updated supp statement B

B. Collection of Information Employing Statistical Methods

1. Universe and Respondents Selection

The objective of this survey is to produce estimates of rates of sexual assault for juvenile correctional facilities that house adjudicated youth. To meet the goals of PREA, the sample is designed to produce facility-level estimates for a national sample of facilities. Under the Act, participation of sampled facilities is mandated, and BJS is required to publicly list any facilities declining to participate. All youth participation is voluntary.

The sampling frame will be the universe of juvenile facilities in the 2006 Census of Juveniles in Residential Placement (CJRP), conducted maintained by the Census Bureau for the Office of Juvenile Justice and Delinquency Prevention. Only adjudicated youth are of interest in this survey, so facilities with no adjudicated youth will be excluded from the sampling frame. Non-adjudicated youth will not be included in the survey because of the practicalities of gaining parental consent given short lengths of stay and pre-adjudication status. If facilities require active consent from the parent, a lead time of approximately 8 weeks is required. Non-adjudicated youth are typically not in facilities for this amount of time. Therefore, the present survey will only attempt to sample adjudicated youth where it is more practical to get parental consent for a significant proportion of the target population.

Survey interest is primarily in state facilities, but we will also sample from large non-state facilities (county, municipal, and other). Thus the sampling frame is to include state facilities with at least 10 adjudicated youth and non-state facilities with at least 121 adjudicated youth, according to the 2006 CJRP.

We next describe the sampling plan for facilities. All facilities with 90 or more adjudicated youth will be sampled with certainty. We desire to have at least one sample facility in each state. Thus, for each state that does not contain a certainty facility, a stratum consisting of only facilities from that state will be formed. If the number of adjudicated youth in such facilities is high, then only the largest facilities will be put into the state stratum. All remaining facilities will be grouped into strata without regard to state, using stratification variables such as percent female, percent Hispanic, type of facility (state or non-state), region, and facility size. Strata may vary in size, so that the number of sample facilities will also vary by stratum. Sampling of facilities within a stratum will be with probability proportional to the estimated number of adjudicated youth.

Sampling of youth within facilities will be with equal probability, except that all females will be sampled, which are a small percentage (10.4 percent) of adjudicated youth in the sampling frame.¹ Females are over-sampled because it is expected that rates of sexual violent victimization is higher than for males.

We wish to produce estimates for individual facilities, except for small facilities where confidentiality may be breached. Although it will be impossible to determine individual survey responses, if the survey-estimated abuse rate implies that there were fewer than three youth who stated that they have been abused, the youth may be at risk of being identified. Therefore, the sample is designed to be able to obtain estimates for larger facilities, where there will be more youth interviewed and a greater chance of meeting the criteria for publishing a facility-level estimate.

To further protect confidentiality and to mask the fact that the survey is solely intended to collect data on sexual assaults, a small percentage of the sample at each facility will get a questionnaire on drug and alcohol use prior to being admitted to the facility rather than on sexual assault. Thus, no one other than the youth answering the questions will know whether a given youth got the sexual assault questionnaire.

Estimates for individual facilities with more than 50 adjudicated youth will generally have reasonable-sized standard errors. In facilities with 240 or fewer adjudicated youth, all youth will be included in the survey, so that standard errors will be as small as possible. For facilities with more than 240 youth, we will select 240. With an expected 60 percent response rate and with 90 percent of the youth getting the sexual assault questionnaire, this will result in about 130 completed sexual assault interviews out of an attempted sample of 240. For an abuse rate as low as 5 percent, a 95 percent confidence interval will not contain zero (assuming a design effect of 1.5 as the result of differential nonresponse adjustment factors within a facility). Note that the response rate will vary according to whether the facility has in loco parentis for all, some, or none of the youth. With in loco parentis, or if the youth is over 18, response rates will be much higher because it will not be necessary to obtain permission from parents or guardians. Thus, somewhat fewer than 240 youth may be sampled in large facilities with all or most youth in these situations. Table 3 provides estimates of designated sample sizes by facility size. The additional sample of female youth resulting from taking all female youth are not included in the table.

Table 3. Designated sample sizes by facility size

Facility size	Number of facilities	Number of youth	Number of sampled facilities	Designated number of sampled youth	Number of completed sexual assault
10-20	111	1,626	13	190	103
21-50	183	5,671	41	1,1326	716
51-89	69	4,773	35	2,546	1,375
90-120	29	2,943	29	2,943	1,589
121-150	33	4,439	33	4,439	2,397
151-164	11	1,712	11	1,712	924
165250	45	9,137	45	9,133	4,931
251-350	30	8,990	30	7,200	3,888
351 or more	19	9,408	19	4,560	2,462
Total	530	48,699	256	34,049	18,385

2. Procedures for Information Collection

The methods proposed for use in data collection are as follows:

1. Facility Recruitment

A sample of 256 juvenile correctional facilities will be selected from a frame of federal, state, and local correctional facilities. The Census of Juveniles in Residential Placement (CJRP), 2006, conducted by the Office of Juvenile Justice and Delinquency Prevention every two years, will serve as the sampling frame. Each sampled facility will be contacted to notify them of their selection and to request participation. A contact person will be designated at each facility and study materials will be mailed to this person (Attachment 1). Facility recruiters will conduct a series of brief phone calls to collect information about facility characteristics and obtain logistic information for planning the survey visit (Attachment 8).

b. Sampling of Youth

Within eight weeks prior to data collection at a facility, the facility will provide a roster of all adjudicated youth who are currently residing there. A random sample of youth (i.e., preliminary sample) will be drawn from the roster. The facility contact person will provide periodic updates prior to the visit to the facility. The final sample will be selected after the field team arrives at the facility.

c. Parent/Guardian Permission

For those facilities requiring parent/guardian permission, procedures to contact the households will be negotiated with the facilities. Either the facility or the study contractor will send written materials containing an explanation of the study and the nature of youth involvement. Depending on the procedures that are negotiated, mail or telephone prompting of parents/guardians that do not respond to the initial mailing will be made.

d. Data Collection

A team of interviewers will visit the facility. They will ask facility staff to bring each sampled youth (with requisite permission) to a private interviewing area. The interviewer will read an assent script (Attachment 4) to the sampled youth and solicit his/her participation. The script is designed to confirm that youth understand the conditions of participation. Any youth indicating a lack of comprehension of these conditions will be excused from the interview, including those with mental health issues or impairment due to medication.

If the youth agrees to participate, the interviewer will give the youth a brief tutorial on answering questions on the touch screen computer and then allow the youth to answer survey questions in complete privacy. In order to allow youth with reading difficulties to participate, the youth will wear a set of headphones and hear the questions being read as they appear on the screen. The youth will enter his/her response by touching a button on the screen – no computer expertise is required. The computer program will randomly pick a series of questions to administer. Most youth will get the series of questions about sexual assault, however, a portion of youth will get a series of questions about alcohol and drug use instead (Attachment 1). No one but the youth will know which series of questions was asked. At the end of the questionnaire, the youth will turn the computer back to the interview. The facility staff person will escort the youth from the interview area and the interviewer will then finish the process by answering a set of debriefing questions about the interview.

In order to determine if there is any bias introduced from non-respondents, facilities will be asked to provide administrative record data for all sampled youth. This will allow researchers to compare characteristics (e.g., demographics, committing offense) of youth who participate and youth who do not participate.

e. Data Quality

Verifying the veracity of the data is complex. Under the strict regulations of confidentiality in collecting such sensitive data from human subjects, we can neither link administrative records with individual reports, nor conduct follow-up interviews with surveyed youth as a measure of consistency.

It is possible that some youth make false allegations to make the facility look bad; it is also possible that despite BJS efforts to assure confidentiality and anonymity, victimized youth will fail to report due to fear or shame.

As demonstrated in the juvenile workshop rollout, the survey does allow an analysis of variables that should be correlated with whether a youth has been victimized (e.g. does a youth report liking staff, yet also make an allegation of forceful sexual abuse by a staff member?) and whether the incident details make sense (e.g. is a male youth reporting sexual activity with a female youth in a facility that is all males?). See the Westat Final Pretest Report for additional information on measures of outliers and consistency.

BJS is very clear in all forms of representation of the data that these are allegations of sexual violence, not incidents which have been investigated or verified.

3. Methods to Maximize Response

Every effort is being made to make the survey materials clear and simple to use. The study is prepared to assist facilities to obtain parent/guardian consent. This includes conducting the mailings, using special mailing procedures (e.g., express delivery), making telephone calls to check on consent packages, obtaining verbal consent by telephone (when approved by the facility). The confidential nature of the data collected is clearly and repeatedly explained in the assent process. The NSYC questionnaire has been designed to maximize respondent comprehension and participation and minimize burden, including an easy-to-use touch-screen interface with the questions simultaneously delivered via headphones.

A Spanish version of the questionnaire will be available for non-English, Spanish-speaking respondents. Field staff from the contractor will be available to answer any questions that respondents may have, including bilingual staff who can answer questions in Spanish.

Arrangements with mental health staff at each facility, or if needed, an on-call person or some other arrangement will be made for delivery of counseling services for respondents interested in obtaining counseling services or assistance following the survey.

Despite these measures, we estimate an overall response rate of 60%, taking into account the reduced rate of participation among facilities that require parental consent. We estimate about half the sampled facilities will decline to utilize in loco parentis given the sensitivity of the survey questions. In the pilot test, about 50% of parents contacted for sampled and eligible youth were no contacts or non-responses (see Westat Final Pretest Report). This rate will vary by type of consent permitted (verbal and recorded over the phone or written and sent through the mail) by the facility. BJS will flag those facilities requiring parental consent when listing facility-level response rates.

BJS is collecting core variables on all sampled inmates such as age, race, and offense, and using this information to make non-response adjustment to the facility-level data and the national-level data.

There may also be some youth who will not be eligible to participate due to an inability to assent, physical ailments (such as blindness, paralysis, etc.), hospitalization, or violent status. Reasons for non-participation will be specifically coded and analyzed at the end of data collection.

4. Test of Procedures or Methods

The interview and data collection procedures were tested in a field test conducted March – July 2007, results of which have been incorporated into the final study design described throughout the supporting statement (Attachment 6; also see Westat’s Final Pretest Report.)

5. Consultation Information

The Corrections Statistics Unit at BJS takes responsibility for the overall design and management of the activities described in this submission, including sampling procedures, development of the questionnaires, and the analysis of the data. BJS contacts include:

Paige M. Harrison, Statistician

Corrections Statistics Unit

Bureau of Justice Statistics

810 Seventh St., N.W.

Washington, DC 20531

(202) 305-0809

Allen J. Beck, Ph.D

Principal Deputy Director

Bureau of Justice Statistics

810 Seventh St., N.W.

Washington, DC 20531

(202) 616-3277

The co-Principal Investigators are:

David Cantor

Associate Director

Westat

1650 Research Blvd.

Rockville, MD 20850

(301) 294-2080

Andrea Sedlak

Associate Director

Westat

1650 Research Blvd.

Rockville, MD 20850

(301) 251-4211

Works cited:

Tourangeau, Roger and Tom W. Smith. (1996). Asking Sensitive Questions: The Impact of Data Collection Mode, Question Format, and Question Context. Public Opinion Quarterly 60: 375-304.

Des Jarles, Don C; Paone, Denise; Milliken, Judith; Turner, Charles E.; Miller, Heather; Gribble, James; Shi, Qiuhu; Hagan, Holly; and Friedman, Samuel. (1999). Audio-Computer Interviewing to Measure Risk Behavior for HIV Among Injecting Drug Users: A Quasi Randomized Trial. The Lancet 353: 1657-1661.

Gribble, James N.; Miller, Heather G.; Rogers, Susan M.; Turner, Charles F. (1999) Interview Measure and Mode of Sexual Behaviors: Methodological Issues. Journal of Sex Research, Feb. 1999.

1. A subsample of females will be necessary in those few facilities with a very large number of females, to avoid a heavy burden on the facility.