Appendix C

Sources and Qualifications of Data from the Inventory of Mental Health Services in Juvenile Justice Facilities

CMHS obtained the universe of names and addresses of facilities from the 1997 Census of Juveniles in Residential Placement from the Office of Juvenile Justice and Delinquency Prevention (OJJDP), U.S. Department of Justice. The OJJDP database contained a broad diversity of facility types. Facilities ranged from correctional facilities solely housing youthful offenders, such as training schools, to places such as shelters that housed not only offenders, but also young people uninvolved in the juvenile justice system, such as those who had been neglected and abused, as well as those with mental, emotional, or behavioral health problems. Although shelters, for example, would more appropriately be called juvenile residential placements than juvenile justice facilities, for consistency in this chapter all facilities and places where youth involved in the juvenile justice system are housed are referred to as juvenile justice facilities.

The universe received from OJJDP included 3,578 facilities. However, this chapter is based on 2,798 facilities. CMHS deleted the following: all substance abuse, independent living, and foster care facilities; facilities that reported no offenders and no facility type; facilities that were found during telephone follow up to be closed or out of scope, or could not be found and were believed to be closed; and facilities with fewer than three juveniles.

On the basis of recommendations by the advisors to the survey (figure C1), two survey forms were developed. According to these experts, the largest single proportion of places called juvenile justice facilities are group homes and halfway houses; however, a relatively small proportion (13 percent) of youth in juvenile justice facilities are in these homes and houses on one day. These are principally small community residences, like homes, privately owned or operated. They have few onsite services and rarely any staff to respond to surveys. Therefore, the form designed for larger, more complex facilities was not appropriate. A shorter version of the long form was developed and sent to group homes and halfway houses.

Facilities were sent the short form or long form on the basis of information about facility type received from OJJDP. Facilities receiving CMHS forms were also asked to self-classify into one of seven facility types. Where a respondent chose "other," CMHS followed OJJDP's convention for assigning type. Where a facility received the long form but then reported a facility type (group home or halfway house) that qualified it for the short form, or vice versa, the facility was moved within the database to be analyzed with its self-described facility type. Although most items on the forms were the same, some differed or were not included on the short form. For analytical purposes, where items were the same, data from both forms were merged, as in chapter tables 18.1 through 18.3. If an item appeared on only one form, the number of responses is smaller. Tables C1 and C2 provide detail on how different denominators were derived. Data in chapter tables are generally based on the following: the entire universe (2,798 facilities), whether or not any mental health services were available to youth; those facilities providing access to at least one mental health service (2,639 facilities); all facility types (except group homes and halfway houses) that provided access to any service and answered the more detailed long form questions (see table C1, column 2, where 1,291–98 = 1,193 facilities); or group homes and halfway houses that provided access to any service and answered the different short form questions (see table C1, column 3; 881 facilities).

Forms were initially mailed to each in-scope juvenile justice facility in June 1998 for response by mail. For a second mailing to facilities that had not responded, the reference date was changed from July 15 to November 18, and the 1-month reference period was moved from July to November. In February 1999, a 1-page certified letter was sent to remaining nonrespondents. At this time, facilities were offered the option to respond by fax instead of mail. Shortly after this mailing, remaining nonrespondents were called for the collection of core data items.

The overall response rate was 69 percent. Response rates, by facility type, are in table C3. To weight the respondents to the totals, missing values were imputed based on the following variables, which were considered to be critical for providing estimates for subcategories (in consultation with OJJDP): type of facility; ownership (either public or private); type of youth in facilities (either offender only or both offenders and nonoffenders); and facility size (3 to 30 juveniles and 31 and over). Remain-ing items were not imputed.

Index | Previous | Next