DCRP Part B_ 01_2016

3

B. Statistical Methods

1. Universe and Respondents Selection

The universe for the prison portion of the DCRP is all deaths of incarcerated persons occurring in the custody of state prisons. Respondents are persons in organizations having custody over the decedent. The potential respondent universe consists of each of the 50 state departments of corrections (DOCs). BJS collects data from persons in the DOC directly.

For each of the three collections under the DCRP (as identified in the introduction to Part A), BJS collects data on the entire universe of deaths that fall under the scope of the DCRP, rather than sampling deaths or agencies. The 2013 Death in Custody Reporting Act legislation (referenced in Part A, Item 1) required all state and federal prisons, as well as local jails, and federal, state and local law enforcement agencies to report data on deaths in custody to the Attorney General. BJS established reporting procedures for the DCRP in response to the DICRA legislation and has maintained these reporting procedures since 2000. In 2001, BJS established procedures for collecting data directly from the 50 state departments of corrections. In doing this, BJS built upon established contacts and procedures for collecting data from these entities for other BJS collections.

BJS obtains the universe of deaths rather than samples for a number of reasons. The principle reason is that the 2013 DICRA requires a full enumeration of deaths occurring in the custody, including those occurring in state prisons. Furthermore, the goal of the DCRP is to track changes in custodial mortality in state prisons, and sampling would complicate this task immensely as well as preclude making important comparisons over time among causes of death and by demographic groups.

For the prison portion of the DCRP, a goal of the program is to track state-level changes in mortality in prisons, not simply to generate a national estimate of prisoner mortality. The only option to achieve this goal by sampling would be to sample death records within a state. Given the relative rarity of some important manners of death (e.g., suicide and homicide), sampling strategies, by design, would be very complicated and potentially burdensome on the respondents because a sampling design would have to be done at the facility rather than the DOC-level. There are an estimated 1,250 state prison facilities in the United States, represented by 50 state-centralized DOCs. In order to generate state-level prisoner mortality estimates, we would have to design a prison sample that was both state and nationally representative. The sample respondents would far exceed the current number of DCRP prison collection respondents (50 respondents representing 50 DOCs).

Sampling deaths to obtain information about the circumstances leading to prisoner deaths would significantly compromise BJS’ ability to accurately report on mortality in prisons. Sampling respondents would not appreciably reduce burden but would greatly diminish the quality and utility of the data.

2. Procedures for Information Collection

Procedures for the DCRP collection. During November of the year prior to the upcoming calendar year collection, BJS’ data collection agent (RTI International) notifies prison respondents of the annual collection cycle. The notification occurs during annual verification calls. In the annual verification calls, RTI verifies that the current collection year respondent will be the respondent for the upcoming collection year and if not, will obtain the new respondent’s contact information. Changes in prison respondents is not common –fewer than three DOC respondents changes annually.

Using the contact information obtained from the verification calls, RTI mails to each DCRP Prison respondent an information packet announcing the start of the current year’s collection cycle (see Appendix C, attachment G). In addition to the announcement letter, the packets contain year-specific envelopes that distinguish the current year’s surveys (NPS-4A for prisons) from the annual summary forms (NPS-4 for prisons).

The packets include a reporting instructions sheet and a DCRP fact sheet. The instruction sheet directs the respondents to the DCRP Web-site and provides them with login information to use in submitting data via the Web-reporting tool. In addition, respondents are reminded to submit an annual summary form and any outstanding death record forms from the preceding year by March 1^st of the current year. The primary mode of collection, the Web-reporting tool, allows respondents to submit individual death record forms on an ongoing basis. Secondary modes of submission of records include email, fax or U.S.P.S. A postage-paid business reply envelope is provided for respondents who choose to submit their data via U.S. mail.

All data for a calendar year are due within 90 days following the end of the year. Follow-up contacts begin in May and continue through July following the collection year to solicit late responses. RTI contacts nonrespondents via phone, fax, email, and mail (for respondents for whom RTI does not have an email address). Concentrated data quality follow-up with respondents runs May through July (see appendix C, attachment H).

RTI International monitors the DCRP data and its respondents throughout the active data collection period. This includes tracking the time to complete, number of contacts made to each agency, and other information (e.g., which form(s) have and have not been returned to date). This information is then used for tailored follow-up prompts to partial or nonrespondents. A similar approach is employed when there is missing or conflicting data within the returned forms. For such outreach activities, agencies are assigned and managed by project team members, each of whom serves as that agency’s liaison for DCRP.

3. Methods to Maximize Response

BJS has consistently achieved a 100% response rate for the prison portion of the DCRP. Since the collection began in 2001, all 50 state departments of corrections have participated annually in the DCRP-Prisons collection.

BJS has consistently maintained a 100% survey response rate and item response rate between 95 and 100% prison portion of the DCRP. Given the high response rates, BJS typically reports only valid responses. The only exception to the missingness rule is in reporting death totals, where cases missing cause of death need to be included to generate an accurate total. BJs do not impute death data missing values. While imputation can be utilized, BJS opts instead to analyze only valid responses because missingness is so small as to not introduce bias, which is a significant concern for mortality studies.

Data collection staff not only use follow-up emails and telephone calls to promote high response rates, but also conduct out-of-range analysis of critical items to check data quality and follow up with respondents. These methods have proved effective in reaching a 100% unit response rate, minimizing item nonresponse, and increasing data validity.

BJS provides multiple modes of data submission, with web as the primary mode for state prison respondents. As stated in Part A of this package, nearly all state prison respondents submit their data electronically, either via the website or through a bulk file. BJS developed a bulk file template in which state prison administrators can submit their prison inmate death records electronically due to the volume of deaths associated with prison submissions. In 2013, 94% of prison respondents submitted their records electronically (86% via the web, 10% via the bulk file).

Through discussions with various stakeholders and respondents for the prison portion of the DCRP, BJS has found general satisfaction with current surveys, that the questions cover topics relevant to prisoner mortality and the instructions for completing the survey are clear.

4. Test of Procedures or Methods

BJS obtains feedback from DOC respondents when significant changes to the existing surveys are implemented, or when a new survey instrument is introduced to the field.

As mentioned in Part A, BJS will conduct work on assessing cause of death reporting. Based on previous work with data from the NDI, the cause of death variable could be improved, particularly in regards to capturing underlying cause of death, which is not currently captured on the DCRP.

5. Contacts for Statistical Aspects and Data Collection

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

6. Contact Information

BJS contacts for the DCRP include the following:

Anastasios ‘Tom’ Tsoutis

Acting Chief

Corrections Statistics Unit

Bureau of Justice Statistics

810 Seventh Street, N.W.

Washington, DC 20531

(202) 514-1062

[email protected]

Margaret Noonan

Statistician and DCRP Project Manager

Corrections Statistics Unit

Bureau of Justice Statistics

810 Seventh Street, N.W.

Washington, DC 20531

(202) 353-2060

[email protected]

3