Minorities Referred at Higher Rates: Analysis of DFYS Data

Minorities Referred at Higher Rates: Analysis of DFYS Data

N. E. Schafer and Richard W. Curtis

"Minorities Referred at Higher Rates: Analysis of DFYS Data" by N. E. Schafer and Richard W. Curtis. Alaska Justice Forum 14(3): 2-4 (Fall 1997). In a study of 28,618 referrals to the Alaska juvenile justice system, the Justice Center found considerable disparity between white and minority youth in the rate at which they were referred, the frequency of their referrals, and the types of offenses which brought them to the attention of juvenile justice personnel. This report analyzes referrals from 1992 to 1995 of youth aged 10 to 17 years old who were categorized in one of three racial groups - white, Alaska Native, and African-American.

The data analysis discussed here was funded by a gift from Cook Inlet Region, Inc.

In a study of 28,618 referrals to the Alaska juvenile justice system, the Justice Center found considerable disparity between white and minority youth in the rate at which they were referred, the frequency of their referrals and the types of offenses which brought them to the attention of juvenile justice personnel.

A referral comprises paperwork which identifies the youth to the Division of Family and Youth Services (DFYS), specifies the offense behavior and may extensively describe the circumstances surrounding the behavior. Youth usually are referred by law enforcement agencies and the referral may include arrest and detention. The police may have responded to a citizen complaint or acted on a warrant; they may, in the case of shoplifting, for example, have accepted custody of the youth from store personnel; or they themselves may have discovered juveniles engaged in illegal activities. In addition, field probation officers refer youth either for new offenses or for violating the conditions of probation.

Each time an incident is reported to DFYS the details are entered into a data base known as PROBER. Though PROBER contains a substantial amount of additional information, in this article we focus on referral events, using data from 1992-1995.

Demographics of Referrals

We confined our analysis to youth who were 10 to 17 years old at their first appearance in the data set and who were categorized in one of three racial groups—white, Alaska Native and African-American. These three races accounted for 90 per cent of all referrals.

The data are incident-based; thus, the number of youth referred is considerably smaller than the number of referrals. The 26, 618 referrals accumulated over a four-year period involved only 14,145 individuals, an average of two referrals for each youth in the data set. This information is displayed in Table 1 by race and gender. Note that males accounted for nearly three-quarters of all referrals but were just two-thirds of all youth referred. Although white youth of both genders were two-thirds of all youth referred, they accounted for less than 60 per cent of all referrals recorded, while a smaller proportion of minority youth accounted for a larger percentage of all referrals. These differences are clearer in an examination of the mean (average) number of offenses per youth. Each Native youth accumulated an average of just over 2.4 referrals and each black youth slightly less. White youth, on the other hand, had fewer than two referrals per person.

Table 1. Referrals - Demographics, 1992-1995

Rates of Referrals

Another way of examining racial disproportionality is to examine the rate at which each race accumulates referrals. We used as a base for computing such rates Alaska Department of Education (DOE) enrollment figures for grades 5 through 12. Figures, broken down by race and area, were available for each of the four years studied. During each year Alaska Natives constituted about 23 per cent of all youth enrolled in grades 5 through 12 and African Americans around 5 per cent.

We computed a rate of referral for each 1000 persons in the racial population in question using the enrollment figures and found them consistently higher for minority youth than for white youth (Table 2). The referral rate for each 1000 Alaska Natives in the youth population ranged from a low of 14.2 in 1992 to a high of 16.8 in 1994. The referral rate for white youth ranged from 8.5 to 8.9 per 1000. Referral rates for African Americans were significantly higher than for either other race, ranging from 17.7 to 20.5. (These rates were computed based on DOE enrollment figures. Dropout estimates provided by DOE for the 1995-96 school year suggest that the enrollment-based rates presented in this study may be understated by as much as 4.1 per cent for Natives, 2.5 per cent for African Americans, and 2.7 per cent for whites.)

Table 2. Referrals and Referral Rates of Youth 10-17 Years Old in Alaska, Statewide, 1992-1995

By this measure, as well as by measures of proportion, minority youth are overrepresented at the initial entry point of the Alaska juvenile justice system. The referral data cannot show us why they are overrepresented, but it can provide information on the reasons for the referrals, i.e., the offenses for which the youth are referred.

Reasons for Referrals

Offense information was entered in the data based on the most serious charge at referral (if there was more than one charge). We initially categorized referral events as crimes against persons, property crimes, public order crimes and other offenses. However, because these categories include both very serious and very minor offenses, we decided to control for offense severity. We therefore based most of our offense-based analyses on selected crimes either because they were the most numerous in their category or because there was particular public interest in them (Table 3).

Table 3. Referral Distribution by Charge Across Primary Racial Groups, 1992-1995

For example, of 4,078 referrals for crimes against persons, only 40 (less than one tenth of one per cent) were for the crimes of murder in the first or second degrees, manslaughter or criminally negligent homicide. The majority (61%) were for misdemeanor assault (assault in the fourth degree) so we used this charge in our analyses of crimes against persons. For crimes against property we focused on burglary (the only felony), criminal mischief and misdemeanor theft: these accounted for two-thirds of all property crime referrals. Referrals for possession/consumption of alcohol were the most numerous in the public order category. We also included referrals for misdemeanor drug possession because drug abuse is perceived as a growing problem.

The row percentages in Table 3 show the proportion of referrals for each offense attributable to each racial group. For offenses against persons, minority youth are represented in proportions greater than their representation in the general youth population.

For referrals on the three selected property offenses of burglary, criminal mischief and theft in the third or fourth degrees, Native youth were overrepresented in the first two, but represented slightly under their proportion in the general population for theft. Misdemeanor theft is the only referral offense for which white youth were represented in proportions which approximated their percentage of the general youth population. They were responsible for approximately 70 per cent of all thefts in the data base, while they constituted 72 to 73 per cent of all youth enrolled in school in 1992, 1993, 1994 and 1995.

Referrals for alcohol offenses were accumulated by Alaska Native youth in greater numbers than for any other racial group. They accounted for 54.5 per cent of all referrals for minor alcohol offenses, while white youth accounted for 44.7 per cent. African American youth were seldom referred for alcohol violations: only 34 of 2,500 referrals of black youth were for this behavior.

There are regional differences in referrals for this offense. Because the regions differ greatly in population, the rates of referral for this offense make a better comparison. In the northern region, where the Alaska Native population is high, some communities exhibit strong legal proscriptions against drinking. In this region the rate of referral per 1000 Native youth in the region ranged from 3.6 to 4.3. In Anchorage, on the other hand, there were only 323 referrals for alcohol in all four years. The rate per 1000 Native youth in Anchorage only went above one per 1000 in 1995 (1.1). In 1992 it was 0.2. Clearly there are regional variations in the types of offenses for which youth are referred as well as differences by race.

Data on Individuals

Because the referral data are incident-based, we have been discussing incidents rather than people. As we have already noted, there are more referrals than people since some individuals accumulate several referrals. Indeed, 116 youth accumulated 15 or more referrals during the period studied. Three had 18 referrals—the highest number listed. Two-thirds of the youth appeared only once in the data set; of these, 70 per cent were white, 65 per cent were black and 60 per cent were Native. We limited our analysis of individuals to those whose full referral history was in the data set by selecting only those individuals whose history indicated no prior record. This left us with a sample of 11,799 youth referred one or more times in 1993, 1994, 1995 and 1996. We examined the data on individuals on the basis of gender as well as race. Table 4 describes these youth and includes the mean number of referrals committed by each individual in that category. For example, the 2,696 white females in the data base accumulated a total of 3,938 referrals—an average of 1.46 referrals per person.

Table 4. Distribution of Persons, Age, and Referrals by Race and Sex, 1992-1995

The gender differences are instructive. The percentage of individual females (34.3%) is greater than the percentage of referrals attributed to them (29.9%). They are presumably more likely than male youth to be referred only once. There are, however, differences by race, with a larger percentage of referrals of Native females (33.0%) than of either African American (25.0%) or white females (29.2%). Indeed, Alaska Native girls amassed a higher average number of offenses (1.89) than white boys (1.79).

Studies conducted in other states have suggested that girls are more likely to be entered into the juvenile justice system for their own protection, while boys are more likely to be referred for the protection of the public. We therefore continued to use gender in our analysis of offense behavior by race. Because 73.6 per cent of the individuals were referred for the first time for the offenses we selected to control for severity, we have confined our analysis to these in Table 5. This table includes only those who had no prior record noted in their PROBER file.

Table 5. First Referral for Selected Offenses by Race (Individuals), 1992-1995

Among Native youth, numerically more girls than boys have been referred for possession/consumption of alcohol. This charge accounts for 35 per cent of all first referrals for Native females. More white girls than Native girls were referred for this offense, but as a percentage of all of their first referrals it was a distant second. For white females, the primary first referral charge was misdemeanor theft; 48.3 per cent of white girls were referred for this. An even larger percentage of African American girls was referred for theft (63.3%). Theft was first for a greater proportion of white (30.9%) and black boys (39.8%) as well, but in neither case do their proportions approach those for girls. We note also that a larger percentage of girls than boys was referred for misdemeanor assault. This difference was found among the minority groups; white boys and girls were referred for assault in the fourth degree in nearly equal percentages.

The data presented in this article cannot be used to assess the degree to which race influences the decision to refer youth to DFYS. Certainly minorities are disproportionately referred. The extraordinary number of referrals of Native youth for possession/consumption of alcohol suggests that the agencies making the referrals view this behavior differently for Native than for white or black youth. The data also show strong differences between highly urban Anchorage and the more rural parts of the state in the types of activities for which youth are referred. While these may result from different law enforcement priorities, they may also result from greater criminal opportunities in the city. The differences by region will be the subject of a future Forum article.

Nancy Schafer is a professor at the Justice Center. Richard Curtis is a research associate with the Center. Cassie Atwell of the Justice Center also contributed to the data analysis presented in this article.