Disproportionate Minority Contact in Indiana: Quantitative Analyses Final Report

Transcription

1 Disproportionate Minority Contact in Indiana: Quantitative Analyses Final Report Russell J. Skiba, Ph.D. Karega Rausch Daniel Abbott Ada Simmons, Ph.D. Submitted to the Indiana Criminal Justice Institute Title II Grant, June 30, 2007

2 Disproportionate Minority Contact: Quantitative Analyses Final Report Amendments to the Juvenile Justice and Delinquency Prevention Act (JJDP) in 2002 dramatically altered how states identify and address ethnic disproportionality in juvenile justice systems. States are now required to address not only the disproportionate confinement of youth of color, but also disproportionality in contact points throughout the juvenile justice system. The purpose of this report is to describe the extent of disproportionate minority contact (DMC) in Indiana in order to meet this federal requirement and to help inform local and state policymaking. It will begin by describing the current DMC federal mandate, and then briefly review the limited national and state data on the extent of DMC. Recent data from Indiana will describe the extent of DMC from seven counties at eight key decision points. Results indicate that overrepresentation of Black youth at arrest makes a strong contribution to DMC in Indiana, with further disproportionate contact at several other points of contact. Federal Involvement in DMC: The Juvenile Justice Delinquency and Prevention Act Federal law has required states to attend to ethnic disproportionality in juvenile justice systems since at least The Juvenile Justice and Delinquency Prevention Act of 1974, as amended in 1988, required states participating in the JJDP formula grants program to address efforts at reducing the disproportionate confinement of minority youth. Since that time, the Act has been modified by making disproportionality one of four core requirements and ties 25 percent of states formula grants allocation to their compliance plans. Further, the 2002 amendments require states to address the disproportionate number of youth from racial/ethnic backgrounds that come into contact with the juvenile justice system more generally, without setting numerical standards or quotas. Addressing DMC occurs through four interrelated phases: 2 First, states must determine if DMC is evident at the state or local county level by identifying and reporting the extent of ethnic disproportionality at multiple contact points of their juvenile justice systems. States are to provide data disaggregated by race and ethnicity, on nine decision points: arrest, juvenile court referral, pre-adjudication detention, pre-petition diversion, petition, transfer/waiver to adult court, adjudicated delinquent, post-adjudicated probation, and post-adjudicated secure confinement. 3 Figure 1 illustrates how these nine contact points fit together, illustrating flow through the juvenile justice system. If identification data reveal DMC to be evident, states must examine how it occurs by assessing and diagnosing underlying factors believed to contribute to DMC. Data from the identification and assessment phase are then used to develop and implement state and local intervention programs. Finally, states monitor and evaluate the effectiveness of intervention activities, and revise their plans and activities as necessary.

3 National and State Data on DMC Since tracking of DMC at multiple decision points has been a federal requirement only since 2002, national data documenting the extent of DMC at each decision point and for each racial group are not extensive. Of the limited available data, national estimates of DMC in the juvenile justice system have consistently found minority overrepresentation, especially for African American youth. 4 Available national studies using data from the 1990 s and for the most recent available data in 2004 find that while making up about 15% of the juvenile population, African American youth have been overrepresented in the composition of arrests (26-28%), referrals to juvenile court (30-31%), detentions (37-45%), cases petitioned (33-37%), adjudications (32-36%), post-adjudication placement out-of-home including residential facilities (36-46%) and waivers to adult court jurisdiction (42-52%). National data on the extent of DMC for Latino youth are scarce and inadequate to date, because reported data do not typically separate counts for Latino and White youth and do not typically distinguish between race and ethnicity. 5 State reports generally find DMC to be widespread and not confined to any region of the country. 6 In a comprehensive review of state reports and compliance plans during the 1980 s and 1990 s, Michael Leiber found that among reports from 43 states and the District of Columbia, minority overrepresentation was evident in every state and region of the country. Further, while the decision point where overrepresentation was greatest varied by state, overrepresentation existed at all decision points included in state reports. Individual state reports on DMC tend to find overrepresentation for both African American and Latino youth, but African American youth overrepresentation tended to be much more consistent and pronounced. 7 For example, Leiber reports that for the detention and secure correction decision points, composition percentages for African American youth are on average about 3 times higher than African American representation in the youth population. Overrepresentation of Hispanic youth was evident, but smaller (i.e., detention 1.4 times higher than their population representation and 1.9 for secure corrections). In one of the few reports specifically devoted to Hispanic DMC, Villarruel and Walker report that in one large county (Los Angeles) and in some states, Hispanic youth are overrepresented in arrests, pre-trial detention, detention centers, and waivers to adult court. 8 Data on DMC in Indiana Data on the extent of DMC in Indiana are very limited, but suggest that Indiana is not immune from DMC. 9 The most recent published data are for the 2001 reporting year, in which data were available for five of the nine decision points. Compared to White youth, African American youth in Indiana were found to be overrepresented in arrests (3.47 times more likely), secure detention (1.40), and confinement in secure juvenile facilities (1.66), and underrepresented in cases diverted (0.33 times as likely). African American youth were found to be roughly proportionately petitioned. No data for Hispanic youth were reported.

4 Summary In summary, the limited national and state evidence on DMC suggests that: Minority groups, African Americans in particular, are typically found throughout the juvenile justice system at rates 2-3 times higher than their representation in the general population. DMC is evident in every state and region, and at each decision point studied to date. Evidence is very limited on the extent of DMC for Hispanic youth, but suggests that they may be overrepresented at some decision points and in some states. Data on DMC in Indiana are very limited. In 2001, African American youth in Indiana were found to be more likely than White youth to be arrested, detained, and confined, less likely to be diverted, and had about an equal chance of being petitioned. Methodological and Measurement Advances in DMC Research Nationally and in Indiana Recent methodological and measurement advances in the study of DMC provide more precision in understanding the extent of DMC, yet also make comparisons with prior data difficult. Prior to 2003, much of the research documenting the extent of DMC used methods no longer recommended by the OJJDP or equity researchers in other areas such as education. 10 Virtually all prior research used the Disproportionate Representation Index (DRI) to measure disproportionality. The DRI compares a racial group s composition at a decision point (e.g. adjudicated delinquent) to the composition of that racial group s representation in the general youth population. This index has subsequently been found to be a poor measure of DMC, primarily because it is highly dependent and affected by the relative size of the minority youth population. By comparing rates at decision points to rates in the general population, the DRI also does not account for disproportionality in previous decision points. 11 Given these significant shortcomings, OJJDP now recommends that states use risk indices in order to calculate the Relative Rate Index (RRI), a measure identical to the relative risk ratio found in other areas of equity research. 12 The RRI allows for a more direct comparison between racial groups (instead of to the racial group s representation in the population), and allows for an examination of DMC at individual decision points after accounting for DMC at previous decision points (see Methodology section for a more detailed description of how the risk index and RRI are calculated). In Indiana, recent technological advances make the study of DMC more tenable than in previous years. As cited in Indiana s most recent federal report to OJJDP, 13 in most counties juvenile records continue to be maintained in non-standardized and non-uniform ways, and without a central repository for data collection and analysis. Complications associated with data collections that rely on a variety of unrelated sources whose primary function is not data collection can introduce undesirable inconsistencies in data collection

5 protocols and definitions of terms across sites. However, some of the larger counties in Indiana, including those with the highest concentrations of youth of color, use the QUEST data management system, a comprehensive computerized case management system. This system contains data on virtually all elements of a juvenile s case, including demographic information such as race/ethnicity, and decision outcomes throughout the juvenile justice system, thus creating a unified data storehouse amenable to analysis of DMC by ethnic group at each critical decision point. Datasets Methodology Data for the analysis of DMC in Indiana are drawn from two sources: the QUEST data management system and the United States Department of the Census. QUEST Data. Data for the movement of youth through the juvenile justice process were drawn from the QUEST database. QUEST is a data management system operated and maintained by Gottlieb & Wertz, Inc. that contains real-time, comprehensive information for each juvenile contacting the juvenile justice system. Gottlieb & Wertz indicated that seven counties using QUEST had sufficient data for analysis. These counties were: Allen, Howard, Lake, Madison, Marion, Tippecanoe, and St. Joseph. Each of the seven counties gave written permission to the Indiana Criminal Justice Institute for Gottlieb & Wertz to extract their data from QUEST for an analysis of DMC. Gottlieb & Wertz extracted these data and put them in a uniform database for analysis. These extracted data contained incident level information for each contact with the juvenile justice system for the 2005 calendar year. Each contact record included information about the juvenile (e.g. race/ethnicity), and outcomes of decision points throughout the system, described below: Arrest: Extracted records contained information on the type of incident for each contact record. Only records noted as arrests (or closely affiliated terms 14 ), were included in the analysis. Detention at Intake: Counts include incidents detained at the initial intake phase, and do not include detentions as a result of disposition decisions. Diversion: Pre-petition diversions or informal adjustments (i.e., counted only if the incident was diverted or informally adjusted, and the incident was not petitioned). Petitioned: Cases in which one or more formal charges were filed against the juvenile, post-arrest. Transfer/Waiver: Cases in which the juvenile court waived its jurisdiction and transferred the incident to adult court, post-arrest. Adjudicated: Cases in which youth were adjudicated delinquent. Probation: Post-adjudicated delinquent cases in which the disposition was probation. Department of Corrections: Post-adjudicated delinquent incidents in which the disposition was a commitment to the department of corrections.

6 Census Data. Census data 15 were used to estimate county-level youth populations in Indiana disaggregated by race and Hispanic origin. Data were drawn from the Population Estimates Program, July 1, 2005 estimate ages Data from seven Indiana counties were used in this analysis: Allen, Howard, Lake, Madison, Marion, Tippecanoe, and St. Joseph. Collectively, these counties account for approximately 37.5% of the estimated youth population in Indiana, and 70.1% of the minority youth population. Racial/ethnic groups used were Black (Not of Hispanic Origin), White (Not of Hispanic Origin), Asian or Pacific Islander (Not of Hispanic Origin), Native American/Indian (Not of Hispanic Origin), Hispanic (Of any single race), and Other or Unknown (which includes youth of multiple races). Data were downloaded from the U.S. Department of the Census website. All data received for the 2005 calendar year were included in the analysis, including cases still in process (cases opened in 2005 but without a final resolution in 2005). Further, consistent with guidance given from OJJDP, this analysis is an incident-level rather than individual level analysis. An incident level analysis means that individual youth contacting the system multiple times are counted for each time they contact the system (duplicated count). Thus, these data represent incident rates (i.e., traffic flow ) through the system rather than individual youth rates. The analysis assumes relatively consistent definitions for each data category across the seven counties and in recording a youth s racial/ethnic group identification. To the degree there is wide divergence in how counties define racial groups and decision points, the aggregated data across all seven counties may not represent an accurate profile. Finally, since these are data from a single year, they represent a snapshot of the process in a single year, rather than an estimate of overall disproportionality in the system. Thus, a finding of no evidence of DMC at a given contact point in these data (e.g., incarceration) does not mean that there is no disproportionality at that point, only that there was no disproportionality added by that point in the process for the year measured. Data Analysis Measures and Methods The data reported represent DMC estimates for each decision point individually, and thus allows for stage specific comparisons after accounting for DMC at the previous step. For example, DMC calculations at the adjudicated delinquent contact point are based on the proportion of youth found delinquent from the proportion of youth petitioned. Measures Multiple indicators are used to quantify DMC, each of which is described below: Risk Index: This measure is the proportion of a given group in a decision point and represents the best estimate of the group s risk for that outcome. Relative Rate Index: In order to interpret the risk index, a ratio of the risk of the target group to one or more groups is constructed. A ratio of one indicates exact proportionality, while ratios above or below 1.0 indicate over- and underrepresentation, respectively. For example, if we compared the African American

7 risk for arrest (e.g. 5%) with the risk index for all other youth (e.g. 1%), these data would yield a RRI of 5.0 (5/1), suggesting that African Americans are five times more likely to be arrested than other youth. For this analysis, the comparison group is all other youth (e.g., African American youth compared to all non-african American youth). Chi-Square Significance Level: Chi-Square is an inferential statistic that tests the degree to which contact at each decision point is dependent on race/ethnicity. The significance level describes the probability that the results are due to chance. For example, if DMC for African Americans at arrest is significant at the p <.001 level, this means that the chances of DMC at this magnitude occurring by chance are less than 1 in 1,000 when there is actually no difference in the youth population. 16 Results Disproportionality data for each of the eight steps in the process, aggregated across all seven counties, are presented in Table 1. Notes on Interpretation. Columns 1 and 2 (Population) represent the total population count and percentage at each step for each racial/ethnic category (note that population count will change: for those arrested, the population is the total population in the counties and for subsequent steps, it is the count of youth from the previous step, such as those who were arrested or, as in the case of probation, those who had been adjudicated). Columns 3 and 4 represent the total number and percentage of each racial/ethnic group coming into contact with that step. Columns 5 and 6 represent the risk index (RI, percent of the group involved at that step) for the target race and for all other races. Note that since all other is being used as the comparison, the size of the RI will change for each target race. The RI for total at each step provides an overall assessment of what percentage of students in total have come into contact with the system at that point. Thus, for Arrest, the Total indicates that, across the seven counties, 5.23% of juveniles were arrested in the data. The clearest indication of disproportionality for each racial/ethnic group is found in the final two columns, which present the relative rate index (RRI) and results of the Chisquare significance test. The RRI is a ratio where the nuerator is the RI for the Target Race and the RI for all other groups combined is the denominator. Thus, across all seven counties, 10.03% of Black juveniles have been arrested, but only 3.8% of youth in all other groups combined. Dividing the first RI by the second yields a relative rate index of 2.64, indicating that Black youth are over 2.5 times as likely to be arrested as other youth. The final column represents the significance of a chi-square test on the data that provides something of a control for sample size, especially when there are small numbers of incidents. Two levels of determination of significance are denoted by highlighting in the table. An RRI greater than 1.3 plus a significant chi-square is bolded as a trend toward significant overrepresentation that may warrant attention, while an RRI of.77 or less plus a

8 significant chi-square is bolded as a trend towards under-representation. An RRI of 1.5 or greater and a significant chi-square is bolded and highlighted in yellow as a significant level of overrepresentation (RRI=.67 or less as an indication of significant underrepresentation). Note that only over- or under-representation in the expected direction is highlighted: Thus the under-representation in arrests of virtually all groups other than Black and Other is not noted. The analyses thus looked for evidence of overrepresentation in the areas of Arrest, Detention at Intake, Petitioned, Adjudicated, Transfer/Waived to Adult Court, and Secure Confinement. Only evidence of underrepresentation was highlighted in the steps of Diversion and Probation. Regardless of the absolute size of the RRI, data with less than 10 individuals in the cell (e.g., 10 Hispanic students arrested in a given county) are disregarded, since the RRI is an unstable estimate for numbers smaller than 10. Results across All Seven Counties The results from Table 1 indicate that, when the data from all seven counties are aggregated, Black youth are significantly over-represented at the stage of Arrest (RRI = 2.64) and Transfer/Waived to Adult Court (RRI = 2.06). They are significantly underrepresented in the step of Diversion (RRI = 0.52). There are also trends in a number of other steps for Black youth, including a trend toward over-representation at the stage of Petitioned/Charged (RRI = 1.39) and a trend toward under-representation in Probation (RRI = 0.75). There were few instances of disproportionality found for any other racial/ethnic group. Youth in the Other/Unknown racial category tended toward over-representation in Arrests (RRI = 1.47) and in Secure Confinement (RRI = 2.03), and tended toward underrepresentation in Diversion (RRI = 0.67). No other race was found to have evidence of disproportionality in the expected direction at any of the stages of contact with the juvenile justice system.

9 Table 1: Disproportionate Minority Contact in Indiana: Summary of All 7 Counties Population Contact Risk Index Relative Rate Index Count % of Total Count % of Total Target Race All Other Target Race / All Other Arrest (B) Native American 1, % % 1.85% 5.23% 0.35 p<.001 Asian or Pacific Islander 6, % % 1.10% 5.28% 0.21 p<.001 Hispanic 43, % 1, % 3.47% 5.39% 0.64 p<.001 Black, Not of Hispanic Orgin 115, % 11, % 10.03% 3.80% 2.64 p<.001 White, Not of Hispanic Orgin 324, % 12, % 3.75% 7.88% 0.48 p<.001 Other/Unknown 13, % 1, % 7.61% 5.16% 1.47 p<.001 Total 504, % 26, % 5.23% Detention at Intake (E) Native American % % 21.74% 30.77% 0.71 ns Asian or Pacific Islander % % 31.34% 30.76% 1.02 ns Hispanic 1, % % 33.98% 30.57% 1.11 p<.01 Black, Not of Hispanic Orgin 11, % 3, % 34.25% 27.78% 1.23 p<.001 White, Not of Hispanic Orgin 10, % 2, % 26.49% 34.12% 0.78 p<.001 Other/Unknown % % 33.30% 30.66% 1.09 ns Total * 24, % 7, % 30.76% Diversion (D) Native American % % 13.04% 8.40% 1.55 ns Asian or Pacific Islander % % 22.37% 8.37% 2.67 p<.001 Hispanic 1, % % 6.84% 8.50% 0.80 p<.05 Black, Not of Hispanic Orgin 11, % % 5.54% 10.66% 0.52 p<.001 White, Not of Hispanic Orgin 12, % 1, % 11.46% 5.79% 1.98 p<.001 Other/Unknown 1, % % 5.68% 8.51% 0.67 p<.01 Total 26, % 2, % 8.41% Petitioned (Charged) (F) Native American % % 52.17% 65.33% 0.80 ns Asian or Pacific Islander % % 38.16% 65.40% 0.58 p<.001 Hispanic 1, % % 61.55% 65.55% 0.94 p<.01 Black, Not of Hispanic Orgin 11, % 8, % 77.49% 55.78% 1.39 p<.001 White, Not of Hispanic Orgin 12, % 6, % 54.45% 74.65% 0.73 p<.001 Other/Unknown 1, % % 64.74% 65.34% 0.99 ns Total 26, % 17, % 65.32% Adjudicated (Guilty) (G) Native American % % 66.67% 72.10% 0.92 ns Asian or Pacific Islander % % 68.97% 72.10% 0.96 ns Hispanic % % 62.14% 72.66% 0.86 p<.001 Black, Not of Hispanic Orgin 8, % 6, % 73.14% 70.95% 1.03 p<.01 White, Not of Hispanic Orgin 6, % 4, % 71.85% 72.24% 0.99 ns Other/Unknown % % 74.46% 72.00% 1.03 ns Total 17, % 12, % 72.09% Transfer/Waived to Adult Court (J) Native American % % 0.00% 0.87% 0.00 ns Asian or Pacific Islander % % 0.00% 0.87% 0.00 ns Hispanic 1, % % 0.27% 0.91% 0.29 p<.01 Black, Not of Hispanic Orgin 11, % % 1.22% 0.59% 2.06 p<.001 White, Not of Hispanic Orgin 12, % % 0.67% 1.04% 0.65 p<.01 Other/Unknown 1, % % 0.20% 0.90% 0.22 p<.05 Total 26, % % 0.87% Probation (H) Native American % % 75.00% 48.20% 1.56 ns Asian or Pacific Islander % % 65.00% 48.19% 1.35 ns Hispanic % % 57.99% 47.74% 1.21 p<.001 Black, Not of Hispanic Orgin 6, % 2, % 41.62% 55.62% 0.75 p<.001 White, Not of Hispanic Orgin 4, % 2, % 56.21% 43.24% 1.30 p<.001 Other/Unknown % % 46.28% 48.30% 0.96 ns Total 12, % 5, % 48.22% Secure Confinement (DOC) (I) Native American % % 0.00% 5.41% 0.00 ns Asian or Pacific Islander % % 0.00% 5.42% 0.00 ns Hispanic % % 3.82% 5.49% 0.70 ns Black, Not of Hispanic Orgin 6, % % 5.90% 4.85% 1.22 p<.01 White, Not of Hispanic Orgin 4, % % 4.43% 6.02% 0.74 p<.001 Other/Unknown % % 10.54% 5.20% 2.03 p<.001 Total 12, % % 5.41% Note: * Total based on six counties because Tippecanoe does not collect data for Detention at Intake Contact Point. Chi-square Significance

10 County-by-County Results. Tables for each of the seven participating counties, formatted in the same manner as Table 1, may be found in Appendix A. Significant and likely disproportionality are noted for each county (Appendix Tables A1 to A7) in the same manner as they are in Table 1. Table 2 presents a summary of county-by-county data for Black and Hispanic youth, in order to illustrate county level variation in disproportionality across the 8 stages. At arrest, there is a high level of consistent disproportionality in the data: Black youth are over-represented in arrests in all seven counties. At other stages of the process, however, there is some variability across counties. At Diversion, three out of seven counties (Madison, Marion, and Tippencanoe) showed under-representation of Black youth. Black juveniles were significantly more likely to be Transferred/Waived to Adult Court in Lake County and Marion County. Finally, Black youth showed a tendency toward more frequent detention at intake than other youth in Madison County, and tended toward over-representation at the charging stage in Howard County. Finally, Hispanic youth were significantly more likely to be placed in Secure Confinement in St. Joseph County, over-represented in Petitioned/Charged in Tippecanoe County, and under-represented in Probation in Allen County.

11 Table 2. Summary of Disproportionate Minority Contact in Indiana Arrest (B) Hispanic African American (Hispanic) (All Other) Rate Index (AA) (All Other) Rate Index Allen 3.24% 4.12% % 2.86% 3.80 Howard 8.16% 6.59% % 5.75% 2.66 Lake 2.64% 3.29% % 2.47% 1.96 Madison 7.27% 8.53% % 7.24% 2.74 Marion 3.75% 6.19% % 3.62% 3.14 St. Joseph 4.75% 6.26% % 4.73% 2.93 Tippecanoe 4.75% 5.24% % 4.66% 4.01 Total 3.47% 5.39% % 3.80% 2.64 Detention at Intake (E) Hispanic African American (Hispanic) (All Other) Rate Index (AA) (All Other) Rate Index Allen 55.90% 53.48% % 48.89% 1.24 Howard 21.74% 16.43% % 16.32% 1.09 Lake 35.29% 37.82% % 33.33% 1.27 Madison 2.04% 6.93% % 6.14% 1.48 Marion 33.74% 30.91% % 30.68% 1.02 St. Joseph 25.11% 22.82% % 20.92% 1.28 Tippecanoe Total * 33.98% 30.57% % 27.78% 1.23 Note: "-" Indicates data not available. Note: " " represents division by zero. Note: * Total based on six counties because Tippecanoe does not collect data for Detention at Intake Contact Point. Diversion (D) Hispanic African American (Hispanic) (All Other) Rate Index (AA) (All Other) Rate Index Allen 31.06% 34.95% % 37.57% 0.81 Howard 8.70% 15.36% % 15.45% 0.90 Lake 0.39% 0.31% % 0.42% 0.46 Madison 14.29% 7.35% % 8.67% 0.42 Marion 2.71% 1.99% % 2.58% 0.63 St. Joseph 1.79% 5.74% % 5.93% 0.79 Tippecanoe 22.52% 19.80% % 21.03% 0.64 Total 6.84% 8.50% % 10.66% 0.52 Petitioned (Charged) (F) Hispanic African American (Hispanic) (All Other) Rate Index (AA) (All Other) Rate Index Allen 39.13% 38.61% % 36.88% 1.12 Howard 13.04% 23.00% % 20.78% 1.41 Lake 81.76% 81.40% % 82.07% 0.98 Madison 38.78% 52.24% % 51.50% 1.04 Marion 71.92% 91.36% % 81.45% 1.19 St. Joseph 38.57% 34.53% % 33.11% 1.14 Tippecanoe 39.64% 30.39% % 30.82% 1.04 Total 61.55% 65.55% % 55.78% 1.39

12 Table 2. Summary of Disproportionate Minority Contact in Indiana (continued) Adjudicated (Guilty) (G) Hispanic African American (Hispanic) (All Other) Rate Index (AA) (All Other) Rate Index Allen 80.95% 72.61% % 74.86% 0.94 Howard 83.33% 84.17% % 84.29% 0.99 Lake 52.76% 57.68% % 53.60% 1.14 Madison 73.68% 65.31% % 65.86% 0.97 Marion 65.41% 76.38% % 76.36% 0.99 St. Joseph 77.91% 79.85% % 81.34% 0.95 Tippecanoe 63.64% 65.35% % 66.44% 0.87 Total 62.14% 72.66% % 70.95% 1.03 Transfer/Waived to Adult Court (J) Hispanic African American (Hispanic) (All Other) Rate Index (AA) (All Other) Rate Index Allen 0.00% 0.07% % 0.00% Howard 0.00% 0.09% % 0.00% Lake 0.00% 0.34% % 0.11% 4.90 Madison 0.00% 0.05% % 0.06% 0.00 Marion 0.00% 1.42% % 0.71% 2.59 St. Joseph 0.00% 0.14% % 0.00% Tippecanoe 3.60% 3.53% % 3.63% 0.81 Total 0.27% 0.91% % 0.59% 2.06 Note: " " represents division by zero. Probation (H) Hispanic African American (Hispanic) (All Other) Rate Index (AA) (All Other) Rate Index Allen 47.06% 62.75% % 61.83% 1.00 Howard 80.00% 73.85% % 73.91% 1.00 Lake 66.36% 60.01% % 61.63% 0.97 Madison 57.14% 74.12% % 73.04% 1.04 Marion 45.55% 36.01% % 41.12% 0.81 St. Joseph 61.19% 67.51% % 68.99% 0.92 Tippecanoe 85.71% 90.16% % 92.31% 0.79 Total 57.99% 47.74% % 55.62% 0.75 Secure Confinement (DOC) (I) Hispanic African American (Hispanic) (All Other) Rate Index (AA) (All Other) Rate Index Allen 0.00% 1.30% % 0.95% 1.70 Howard 0.00% 2.75% % 3.11% 0.52 Lake 0.45% 1.60% % 1.08% 1.70 Madison 0.00% 4.34% % 4.52% 0.75 Marion 4.71% 6.62% % 6.36% 1.05 St. Joseph 17.91% 9.15% % 9.15% 1.17 Tippecanoe 0.00% 0.63% % 0.33% 6.80 Total 3.82% 5.49% % 4.85% 1.22 Note: "-" Indicates data not available. Note: " " represents division by zero.

13 Summary and Discussion In order to provide data for Indiana s reporting requirements under the Juvenile Justice Delinquency and Prevention Act, an analysis of data from the QUEST database was conducted. The purpose of the analysis was to identify levels of disproportionate minority contact at each of the eight contact points in the Indiana Juvenile Justice system. Since the estimates at each point of contact are based on those in the system at the previous phase (e.g., estimates of Detention at Intake are drawn from the pool of those who have been arrested), these data might be regarded as an estimate of the unique contribution to DMC of each stage. Clearly, the most significant contribution to DMC in Indiana in the data is at the stage of Arrest. Black youth were 2.64 times as likely as all other youth to be arrested, and this significant finding held across all seven counties in the database. Black youth were also over-represented in Transfer/Waived to Adult Court and Petitioned/Charged. Evidence of under-representation of Black youth was found at the points of contact of Diversion and Probation. Aside from the category Other/Unknown, there was no no evidence of disproportionality at the summary level for any other racial/ethnic group. County by county results were consistent with these findings, although there was some variability across counties. Over-representation in Arrests was found in all seven of the counties. Between one and three counties were found to have Black over-representation at the steps of Transfer/Waived to Adult Court and Detention. A number of counties also showed under-representation of Black youth at the stage of Diversion. Hispanic youth were found to be disproportionately represented in secure confinement, petitioned/charged, and probation in one county each. It is important to bear in mind that these results represent analyses of the process for a single year. Failure to find evidence of DMC at a given point does not mean that there is proportionality in who makes contact with the juvenile justice system in general, and ultimately in the composition of disposition outcomes. For example, finding a lack of disproportionate minority contact at the point of Secure Confinement in no way means that there is not disproportionality in the juvenile detention population of the state of Indiana. Rather, these analyses identify which points of contact show evidence of DMC, thus contributing to racial/ethnic disproportionality in juvenile justice in the state of Indiana. It is likely that the contributions made to DMC at the various points of contact add up over time to account for the current disproportionality in incarceration for youth of color in the state of Indiana. The current analyses show that DMC at Arrest is clearly a key contributor to overall disproportionality of Black youth in the juvenile justice system, and that a number of other points of contact appear to add to that initial disproportionality. Why disproportionality exists at the various points of contact, or why there is county-bycounty variability in DMC remains to be addressed in further analyses.

14 Endnotes 1 Office of Juvenile Justice and Delinquency Prevention (OJJDP) (2006). Disproportionate minority contact technical assistance manual (3 rd edition). Washington, D.C.: U.S. Department of Justice. 2 Leiber, M.J. (2002). Disproportionate minority confinement (DMC) of youth: An analysis of state and federal efforts to address the issue. Crime & Delinquency, 48, 3-45; OJJDP, Results are only reported for 8 of the 9 different decision points. Data provided by the QUEST data management system (see methods section) did not include variables enabling a distinction between the arrest (B) and juvenile court referral (C) decision points. Further, guidance given by local practitioners indicated that decisions made at the juvenile court referral stage actually occur during an intake process shortly after arrest. Such guidance also suggested that youth who were arrested virtually always went through the intake process and thus, in Indiana it may be better to conceptualize the arrest and juvenile court referral (intake) as one step. 4 Much of the literature used in this report comes from the following sources: Bishop, D., & Frazier, C. (1988). The influence of race in juvenile justice processing. Journal of Research in Crime and Delinquency, 22, ; Hawkins, D.F., & Kempf-Leonard, K. (Eds.) (2005).Our children, their children: Confronting racial and ethnic differences in American juvenile justice. Chicago: University of Chicago Press; Kempf-Leonard, K., Pope, C.E., & Feyerherm, W.H. (Eds.) (1995). Minorities in juvenile justice. Thousand Oaks: Sage Publications; National Research Council and Institute of Medicine (2001). Juvenile crime, juvenile justice (chapter 6). Joan McCord, Cathy Spatz Widom, & Nancy Crowell (Eds). Washington, D.C.: National Academy Press; OJJDP (2001). Disproportionate minority confinement: A review of the research literature from 1989 through Washington, D.C.: U.S. Department of Justice; Pope, C.E., & Feyerherm, W. (1992). Minorities in the juvenile justice system: Full report. Rockville, MD: U.S. Dapartment of Justice; Poe-Yamagata, E. & Jones, M.A. (2000). And justice for some. Washington, D.C.: Building Blocks for Youth; Snyder, H.N. & Sickmund, M. (1999) Juvenile Offenders and Victims: 1999 National Report. Washington, DC: U.S. Department of Justice; Stahl, A.L., Sickmund, M., Finnegan, T.A., Snyder, H.N., Poole, R.S., & Tierney, N. (1999). Juvenile Court Statistics Washington, D.C.: Office of Juvenile Justice and Delinquency Prevention; Stahl, A., Finnegan, T., and Kang, W. (2007). Easy access to juvenile court statistics: Online. Retrieved January 4, 2007 at: 5 Villarruel, F. & Walker, N.(2002). Dónde está la justicia? A call to action on behalf of Latino and Latina youth in the U.S. justice system. Washington, D.C.: Building Blocks for Youth. 6 Leiber (2002). 7 Leiber (2002). 8 Villarruel & Walker (2002). 9 W. Haywood Burns Institute (2007). Indiana s plan for reducing disproportionate minority contact. Online. Retrieved January 4, 2007 at

15 10 Donovan, M. S., & Cross, C. T. (Eds.). (2002). Minority students in special and gifted education. Washington, D.C.: National Academies Press; OJJDP (2006); WESTAT (2005) Methods for assessing racial/ethnic disproportionality in special education: A technical assistance guide. Washington, D.C.: US Department of Education Office of Special Education Programs. Online. Retrieved October 5, 2006 from e.pdf. 11 Feyerherm, W. & Butts, J. (n.d). Proposed methods for measuring disproportionate minority contact (DMC). Presentation for the Office of Juvenile Justice and Delinquency Prevention. Online. Retrieved November 3, 2006 from 12 Donovan & Cross (2002); OJJDP (2006); WESTAT (2005). 13 Indiana Criminal Justice Institute (2006). Indiana s three-year delinquency prevention and systems improvement plan update and application. Online. Retrieved April 1, 2007 from 14 Terms that clearly denoted contact with law enforcement, such as gang/abuse activity, were also included in the analysis. 15 U.S. Department of the Census, (2006). Population estimates program, county population estimates by age, sex, race and Hispanic origin: April 1, 2000 to July 1, Washington, D.C.: Author. 16 Probability values were associated with the following Chi-square values: Chi-square value between 3.84 and 6.62 yielded p<.05; yielded p<.01; values above yielded p<.001. Chi-square values less than 3.84 were considered not significant, meaning there was greater than a 5 in 100 chance that the results could be due to chance.

16 Appendix A Disproportionate Minority Contact in Indiana: County by County Results A1 Allen County A2 Howard County A3 Lake County A4 Madison County A5 Marion County A6 St. Joseph County A7 Tippecanoe County

17 Appendix A1. Disproportionate Minority Contact in Indiana: Allen County Population Contact Risk Index Relative Rate Index Count % of Total Count % of Total Target Race All Other Target Race / All Other Chi-square Significance Arrest (B) Native American % % 1.57% 4.07% 0.39 p<.05 Asian or Pacific Islander 1, % % 2.53% 4.09% 0.62 p<.01 Hispanic 4, % % 3.24% 4.12% 0.79 p<.01 Black, Not of Hispanic Origin 11, % 1, % 10.89% 2.86% 3.80 p<.001 White, Not of Hispanic Origin 57, % 1, % 2.83% 7.50% 0.38 p<.001 Other/Unknown 2, % % 3.15% 4.09% 0.77 p<.05 Total 77, % 3, % 4.06% Detention at Intake (E) Native American % % 75.00% 53.57% 1.40 ns Asian or Pacific Islander % % 48.48% 53.65% 0.90 ns Hispanic % % 55.90% 53.48% 1.05 ns Black, Not of Hispanic Origin 1, % % 60.63% 48.89% 1.24 p<.001 White, Not of Hispanic Origin 1, % % 47.66% 59.84% 0.80 p<.001 Other/Unknown % % 59.21% 53.46% 1.11 ns Total 3, % 1, % 53.60% Diversion (D) Native American % % 50.00% 34.73% 1.44 ns Asian or Pacific Islander % % 42.42% 34.67% 1.22 ns Hispanic % % 31.06% 34.95% 0.89 ns Black, Not of Hispanic Origin 1, % % 30.55% 37.57% 0.81 p<.001 White, Not of Hispanic Origin 1, % % 38.67% 30.63% 1.26 p<.001 Other/Unknown % % 25.00% 34.99% 0.71 ns Total 3, % 1, % 34.75% Petitioned (Charged) (F) Native American % % 50.00% 38.62% 1.29 ns Asian or Pacific Islander % % 36.36% 38.66% 0.94 ns Hispanic % % 39.13% 38.61% 1.01 ns Black, Not of Hispanic Origin 1, % % 41.26% 36.88% 1.12 p<.05 White, Not of Hispanic Origin 1, % % 36.02% 41.39% 0.87 p<.01 Other/Unknown % % 50.00% 38.36% 1.30 p<.05 Total 3, % 1, % 38.64% Adjudicated (Guilty) (G) Native American % % 50.00% 73.08% 0.68 ns Asian or Pacific Islander % % 66.67% 73.10% 0.91 ns Hispanic % % 80.95% 72.61% 1.11 ns Black, Not of Hispanic Origin % % 70.61% 74.86% 0.94 ns White, Not of Hispanic Origin % % 74.70% 71.52% 1.04 ns Other/Unknown % % 71.05% 73.10% 0.97 ns Total 1, % % 73.04% Transfer/Waived to Adult Court (J) Native American % % 0.00% 0.06% 0.00 ns Asian or Pacific Islander % % 0.00% 0.06% 0.00 ns Hispanic % % 0.00% 0.07% 0.00 ns Black, Not of Hispanic Origin 1, % % 0.16% 0.00% ns White, Not of Hispanic Origin 1, % % 0.00% 0.13% 0.00 ns Other/Unknown % % 0.00% 0.06% 0.00 ns Total 3, % % 0.06% Probation (H) Native American % % 0.00% 61.93% 0.00 ns Asian or Pacific Islander % % 75.00% 61.74% 1.21 ns Hispanic % % 47.06% 62.75% 0.75 p<.05 Black, Not of Hispanic Origin % % 61.89% 61.83% 1.00 ns White, Not of Hispanic Origin % % 62.47% 61.27% 1.02 ns Other/Unknown % % 77.78% 61.36% 1.27 ns Total % % 61.86% Secure Confinement (DOC) (I) Native American % % 0.00% 1.23% 0.00 ns Asian or Pacific Islander % % 0.00% 1.24% 0.00 ns Hispanic % % 0.00% 1.30% 0.00 ns Black, Not of Hispanic Origin % % 1.62% 0.95% 1.70 ns White, Not of Hispanic Origin % % 0.92% 1.53% 0.60 ns

18 Appendix A2. Disproportionate Minority Contact in Indiana: Howard County Population Contact Risk Index Relative Rate Index Count % of Total Count % of Total Target Race All Other Target Race / All Other Chi-square Significance Arrest (B) Native American % % 1.92% 6.66% 0.29 ns Asian or Pacific Islander % % 1.14% 6.70% 0.17 p<.01 Hispanic % % 8.16% 6.59% 1.24 ns Black, Not of Hispanic Origin 1, % % 15.30% 5.75% 2.66 p<.001 White, Not of Hispanic Origin 14, % % 5.85% 10.65% 0.55 p<.001 Other/Unknown % % 1.89% 6.77% 0.28 p<.001 Total 17, % 1, % 6.64% Detention at Intake (E) Native American % % 0.00% 16.65% 0.00 ns Asian or Pacific Islander % % 0.00% 16.67% 0.00 ns Hispanic % % 21.74% 16.43% 1.32 ns Black, Not of Hispanic Origin % % 17.79% 16.32% 1.09 ns White, Not of Hispanic Origin % % 16.14% 18.01% 0.90 ns Other/Unknown % % 11.11% 16.68% 0.67 ns Total 1, % % 16.64% Diversion (D) Native American % % 0.00% 15.12% 0.00 ns Asian or Pacific Islander % % 0.00% 15.13% 0.00 ns Hispanic % % 8.70% 15.36% 0.57 ns Black, Not of Hispanic Origin % % 13.83% 15.45% 0.90 ns White, Not of Hispanic Origin % % 16.03% 12.54% 1.28 ns Other/Unknown % % 0.00% 15.22% 0.00 ns Total 1, % % 15.10% Petitioned (Charged) (F) Native American % % 0.00% 22.63% 0.00 ns Asian or Pacific Islander % % 0.00% 22.65% 0.00 ns Hispanic % % 13.04% 23.00% 0.57 ns Black, Not of Hispanic Origin % % 29.25% 20.78% 1.41 p<.01 White, Not of Hispanic Origin % % 21.37% 26.05% 0.82 ns Other/Unknown % % 11.11% 22.70% 0.49 ns Total 1, % % 22.61% Adjudicated (Guilty) (G) Native American % % % Asian or Pacific Islander % % % Hispanic % % 83.33% 84.17% 0.99 ns Black, Not of Hispanic Origin % % 83.78% 84.29% 0.99 ns White, Not of Hispanic Origin % % 84.24% 83.95% 1.00 ns Other/Unknown % % % 84.09% 1.19 ns Total % % 84.15% Transfer/Waived to Adult Court (J) Native American % % 0.00% 0.09% 0.00 ns Asian or Pacific Islander % % 0.00% 0.09% 0.00 ns Hispanic % % 0.00% 0.09% 0.00 ns Black, Not of Hispanic Origin % % 0.40% 0.00% ns White, Not of Hispanic Origin % % 0.00% 0.32% 0.00 ns Other/Unknown % % 0.00% 0.09% 0.00 ns Total 1, % % 0.09% Probation (H) Native American % % % Asian or Pacific Islander % % % Hispanic % % 80.00% 73.85% 1.08 ns Black, Not of Hispanic Origin % % 74.19% 73.91% 1.00 ns White, Not of Hispanic Origin % % 73.55% 75.00% 0.98 ns Other/Unknown % % % 73.87% 1.35 ns Total % % 73.99% Secure Confinement (DOC) (I) Native American % % % Asian or Pacific Islander % % % Hispanic % % 0.00% 2.75% 0.00 ns Black, Not of Hispanic Origin % % 1.61% 3.11% 0.52 ns White, Not of Hispanic Origin % % 3.23% 1.47% 2.19 ns

19 Appendix A3. Disproportionate Minority Contact in Indiana: Lake County Population Contact Risk Index Relative Rate Index Count % of Total Count % of Total Target Race All Other Target Race / All Other Chi-square Significance Arrest (B) Native American % % 1.72% 3.18% 0.54 ns Asian or Pacific Islander % % 1.03% 3.19% 0.32 p<.001 Hispanic 19, % % 2.64% 3.29% 0.80 p<.001 Black, Not of Hispanic Origin 31, % 1, % 4.85% 2.47% 1.96 p<.001 White, Not of Hispanic Origin 54, % 1, % 2.30% 4.04% 0.57 p<.001 Other/Unknown 2, % % 6.28% 3.11% 2.02 p<.001 Total 108, % 3, % 3.17% Detention at Intake (E) Native American % % 0.00% 37.49% 0.00 ns Asian or Pacific Islander % % 0.00% 37.55% 0.00 p<.05 Hispanic % % 35.29% 37.82% 0.93 ns Black, Not of Hispanic Origin 1, % % 42.48% 33.33% 1.27 p<.001 White, Not of Hispanic Origin 1, % % 31.88% 40.58% 0.79 p<.001 Other/Unknown % % 43.08% 37.22% 1.16 ns Total 3, % 1, % 37.45% Diversion (D) Native American % % 0.00% 0.32% 0.00 ns Asian or Pacific Islander % % 20.00% 0.26% p<.001 Hispanic % % 0.39% 0.31% 1.28 ns Black, Not of Hispanic Origin 1, % % 0.19% 0.42% 0.46 ns White, Not of Hispanic Origin 1, % % 0.24% 0.36% 0.67 ns Other/Unknown % % 0.77% 0.30% 2.55 ns Total 3, % % 0.32% Petitioned (Charged) (F) Native American % % % 81.43% 1.23 ns Asian or Pacific Islander % % % 81.40% 1.23 ns Hispanic % % 81.76% 81.40% 1.00 ns Black, Not of Hispanic Origin 1, % 1, % 80.70% 82.07% 0.98 ns White, Not of Hispanic Origin 1, % 1, % 83.09% 80.53% 1.03 ns Other/Unknown % % 71.54% 81.84% 0.87 p<.01 Total 3, % 2, % 81.45% Adjudicated (Guilty) (G) Native American % % 25.00% 57.00% 0.44 ns Asian or Pacific Islander % % 60.00% 56.94% 1.05 ns Hispanic % % 52.76% 57.68% 0.91 ns Black, Not of Hispanic Origin 1, % % 61.12% 53.60% 1.14 p<.001 White, Not of Hispanic Origin 1, % % 53.68% 58.85% 0.91 p<.01 Other/Unknown % % 56.99% 56.95% 1.00 ns Total 2, % 1, % 56.95% Transfer/Waived to Adult Court (J) Native American % % 0.00% 0.29% 0.00 ns Asian or Pacific Islander % % 0.00% 0.29% 0.00 ns Hispanic % % 0.00% 0.34% 0.00 ns Black, Not of Hispanic Origin 1, % % 0.52% 0.11% 4.90 p<.05 White, Not of Hispanic Origin 1, % % 0.00% 0.45% 0.00 p<.05 Other/Unknown % % 1.54% 0.24% 6.38 p<.01 Total 3, % % 0.29% Probation (H) Native American % % % 60.86% 1.64 ns Asian or Pacific Islander % % 33.33% 60.99% 0.55 ns Hispanic % % 66.36% 60.01% 1.11 ns Black, Not of Hispanic Origin % % 60.08% 61.63% 0.97 ns White, Not of Hispanic Origin % % 60.47% 61.11% 0.99 ns Other/Unknown % % 56.60% 61.04% 0.93 ns Total 1, % % 60.89% Secure Confinement (DOC) (I) Native American % % 0.00% 1.44% 0.00 ns Asian or Pacific Islander % % 0.00% 1.44% 0.00 ns Hispanic % % 0.45% 1.60% 0.28 ns Black, Not of Hispanic Origin % % 1.83% 1.08% 1.70 ns White, Not of Hispanic Origin % % 1.44% 1.44% 1.01 ns

20 Appendix A4. Disproportionate Minority Contact in Indiana: Madison County Population Contact Risk Index Relative Rate Index Count % of Total Count % of Total Target Race All Other Target Race / All Other Chi-square Significance Arrest (B) Native American % % 4.35% 8.50% 0.51 ns Asian or Pacific Islander % % 0.96% 8.53% 0.11 p<.01 Hispanic % % 7.27% 8.53% 0.85 ns Black, Not of Hispanic Origin 2, % % 19.80% 7.24% 2.74 p<.001 White, Not of Hispanic Origin 22, % 1, % 7.02% 16.92% 0.41 p<.001 Other/Unknown % % 19.56% 8.30% 2.36 p<.001 Total 26, % 2, % 8.50% Detention at Intake (E) Native American % % 0.00% 6.83% 0.00 ns Asian or Pacific Islander % % % 6.78% p<.001 Hispanic % % 2.04% 6.93% 0.29 ns Black, Not of Hispanic Origin % % 9.09% 6.14% 1.48 p<.05 White, Not of Hispanic Origin 1, % % 6.44% 7.75% 0.83 ns Other/Unknown % % 2.25% 7.02% 0.32 ns Total 2, % % 6.83% Diversion (D) Native American % % 0.00% 7.51% 0.00 ns Asian or Pacific Islander % % 0.00% 7.51% 0.00 ns Hispanic % % 14.29% 7.35% 1.94 ns Black, Not of Hispanic Origin % % 3.68% 8.67% 0.42 p<.001 White, Not of Hispanic Origin 1, % % 8.49% 5.17% 1.64 p<.01 Other/Unknown % % 8.99% 7.44% 1.21 ns Total 2, % % 7.50% Petitioned (Charged) (F) Native American % % 50.00% 51.95% 0.96 ns Asian or Pacific Islander % % 0.00% 51.97% 0.00 ns Hispanic % % 38.78% 52.24% 0.74 ns Black, Not of Hispanic Origin % % 53.38% 51.50% 1.04 ns White, Not of Hispanic Origin 1, % % 52.32% 51.06% 1.02 ns Other/Unknown % % 44.94% 52.24% 0.86 ns Total 2, % 1, % 51.94% Adjudicated (Guilty) (G) Native American % % % 65.42% 1.53 ns Asian or Pacific Islander % % % Hispanic % % 73.68% 65.31% 1.13 ns Black, Not of Hispanic Origin % % 64.13% 65.86% 0.97 ns White, Not of Hispanic Origin % % 65.56% 65.18% 1.01 ns Other/Unknown % % 67.50% 65.37% 1.03 ns Total 1, % % 65.45% Transfer/Waived to Adult Court (J) Native American % % 0.00% 0.05% 0.00 ns Asian or Pacific Islander % % 0.00% 0.05% 0.00 ns Hispanic % % 0.00% 0.05% 0.00 ns Black, Not of Hispanic Origin % % 0.00% 0.06% 0.00 ns White, Not of Hispanic Origin 1, % % 0.06% 0.00% ns Other/Unknown % % 0.00% 0.05% 0.00 ns Total 2, % % 0.05% Probation (H) Native American % % % 73.77% 1.36 ns Asian or Pacific Islander % % % Hispanic % % 57.14% 74.12% 0.77 ns Black, Not of Hispanic Origin % % 76.27% 73.04% 1.04 ns White, Not of Hispanic Origin % % 73.92% 73.52% 1.01 ns Other/Unknown % % 62.96% 74.21% 0.85 ns Total % % 73.80% Secure Confinement (DOC) (I) Native American % % 0.00% 4.26% 0.00 ns Asian or Pacific Islander % % % Hispanic % % 0.00% 4.34% 0.00 ns Black, Not of Hispanic Origin % % 3.39% 4.52% 0.75 ns White, Not of Hispanic Origin % % 4.13% 4.57% 0.90 ns