FINDINGS FROM THE 1991 NATIONAL SURVEY



of JUVENILE CORRECTIONS



A Preliminary Report of the NGCRC



PRINCIPAL INVESTIGATOR: George W. Knox, Ph.D.(1)

Director, NGCRC

RESEARCH ASSOCIATES:

Bridgette Adams

Michael Adekale, B.S.

Bernard Akpan, B.S., M.S.

Robert Bonner

Raimondo Brown, B.S.

Ne'Cole Bryson

Nora Bush, B.S.

Rochelle Carey

Lashoney Carter

Cecelia Crenshaw, A.A., B.S.

Willie D. Cross, B.S., M.S.

Rolando Curington, B.S., M.S.

Harold Dade

Renee Delvecchio, B.S.

Mike Dooley

Linda Dozier

Yvette Gilmore-Hayes

Trenton Fedrick

Keith Griffin, B.S., M.S.

Abubakr Hamad, LL.B., LL.M, M.S.

Denise M. Hardin, B.S.

Paul Hoffman, M.S.W.

Carlton Jackson

Levelle Kimble, A.A., B.A.

Sam Lacey

David Laske, Ph.D.

Clara Massey, B.S.

Arturo Mota

Fitzgerald Mullins, B.S.

Willie Nance

Benita Parker, B.S.

Odessa Randle

Romelio Rogel, B.S.

Elvis Slaughter

Emile Spearman

Barbara Vasquez

David R. Williams, B.S., B.A.

Yolande Williams, B.S.

Deborah Y. Zackery, A.A.

Carolyn Zook, A.A.















ABSTRACT



The findings reported here are from a national random sample of N = 274 short and longterm juvenile correctional facilities. This represents a 42.7 percent return rate for the sample frame of N = 641 facilities and includes responses from 44 states and the District of Columbia. Conducted in October, 1991 this survey sought to replicate some of the variables analyzed a year earlier (Knox, 1990) and include new additional variables. Presented here are descriptive findings and some preliminary findings dealing with many issues including: overcrowding, gangs, conflict, programs, services, level of security, use of clergy, and institutional other and confined juvenile characteristics.





- ii -



FINDINGS FROM THE FALL 1991 JUVENILE

CORRECTIONS SURVEY:

A Preliminary Report





INTRODUCTION

It is fair to say that the amount of research conducted on juvenile corrections is not proportional to the importance attached to juvenile justice generally. Much more research, simply put, is focused on adult corrections. Part of this has to do with the greater difficulty of carrying out research involving juveniles when the parens patriae doctrine adds greater restrictions in access to these youths compared to adult offenders.

The present study has had an eclectic goal of studying a wide variety of issues and problem areas in juvenile corrections. It is, after all, a collective research project lead by the principal investigator and contributed to in a myriad of ways by the many research associates. These research associates are primarily graduate students in criminal justice. It is therefore a study which was born of volunteerism, which was carried out by our own labor, and which had no government or foundation financing. We are thankful for the cooperation from the responding juvenile correctional facilities.



METHODOLOGY

This is the second national juvenile corrections research project undertaken and reported by the author. The first analysis (Knox, 1990) was based on N = 155 long term juvenile correctional institutions, primarily state training schools, in 49 states.(2) The present effort sought to include every known juvenile correctional facility, residential or non-residential (day center, etc), long term (state training schools, camps, etc) and short term (temporary detention centers, etc) facilities, public and private. This represented a sample frame of N = 641 facilities.

As a collective research enterprise, every research associate had at least one question in the survey instrument and contributed to the labor intensive effort of this large mail questionnaire project. The costs of this research project were therefore planned to be equally shared as well in terms of photoduplication and postage. This project arose in conjunction with two courses being taught on Juvenile Correctional Institutions where the student could elect to either carry out an individual report or become involved in this collective research project.

The questionnaire is entitled "The Fall 1991 Juvenile Corrections Survey" and consists of six pages of questions. It was mailed to the N = 641 juvenile facilities on October 1, 1991. Those responding by November 27, 1991 constitute the sample used for this report. This involves a national random sample of N = 274 short and long term juvenile correctional facilities and institutions distributed in 44 different states and the District of Columbia. The unit of analysis here is therefore the individual type of juvenile correctional facility or the views of its administrator.



DESCRIPTIVE FINDINGS

Some 241 facilities had a total male juvenile count ranging from as few as two to a high of 1,666. Some 154 facilities had a total female juvenile count ranging from a low of one to a high of 169. For all of the juvenile facilities reporting a current population count on their male and female juveniles, a total of N = 18,657 males and N = 1,736 females were held in the facilities surveyed.

For the most part, most of these juvenile facility/program administrators knew whether theirs was a short term or a long term facility. That is, temporary juvenile (typically county level) detention centers were coded as "short term", these are facilities where youths are awaiting a court determination of their future placement. State training schools, forestry camps, state secure correctional centers for adjudicated youths, etc, also generally knew they were considered "long term" facilities. However, additional items of information were requested which allowed for a quantitative and objective classification into three such categories: (1) short term, (2) long term, and (3) other. If the facility had an average length of stay for the juveniles of three months or more it was coded as a long term facility. If less than three months, it was coded as a short-term facility. Some facilities were combined centers housing both and were classified as "other".

This data shows that 48.9 percent (N = 132) were classified as short term facilities. A nearly equal group was classified as long term care or custody (N = 125, 46.3%). A small number (N = 13, 4.8%) were classified as other.

One question early in the survey instrument was both fiscal and philosophical in nature. It asked "what percentage increase in your operating budget would guarantee a reduction in recidivism?" and therefore nearly half of the sample astutely remarked next to the item "there are no GUARANTEES". Among those facility administrators who did respond, half indicated they would need their budgets increased over 25 percent to accomplish such reduced recidivism. Juvenile corrections, like its adult counterpart, throughout the United States appears to suffer from budget cutbacks as well based on the profuse hand-written comments at various points in this survey.

Most of the sample therefore constituted residential facilities (N = 249, 92.2%). Only 7.8 percent (N = 21) were non-residential (e.g., day treatment centers, etc).

No current typology of juvenile correctional facilities and programs has been suggested in the literature. Rather, the "type" of facility is often a political attribution based on local enabling legislation and statutes. A list of 13 such possible types of facility were provided in a check off survey item. This distribution is provided below. Generally, some respondents felt that "state training school" and "state juvenile correctional center" were equivalent in nature, it was a difference only in name not in form, structure, or function. A facility that was specifically restricted to observation and assessment (e.g., diagnostic center) was classified as a reception and classification center.



TABLE 1

DISTRIBUTION OF SAMPLE BY TYPE OF FACILITY



Type of Facility N %

State Training School 58 21.4

State Juvenile Correctional Center 51 18.8

Boot Camp 0 0.0

Halfway House 30 11.4

Day Treatment Center 18 6.6

Forestry/Conservation Camp 9 3.3

Reintegration Center 5 1.8

Group Home 33 12.2

Temporary Detention Center 32 11.8

Reception/Classification Center 9 3.3

Shelter Care Program 2 0.7

Youth Development Center 21 7.7

Medical/Psychiatric Facility 3 1.1

***** *****

TOTAL 271 100.0



As seen in Table 1 above, state training schools and state juvenile correctional centers together constituted over a third of the entire sample (40.2%). Group homes and halfway houses together constituted over a fifth of the sample (23.3%).

When asked if their facility was community-based, about half (52.2%, N = 141) indicated it was. That is, some 47.8 percent indicated, conversely, that their facility was not community-based. The term "community-based" was not predefined for the respondents here.

These facilities still represent the long standing historical trend of being, to a large extent, geographically located in non-urban settings. Some 45.4 percent (N = 123) indicated they were located in a rural area. Some 43.5 percent (N = 118) indicated they were located in an urban area. And some 11.1 percent (N = 30) indicted they were "other", which was explained in writing as indicating they were suburban based, or in one case located on a military base.

These facility administrators were asked "generally, is overcrowding a problem in your facility?". About a third (N = 91, 33.7%) indicated "yes". About two-thirds (66.3%, N = 179) indicated that overcrowding was not a problem in their facility.

While the concept was not defined precisely, nor did it capture the difference between informal services and formally staffed programs, these facility administrators were asked "Does your facility have a planned re-entry program designed to provide continuity after release in terms of aftercare services?". Some 70.3 percent (N = 192) indicated in the affirmative. Just over a fourth (29.7%, N = 81) indicated they did not have such a planned re-entry program for their juveniles.

Approximately two-fifths of these juvenile facilities (N = 107, 40.1%) indicated they were currently accredited by the American Correctional Association. Thus, over half were not ACA accredited (N=160, 59.9%). A few (N = 26) of these facilities indicated their agency was accredited through another association.

These juvenile facilities reported on a two-part question about the age range for the juveniles committed to their facility. The minimum age showed a range between 6 years old and 17 years old. The maximum age showed a range between 12 and 22 years old.

The mean (arithmetic average) minimum age was 12.5 and the mean maximum age was 18.3.

The typology of personnel assumed by the American Correctional Association in its Standards issued by the Commission on Accreditation for Corrections, describes five categories of staffing: (1) clerical/support, (2) support (regular daily contact, (3) professional specialist, (4) child care/supervision, and (5) administrative/management personnel (ACA, 1983: p. 22). Technically, ACA accreditation guidelines require an undergraduate degree (or its equivalent) for the facility administrator (ACA, 1983: p. 16), but specific higher education is not delineated for other non-management staff. The present survey sought to determine these minimum educational requirements for employment among non-management staff.













TABLE 2



DISTRIBUTION OF MINIMUM EDUCATIONAL

REQUIREMENTS FOR NON-MANAGEMENT STAFF



Educational Level Required N %

No educational requirement 26 9.7

G.E.D. 95 35.6

High school degree 91 34.1

Less than an A.A. degree 10 3.7

A.A. degree 15 5.6

B.A. degree 30 11.2

***** *****

TOTAL 267 100.0



As seen in Table 2 above, four out of every five of these facilities (79.4%) required less than or equal to the equivalent of 12 years of education (GED, or high school degree) for non-management staff. Some 11.2 percent (N = 30) required a bachelors degree for non-management staff. None of the N = 169 juvenile correctional facilities responding to this survey required a Master's degree for non-management staff. This distribution was virtually identical to that of the same data collected a year ago (Knox, 1990).

The work of Irving Goffman and others on total control institutions showed much in common between mental hospitals and prisons. Thus, in both it is not uncommon among line staff to find a belief that when there is a full moon that residents in their custody "tend to act up" or "tend to be more assaultive". Both the 1990 and the 1991 surveys of juvenile correctional facilities included, therefore, the question to facility administrators if they believe there is any basis to this belief often shared among line staff that juveniles tend to be more assaultive when there is a full moon. Table 3 below provides the comparative findings for the Fall, 1990 survey and the present research.



TABLE 3



DISTRIBUTION OF BELIEFS ABOUT THE FULL MOON

AND JUVENILE ASSAULTIVENESS





Believe juveniles are

more assaultive when Fall 1990 Fall, 1991

there is a full moon? N % N %

NO 79 51.0 147 54.4

YES 23 14.8 31 11.5

MAYBE 33 21.3 62 23.0

NOT SURE 20 12.9 30 11.1



Apparently, from Table 3 there is remarkable consistency in the trend comparing the 1990 survey with the 1991 survey that about half of these juvenile correctional facility administrators do not outright reject the notion attributed to line staff that juveniles are more assaultive when there is a full moon.

Using a checklist approach, these juvenile correctional facility administrators were also asked to indicate if their facility had any of a list of specific service programs. Unfortunately, the term "program" can often be construed as any informal effort, as is likely here. That is, ideally we need to know if these are formalized, funded programs with full-time staff and resources allocated to these types of service programs. Such an analysis will have to wait for the 1992 survey. What we can present here is a comparison of the 1990 and the present 1991 surveys as shown in Table 4 below.

TABLE 4



DISTRIBUTION OF SERVICE PROGRAMS

"yes" have the Program

1990 Survey 1991 survey

Type of Program: N % N %

Sex education 132 88.6 237 89.8

Drug rehab./counseling 144 94.1 237 90.5

Job readiness skills training 126 85.7 221 86.3

Treatment for depression 109 75.7 169 68.7

To enhance self-esteem 135 91.2 249 94.7

To increase life coping skills 143 94.1 249 94.0

To decrease propensity to

commit self-destructive acts 109 74.7 204 80.3

Gang deprogramming 24 16.6 57 23.8

AIDS/HIV awareness (not used) 259 96.6





In comparing the 1990 and the 1991 juvenile corrections surveys in Table 4 above, and elsewhere in this report, the reader should note simply that the 1991 annual survey is more encompassing and therefore includes more short-term facilities and more non-residential facilities. The 1990 survey tended to be restricted to long term state training schools and state juvenile correctional institutions. Also, the AIDS/HIV awareness program was not listed in the 1990 survey. Still, as seen in Table 4, there is remarkable consistency between the two different research findings.

In the 1990 survey the "gang deprogramming" service program was included, initially, as a validity factor. That is, as the principal investigator had recently published the first full text book on gangs (Knox, 1991) the research associates felt that one way to determine whether the survey was properly completed or not was to include a type of program that was not indicated anywhere in the literature (e.g., "gang deprogramming"). Thus, when the 1990 survey revealed some 16.6 percent reported having such a "gang deprogramming" service program in their juvenile correctional institution calls were made to these facilities to inquire about these unique initiatives. Generally, what this revealed was the belief that is empirically consistent with the drug-gang research literature, that secondary and tertiary prevention/intervention initiatives can be undertaken to erode "gang ties", to "dismember" gangs, to "coopt" those who are not hard core gang members, and to basically "siphon off" gang members whose commitment level is not very high. But these are informal, not formal, initiatives.

As in the 1990 survey, a question was included in the 1991 survey to basically gauge the recidivism variable most commonly referred to as the "repeater" or "revolving door" factor of juvenile delinquency relapse. The question read "in your best estimate, what percentage of juveniles confined in your facility have served time with you before?". This produced a range between a low of "zero percent" (N = 52, 19.9%) to a high of 98 percent. Generally, over half (56.7%) indicated such a recidivism rate of 10 percent of higher. As in adult corrections, statutory guidelines for juvenile correctional facilities do not require the mandatory reporting of recidivism statistics; and no national group has yet to take full responsibility for codifying, standardizing, and reporting these measures of organizational effectiveness.

As in adult corrections, it is an area of "non accountability". That is, generally, these juvenile correctional agencies are not expected or required by statute to collect and report such information. And the social science research literature is also very fragmented on this matter. Clearly, more recidivism research has been directed at the adult end of the correctional spectrum than on juveniles. In fact, some of the most important juvenile corrections recidivism research undertaken has been relegated to the fugitive literature or is simply never released (see Knox, 1991: pp. 111-113).

Another set of variables measured in both the 1990 and the 1991 juvenile corrections surveys dealt with the matter of gang force strength. The reader is simply cautioned that these are estimates of what percentage of the juveniles in their correctional facilities are gang members which were provided by agency administrators. We might reasonably expect some differences, for example, in comparing the estimates of administrators with line staff regarding what percentage of their juvenile charges are gang members. One dramatic research finding that cannot be reported here(3) (Knox and Tromanhauser, 1991) in specific detail is this: there is a much higher rate of gang membership when we directly ask the confined juveniles themselves through survey research than appears through agency administrator surveys.

The 1991 survey shows that for male juveniles the estimate of gang affiliation ranges between a low of zero percent (15.9%) to a high of 100 percent. And, similarly, the estimate for confined females ranges from a low of zero percent (58.3%) to a high of 100 percent. The 1991 means for the current estimate of what percentage gang members are of the total population of these juvenile facilities was 20.8 percent for males and 4.9 percent for females(4). This compares with 17.7 percent for males and 4.4 percent for females from the 1990 survey. Obviously, from this kind of survey research methodology we cannot speak to whether there are objective increases in gang membership, per se, within American juvenile correctional institutions; because the increase in the 1991 survey may simply reflect a more open admission of the problem, overcoming agency denial, or a host of plausible rival hypotheses. Further research on this problem is urgently needed, but unfortunately the current research agenda on gangs in correctional institutions by the National Institute of Justice (NIJ) does not include juvenile corrections, and focuses like the Camp and Camp (1985) research strictly on adult facilities. Regardless, further detailed analysis of this issue is anticipated in the near future based on the present research.

A similar gang finding trend is found in comparing the 1990 and the 1991 surveys with regard to the question "do whites have a separate gang in your facility?". From the present 1991 survey, about a fourth (25.4%, N = 57) indicated yes, that among whites who are confined in juvenile correctional facilities, there exist separate basically "white gangs". In the 1990 survey some 21.7 percent of the institutions reported the existence of such separate white gangs.

As in the 1990 survey, the 1991 survey included the question "to what extent are gang members involved in extremist political beliefs?". The response mode included the range between zero (not at all) and 10 (a great deal). Findings from the 1991 survey show that 44.5 percent indicated zero; or no such gang involvement with political extremism. But some 14.4 percent rated this factor at "five" or higher.

Another series of questions on the 1991 Survey dealt with gang violence inside juvenile facilities, again replicating the 1990 research(5). When asked to "estimate the number, within the last one year period, how many serious injuries occurred in your facility from attacks/fights/assaults involving gang members", nearly three-fourths (74.6%, N = 194) indicated none. But the range was between zero such gang violence incidents and a high of 22.

When asked, "have gang members been a problem in terms of assaults on your staff?", some 92 percent (N = 242) indicated no for the current 1991 survey. And only 8 percent (N = 21) of these responding juvenile facilities indicated "yes" for such problems of gang assaults upon staff.

While 83.7 percent of the responding facilities indicated that no such assaults or attacks had been made on staff by gang members during the last one year period, some 16.3 percent indicated such attacks had in fact taken place. Among fourteen of the responding institutions these attacks on staff by gang members were serious enough to require hospitalization of the staff members involved. Further, some 8.2 percent of the responding juvenile facilities indicated that the fear of violence represented by gangs, in their view, contributes to staff turnover.

When asked "to what extent are gang members responsible for damage to property in your facility", using a response mode that ranged from zero (none) to 10 (chiefly responsible), the present 1991 findings are similar to the previous year (see Knox, 1991). Some 53.9 percent indicated zero, that is, gang members are not at all responsible for damage to such government property. Clearly, then, somewhat less than half indicated that this was to some extent true.

Another question was designed regarding the gang problem and corrections, and it was this "would your facility function better if gang members could be transferred to a central-national federal unit?". Obviously, no such "gang alcatraz" exists, nor has anyone proposed such a measure. What this really tends to measure, then, is the frustration with the gang problem and the tendency to want to "dump the problem" on someone else in the same way that gave rise to the gang presence within juvenile correctional institutions generally. Nearly a fifth, some 19.9 percent indicated "yes" they would like to transfer their youthful gang members to a central national unit(6).

As in law enforcement and adult corrections the term "gang training" can mean a lot of things. Some consider this equivalent to "group disturbance" training. As if all the gang problem implied was some factor of "potential disruptiveness" during correctional confinement. Like racial conflict, it is a problem found in many adult and juvenile correctional institutions, but it is not a factor that carries any guidelines from national groups like the American Correctional Association.

The Correctional Gang Training Curriculum under development by the Gang Crime Research Center includes training in thirty-two different content areas ranging from Gang Language and Argot to Gang Theory II: Advanced. One cannot simply obtain traditional "riot control" training and assume this is equivalent to the highly specialized and substantially different problem represented by modern gangs and their presence inside correctional institutions or facilities. The simple fact is that gangs differ along the social organizational dimension from being informal social groups to formal organizations (see Knox, 1991, An Introduction to Gangs).

The present 1991 survey showed that nearly a third (33.8%, N = 88) indicated their staff do receive such formalized training in dealing with the gang problem. Which means, of course, that about two-thirds of these same juvenile correctional facilities do not provide such staff training. A new published report (Knox, Mc Currie, and Tromanhauser, 1991) shows how juvenile institutions from the 1990 survey would prefer to deal with the gang problem. Analysis of this gang training factor for the present 1991 survey reveals that several factors account for significant differences. Table 5 shows that long term facilities are much more likely to report such formal gang training. Similarly, those that are community-based are the least likely to provide such formal gang training as seen in Table 6.

TABLE 5



FREQUENCY DISTRIBUTION FOR

TYPE OF JUVENILE CORRECTIONAL FACILITY

(SHORT TERM, LONG TERM, OTHER)

BY FORMAL GANG TRAINING



Do Their Staff Receive

Formal Gang Training?

**********************

NO YES

Type of Facility: ********** *********

Short Term 95 28

Long Term 68 53

Other 7 5



Chi square = 12.4, p = .002







TABLE 6



FREQUENCY DISTRIBUTION OF WHETHER THE JUVENILE

FACILITY IS COMMUNITY BASED

BY FORMAL GANG TRAINING



Do Their Staff Receive

Formal Gang Training?

**********************

NO YES

Whether Community-Based: ********** ********

Not Community-based 71 54

Community-based 98 34



Chi square = 8.67, p = .003



Closely paralleling the 1990 survey, the present 1991 survey again showed that about half of these juvenile correctional facilities report that racial conflicts are a problem among juveniles confined in their facilities (51.9%, N = 137). But again, like the gang problem, this is not a problem for which ACA standards exist for either juvenile or adult correctional facilities in spite of the scope and extent of the problem and its potential liabilities. The prominent under coverage of these two issues, and the significant relationship between racial conflict and the gang problem, in ACA publications might suggest they are taboo topics nationally.

A factor previously identified as being significantly associated with racial conflict is that of overcrowding. Here again overcrowding for all facilities survey significantly differentiated racial conflict (p < .05). Among those facilities that were residential (eliminating non-residential programs), again those facilities with overcrowding are more likely to report a problem of racial conflict (Chi square = 4.98, p = .02).

Outside of the 1990 survey of juvenile corrections and the 1991 survey of adult prison wardens (see Knox, 1991), no national comparative research has been reported on gangs and how correctional agencies "handle the gang problem" since the time of Camp and Camp (1985). The Camp and Camp (1985) study was actually conducted in 1983 and involved the state itself as the unit of analysis, aggregating all adult correctional agencies by state into a single observation for ease of data collection and reporting. Obviously, such a research method has many flaws, most noticeably that of the "aggregation problem", but still it was a very important contribution being the first federally funded research project for a national comparative analysis. Only in the summer of 1991 did the National Institute of Justice undertake a similar assessment of gangs in corrections research initiative. The current NIJ research is intended to be of practitioner value and is therefore being conducted by ACA. Research such as the present 1991 survey is apparently regarded in some federal criminal justice research funding agencies as "academic research" having very limited value to practitioners; as if social policy should be developed not from social science, but from hunches, intuition, and having no grounding in the literature or wider body of social science theory.

We are pleased in this regard, then, to be able to say our viewpoints in this analysis are truly unblemished with regard to "letting the chips fall where they may". If we were to include such a traditional "acknowledgements" section, it would probably have to read something like this: "we want to thank federal funding agencies and private foundations for not contributing one cent towards this research initiative, because it has truly allowed us to pursue intellectual honesty and an open investigation of the problems, not encumbered by such fiscal relationships or concerns about subsequent funding". In fact, we do consider this a "practitioner" oriented research initiative. But social science need not be independent of such practical concerns either.

The textbook on gangs recently published by Knox (1991) compared the Camp and Camp (1985) strategies for dealing with gangs in corrections with juvenile facilities. The present 1991 survey of juvenile corrections has also replicated the Camp and Camp (1985) gang control strategy research. As shown below in Table

7, the "case by case" method is the most frequently cited strategy for dealing with gangs in juvenile correctional facilities. Genuinely, this "case by case" method simply means a catch all category of something approximating "deal with it as it arises". Or deal with it on an individual basis.

As seen in Table 7 below, transferring gang members is the second most frequently cited method for dealing with the gang problem. Intercepting communications presumably relates to a variety of internal and external informational processes affecting the facility. It was interesting to see that "balancing the number of rival gang members living in the same unit" enjoyed such a prominent position in the rank ordering of methods for dealing with gangs in juvenile correctional facilities.

TABLE 7



FREQUENCY DISTRIBUTION OF METHODS FOR DEALING WITH

THE GANG PROBLEM IN JUVENILE

CORRECTIONAL AGENCIES





Method of Gang Control: N

Case by case 130

Transfer 59

Intercept Communications 54

Balance the number of

rival gang members

living in the same unit 37

Isolate Leaders 34

Lockdown 25

Displacing members to

different facilities 28

Ignoring their existence 16

Segregation 19

Use of informers 19

Meeting with gang leaders

on "as needed" basis 23

Prosecution 19

Joint meetings between

various gang leaders 13

Coopting others to

control gangs 9

Infiltration 3



What this "balancing" option is often called is the "set off" principle, meaning in a particular ward, cell house, or living unit, provide a "balance of power" by basically putting in just as many "crips" as "bloods", essentially a "balance the opposition" under some conception of the famous divide and conquer thinking. As a way of dealing with the gang problem, however, this may not be very well thought out at all. In fact, such a strategy could conceivably increase the gang problem and induce conflict, including violence, within a juvenile correctional setting.

Let the record reflect the social scientific truth here: there has been absolutely no evaluation research reported anywhere regarding which, if any, of these gang control methods is effective, counter-productive, or having "null effect". Those working in juvenile corrections, as elsewhere who have to face the gang problem, are simply advised to "get the best facts". And the best facts are not necessarily from other practitioner points of view. The best facts are those that are based on hard research and an appraisal of the social science literature. Unfortunately for juvenile and adult correctional administrators today, the current federal gang research funding initiative gives no attention to evaluating these problems.

But the literature is suggestive to correctional administrators. The study by Brobrowski (1988) concluded that the rise of "gang nations" (confederations of previously oppositional gangs, e.g., "folks" versus "peoples" in Illinois prisons) was due entirely to a misdirected and misguided prison policy of allowing gang members to meet as groups, recognizing them as groups, and official negotiation with their gang leaders. It is fair to alert the reader that there is more than some level of legal liability implied by the use of some of these methods of gang control from a prisoners rights viewpoint. Mostly, if taken to its extreme logical conclusion the tendency to want to "do what others are doing about the problem" runs the serious risk that if the others truly are doing the wrong thing there is more than some social policy jeopardy represented by simply imitating or adopting procedures with unknown possible long term effects. They are unknown effects because none of these strategies have been evaluated.

The penal sanction for juveniles is in many respects viewed as the placement of last resort in the juvenile justice process. However, it bears no necessary relationship to what is most needed. As illustrated before in this report, most of these juvenile facilities continue to be located in rural areas far from the neighborhood of the family of orientation for these youths. Thus, a question in the survey sought to ascertain what proportion of the youths, in the opinion of the facility superintendent, could benefit more from a community-based program. Some 14.5 percent (N = 32) felt that less than 5 percent of their youths could benefit more from a community-based program. Some 28.5 percent (N = 63) felt this might help from between 5 to 25 percent of their youths. Some 23.5 percent (N = 52) felt it could benefit from 25 to 50 percent of their youths. And the single largest response group (33.5%, N = 74) felt it could benefit more than half of the youths in their facility.

One of the questions on the survey was based on the growing problem of sexually transmitted diseases within correctional institutions and inmate sexual behaviors. It asked "what percentage of the juveniles would you estimate are homosexual?". The data show a range between a low of zero (15.7%) and a high in one case of 99 percent; which happened to be a sex crimes treatment program. Eliminating the straggler case which reported the high of 99 percent homosexual leaves a range between zero and 35 percent. Some 50.8 percent reported a factor of five percent or more of their juveniles as homosexual. The mean was 5.5 percent for this factor.

A political-philosophical question was also included and it read "Do you feel that the juvenile justice system is too lenient with serious juvenile offenders?". Some 62.1 percent indicated "yes". Just over a third (37.9%, N = 99) rejected this notion by responding "no".

The respondents were also asked to rank order six options in terms of their ascending potential effectiveness as goals for juvenile corrections. They were asked to rate these six goals of juvenile justice: rehabilitation, deterrence, incapacitation, reintegration, reconciliation, and restitution. They were further asked to rate them in ascending order, where 1 = most effective, and 6 = least effective.

Rehabilitation received the highest ratings with 56 percent rating it number one (with a score value between 1 thru 6, where 1 indicates the most effective goal for juvenile corrections, the mean was 1.7). Deterrence was rated as number one (most effective) by 24.6 percent of the respondents (mean = 3.1). Reintegration was rated most effective by 12 percent (mean = 3.1). Incapacitation was rated most effective by 6.7 percent (mean = 4.8). Restitution was rated most effective by 4.7 percent (mean = 3.7). And reconciliation(7) was rated as most effective by only .4 percent (N = 1) with a mean score of 4.

The recent work of Jerome G. Miller (Last One Over the Wall: The Massachusetts Experiment in Closing Reform Schools) noted how rehabilitation as a goal is virtually impossible and extremely untenable from a recidivism evaluation point of view for traditional state training schools. However, among the N = 55 state training schools that ranked the effectiveness of rehabilitation as a goal, some 65.5 percent ranked it as number one: the most effective goal. And some 51.1 percent of the State Juvenile Correctional Centers (N = 45) ranked it number one.

An effort was made to ascertain the racial distribution of the youths confined in these facilities. The respondents were asked to complete four separate estimates for the percentage of youths who were: white, Black, Hispanic, and other. This showed that for whites the percentage ranged from a low of zero percent to a high of 100 percent. For Blacks this ranged from zero to 99 percent. For Hispanics this ranged from zero to 100 percent. And for others (typically Native American Indians or Asians) this ranged from a low of zero to a high of 100 percent. Overall, the mean percentages were: 42.4 percent white, 41.9 percent Black, 11.9 percent Hispanic, and 3.1 percent other(8).

Just over a third of these juvenile corrections administrators completing the survey (35.1%, N = 94) reported that their facilities make use of literacy-training volunteers.

Three measures sought to ascertain how many members of the clergy (religious leaders, chaplains, priests, etc) were available to the youths in these facilities. We must assume that this item was problematic and misinterpreted by the respondents. Because otherwise we must assume there is a very minimal official commitment to the extent of religious programming for confined youths in America. There were a total of only 46 full-time clergy represented as employed in these facilities. There were an additional 46 part-time clergy. And there were 290 religious volunteers reported(9).

Most of these facilities reported that their facility provide the youths with the opportunity to exercise in an equipped gymnasium at least once a week (80.7%, N = 218).

Many of these facilities (61.2%, N = 164) also report that their facility makes use of a "point system" for behavioral modification.

A question directed at the family's of these confined juveniles read "How would you rate the quality of degree of supportiveness in the families to which the youths in your facilities must ultimately return?". The response mode continuum allowed for ratings between a low of zero (not supportive) to a high of ten (very supportive). It is of some interest to note that not a single respondent indicated the highest level of "very supportive" (e.g., 10). Rather half of the respondents (50.4%, N = 136) estimated a supportiveness level of three or below.

The cost factor of juvenile corrections was also measured here in terms of the daily cost for one youth. This ranged from a low of $16 to a high of $300 in terms of daily costs(10). The mean was $102.24 for the N = 238 respondents who provided this information. It is fair to assume that some significant variation exists with regard to the cost of care factor when controlling for type of facility. For example, an analysis of variance test showed a significant difference (F = 6.04, p = .01) in comparing the costs in terms of whether these were community-based facilities. The average cost for community based facilities was $94 compared to $110 for those facilities that were not community based. Community based juvenile facilities therefore reported significantly lower daily costs. No significance emerged in contrasting short term and long term facilities or whether the facility was located in rural or urban areas.

Slightly under two-thirds of these facilities (63.6%, N = 173) reported that they have an Advisory Board which includes community members. Which is to say, that over a third (36.4%, N = 99) do not have this type of community input, oversight, or policy/advisory involvement.

The staffing pattern was also measured in terms of how many full-time staff members are males and females. For females this ranged from a low of 0 to a high of 256. For males this ranged from a low of zero to a high of 541.

The average length of stay for juveniles in these facilities was also a self-reported factor by the facility administrator. In one case, however, the responding institution was able to produce a computer printout of the average length of stay which was included along with the survey response. Overall, regarding length of stay in these facilities, this variable ranged in terms of days between a low of five to a high of 720 days. The mean was 185.9 days.

These are apparently youths estimated to have been, to a large extent, sexually abused as children according to the present data. The range of estimates for what percentage were sexually abused as children ranged from a low of one percent to a high of 100 percent. In fact, in over half of these institutions the sexual abuse estimate was 30 percent or higher. In over a third of the facilities the estimate was 50 percent or higher. The mean was 37.8 percent.

We were interested in to what extent the youths in these facilities were actually receiving portable skills training that might be useful in terms of the primordial stated goal of rehabilitation. We asked if they were provided with on-site training in the following occupational skills: carpentry, welding, bricklaying, plumbing, auto repair, and computer programming(11). Some 88 reported providing carpentry training. Some 73 reported computer programming training. Some 51 reported auto repair. Some 40 provided welding training. Some 27 provided bricklaying training. And some 26 provided training in plumbing.

The superintendents and directors of these juvenile correctional facilities were also asked to estimate what percentage of the youths in their facility were adjudicated for drug-related offenders. This ranged from a low of zero percent to a high of 100 percent. However, over half of the sample indicated that this factor of "drug related" offenders in their population was 30 percent or higher. Over a third indicated a this factor was 45 percent or higher. The mean or average was 36.4 percent for the N = 264 facilities responding to the question.

Without defining it, but obviously with general child abuse being more encompassing than child sexual abuse, the respondents were also asked to estimate what percentage of the juveniles in their facility have a history of child abuse. This showed, again, a range between zero to a high of 100 percent. However, 60 percent of these juvenile correctional facilities reported a level of 50 percent or higher. And a fourth of the respondents indicated a level of 80 percent or higher. The mean or average was 52.5 percent for those responding to this question (N = 267).

The security level was also reported by these juvenile correctional facilities and is shown in Table 8.

TABLE 8

DISTRIBUTION OF SECURITY LEVELS

N %

No security 47 17.3

Maximum security 52 19.2

Medium security 75 27.7

Minimum security 97 35.8

According to data from Miller and Ohlin (1981: p. 460) few strong differences emerged in comparing secure and nonsecure juvenile correctional settings by the views of staff and confined youths on program services/characteristics. While limited to Massachusetts, however, the Miller and Ohlin (1981) research was very important in suggesting strategies of advocacy for greater juvenile services such as aftercare. However, our own data do indeed show some significant differences in comparing the level of security of the facility by program services and program characteristics which tend to suggest, as might be expected from theory, that generally the more secure facilities are more problematic.

First of all, security level is a factor that is significantly related to whether the juvenile correctional facility is accredited by the American Correctional Association (ACA). As seen in Table 9, the maximum security facilities are less likely to be accredited by ACA; and, vice versa, the minimum security facilities and "no security" or open facilities are more likely to be accredited by ACA.

TABLE 9

FREQUENCY DISTRIBUTION OF WHETHER THE FACILITY

IS ACCREDITED BY THE AMERICAN CORRECTIONAL ASSOCIATION

IN TERMS OF FACILITY SECURITY LEVEL



Security Level of The Juvenile Facility

None/Open Maximum Medium Minimum

********* ******* ****** *******

ACA Accredited? NO 19 34 53 51

YES 24 17 21 45

Chi square = 11.4, p = .01







Another finding of some import is that there is a direct and proportionate relationship between security level and use of a "point system" for behavioral modification. The higher the security level, the greater the likelihood of using a point system for behavioral modification in the facility administration is the finding that emerges here. As seen in Table 10, this is a difference in terms of using a "point system" for behavioral modification from 76.4 percent among maximum security youth facilities to 45.6 percent for no security or open facilities. Security level shows a significant relationship to the use of a "point system" for behavioral modification.









TABLE 10



FREQUENCY DISTRIBUTION OF "POINT SYSTEM" FOR

BEHAVIORAL MODIFICATION BY

FACILITY SECURITY LEVEL



Security Level of The Juvenile Facility

None/Open Maximum Medium Minimum

Report use of a ********* ******* ****** *******

"point system" for

behavioral

modification? NO 25 12 26 39

YES 21 39 48 56



Chi square = 10.3, p = .01





The analysis undertaken here on security level at a modest level of effort reveals this factor to be significantly related to a number of different factors. Figure 1 below summarizes these factors shown to be significantly differentiated by security level. In Figure 1, security level has been recoded to reflect a dichotomous variable of "low" (no security, open, minimum) versus "high" (medium, high security).































FIGURE 1



FACTORS SIGNIFICANTLY RELATED TO LOW/HIGH SECURITY

LEVELS IN JUVENILE CORRECTIONAL FACILITIES



Factor Description Nature of the Relationship Chi-square Prob.

Suicide attempts(12) Higher the security level,

higher the suicide attempts. 15.8 <.001



% Drug Related(13) Higher the drug-related offenses,

the lower the security level. 4.22 .04



Clergy Access(14) Higher the security level,

the higher the clergy access. 8.31 .004



Gang Damage(15)

Higher the security level,

the higher the gang damage. 5.44 .02













FIGURE 1 CONTINUED



Factor Description Nature of the Relationship Chi-square Prob.

Gang Assaults(16) Higher the security level,

the higher the assaults. 10.46 .001



Recidivism(17) Higher the security level,

the higher the recidivism. 26.60 <.001



Community-based Community-based facilities

have lower security levels. 49.1 <.001



Overcrowding Higher the security level,

the higher the overcrowding. 14.5 <.001



ACA accreditation ACA accredited facilities have

the lower security level. 10.1 .001



White gangs present Higher the security level, more

likely white gangs present. 6.90 .009



Use "point system" Higher the security level,

higher use of "point system". 6.29 .01



Tear Gas Policy(18) Higher the security level,

higher use of tear gas. 13.27 <.001





These juvenile correctional administrators were also asked to separately estimate, among the juveniles in their facility, what percentage have religious beliefs and secondly what percentage actually practice those beliefs. For the estimate of those juveniles having religious beliefs this ranged from a low of zero to a high of 100 percent; with a very normal distribution, implying that 50 percent of the facilities estimated 50 percent or more of their youths had religious beliefs; with a mean of 54.5 percent. However, the ranges were more constricted with regard to estimates of practicing these religious beliefs; ranging from a low of zero percent to a high of 95 percent; with a mean of 23.1 percent. In fact half of the facilities estimated 15 percent or less of the youths practiced their religious beliefs. This is consistent with the earlier findings on the less than conspicuous presence of members of the clergy in these facilities.(19)

When asked how many suicide attempts among juveniles have occurred in these facilities during the last twelve month period, this showed a range between zero to a high of 175(20). The important finding is that just under half of the facilities (46.1%) reported one or more such attempts during the preceding year. Over ten percent indicated five or more such attempts in their facility during the last twelve months.

Another variable measured on the survey involved the estimated percentage of sexual or physical abuses between juveniles that go unreported in these facilities. This ranged from a low of zero (35.6%) to a high of 85 percent. Slightly under half of the facilities indicated this factor was five percent or more. And over a tenth indicated this was twenty five percent or more. The mean or average was 8.1 percent.

Only a fourth (25.7%, N = 67) approved of the use of the death penalty for juveniles convicted of capital offenses. This general opposition to the death penalty for juveniles was further revealed when the respondents were asked "in your opinion, what should be the absolute minimum age for persons receiving the death penalty?". This showed a low range of 8 years old to a high of 150 years old; with several indicating 100 years or above --- an explicit further declaration of their opposition, perhaps to the death penalty itself. The fact that a number refused to answer this and wrote in such comments also underscores this trend. Still, over a third (36.2%) agreed to using the death penalty at ages of 16 or below.

Respondents were also asked what percentage of the juveniles in their facility would be considered "habitual juvenile offenders" (having three or more previous adjudications for offenses which if prosecuted as an adult would be serious felonies). The range was from a low of zero percent to a high of 100 percent. This showed a very normal distribution, with about half of the facilities indicating a level of 50 percent, etc. The mean or average was 48.8 percent.

Only 9.6 percent of these facilities were authorized to use chemical agents to quell disturbances.

Among the facilities with female juveniles (N = 106), when asked what percentage of these youths have had children, nearly half (47.2%) indicated ten percent or more have had children. The range was between zero to 100 percent, with a mean of 18.4 percent overall.

SUMMARY AND CONCLUSION

The value of this preliminary report is the type of information it provides on specific facility factors and characteristics of the confined youths in the United States. As Krisberg and Schwartz said "little is known about the personal characteristics of the nearly half million children who are annually confined in public juvenile correctional facilities" (1983: p. 362). Specific services, programs, and characteristics of these facilities are also not well documented in the literature. Thus, the present study sought to focus on various aspects of juvenile corrections which have surfaced in the literature.

The promise made to the respondents is that we would provide them with a preliminary report. We believe we have fulfilled the promise of rapid feedback in as much as this report is being issued two months after the research project was implemented. Any comments, suggestions, or criticisms would be welcome. Once again, we want to thank the 274 superintendents/directors who took the time to respond to this survey.

Further analysis and follow-up research on some of the issues raised in this report is certainly intended by the author who is currently working on the first full textbook on juvenile corrections.









REFERENCES

Bartollas, Clemens and Christopher M. Sieverdes

1982 "Juvenile Correctional Institutions: A Policy Statement", Federal Probation (46): 22-26.

Bessette, Joseph M.

1989 Children in Custody, 1975-1985. Census of Public and Private Juvenile Detention, Correctional, and Shelter Facilities. May, 1989. U.S. Department of Justice. Bureau of Justice Statistics. Washington, D.C.

Blomberg, Thomas G. and Sherry L. Caraballo

1979 "Accelerated Family Intervention in Juvenile Justice", Crime and Delinquency (25)(4): 497-502.

Breed, Allen F.

1990 "America's Future: The Necessity of Investing in Children", Corrections Today (February): pp. 68-72.

Camp, George M. and Camille Graham Camp

1985 Prison Gangs: Their Extent, Nature and Impact on Prisons. Washington, DC: U.S. Government Printing Office.



Carter, Robert M.; Daniel Glaser; and Leslie T. Wilkins

1972 Correctional Institutions. Philadelphia: J.B. Lippincott Company.

Colson, Charles and Daniel Van Ness

1989 Convicted: New Hope for Ending America's Crime Crisis. Westchester, IL: Crossway Books.

Feld, Barry C.

1977 Neutralizing Inmate Violence: Juvenile Offenders in Institutions. Cambridge, Mass: Ballinger Publishing Co.

Frasca, Michael A.

1989 "Don't Blame The Moon", Sky & Telescope (Oct), p. 340.

Grissom, Grant R. and William L. Dubnov

1989 Without Locks and Bars. New York: Praeger.

Henderson, James D. and Rich Phillips

1989 "Firearms, Gas and Use of Force", Corrections Today (December): pp. 126,128,130, 132.

Hodges, Ben and David Mahrer

1981 "Competency Based Education: Implications for Education in Juvenile Corrections Settings", Journal of Correctional Education (32)(3)(Sept): 14-19.

Jacobs, James

1974 "Street Gangs Behind Bars", Social Problems (21)(8): 395-408

1977 Stateville. Chicago: University of Chicago Press.

Knox, George W.

1989 "Family Services in Corrections", The State of Corrections, Proceedings of the Annual Conferences of The American Correctional Association, pp. 179-182.

1990 National Survey of Juvenile Corrections. Unpublished report. Chicago State University.

1991 An Introduction to Gangs. Berrien Springs, Michigan: Vande Vere Publishing.

Knox, George W.; Thomas Mc Currie; and Edward Tromanhauser

1991 "Training Issues Regarding Gangs in Juvenile Correctional Institutions", Journal of Correctional Training (in press).

Knox, George W. and Edward Tromanhauser

1991 A Survey of 1,861 Confined Juveniles on Health Risks. (forthcoming publication: not yet approved for release).

Krisberg, B. and I. Schwartz

1983 "Rethinking Juvenile Justice",l Crime and Delinquency (July): 333-364.

Lane, Michael

1989 "Inmate Gangs", Corrections Today (51)(4)(July): 98, 99, 126-128.

Linster, Richard L.; Pamela K. Lattimore; and Christy A. Visher

1990 Predicting the Recidivism of Serious Juvenile Offenders National Institute of Justice Discussion Paper, Washington, D.C.

Maleson, Franklin G.

1981 "Dilemmas in the Evaluation and Management of Religious Cultists", American Journal of Psychiatry (138)(7)(July): 925-929.

Mardon, Steven

1991 "On Board, Not Behind Bars", Corrections Today (February): 32-38.

Miller, Alden D. and Lloyd E. Ohlin

1981 "The Politics of Secure Care in Youth Correctional Reform", Crime and Delinquency (27)(4)(Oct): 449-467.

Miller, Jerome G.

1991 Last One Over the Wall: The Massachusetts Experiment In Closing Reform Schools. Columbus, OH: Ohio State University Press.

Miller, Stuart J.

1971 Post-Institutional Adjustments of 443 Consecutive TICO Releases, Ohio State University, Ph.D. dissertation.

Nidorf, Barry J.

1989 "Community Corrections: Turning the Crowding Crisis Into Opportunities", Corrections Today (Oct): 82-88.

Oberst, Margaret

1988 Inmate Literacy Programs: Virginia's 'No Read, No Release' Program. Lexington, KY: Council of State Governments.

Palmer, T. and Wedge, R.

1989 "California's Juvenile Probation Camps: Findings and Implications", Crime and Delinquency (35): 234- 253.

Pasternack, Robert; Robert Portillos; and Henry Hoff

1988 "Providing an Appropriate Education to Adjudicated and Incarcerated Juvenile Delinquents: The Challenge to Correctional Education Administrators", Journal of Correctional Education (39)(4)(Dec): 154-159.

President's Commission

1967 Task Force Report: Corrections. The President's Commission on Law Enforcement and Administration of Justice. Washington, D.C.: U.S. Government Printing Office, pp. 141-149.

Rakis, John

1984 "Suicide Prevention Measures in Urban Detention Facilities Throughout the United States", Journal of Prison & Jail Health (4)(2)(Fall): 91-95.

Rowan, Joseph R.

1989 "Suicide Detection and Prevention: A Must for Juvenile Facilities", Corrections Today (August): Pp. 218,220,226.

Shannon, Michael J.

1987 "Officer Training: Is Enough Being Done?", Corrections Today (April): pp. 172-175.

Short, James F., Jr. and Fred L. Strodtbeck

1974 Group Process and Gang Delinquency. Chicago: The University of Chicago Press.

Smith, Melinda

1990 "New Mexico Youths Use Mediation to Settle Their Problems Peacefully", Corrections Today (June): 112-114.

Snyder, Howard N.

1990 Growth in Minority Detentions Attributed to Drug Law Violators. OJJDP Update on Statistics. U.S. Department of Justice, Office of Justice Programs, Office of Juvenile Justice and Delinquency Prevention. Juvenile Justice Bulletin. Washington, D.C. (March).



Steinhart, P.

1989 "Pull of the Moon", Audubon (91): 32-34.

Stephens, Ronald D.

1988 Gangs in Schools: Breaking Up is Hard to Do. National School Safety Center, Pepperdine University: Malibu, CA. Ungerleider, J. Thomas and David K. Wellisch

1979 "Coercive Persuasion (Brainwashing), Religious Cults, and Deprogramming", American Journal of Psychiatry (136)(3)(March): 279-282.

van Wormer, Katherine

1990 "Norway's Juvenile Justice System: A Journey Back in Time", Corrections Today (August): pp. 106-113.



1. National Gang Crime Research Center.

2. Previous research by the President's Commission on Law Enforcement and Administration of Justice (1967), Task Force Report: Corrections reported in Carter, Glaser and Wilkins (1972: p. 57) involved a sample size of N = 220 "state-operated juvenile institutional facilities in all States, Puerto Rico, and the District of Columbia" for a total of 52 such "jurisdictions".

3. This research was undertaken using the juveniles themselves as the unit of analysis in an anonymous "self-report" survey for 45 short and long term juvenile facilities in five states. The research agreement does not allow dissemination of the research report until the final version has been approved by the funding source. Thus, we can only relate a general trend here, rather than specific data.

4. By multiplying the population count by these percentages of gang affiliation, a total of N = 4,736 male gang members and N = 514 female gang members would be facility administrator estimate for total gang force strength in this sample.

5. Earlier work used the phrase "socialized delinquent" as being ideal candidates for long-term secure care facilities and made specific reference to gang members: "the ideal candidate is the older socialized delinquent: an extroverted gang member who is at least 16 years of age" (Grissom and Dubnov, 1989: p. 60). To speak of the over 16 year old gang member as a preferred correctional client is an anomaly in the literature.

6. Through the Gang Crime Research Center at Chicago State University there is comparative information on how adult prison wardens and others respond to this same question and others reported here.

7. See the Colson and Van Ness (1989) concept of restorative justice.

8. These mean percentages do not add to 100 percent because of some facilities not providing full distributions.

9. The respondents were asked to indicate the "number" of such categories of clergy in their facilities. Someone who simply placed a "check mark" in one or other of the three categories was treated as non-responsive, or as missing data. A secondary analysis is intended for this item. However, these results are not complimentary towards the rights of religious worship of confined juveniles.

10. Mardon (1991: p. 33) reports a correctional program, a "marine/outward bound" type of program, that averaged $36 per day per client. Clearly, some of those at the lower end of the cost spectrum were of this type of correctional program variety: e.g., day centers, etc.

11. These are the vocational skill training areas often cited in public relations documents and some of the juvenile correctional literature (see Grissom and Dubnov, 1989: p. 62).

12. Suicide attempts was recoded zero if no suicide attempts were reported in the last twelve months; suicide attempts was recoded one if one or more suicide attempts were reported in the last twelve months.

13. "% drug related" was coded as zero if less than 30 percent of the youths in the facility were estimated to have been adjudicated for drug-related offenses. It was coded as one if 30 percent or higher were adjudicated for drug-related offenses.

14. "Clergy access" was coded as zero if there were no full time, part time, or volunteer religious staff/volunteers assigned to the facility. Clergy access was coded as one if one or more full time, part time, or volunteer clergy members were reported in use at the juvenile facility.

15. "Gang damage" was coded as zero if there was no level of reported damage to government property by gang members in these juvenile facilities. It was coded as one if the level of involvement on a scale of one through ten was one or higher.

16. "Gang assaults" was coded as zero if no such serious injuries occurred in the juvenile correctional facility from attacks/fights/assaults involving gang members within the last one year period. It was coded as one if one or more such gang assaults occurred during this period.

17. "Recidivism" was coded as zero if less than or equal to eight percent of the juveniles were reported as having previously served time in the facility. It was coded as one if greater or equal to ten percent had served time in the facility.

18. "Tear gas policy" was coded as one if the facility staff were authorized to use chemical agents to quell disturbances; it was coded as zero if they were not authorized to do so.

19. Future research will seek to clarify this issue considerably and provide some comparison with adult corrections data as well.

20. One respondent indicated 175 such attempts. This stood out mathematically and may have represented a psychiatric facility.