As part of the work of the Alaska Supreme Court Advisory Committee on Fairness and Access, the Alaska Judicial Council designed and implemented a limited examination of felony probation revocation cases. One purpose of the study was to determine if Alaska Native offenders whose probationary status had been revoked received different dispositions on revocation than persons from other ethnic backgrounds. The study was extremely limited, involving only 154 cases from a three-year time span; but the examination of the data indicated that petitions to revoke probation were not filed against minority defendants for different reasons than they were filed against Caucasian offenders, nor did Alaska Natives and other minorities receive harsher sentences when probation had been revoked. The study did show some differences by ethnicity for the original conditions of probation.
The sample studied included only male offenders convicted of selected B or C felonies, from five Alaska communities—Anchorage, Bethel, Fairbanks, Juneau and Nome. All offenders had their probation revoked in 1994, 1995 or 1996.
Altogether, 154 cases were collected: 50 from Anchorage, 32 from Juneau, 30 from Fairbanks, 26 from Bethel, and 16 from Nome. Of the cases in the Bethel court, 85 per cent involved defendants who lived outside Bethel. Of the cases in Nome, 94 per cent involved defendants who lived outside Nome.
The sample was intentionally skewed to include more Natives proportionately than are found in the corrections population. Thus, a total of 89 defendants were Alaska Native or American Indian, 16 were African American, and 44 were Caucasian. The remaining five were characterized in the pre-sentence reports as one Asian, two of "other" ethnic origin, and one "unknown."
All of the Bethel and Nome defendants were Native. The African-Americans had case files in Anchorage (11), Fairbanks (4), and Juneau (1). Caucasians had case files in Anchorage (22), Fairbanks (12), and Juneau (10).
For each defendant, researchers compiled information about marital status, years of education, city of permanent residence, city of current residence, number of prior adult convictions, presence of a juvenile record, and number of prior probation revocations. The resulting portrait was of a single man without significant educational achievement who had one or more prior adult convictions but who did not necessarily have prior probation revocations.
Two-thirds of the defendants were single, 15 per cent were married, and 9 per cent were divorced (marital status was unknown for 8 per cent). About 56 per cent of the defendants had a high school diploma, GED, or some higher education; 39 per cent did not. About 27 per cent of the defendants had no adult prior record of either felonies or misdemeanors. Twenty-nine per cent had one to three prior convictions, and nearly half (44%) had four or more priors. Nearly three-quarters (71%) had no juvenile record. A little over half (55%) of the defendants had no previous probation revocations. About one-third (31%) had one earlier probation revocation, and 14 per cent had two or more.
Each of these defendant characteristics was cross-tabulated with the ethnic origin variable. The cross-tabulation failed to reveal any significant differences in defendant characteristics by ethnic origin.
The offenses of which the defendants were originally convicted included Sexual Assault II (2), Sexual Abuse of a Minor II (27), Theft II (24), Criminal Mischief II (15), Burglary I (11), Burglary II (29), Assault II (4), Assault III (28), and Other (14). For purposes of analysis, these were recoded into "violent," "property," and "other."
Offense of conviction was one variable that did show statistically significant differences by ethnic origin. About 52 per cent of the Native defendants in this group had been convicted of a violent B or C felony, as compared to 25 per cent of the African-Americans and 29 per cent of the Caucasians. This relationship is consistent with findings from other studies that Alaska Natives have a higher rate of conviction for violent offenses than do other ethnic groups. However, it must be noted that the actual percentages in this sample may not be the same as in the overall population of convicted offenders because we selected only males, in specific communities, and only those with B and C felonies.
All defendants had received sentences for class B or C felonies. The sentences recorded in the pre-sentence reports vary somewhat by type of offense and other offenses sentenced at the same time. Sentence length on the original offense ranged from no time to serve (21% of the sample) to 108 months, with 32 per cent receiving one to six-month sentences ("short"), 22 per cent receiving seven to twelve-month sentences ("medium"), 20 per cent receiving 13 to 36-month sentences ("long"), and 5 per cent receiving 37 to 108-month sentences ("longest").
Information also was collected on conditions of original probation. Frequently imposed conditions included alcohol treatment, drug treatment, mental health treatment, sex offender treatment, anger management treatment, restrictions on drinking, and restrictions on movement (e.g., do not contact victim; stay away from certain location or community). No-drinking restrictions were imposed on about 58 per cent of defendants, while movement restrictions were imposed on approximately 36 per cent and anger management treatment was imposed on about 14 per cent.
Violations for which Probation was Revoked
Nearly half (45%) of the current probation revocation petitions had stemmed from a new offense, usually a misdemeanor. The other three primary reasons given by probation officers were: failure to report to the probation officer (20%), alcohol or drug use (16%), and lack of compliance with treatment conditions (15%). Table 2 shows the reasons for probation revocations; Table 3 gives the same information sorted by ethnicity of the defendant.
To get a better sense of the statistical significance of differences among groups, much of the remaining analysis was done only with the subset of all Alaska Native, African American, and Caucasian defendants (N=149). Five defendants with other ethnic origins were excluded. A review of the current violations of these 149 defendants showed no statistically significant differences by ethnic origin. African Americans were somewhat more likely to have a new offense (63%, compared to 41% of Natives and 48% of Caucasians). African American offenders in this group were less likely to have been revoked for not complying with treatment conditions or using alcohol or drugs than were Natives or Caucasians, but again, these differences were not statistically significant.
Conditions of Original Probation
Data on defendants' ethnic origin were compared to frequently imposed probation conditions to see if minority defendants received different kinds of probation conditions than Caucasians. Three probation conditions were significantly related to ethnicity, while two fell just short of significance and two were not significant.
The conditions that did not differ significantly were drug treatment (required for 29% of Native defendants, 43% of Caucasian defendants and 44% of African American defendants), and mental health treatment (required for 24% of the Native defendants, 23% of the Caucasian defendants, and 6% of the African American defendants).
The two conditions that fell just short of statistical significance were alcohol treatment and sexual offender treatment. Alcohol treatment was required for 70 per cent of the Native defendants, 44 per cent of the African American defendants and 55 per cent of the Caucasian defendants. No African American defendants were required to obtain sexual offender treatment. Judges required it for 9 per cent of the Caucasian defendants and 19 per cent of the Native defendants.
Three probation conditions in this group of cases studied showed statistically significant relationships with the ethnic background of the defendant. (It should be noted again that these cases were selected to contain non-representative numbers for certain types of offenses, for ethnic background of offenders, and for certain communities. The data cannot be used to prove that the same conditions exist throughout the general offender populace. However, the offenses selected are among the most common B and C felonies, and the offenders selected do resemble the overall offender population in prior records, education and marital status.) First, more African Americans were required to attend anger management training than were Caucasian defendants or Native defendants. Second, judges imposed no-drinking restrictions on Native defendants more often than on other defendants. Third, movement restrictions were imposed more often on Native defendants than on defendants of other ethnicities.
Anger Management Treatment
Only 21 of the cases reviewed contained an order to obtain treatment on anger management. However, about 38 per cent of African American defendants were ordered to obtain the treatment, compared to only 16 per cent of Caucasians and 8 per cent of Natives.
Cross-tabulations of type of offense by the anger management probation condition for each ethnic group showed noticeable differences among ethnic groups. For example, very few Native defendants were required to obtain anger management treatment, even for violent offenses, as compared to African American defendants, about one-third of whom were required to get anger management treatment for property offenses and 50 per cent of whom were required to obtain it for violent offenses. However, the strength of this finding is undermined somewhat by the small number of African American offenders in this group.
Anger management probation conditions also were analyzed in relation to location of the offense. This cross-tabulation showed that no Bethel-area defendants and only one Nome defendant were ordered to attend anger management. In contrast, about 20 per cent of offenders from Anchorage and Fairbanks received anger management orders.
Sixty-seven per cent of the Native defendants were prohibited from consuming alcoholic beverages, versus 38 per cent of the African American defendants and 46 per cent of the Caucasian defendants. The no-drinking restriction applied comparably to the three ethnic groups for those offenders convicted of violent offenses, but was imposed much more frequently on Native defendants convicted of property offenses (65%), as compared to African American defendants (25%) or Caucasian defendants (42%).
Analysis of the "no drinking" restriction by location of the case showed statistically significant differences among the communities. The restriction was most common in communities with larger numbers of Native residents. Thus, 40 per cent of Anchorage cases and 47 per cent of Fairbanks cases had this restriction, as compared to 92 per cent of Bethel cases, 63 per cent of Nome cases and 66 per cent of Juneau cases.
Fifty-one per cent of the Native defendants, 18 per cent of the Caucasian defendants, and 13 per cent of the African American defendants had movement restrictions (e.g., do not contact victim; do not go to place where alcohol is served) imposed on them by the court. For violent offenses, judges applied movement restrictions to 67 per cent of Native defendants, 50 per cent of African American defendants and 39 per cent of Caucasian defendants. For property offenses, the movement restrictions applied to 38 per cent of the Native defendants, none of the African American defendants, and 10 per cent of the Caucasian defendants.
The movement restrictions appeared to be related to the offense of conviction. While 57 per cent of the defendants originally charged with a violent offense had a movement restriction, only 21 per cent of those charged with a property offense had movement restrictions.
These data should be interpreted in the context of the earlier variable reported—reason for current probation revocation petition. Nearly all of the probation revocations in this group were granted, so we cannot use the data to test whether the presence of certain types of conditions made a difference in the likelihood that one group or another would get probation revoked. Within the group studied, staff cross-tabulated type of offense by ethnic group and by reason for the current probation revocation. Few large differences appeared.
Across the board, defendants originally convicted of a property offense were more likely than those convicted of a violent offense to have committed a new felony or misdemeanor. Caucasian defendants were somewhat more likely than Native defendants to have failed to comply with treatment conditions, but the difference was not statistically significant. Offenders convicted of violent crimes were less likely to have complied with treatment conditions than those convicted of property offenses where there was no new criminal offense. Caucasian and Native defendants were more likely than African American defendants to have the primary reason for the revocation listed as "used alcohol or drugs." The data available in this study, then, suggest that even though some types of conditions may be imposed more frequently on one ethnic group than another, the differences may not affect the likelihood of a certain type of probation violation.
Disposition of Petition to Revoke Probation
Virtually all of the probation revocation petitions in this group of cases were granted by the court. At sentencing on the probation revocation, the judge could have imposed additional time to serve, continued the probation supervision, imposed additional conditions of probation, combined these actions, or taken no new action. For the majority of defendants, the judge imposed additional incarceration (68% of Caucasian defendants, 81% of African American defendants, and 82% of Native defendants; differences not statistically significant). For a small number of defendants, judges required residential treatment in connection with the probation revocation. In fourteen cases, the residential treatment appeared to be in lieu of incarceration, and in seven cases, it appeared to be in addition to incarceration.
As the final step in the analysis, staff recoded the number of months of incarceration imposed after the probation revocation into five categories (none, 1-6 months, 7-12 months, 13-24 months, and over 24 months), and cross-tabulated the number of months against the offense of conviction and ethnic background of the defendant. Among ethnic groups and types of offenses, some differences appeared, but none tested as statistically significant. Thirteen per cent of Native defendants received no incarceration; 19 per cent of African American defendants did not serve additional time after probation revocation; and 27 per cent of Caucasian defendants did not. Native defendants (28%) appeared more likely to receive short terms of incarceration (1 to 6 months) than did African Americans (13%) or Caucasians (16%). Very few notable differences appeared in the overall analysis by type of offense.
Only one area showed statistically significant differences within an ethnic group. Native defendants convicted of violent offenses were less likely than Native defendants convicted of property offenses to have additional incarceration imposed after a probation revocation. Among Caucasian defendants, the opposite pattern occurred: those convicted of property offenses were less likely to receive incarceration after a probation revocation (32%) than those convicted of a violent offense (17%).
This study described a group of 154 defendants for whom probation revocation petitions were filed in 1994, 1995 and 1996. Over half had a high school education or better and about two-thirds were single. Most (73%) had a prior record of felonies or misdemeanors, but fewer than half (45%) had a record of prior probation revocations.
The most significant differences among groups appeared in the offense of original conviction and the conditions of the original probation. Native defendants (52%) were significantly more likely to have been convicted of a violent offense than were African American defendants (25%) or Caucasian defendants (29%). African American defendants were significantly more likely to have anger management imposed as a condition of probation, while Native defendants were significantly more likely to have a "no-drinking" or a "movement" restriction imposed. One hypothesis to explain this relationship is that judges applied these conditions of probation instead of sentencing Native defendants to treatment programs because such programs are not available in most villages.
The study did not support the hypothesis that petitions to revoke probation are filed against minority defendants for different reasons than they are filed against Caucasian defendants. Nearly half of the probation revocation petitions listed a new offense (most likely a misdemeanor) as the reason for the petition. Nor did the study support the hypothesis that judges imposed harsher sentences against minority defendants who violated probation than against their Caucasian counterparts. Judges imposed additional incarceration after the probation revocation for most defendants, along with some residential treatment requirements and other conditions.
This article is based on a study conducted by the Alaska Judicial Council in 1997. A complete copy of the study is included in the Report of the Alaska Supreme Court Advisory Committee on Fairness and Access.