III. Type of Facility
As mentioned earlier, we discovered problems with the current classification of facilities during the preliminary review of 1993 census forms. The "Type of Facility" question looks like this:
IV.A. Facility Type|
This facility is primarily a --
Mark (X) the one box that best describes the facility.
Not all facility respondents can fit their facilities into one of the predetermined types listed for this question. Others can fit their facilities into one of the predetermined types, but would prefer to call the facility something else. Still others are in facilities that have multiple programs that would fit into more than one type of facility. In addition, there are some respondents who wanted some qualifying adjective, such as "residential" or "treatment" added to the facility type to describe their facilities.
We got a variety of responses to the question, "What type of facility is this?"5
It appears from the above findings that the current facility type question and response categories are not fully capturing the range of actual facility types.
- Eleven respondents gave an answer that exactly matched the facility type we used in our allocation. These included:
- 1 (of 2) multi-purpose facilities,
- 3 (of 6) detention centers,
- 1 (of 4) training schools,
- 1 (of 2) reception or diagnostic centers,
- 1 (of 5) shelters,
- 2 (of 4) residential treatment facilities, and
- 2 (of 11) group homes.
- Nine respondents gave an answer that was included in the list of facility types used for our allocation, but differed from the facility type identified for them on the previous census form and in the sample. The following list shows the allocated facility types first and the actual facility types as reported by respondents second. The respondent(s) at:
- 1 shelter and 1 residential treatment center identified their facilities as multi-purpose facilities,
- 1 multi-purpose facility identified the facility as a reception center,
- 3 halfway houses or group homes identified their facilities as residential treatment facilities,
- 2 halfway houses or group homes identified their facilities as residential treatment centers, and
- 1 halfway house or group home identified the facility as a residential treatment program.
- Six respondents gave answers similar to the facility type used in our allocation, but added some qualifier(s). These included:
- Secure detention (2),
- Emergency Shelter (1),
- Secure Diagnostic and Treatment Center (1),
- Non-secure Detention Group Home (1),
- Probation Camp or Drug Treatment Boot Camp (1), and
- Juvenile Shelter Hall (1).
- Six other respondents initially gave us a reply that was not included on our list of facility types, but after being read the list we had, could easily place their facilities into one of the categories.
- Short-Term Offenders Program (could be classified as a halfway house),
- Program for Runaway and Homeless Youth (could be classified as a shelter),
- Behavioral Assessment Center (could be classified as a reception or diagnostic center),
- Adolescent Assessment Center (could be classified as a diagnostic center), and
- Residential and Educational Center (could be classified as a group home).
- Community Residential Placement Program (could be classified as a halfway house or group home).
- Three respondents initially gave us a reply that was not included on our list of facility types, but after being read the list could, if forced, place their facilities into one of the categories. These respondents were not comfortable with the classification but wouldn't object to it either.
- Juvenile Rehabilitation Institution (could fit into training, but says the term is outdated),
- Residential Care Facility (the ranch and group home classifications fit loosely, but not exactly), and
- Private Wilderness Facility (could classify as a ranch if there was no other choice).
- One respondent initially gave us a reply not included among the facility types used in our allocation, and after being read the list, could not place his facility into one of the categories.
- Short-Term Offenders Program
- There were respondents who wanted to check more than one category on the list of facility types. Some of the facility types were already listed, while others were not. In general, these facilities had multiple programs within the same facility. These facilities included:
- Long-Term Secure Treatment (the respondent would need a combination of training school and reception/diagnostic center),
- detention and training,
- shelter, residential treatment, and detention,
- residential treatment, diagnostic, boarding homes, and group homes, and
- emergency shelter, drug and alcohol program, independent living program, group home facilities (entire facility is classified as a residential treatment facility).
Attributes of Facilities
We asked each respondent to describe the attributes that fit his/her facility's type. The specific characteristics respondents used to classify facilities are included in Attachment C. Respondents from some types of facilities use the facility type designated on their state operating licenses. State licensing requirements may therefore be an important factor in how respondents decide what to mark for facility type. Juvenile Rehabilitation Institution and Secure Treatment Facility are two examples of facility types that are licensed in some states.
Some of the facility types have been called outdated by some respondents. One respondent from a training school (as identified in the census) says that the term "training school" is long outdated. She preferred to call her facility a "residential rehabilitation center" because this more closely fit. Others said that the term "halfway house" applies to facilities for adults, not those for children.
After compiling all of the statements from various respondents regarding facility type, we have a few suggestions for changes to be made to the questionnaire. First, there are a few facility types that may be added to the current facility type list: "Boot Camp", "Residential Treatment Center" (RTC) and/or "Residential Treatment Facility" (RTF). The reason we have both RTC and RTF is that in at least one state, RTC and RTF mean different things. In New York, an RTF is a mental health facility, while RTC refers to non-mental health treatment programs.
Also, there should be a method by which respondents can indicate the multi-purpose composition of their facilities. In our study, we identified some facilities with multiple programs, falling under separate categories, such as:
One approach to dealing with this issue is to send multiple forms and ask the respondent to fill out a separate form for each program within the same facility. Some of the facilities described above did receive multiple forms, some did not. Filling out multiple 13-page forms becomes burdensome for many respondents; they would be completing budget and personnel information multiple times. Some respondents cannot separate out these aspects of their facilities. Others currently receive multiple forms, but fill out only one, trying to fit in as much information as they can about all programs. As they do so, however, they become frustrated trying to decide which programs to include for certain questions.
- A facility that has a detention center and a boot camp (the respondent would classify the boot camp as a training school if he had to choose from the current list). The programs had different security levels and different types of children (one had all adjudicated children, the other had all detained children). The respondent became frustrated when he realized that the questionnaire only allowed him to respond for one program at the facility.
- A multi-purpose facility that offers a detention center, residential treatment center and group homes as three separate programs. The programs are similar in having about 70 children each, but differ in terms of the age, sex and types of juveniles as well as in security levels. As a result, the respondent had a difficult time deciding how to fill out the facility-oriented questions which assume that there is just one undifferentiated program. We feel that this type of respondent should be given the opportunity to represent the varying aspects of the different programs in order to give more accurate data.
- A facility complex that offers distinct programs with different security levels to different types of children. Some of these programs are even run by separate organizations. One is these is not a program but an ad hoc arrangement due to overcrowding: a joint administrative-service area in the complex not intended to provide living quarters currently holds more than 20 detention children in the gym on cots. There is a question whether these distinct programs and the overflow area should be classified as one facility or as separate facilities in the same building.
- A multi-purpose facility that includes a residential treatment program, several diagnostic programs, a group emergency foster care program and a family support center on a common campus, as well as several boarding homes in the local community. According to the respondent, the boarding homes have minimum security while the others are medium security. The diagnostic programs have fewer adjudicated children, while the residential treatment programs have more. Children in the residential treatment program stay in the facility for 9-12 months, while those in the group emergency foster care program, on average, stay for just 60 days. The respondent wants to combine all of these programs and juveniles on one form, rather than filling out the multiple forms sent by GOVS.
- One facility respondent thought the easiest way to categorize the children in his facility was by "status" category or the reason why a juvenile is staying in the facility, even though he thought this may not be applicable to most other facilities. He can divide the children by program type, but there are slight differences in the security levels within each.
The plan to develop a new data collection instrument focused primarily on child-
level data means that only a few critical questions on facility level classification will be needed in the new form. This will greatly reduce the burden on respondents in multi-purpose facilities by eliminating the need to complete a 13-page form for each functional unit. We suggest OJJDP might consider asking the respondent near the beginning of the new questionnaire, something like, "Does this facility have one program or more than one program? Examples of programs include: shelters, detention center, group home, residential treatment facility or center, etc...." Respondents marking the "more than one" category would be asked to fill out additional blocked sections on the same form, or attachment sheets, each with a very brief list of classification questions, followed by the roster of children in that subdivision of the facility. For example, one respondent would be able to answer classification questions (e.g., security level, custodial authority, legal status of children) and roster questions for his detention center and boot camp separately.
Allowing respondents to list all of the children by program might make it easier for them to complete this task. This approach would enable the census to obtain specific information on each of the programs without adding substantial respondent burden. If this is done, there should be some definition of program given, or an algorithm developed in the questionnaire processing to determine what constitutes separate programs. This way, the respondents could fill out the questionnaire from their perspectives, and GOVS could decide afterwards (using an algorithm) how to classify the various programs within the facility.
Universe coverage is another issue related to facility type. Respondents mentioned several types of facilities or parts of facilities that may or may not currently be included in the census. Some examples are small family homes with 2-4 beds, juveniles staying in facility gyms and detention foster homes because of overcrowding and facilities that run programs that are "off campus" in several states. Since there are no instructions specifically relating to this, it is unclear whether respondents for these types of facilities receive a form, and if they do, whether they include all of the relevant children. At least four facilities had independent living programs in the community that they have been including on their census forms in the past that probably should be considered out of scope. We suggest OJJDP might want to decide which of these newly identified facility types should be considered in scope so that clear directions can be given to respondents on the types of residential arrangements to include.
An alternative to a facility type check-off response is to develop and use a list of facility criteria. Respondents would mark each attribute fitting the facility and an electronic algorithm would classify the facility. It is not clear that this very new approach would provide a better way of classifying facilities. We recommend expanding the list of facility types instead.