GQ Correctional Facilities Plan

The Census Bureau plans to conduct additional research under the generic clearance for questionnaire pretesting research (OMB number 0607-0725). We will conduct semi-structured interviews with inmates in correctional facilities. This is part of a larger research project that aims to collect information on the 2010 Census Group Quarters (GQ) enumeration, and document and understand the issues that may arise if a second enumeration of GQs is included as part of the Census Bureau’s coverage measurement program in 2020.

The purpose of the larger research project is to help develop the most feasible method to evaluate the accuracy of the census count in living arrangement known as “group quarters,” defined as places where people live or stay, in a group living arrangement that is owned or managed by an entity or organization providing housing and/or services for the residents. Group quarters can include, but are not limited to, federal detention centers, residential treatment centers, college/university student housing, domestic violence shelters, group homes, and military barracks. The Census Bureau conducts numerous surveys for the Bureau of Justice Statistics (BJS) with populations residing in correctional facilities. The American Community Survey has been conducing surveys with the prison population since 2003.

This new study will build upon 12 months of previous decennial ethnographic research in three correctional facilities (two prisons and one jail) in a large western state conducted by Professor Barbara Owen. (The letter requesting approval for that research was submitted to OMB on March 30, 2010.) In 2010, Census enumerators used administrative records to enumerate all three study sites. While efficient, the use of administrative records raises some critical concerns regarding the availability and accuracy of key information required by the Individual Census Report (ICR). In particular, Owen (2010) found that the response categories provided in the ICR for the race and ethnicity items do not match those used by the state records. In addition, the alternate address requested on the ICR may not be available and could be outdated or inaccurate if administrative records are used. The major goal of this new research is to conduct a small-scale test to evaluate the accuracy of the data provided by the administrative records for decennial enumeration and to help develop a new instrument ideal for coverage measurement study in such group quarters. Specific goals of these interviews include determining what the correct question wording is to collect an alternate address from prisoners, whether prisoners will be willing and able to fill out the form (nonresponse due to literacy vs. refusals) and how whether the responses to the ICR liked-questionnaire matched those provided from administrative records

In April 2011, a maximum of 80 semi-structured interviews will be conducted with inmates in the state of California by Professor Owen under contract. Dr. Owen has 15 years of experience conducting research in correctional facilities related to female prisoners, repeat offenders, and prison security. She is well-known in the field of criminology and has a good working relationship with Allen Beck, the Chief of Corrections Statistics Program at the BJS.

One women’s prison and one men’s prison have been selected for this study. The two sites were chosen because Dr. Owen has security clearance for all correctional facilities in the State of California, and she has ongoing work relationships with these facilities. Each prison is divided into yards,” which are clusters of buildings separated by walls and fencing. Only yards housing general populations (inmates that are allowed to move around freely without staff escort) will be eligible for the study.  During the first selection phase, one eligible yard will be randomly selected from each prison. One building will be randomly selected from the clusters of buildings in the identified yard.  Approximately 40 inmates will be randomly selected (not recruited) from the roster of inmates housed in the selected building. They will be called out individually to the day room, where they will be interviewed in private. Each inmate will only be asked questions about him/herself, not about any other prisoners. Dr. Owen has human subjects approval for multiple studies with prisoners, and has never had any issues with conflict related to inmates informing on each other.

Although the interview participants will be randomly selected from the inmates in the prison, this sample cannot be used to generalize to the larger population in correctional facilities since state prisons are different across states and are different from Federal prisons and other correctional facilities both within and across states.

The semi-structured interviews will consist of two sets of questions. First, if possible, respondents will be asked to complete a questionnaire with questions that are identical to the ones used in the 2010 Individual Census Report (ICR) to enumerate GQs. Otherwise, these questions will be administered in-person by the researchers. These questions collect the demographic information of the inmates and their alternate addresses where they will live or stay if they reported that they do not live at the facilities most of the time. A copy of the ICR is attached. Next, the researcher will administer a set of debriefing questions regarding inmates’ perspectives on the form. Lastly, the researchers will administer probes previously used in the 1995 Living Situation Survey Study designed to study people who are tenuously attached to households. They will probe for an address where someone may report the inmates on their census forms and ask questions about the name and relationship of one person with the inmates from that elicited address. A copy of the interview protocol is attached.

The demographic information (name, sex, age, date of birth, and race) collected from the ICR questions will be compared to the administrative records obtained from the facilities to examine the quality of the records. This information is available electronically. The qualitative data from the narrative accounts of the concept of addresses will aid the Census Bureau in designing approaches to enumerating inmates given changes in many state laws requiring enumeration in home communities and the need for coverage measurement to estimate erroneous enumeration (e.g. duplication and enumeration at the wrong place) of prisoners. According to the Prison Policy Initiative (PPI), the states of Maryland, Delaware and New York have passed laws that will require the census to count incarcerated people at home for state and local redistricting purposes. Unlike other demographic information, prisons in the state of California do not routinely collect alternate addresses for their inmates. When the addresses are kept, they are in paper format. Little is known about the availability and accuracy of the addresses provided by the administrative records for these inmates. This study is the first step towards understanding the issues when using administrative records to enumerate inmates and how best to design a coverage measurement instrument to evaluate the census count.

Respondents will be told that their participation is voluntary and that the information they provide will be confidential. A copy of the consent form they will be asked to sign is attached. The respondents will not receive monetary incentives for participating in this research.

The estimated time for completion of an interview is about one-half hour. Thus, the maximum estimated burden for this research is 40 hours (80 participants X half hour).

The contact person for questions regarding this data collection is:

Anna Y. Chan

Research Social Scientist

Statistical Research Division

U.S. Census Bureau

Washington, D.C. 20233

(301) 763-8462

anna.y.chan@census.gov