January 27, 2012

Methodology

This report is based in part on research conducted by Human Rights Watch in nine states during 2011. We visited 20 prisons in California, Colorado, Georgia, Mississippi, New York, Ohio, Rhode Island, and Washington, and talked with senior headquarters-based corrections officials as well as prison-based staff ranging from wardens to correctional officers. We also interviewed—mostly but not always in privacy—men and women of various ages who were incarcerated in the facilities we visited. Most, but not all, of the facilities or specific units we visited contained a large percentage of older prisoners. We also visited with senior corrections medical personnel and other state officials in Connecticut. In addition, throughout the year we also consulted with numerous correctional and gerontology experts, as well as conducted extensive research in the academic literature on aging and corrections.

This report also includes statistical data obtained from different sources.

Our data on the number of sentenced state and federal prisoners and the number of prisoners by age was obtained from the US Department of Justice Bureau of Justice Statistics’ “Prisoners Series” for the years 1995 to 2010. Each of the annual reports for those years is available online at the Bureau of Justice Statistics (BJS) website (http://bjs.ojp.usdoj.gov). Using the methodology described in its reports, BJS estimates the number of prisoners in different age categories. Human Rights Watch calculated percentages and trends of state and federal prisoners by age using the BJS data. Unless otherwise indicated, all references to federal and state prisoners by age obtained from BJS reports are sentenced prisoners under the jurisdiction of state and federal correctional authorities.

Our data on the number and age of new court commitments to state prison (almost all of which are admissions into prison of offenders convicted and sentenced by a court, usually to a term of more than one year) from 1995 to 2009 (the most recent year for which such information was available at this writing) was obtained from the annual statistical tables prepared by the Bureau of Justice Statistics as part of its National Corrections Reporting Program Series. The tables are available online (at http://bjs.ojp.usdoj.gov/index.cfm?ty= pbdetail&iid=2174). Human Rights Watch calculated trends over time and percentage increases in new court commitments by age using these tables. See below for a description of the National Corrections Reporting Program.

We obtained data on the age of federal prisoners, their age at entry to prison, and the length of their sentences by accessing information on defendants processed in the federal criminal justice system through the Federal Criminal Case Processing Statistics (FCCPS) of the Bureau of Justice Statistics (available at http://bjs.ojp.usdoj.gov/fjsrc/index.cfm). The FCCPS enables members of the public to generate statistics online, including the construction of tables and trends by frequency and percentage of persons in or entering federal prison in selected years, their age, and sentence lengths. The most recent year for which FCCPS provides data is 2009.

Finally, our report includes a detailed analysis undertaken by Dr. Patrick Vinck, consultant to Human Rights Watch, of the state prison population and admissions data for 2009 compiled under the National Corrections Reporting Program (NCRP) of the United States Department of Justice Bureau of Justice Statistics. The NCRP collects administrative records information on prisoners admitted to prison, released from prison, discharged from parole, or in prison at year-end from participating states. The number of participating states varies by year and by the category of data being provided. Thirty states participating in the NCRP submitted prison admissions data for 2009 and 24 submitted year-end prison population data for 2009, the most recent year for which NCRP data is available.

Dr. Vinck’s analysis was conducted with the software Statistical Package for the Social Sciences (SPSS) under a restricted data use agreement with the Inter-university Consortium for Political and Social Research (distributor of the NCRP data).

Several methodological elements need to be highlighted:

  • Age: The NCRP data includes date of birth for each prisoner. We used the date of birth to determine prisoners’ ages as of year-end 2009. We computed the age at admission by comparing the date of admission with the date of birth. When the date of birth or date of admission was incomplete, but the available data clearly indicated that the individual was above or below 55 at admission, the record was categorized accordingly. We were unable to compute the age at admission for 2,742 prisoners (0.52 percent of the total number for whom we had records); we were also unable to compute the age in 2009 of 1,749 individuals (0.2 percent of the total number for whom we had records).
  • Offense Categories: For the purpose of the analysis, we categorized offenses in 5 groups according to the offense codes used in the National Corrections Reporting Program. The groups are as follow: (1) violent offense (code 010-180), (2) property offense (190-335), (3) drug offense (340-450), (4) public order offense (460-672), and (5) other/unknown if the offenses were unspecified or missing. In addition, within violent offenses, sexual offenses are distinguished as a sub-category. This includes the following offenses: rape (force), rape (statutory, no force), sexual assault (other), lewd act with children, and forcible sodomy.
  • Type of admissions: Persons can be admitted to prison for various reasons. In this report, we include in the category of “new admissions” new court commitments, parole revocations with new sentences, mandatory parole releases with new sentences, and probation revocations with new sentences. We grouped all other admissions, including returns for technical parole violations and admissions where the status of the new sentence was unknown, as “Other Admissions.”
  • Limitations: There are a number of limitations associated with the 2009 NCRP data which should be kept in mind. First, we do not know to what extent non-reporting states may differ from reporting states. However, the states reporting in 2009 have three-quarters of all state prisoners and we believe our findings regarding older prisoners based on those states are likely to be representative of state prisoners nationwide. Second, the administrative records include a number of item-specific missing data (for example, gender not recorded, sentence not recorded). For that reason, the number of cases included in each analysis may vary. More generally, the reliability of the data contained in the NCRP database cannot be assessed. The NCRP database is based on a structured questionnaire completed annually on the basis of official prison records of prisoner population movement. After the questionnaires are processed by the Census Bureau, state tallies are sent to state officials for verification and comment. Limitations and information on data processing are provided in the NCRP codebook associated with the data.