1	The Impact of Undercounting Hispanics on DMC Studies Mark Myrent Illinois Criminal Justice Information Authority JRSA 2007 National Conference October 12, 2007
2	Minority groups other than African Americans have received insufficient attention in DMC research Across the 34 studies reviewed in OJJDP meta-analysis, 27 (79%) included African Americans, while 11 (32%) included Hispanic/Latino.
3	Why focus on Hispanic Youth? Hispanic population in U.S. grew 58% between 1990 and 2000 A rapidly growing number of Hispanics now are living in many areas that, historically, did not have substantial Hispanic populations. Hispanic population is expected to continue to grow at more than 3 times the rate of the total U.S population.
4	Why focus on Hispanic Youth? The limited data on Hispanic still indicates disproportionate representation at multiple stages of the JJ system The justice system often relies on assumptions about which youth are involved in gangs based on stereotypes about Hispanic youth. Hispanics tend to be undercounted to a much greater extent than African Americans
5	Hispanics are undercounted in the population and in the juvenile justice system Researchers at MSU found that states use different methods for collecting and presenting on Hispanic populations in the justice system, including varying definitions of the terms Latino and Hispanic.
6	Hispanics are undercounted in the population and in the juvenile justice system Researchers at MSU found that systems for gathering data in many states do not have a Latino or Hispanic category. Others fail to separate ethnicity from race. In both instances Hispanic youth are often counted as White.
7	NCCD: “And Justice for Some” measures of Hispanic representation Arrests no Referrals to court no Detention yes Formal processing no Disposition no Waiver to adult court no Incarceration/juvenile corrections yes Incarceration/juvenile corrections yes
8	Hispanics are undercounted in the juvenile justice system Undercounting of Hispanics by JJ agencies drives down RI of Hispanic youth, and overstates the RI of White youth.
9	Story Time
10	Conclusions drawn from probation data system experience Race and ethnicity data transposed from arrest fingerprint cards and many police reports is inaccurate because it is derived from either: a single “race” field that does not allow for a “Hispanic” code (arrest f-print card), or a single “race” field for which Hispanic ethnicity has been determined inconsistently (arrest report)
11	Impact of Hispanic Misclassification on DMC
12	Impact of Hispanic Misclassification on DMC Measures - RI
13	Impact of Hispanic Misclassification on DMC Measures - RRI
14	Conclusions drawn from probation data system experience Race and ethnicity should be self-reported by clients Program staff need to have face-to-face contact with clients before completing client intake form
15	Conclusions drawn from probation data system experience Staff must be trained on questioning techniques for obtaining race and ethnicity data from clients as well as alternative strategies pursuant to respondent resistance.
16	Self-reporting race and ethnicity: How does the Census Bureau do it? Separate race from gender In 2000: ethnicity was determined prior to race ethnicity terminology expanded to “Hispanic or Latino” Hispanics could specify sub-group citizens could report multiple races but not multiple ethnicities.
17	How does the Census Bureau define Hispanic/Latino? Census 2000 adheres to the federal standards established by OMB in 1997 Statistical Policy Directive 15 Hispanic or Latino is “person of Cuban, Mexican, Puerto Rican, South or Central American, or other Spanish culture or origin regardless of race.
18	Separating race and ethnicity: Do new problems emerge? Anthropologists weigh in “Although popular connotations of race tend to be associated with biology and those of ethnicity with culture, the two concepts are not clearly distinct from one another.”
19	Separating race and ethnicity: Do new problems emerge? Anthropologists weigh in Today’s’ ethnicities are yesterday’s races Individuals perceive of their racial and ethnic identities as fluid The outcomes of race and ethnicity are defined less by self-definition than by the attribution of those labels by others
20	Separating race and ethnicity: Do new problems emerge? Researchers weigh in Prior to implementation of Directive 15, OMB undertook comprehensive review of categories for data on race and ethnicity. Public comment Research and testing
21	Interagency Committee’s Research Working Group How to collect data on persons who identify as “multi-racial”? Combine race and ethnicity in one question or separate questions? Combine concepts of race, ethnicity, and ancestry? Change terminology? Add new categories?
22	Interagency Committee’s Research Working Group: National Tests Bureau of Labor Statistics: Supplement on Race and Ethnicity to the Current Population Survey Census Bureau: National Content Survey Census Bureau: Race and Ethnic Targeted Test (RAETT)
23	Cognitive pretesting suggests that respondents generally define all the concepts in similar terms Many respondents said that “ethnic group” meant the same thing as “race” Respondents tended to define both race and ethnicity in terms of family origins Many respondents are unfamiliar with the term “ethnicity”
24	Are race categories useful for Hispanics? The overwhelming majority (97%) of the 15.4 million people who reported “Some other Race” as their sole race were Hispanic.
25	Cognitive pretesting suggests that respondents generally define all the concepts in similar terms. Multiethnic and multiracial identifications are frequently not distinguished Some who defined themselves as “multiracial” offered only ethnic groups to explain their backgrounds
26	Are race categories useful for Hispanics? About 6 percent of all Hispanics reported 2 or more races, compared with less than 2% of non-Hispanics. Among those Hispanics who reported more than one race, 81% reported two races – one of which was “Other”
27	AAA Recommendations Combine “race” and “ethnicity” categories into one question until the planning of 2010 Census begins Further research needed to determine the term that best delimits human variability as conceptualized by the American people.
28	If not self-report: Is there consistency in the way agency officials determine race and ethnicity of clients? A study sponsored by the Bureau of Transportation Statistics and Bureau of Justice Statistics found the degree of inter-rater agreement on race and ethnicity classification to be 77%.
29	Factors inhibiting police recording of ethnicity Police attention focused on physical descriptors – gender, hair/eye/skin color, scars/marks/tattoos, Hispanic ethnicity an imprecise indicator of physical appearance
30	Law enforcement attention to ethnicity: emerging Immigration-related factors cited by IACP Victim vulnerability Violence against Women Human trafficking Identity theft crimes Gang violence