commitment. At the time this was initiated this represents the most recent data available from the Department s computerized system, Information System for Youth Services (ISYS). The juvenile population covered in the study are 11 through 17 years of age. The minority youth included in the study are African-Americans. The other race category are not addressed in this study, because the total 11 through 17 population and the DJJ referrals had few persons in this category, 7% and 2% respectively. Unlike other studies, this study uses two different indices to calculate the disproportionate representation of minority youth in the juvenile justice system. A separate profile of disproportionate representation is constructed for each small area. It was possible to provide a complete picture of the comparison between African-American and white youth representation in the Maryland juvenile justice system, because the data analysis was based on youth and youth related information (as opposed to case analysis). The study also includes information on the current alleged offenses of youth referred to intake. Due to data limitations this study does not include all the factors often associated with dispropotionate representation, i.e., family composition, economic, and social factors. 2

12 MARYLAND JUVENILE JUSTICE SYSTEM PROCESSING The Maryland Juvenile Justice System includes a number of integrated processes, procedures, services, and agencies that work to control and alleviate the problem of juvenile delinquency in the state. DJJ is a pivotal part of the system with police, schools, state s attorneys, juvenile courts, and other public and private services. DJJ processes almost fifty thousand (50,000) cases a year. Figure 1 illustrates the case flow of the referrals in the Juvenile Justice System for FY This provides a glimpse of the activity surrounding the cases referred to DJJ and the outcomes of decisions by the Department, the State s Attorney s Offices, and the Juvenile Courts in the State. The chart shows the partnership of the various parts of the system as well as the way referrals enter and proceed through the system. It is to be noted that the numbers provided in the chart are cases and not youths. The Department s referral sources include the various law enforcement agencies throughout the state, the Departments of Education and Social Services, citizens, and parents. As defined by the General Assembly, there are four possible decision alternatives at intake: disapproved, closed at intake, informal supervision and formalization. In FY 1992, the Department s Intake Officers, working out of offices located throughout the state, screened and evaluated 45,824 complaints against youths. Police referrals were the source for 41,481 (91% of total). In Figure 1, the percentages for African-Americans are shown in Parentheses and the percentages of whites are shown without parentheses. Figure 1 indicates that white youth cases were closed, disapproved, or adjusted through informal supervision at a higher percentage than African-American youths. In contrast, approximately 63% of the formalized cases involved African-American youths. When the Department authorizes petitions, the chart shows the activity of the State s Attorney s office either processing the case for court or denying/withdrawing the petition. Twelve percent (2,202) of the total formal cases were denied by the State s Attorney. Nearly all of the remainder cases were heard in court where they can be dismissed, denied, waived to adult jurisdiction, continued or stetted. If the charges are sustained and if a youth is found delinquent, the major court dispositions are probation and commitment for placement, either in the community or in an institutional setting. The boxes under court dispositions show the various outcomes. By far the most frequent Court disposition is probation, which entails providing community services and the case management oversight for adjudicated delinquents who continue to live in their own homes. Physical and legal custody of the youth remains with the parents or guardians. In FY 1992, a little over one fourth of all youth who were adjudicated delinquent were placed on probation (27%). Youths are committed to DJJ by an act of a judicial officer ordering that they be placed in an institutional or residential facility, for a specific reason authorized by law. In this case, there is a transfer of custody to the Department. The Department provides a range of programming for committed youth, from enhanced supervision to secure residential custody. In FY 1992, approximately 11% of the formalized cases were committed to DJJ for placement. 3

13 Figure 1 Flow Chart of Case Referrals in the Maryland Juvenile Justice System FY 1992 Police 41,481 (51%) 46% Dept. of Ed./DSS 1,225 (71%) 28% Parent\Relative 1,278 (55%) 39% Citizen 1,079 (54%) 42% Court 305 (54%) 43% Others 456 (49%) 45% DJJ Intake 45,824 (51%) 47% Decision Code Missing 185 (58%) 41% Disapproved 1,431 (44%) 53% Formal Cases 17,785 (63%) 35% Closed at Intake 18,972 (43%) 54% Informal Cases 7,451 (47%) 50% Petition Denied by States Attorney 2,202 (73%) 25% Petition Withdrawn 45 (48%) 52% Court Dispositions 15,538 (60%) 38% Probation/Protective Supervision 4,827 (51%) 46% Dismissed/Closed 3,930 (71%) 28% Committed to DJJ for Placement 1,885 (70%) 27% Continued Case/Stet 1,735 (49%) 49% Jurisdiction Waived 782 (73%) 26% Inter-Region/State 166 (55%) 42% Nol Pros 307 (49%) 50% Others 97 (51%) 45% Disposition Pending* 1,460 Writ Pending 224 (72%) 20% Transfer 98 (45%) 48% Sub Curia 27 (29%) 71% Note: Percentages for African-Americans are shown in parentheses. Percentages for Whites are shown without parentheses. *Pending the next fiscal year. 4

14 MARYLAND STATISTICAL DATA FOR FY 1992 The most recent data at the time this study was initiated, i.e., FY 1992 indicates that DJJ handled 32,736 youths at intake. Of the total intake youths, 47% were white, 50% were African-American and 2% were other race group. In FY 1992, the white population in the State of Maryland, 11 through 17 years comprised 59%, African-Americans 34%, and other race group 7%. 1 Figure 1 shows the percentages of population, and the percentages of youths seen at the various decision points in the juvenile justice system. Percent 100 Figure 2 Percentage of Population and Representation at Various Decision Points by Race State of Maryland: FY % % 51% 47% 63% 53% 45% 69% 49% 49% 40 34% 35% 28% 19% 20 7% 2% 2% 2% 3% 2% 2% 0 Population Referral Formalization Probation Detention Residential Secure Placement Commitment African-American White Other Maryland FY 1992 data reveals that although the African-American youth account for only 34% of the statewide population they represent: 51% of the total youths referred to DJJ, 63% of the formalizations, 53% of the total youth placed on probation, 69% of the total detained youths, 49% of the total residential placements, and 79% of the total secure commitments. The proportion of African-American youths detained or committed to a secure facility is twice as high as their population size. 1 The other race category is not addressed in this study, because the total 11 through 17 years old youth population and the DJJ referrals had few persons categorized in an other race group, 7% and 2% respectively. 5

15 In Table 1 and Figures 3 and 4 are shown the percent of population and representation of youth at the various decision points by race-gender for the average of FY The African-American male population accounted for 17% of the statewide population. By contrast, 74% of the total secure residential commitments to Charles H. Hickey, Jr. School were African-American males. The ratio of African- American males was (74%/17%=4.23). This means that African-American males in Hickey School were represented, at a rate which was 4.23 times more than their rate of representation in the state population. Conversely, white male youths accounted for only 20% of the total Hickey school commitments, but constituted 34% of the state youth population. The ratio of white males was.59 (20%/34%=.59). This shows that white males in Hickey school were represented, at a rate which was.41 less (1.0 minus.59=.41) than their rate of representation in the state population. However, whether this disproportionate representation can be considered statistically significant requires controlling for age, gender, and offenses. It is important to establish whether African-Americans are more likely to be referred to intake and committed to detention/secure facilities than whites with similar offenses and offense histories, age, and gender. Table 1 Percentage of Population and Representation of Youth in the State of Maryland by Race-Gender: FY 1992 Race/Gender % Population % Referral % Formalization % Probation % Detention % Residential % Secure Residential African-American Male African-American Female White Male White Female

16 Figure 3 Percentage of Population and Representation of African-American Males versus White Males State of Maryland: Average of FY Percent 80 74% 70 66% % 39% 38% 52% 33% 47% 41% 49% 37% % 24% 20% 10 0 Population Referral Formalization Probation Detention Residential Secure Placement Residential African-American Males White Males Figure 4 Percentage of Population and Representation of African-American Females versus White Females State of Maryland: Average of FY Percent 35 32% % % 12% 9% 6% 6% 6% 6% 4% 7% 8% 5% 2% 0 Population Referral Formalization Probation Detention Residential Secure Placement Residential African-American Females White Females 7

17 MARYLAND DATA: FY Tables 2 through 4 provide data on intake referrals, detention, and secure commitments (secure commitments include youths admitted to Hickey and Cheltenham Young Women facilities only) for FY 1990 through African-American male youths show the highest percent at every decision point for all three years. As mentioned earlier, further analysis by age, race, gender, and offense history is required to measure the extent and significance of variations. Table 2 shows a difference of two to three percent between African-American and white male referrals to the Department. Males accounted for 78% of the total referrals. Table 2 Number & Percent of Intake Referrals: FY FY 1992 ============================================================================== FY 1990 FY 1991 FY 1992 Race/ # Youth % Youth # Youth % Youth # Youth % Youth Gender Referrals Referrals Referrals Referrals Referrals Referrals ============================================================================== African-American % % % Male % % % Female % % % White % % % Male % % % Female % % % State Total % % % Male % % % Female % % % 8

18 Table 3 shows the number and percent of delinquent youths in detention facilities. Table 3 reveals that 70% of the total detained youths were African-Americans. Eighty nine percent (89%) of the total detained youths were male. Table 3 Number & Percent of Delinquent Youths in Detention Facilities: FY FY 1992 ============================================================================== FY 1990 FY 1991 FY 1992 Race/ # % # % # % Gender Detention Detention Detention Detention Detention Detention ============================================================================== African-American % % % Male % % Female % % % White % % % Male % % % Female % % % State Total % % % Male % % % Female % % % 9

19 Table 4 includes the number and percent of youths committed to the secure programs. This table indicates that 80% of the total youths committed to the secure facilities were African-Americans. The proportion of white males remained the same. Males accounted for 92%. Table 4 Number & Percent of Youths in Secure Residential Placements FY FY 1992 ============================================================================== FY 1990 FY 1991 FY 1992 Race/ # Secure % Secure # % # % Gender Commit. Commit. Commit. Commit. Commit. Commit. ============================================================================== African-American % % % Male % % % Female % % % White % % % Male % % % Female 4 0.8% 7 1.3% % State Total % % % Male % % % Female % % % 10

20 NATIONAL STUDIES During the last three decades, numerous research studies have been done on the topic of disproportionate representation in the juvenile justice system. One of the researchers, Carl E. Pope reviewed many of the issues and literature regarding African-American youth crime. Carl E. Pope and William Foyerherm have provided an extensive review of literature in their report titled Minorities and the Juvenile Justice System published by the Office of Juvenile Justice and Delinquency Prevention (OJJDP) in November, A summary of their review follows. The majority of research had compared African-American and white offenders with regard to processing decisions or, in some instances, white versus non-white offenders. The journal, Crime and Delinquency had dedicated two volumes to the issue of selection bias. The first, edited by Roland Chilton and Jim Galvin, published in January 1985, focused on race, crime and criminal justice. The second, edited by Barry Krisberg, published in April 1987, focused on minority youth incarceration and crime. Krisberg and his colleagues (1987) in their analysis of census data (Children in Custody) revealed much higher rates of confinement for African-American males and females and Hispanic males. Utilizing juvenile justice records for the year 1983, Fagan, et al., observed that the type of counsel (public defender) provided to minority youth could explain why more minority youth are adjudicated compared to whites. They also found that minority youth adjudicated for minor offenses were committed to corrections more often than their white counterparts (Fagan, et al., 1987) 2. Some authors have noted the importance of different jurisdictions and organizational settings. For example, Aday (1986) found distinct patterns of court dispositions depending upon whether the juvenile court adopted a more traditional or due process approach 3. Bishop and Frazier (1988) examined the case records of all youths precessed in the state of Florida at five different processing stages for a three year period. They also included race, gender, age, offense seriousness and prior records as variables in their analysis. They found that while legal factors were more pronounced, race also influenced decisions. African-American youth were found to be more at risk of severe sanctions than white youth as they were processed through the system 4. Recent studies conducted in Missouri, California, Michigan, Florida, Pennsylvania, Georgia, and Iowa reveal that there are disproportionate issues in the manner in which the minority youths are processed. All these studies reveal that race differences were the greatest at the early stages of intake and detention 5. Maryland s trend data supports this view and in addition, also confirms that the race differences were the greatest at the secure commitment point, i.e., the deep-end of the system. 2 Fagan, J., Slaughter, S., & Hartstone, E. (1987) Blind Justice: The Impact of Race on the Juvenile Justice Process. Crime and Delinquency,33(2) April: Aday, D.P., Jr (1986) Court Structure, Defense Attorney Use, and Juvenile Court Decisions. The Sociological Quarterly, 27(1): Bishop, D.M., & Frazier, C.E. (1988). The Influence of Race in Juvenile Justice Processing. Journal of Research in Crime and Delinquency, 25(3): Carl E. Pope (1994). Racial Disparities in Juvenile Justice System. Overcrowded Times, Solving the Prison Problem, Volume 5, Number 6, December:

21 12

22 CHAPTER TWO METHODOLOGY The objectives of this study are: 1. Identify the race-gender differences by small areas, from the point of intake to secure commitment. 2. Construct profiles of the characteristics of small areas, which include for each race-gender group the total number and percent of 11 through 17 year old population, the index values from intake to secure commitment, the referral offenses, and age data. 3. Compare the differences in decision points between African-American and white youths, while controlling for current and past offenses, race, gender, and age. 4. Determine priorities for the allocation of resources and provide plans in mobilizing the resources according to need in identified areas. Discussion of the objectives: Objective 1: Disproportionate representation of youth are identified by small areas from the point of intake to secure commitment. Youth are also identified by age and race-gender for specific offenses. The first step in analysis is the computation of ratios which identify the over or underrepresentation. The study uses several measures to calculate the disproportionate representation of youth in the juvenile justice system. Objective 2: A profile of the small areas consist of the following: a. Total Population, ages b. Race-gender index values at each decision making point. c. Age data by small areas. d. Current offenses data by offense level. Objective 3: Objectives 1 and 2 measure aggregate data for each small area by race-gender and referral offenses, which identify the disproportionate representation at various decision points. Objective 3 measures individual data. While controlling for the seriousness of offense, race, gender, and age, each African-American youth is compared to a white youth. For example, an African-American and a white youth, who rank the same on the seriousness of offense scale, are compared to determine if there is a disparity in sanction. Objective 4: Profiles will reveal the ratios of disproportionate representation and determine the priorities. This will enable the Department to plan and mobilize resources towards reducing the disproportionate representation. 13

23 Research Methodology The research methodology involves several steps: A. Acquisition of data at each decision point and population data. B. Determination of small areas. C. Computation of race-gender indices and rates per 1,000 youth population. D. Comparison of indices and rates between small areas. A. Data Acquisition Data are gathered in field offices and facilities and recorded from the Department s computerized system, ISYS. These data sets collected include race, gender, date of birth, location of residence (including ZIP code and county of residence), county and ZIP code of the offense, the sources of referral, the intake decision, alleged and adjudicated offense(s), court disposition, and placement information. Population data for 1990 by geographic area was obtained from the U.S. Bureau of the Census. Population projections for the years 1991 through 1992 were obtained from the Maryland Office of State Planning. B. Determination of Small Areas in the State For rural portions of the state, the small areas are determined by counties. The urban small areas are formulated to adequately account for demographic diversity and referral patterns within the counties. For example, Baltimore City and Baltimore County are divided into six small areas by grouping adjoining ZIP codes. Anne Arundel, Montgomery and Prince George s Counties are each divided into two small areas based on the Department's delineation of a Field Service Area. (See Appendix A). C. Computation of race-gender indices and rates per 1,000 youth population The study includes several measures to confirm the existence and degree of disproportionate representation of African-American youth in the juvenile justice system. They are: 1. Calculation of the minority proportion population index : This study uses two different approaches to calculate the minority proportion index in the juvenile justice system. The first approach is a population-based approach, which is within the mandate prescribed by the Federal Office of Juvenile Justice and Delinquency Prevention (OJJDP). To construct profiles of youth by small areas, population-based index values are calculated for each decision point (Intake, formalization, probation, detention, residential placement and secure commitment). This index value is calculated based on the procedure developed by the OJJDP. An index of 1.0 means that the percentage of intake referrals and the percentage of statewide population for a particular race-gender group is the same. An index that is over 1.0 is considered to be an overrepresentation and an index that is less than 1.0 means underrepresentation. 14

24 An index value consists of a numerator and a denominator. The computation of the numerator involves two steps. For example, to calculate the percentage of African-American males at intake, first, the number of youth referred to intake are averaged for the selected three Fiscal Years (FY ) for each small area. This is shown in Table 5 for State of Maryland, under the column N (10,351 African-American referrals). Next, N is divided by the total number of referrals and multiplied by 100 to get the percentage. See column 2. The denominator is the percentage of African-American male youth in the total population for FY This midpoint of FY 1990 through 1992 is chosen for the calculation, because the population had a steady increase during this period. Refer to Table 5 under Population Percentage column. African- American males accounted for 17% of the state population. An index value is calculated by dividing the percentage of African-American male referrals at intake (39%) by the percentage of African-American male youth population (17%). This calculation yields an index value of 2.22 (39/17) for African-American males at intake. (See population index column, Table 5). At the formalized decision point, the African-American male index value is 3.01, i.e, the percentage of African-American male formalizations divided by the percentage of African-American male youth population (52/17=3.01). Using the 11 through 17 years old population as a constant denominator, this process is repeated for each race-gender group at each decision point. 2. Calculation of the minority disproportion referral index : While the population index measures the disproportionate representation of minority youth in the general population, the referral index measures the disproportionate representation in relation to intake referrals. The Department has no control over law enforcement officers arrest discretion nor any disproportionate representation that exists when the arrest referrals reach intake. Therefore, it is also essential for the Department to measure the ratio of disproportionate representation based on its intake referrals as opposed to general population. To reflect this, the intake referral index is established to determine the degree of selection bias within the juvenile justice system. Referral index is calculated by using the percentage of the intake referrals as the denominator. This index measures the representation at various decision points beyond intake. For example, African- American males account for 39% of the total intake referrals (10351/26785=39%) and 52% of formalizations (5520/10527=52%). Therefore the referral index value for African-American males at the formalization point is 1.36 (52%/39%=1.36), a rate which is 1.36 times higher than their size at the intake level. For white males, this rate is.85 (33%/38%), a rate which is.15 less than their proportion at the intake level. At the detention level, the African-American males' index value is 1.70, a rate which is 1.70 times higher than their rate of representation at intake. White males show an index value of 0.64, a rate which is.36 times less than their rate of representation at intake. (See Table 5, referral index column). 3. Rates per 1,000 youth population: This serves as a supplemental measure to index values, because the index values may be misleading in some cases. For example, in a sparsely populated area, it is possible for the index value to be high and the rate per 1,000 youth population to be low. Rates per 1,000 youths are computed by dividing the number of youth in a particular race-gender group (for example, African-American male) by the number of youth in that race-gender group, and multiplying by 1,000. For example, in Table 5, the referral rate per 1,000 African-American males for the entire 15

25 state is It is the number of African-American male referrals (10,351) divided by the total African-American male population (71826), multiplied by 1, Measures of disproportionate representation after application of statistical controls: Much of the data presented in the report is aggregate data based on youth count at multiple decision points. Individual youth information, such as age, gender, race, offense history including current offenses, and disposition status are analyzed using multiple regression. For example, an African-American and a white youth who scored the same on the 'Seriousness of Offense' scale are compared to determine if there is a disparity in the sanctions. To control for offenses, offense history scores including the current and past alleged offense(s), are computed by adding the scores of the total offenses committed by each youth. Offenses are grouped into twelve categories and ranked from the least to the most serious from one to nine (see Appendix C). The scores assigned to each category are in keeping with the Uniform Crime Index (UCD). Examples of the severity ratings include: violent person-to-person-type 1 (9 score), which includes murder, manslaughter, sex first and second degrees, rape, and arson. Violent person-to-person - Type 2 (8 score) which includes aggravated assault and assault and/or robbery with a deadly weapon. Court disposition variables are coded from low to high on a scale from one to four. They are dismissed and closed offenders (1) probation (2), residential/secure commitments (3), and jurisdiction waivers (4). The age variable is also ranked from low to high, with an age of less than 10 being given a score of 1 and age 17, a score of Grouping of offense history scores: This is another procedure used to test whether there is a statistically significant relationship between offense history scores and court dispositions to show which group received the more severe sanctions. Offense history scores are grouped according to intervals of 10 (for example 1-10, 11-20,... up to and 70 and above). For example, all African-American and white male youths who scored less than 10 on their offense history scores are compared to their court disposition status. For each interval, a Chi Square Test is applied to test the significance of court dispositions to offense scores at P <.01 level. The significance at P<.01 level indicates that there is a one in one hundred probability that the difference in disposition status between African-American and white youth could have occurred by chance alone. 16

26 TABLE 5 The Disproportionate Representation of African-American Youth in the Maryland Juvenile Justice System State of Maryland: Average of FY Population Referral Rate Per Number Percent Index Index 1,000 Pop. POPULATION: White Male % White Female % African-American Male % African-American Female % Total % INTAKE REFERRALS White Male % White Female % African-American Male % African-American Female % Total % FORMALIZATION White Male % White Female 643 6% African-American Male % African-American Female 933 9% Total % PROBATION White Male % White Female 198 6% African-American Male % African-American Female 224 6% Total % DETENTION White Male % White Female 99 4% African-American Male % African-American Female 177 6% Total % RESIDENTIAL White Male % White Female 94 8% African-American Male % African-American Female 78 7% Total % SECURE COMMITMENT White Male % White Female 11 2% African-American Male % African-American Female 28 5% Total %

27 18

28 CHAPTER THREE SECTION I: DATA ANALYSIS Format Description This Chapter is divided into two sections. The first section analyzes the data using the race-gender indices described in the methodology chapter. Data are grouped by small areas. Population and Referral indices are given for the small areas and state. Although two indices are used to measure the disproportionate representation by race-gender and small areas, this study analyzes the results of the data using the population-based approach only, as mandated by the Federal Office of Juvenile Justice and Delinquency Prevention. The results of the referral indices are provided in Appendix D. Wherever necessary a discussion is provided for referral index values. Each table in Appendix D also contains information on rates per 1,000 youth population and age data for 11 through 17 years. The second section of this chapter discusses the results of multiple regression analysis and Chi Square test. This study was primarily conducted to determine, whether or not, there is disproportionate representation after controlling for race, gender, age, and offense history. Individual youth information are used to perform a multiple regression analysis. While analyzing the data, an important consideration is given to the magnitude of the problem in terms of the actual number of youths who entered each decision point in some rural areas. This is evident when examining the actual number of youths in certain small areas where the absolute number is very small. For example, the African-American population is very small in Western Maryland and Eastern Shore areas. A high index value in these counties can be due to the relatively small population. Such small numbers, youth population count of less than five (5) at any decision point should not be considered significant or valid for this analysis. Therefore, the rate per 1,000 population measure is used to verify the magnitude of the significance. It is possible to have a high rate of index value when the numbers are small, but the rate per 1,000 population can be relatively low. The following tables show the results of disproportionate representation in terms of race-gender indices at each decision point. As mentioned earlier, for each race-gender, an index value that exceeds 1.0 indicates overrepresentation and an index value under 1.0 indicates underrepresentation. A neutral index value is 1.0, i.e., the population rate and the rate at the decision point are the same. An asterisk next to the index value indicates that the disproportionate representation is not considered significant due to small numbers in that sub-group. 19

29 SUMMARY OF RESULTS: STATEWIDE DATA African-American male youth were consistently overrepresented at each decision point. Their disproportionate representation was pronounced at intake, detention, and secure commitment decision points. Table 6 Population Index Values: State of Maryland Average of FY ========================================================================= White African-American White African-American Decision Point Male Male Female Female ========================================================================= Referral Formalization Probation Detention Residential Placement Secure Commitment Table 6 shows that African-American male youths were overrepresented at each decision point in the juvenile justice system. Their representation at intake was two times more than their representation in the state population. At the secure commitment stage, the deep end of the system, African-American male youth were overrepresented 4.23 times their rate of representation in the state population. The difference in index values between the intake and secure commitment points was 2.01 ( = 2.01). This means that the juvenile justice system accounted for this share of disproportionate representation between these two decision points. The disproportionate representation of African-American male youth was greater at the intake, detention, and secure commitment points than at the formal, probation and residential placement points. Intake and detention are the earlier decision points and the secure commitment is the last point in the juvenile justice system. White males were underrepresented at the following decision points: formalization (0.96), detention (0.72), and secure commitments (0.59). At intake, probation, and residential placement decision points, the indices were 1.13, 1.21, and 1.09, respectively. This shows that white males are less likely to go through the severe sanctions, and if adjudicated, are most likely to receive probation or commitments to residential placement. Female youth were underrepresented at all decision points and their indices were well below their rate of representation in the state population. Their rate of representation was less than males, although African-American females showed a higher value than white females. The referral index showed disproportionate representation for African-American males at each decision point and underrepresentation for all other race-gender groups. The referral index for African-American males at the formalization point was 1.36 and rose to 1.91 at secure commitment. This means that at the formalization point, the African-American males were represented at a rate which was.36 times higher 20

30 than their intake referrals, and at secure commitment their representation was.91 times higher than their intake referrals. The intake rate per 1,000 youth population was 144 for African-American males, 74 for white males, 24 for white females, and 43 for African-American females. The state average intake rate was 65 per 1,000 youth population. For details of rates per 1,000 youth population at other decision points see Appendix D. Figure 4 shows the comparison of index values for white males and African-American males. Figure 5 compares the index values of white females with African-American females. Figure 5 Comparison of Index Values Between African-American and White Males at Various Decision Points State of Maryland Index Values Intake Formalization Probation Detention Residential Secure Plac ement Commitment African-American Males White Males > 1.0 means overrepresentation < 1.0 means underrepresentation 21