U.S. Department of Justice Office of Justice Programs Bureau of Justice Statistics ******************************************************** This file is text only without graphics and many of the tables. A Zip archive of the tables in this report in spreadsheet format (.csv) and the full report including tables and graphics in .pdf format are available on BJS website at: http://www.bjs.gov/index.cfm?ty=pbdetail&iid=5640 This report is one in a series. More recent editions may be available. To view a list of all in the series go to http://www.bjs.gov/index.cfm?ty=pbse&sid=5 ********************************************************* Bulletin Correctional Populations in the United States, 2014 Danielle Kaeble, Lauren Glaze, Anastasios Tsoutis, and Todd Minton, BJS Statisticians At yearend 2014, an estimated 6,851,000 persons were under the supervision of U.S. adult correctional systems, a decline of about 52,200 from 6,903,200 at yearend 2013 (figure 1). After peaking at 7,339,600 persons in 2007, the correctional population decreased each year by an average of 1.0%. By yearend 2014, the population declined by 0.8% to the lowest level observed in more than a decade (6,886,800 in 2003). About 1 in 36 adults in the United States was under some form of correctional supervision at yearend 2014. This was the lowest rate observed since 1996 (5,531,300) when about 1.3 million fewer offenders were under correctional supervision (not shown). This report summarizes data from several Bureau of Justice Statistics (BJS) correctional data collections to provide statistics on the total population supervised by adult correctional systems in the United States. (See Methodology for sources.) These systems include offenders living in the community while supervised by probation or parole agencies and those under the jurisdiction of state or federal prisons or held in local jails. (See Terms and definitions for more information.) ************************************************************** ************ HIGHLIGHTS ************* * Adult correctional systems supervised an estimated 6,851,000 persons at yearend 2014, about 52,200 fewer offenders than at yearend 2013. * About 1 in 36 adults (or 2.8% of adults in the United States) was under some form of correctional supervision at yearend 2014, the lowest rate since 1996. * The correctional population has declined by an annual average of 1.0% since 2007. * The community supervision population (down 1.0%) continued to decline during 2014, accounting for all of the decrease in the correctional population. * The incarcerated population (up 1,900) slightly increased during 2014. * Between 2007 and 2014, about 88% of the decrease in the correctional population (down 488,600 offenders) was attributed to the decline in the probation population. * Seven jurisdictions accounted for almost half (48%) of the U.S. correctional population at yearend 2014. * Nearly all (47) jurisdictions had a larger proportion of their correctional population supervised in the community at yearend 2014 than incarcerated in prison or local jail. ************************************************************** *********************************************** The community supervision population declined during 2014, accounting for all of the decrease in the correctional population *********************************************** From 2013 (6,903,200) to 2014 (6,851,000), the total correctional population declined by 0.8% (table 1). (See appendix tables 1, 2, and 3 for correctional population estimates by jurisdiction and sex.) About 7 in 10 persons under correctional supervision at yearend 2014 were supervised in the community either on probation (3,864,100) or parole (856,900).***Footnote 1 The total correctional population, total community supervision population, and total incarcerated population exclude offenders with multiple correctional statuses to avoid double counting. For this reason, the sum of the community supervision and incarcerated populations, and the change in the populations, will not equal the total correctional population. See table 6 and Methodology.*** In comparison, about 3 in 10 offenders (2,224,400) under correctional supervision were under the jurisdiction of state or federal prisons (1,561,500) or held in local jails (744,600). The 52,200 decrease in the number of persons under correctional supervision during 2014 was attributed to a decline in the community supervision population (down 1.0%), as the change in the incarcerated population during the year was small (up 0.1%). All of the decrease in the community supervision population during 2014 was accounted for by the decline in the probation population (down 46,500), as the parole population increased slightly during the year (up 1,700). After reaching a high of 5,119,000 persons in 2007, the community supervision population declined by annual average of 1.2%. The downward trend in the probation population over the past 7 years was consistent with that of the community supervision population. Since 2007, the probation population declined by an annual average of 1.5%, the largest rate of decline across all correctional populations. In comparison, the parole population grew by an annual average of 0.5% since 2007. During 2014, the number of inmates incarcerated in state or federal prisons or local jails increased slightly (up 1,900), reversing a 5-year decline since 2008. While the jail population grew by 1.8% during 2014, the U.S. prison population dropped by 1.0%. The decrease in the U.S. prison population resulted from a decline in the state (down 10,100) and federal (down 5,300) prison populations. This was the second consecutive decline in the federal prison population after peaking in 2012 (217,800). *********************************************** During 2014, the correctional supervision rate fell for the seventh consecutive year *********************************************** By yearend 2014, about 2,780 offenders per 100,000 U.S. adult residents were under some form of correctional supervision, down from 2,830 per 100,000 adults at yearend 2013 (figure 2). More than half (56%) of the decline in the correctional supervision rate was attributed to the increase in the size of the U.S. adult resident population during the year, while a smaller share of the decline (44%) resulted from the decrease in the correctional population. (See Methodology.) After peaking at 3,210 per 100,000 U.S. adult residents in 2007, the correctional supervision rate fell steadily each year. Since 2007, the trend in the correctional supervision rate diverged from the trend in the number of persons under correctional supervision. The number of persons supervised by adult correctional systems decreased by an annual average of 1.0% from yearend 2007 to 2014. In comparison, the average annual decline in the correctional supervision rate (down 2.1%) was twice as fast during the same period. However, more than half (52%) of the decrease in the correctional supervision rate resulted from the increase in the U.S. adult resident population since 2007, compared to 48% of the decline attributed to the decrease in the number of offenders under correctional supervision. From 2013 to 2014, the rate of offenders under community supervision declined from 1,950 to 1,910 per 100,000 adults, continuing a downward trend since 2007 (table 2). The decrease in the community supervision rate over the past 7 years accounted for about three-quarters of the decline in the correctional supervision rate during the period. The incarceration rate also dropped slightly by yearend 2014, from 910 per 100,000 at yearend 2013 to 900 per 100,000. The incarceration rate has declined steadily each year since 2008. ************************************************************** *********************************************** The difference between measures of the incarceration rate and the imprisonment rate *********************************************** The incarceration rate and the imprisonment rate are two different statistics that BJS reports, depending on the correctional population of interest. The incarceration rate describes the incarcerated population that consists of inmates under the jurisdiction of state or federal prisons and inmates held in local jails. In comparison, the imprisonment rate describes the prison population under the jurisdiction of state or federal prisons and sentenced to more than 1 year. The imprisonment rate excludes prisoners who are unsentenced, those with sentences of less than 1 year, and all local jail inmates. Given these differences, the incarceration rate will always be higher than the imprisonment rate because the imprisonment rate includes only a subset of the population accounted for in the incarceration rate (table 3). This report focuses on the total correctional population, which consists of the community supervision (i.e., probation and parole) and incarcerated (i.e., prison and local jail) populations. Therefore, except for table 3, rates presented in this report are incarceration rates because they describe the total incarcerated population. BJS reports on the imprisonment rates in its annual report on the prison population. For more information on imprisonment rates, see Prisoners in 2014 (NCJ 248955, BJS web, September 2015). ************************************************************** *********************************************** Since 2007, compositional changes in the correctional population were small despite the decrease of 488,900 offenders *********************************************** Despite the overall decline in the correctional population over the past 7 years (down 488,900 offenders), the changes in the composition of the population were small. Probationers continued to account for the majority (56%) of offenders under correctional supervision at yearend 2014 (table 4). In 2014, probationers accounted for a slightly smaller portion of the correctional population compared to 2007 (58%), as the number of probationers decreased each year during the period. Prison and parole populations grew slightly as a share of the total correctional population between 2007 and 2014. Prisoners accounted for 23% of offenders under correctional supervision at yearend 2014, up slightly from 22% in 2007. The parole population accounted for 13% of the correctional population at the end of 2014, up slightly from 11% in 2007. Inmates incarcerated in local jails represented the smallest shares of the correctional population in 2007 and 2014 (11% each). *********************************************** The decline in the probation population from 2007 to 2014 accounted for 88% of the decrease in the correctional population *********************************************** Probationers represented the majority of offenders under correctional supervision from 2007 to 2014, and the decline in this population contributed significantly to the decrease in the correctional population. From 2007 to 2014, the number of probationers decreased by 428,800, representing about 88% of the total decline in the correctional population since 2007—the largest decline among all correctional populations (table 5). The prison and local jail populations also declined between 2007 and 2014. However, they accounted for a significantly smaller portion of the decrease in the correctional population compared to probationers. From 2007 to 2014, the number of inmates in prison declined by 35,300 offenders and the number in local jails fell by 35,600, accounting for equal shares of the decline in the correctional population (down 7% each). The parole population was the only correctional population to increase from 2007 to 2014. About 30,800 more parolees were supervised in the community in 2014 compared to 2007, partially offsetting the overall decline in the correctional population during the 7-year period. *********************************************** Seven jurisdictions accounted for almost half of the U.S. correctional population at yearend 2014 *********************************************** At yearend 2014, the size of the correctional population by jurisdiction varied, from a low of 8,400 offenders to a high of 699,300 (figure 3). Including the federal system but excluding the District of Columbia, the average number of offenders under the supervision of adult correctional systems by jurisdiction was about 133,400. Seven jurisdictions had correctional populations of 300,000 or more offenders, including Texas (699,300), California (589,600), Georgia (579,600), Florida (382,600), Pennsylvania (360,800), the federal system (338,000), and Ohio (326,300). These seven jurisdictions made up almost half (48%) of the U.S. correctional population at the end of 2014. Excluding the federal system, four of the same six jurisdictions had more than 3,000 per 100,000 U.S. adult residents under some form of correctional supervision at yearend 2014 (figure 4). The other two states, Florida (2,390 per 100,000 U.S. adult residents) and California (1,980 per 100,000), had correctional supervision rates that were less than 2,500 per 100,000. *********************************************** At yearend 2014, almost all jurisdictions had a larger portion of their correctional population supervised in the community than incarcerated *********************************************** While the distribution of the correctional population varied by jurisdiction, almost all (47) jurisdictions had more than half of their correctional population supervised in the community on probation or parole at the end of 2014. Including the federal system but excluding the District of Columbia, jurisdictions ranged from a low of 38% of their correctional population supervised in the community at yearend 2014 to a high of 88%, with a national average of about 66% (figure 5). The proportion of the correctional population incarcerated in state or federal prisons or local jails ranged from a low of 12% to a high of 62%, with a national average of about 34%. Of the seven jurisdictions that constituted almost half of the U.S. correctional population at yearend 2014, six had at least 60% of their correctional population supervised in the community rather than incarcerated. These included Georgia (84% of total correctional population in the community), Ohio (78%), Pennsylvania (77%), Texas (69%), California (65%), and Florida (60%). One of the seven jurisdictions, the federal system (62%), had more than 60% of its correctional population incarcerated rather than supervised in the community at the end of the year. Terms and definitions ************************* Adult--persons subject to the jurisdiction of an adult criminal court or correctional agency. Adults are age 18 or older in most jurisdictions. Persons age 17 or younger who were prosecuted in criminal court as if they were adults are considered adults, but persons age 17 or younger who were under the jurisdiction of a juvenile court or agency are excluded. (See Methodology for more information on prison and local jail inmates age 17 or younger.) Annual change—change in a population between two consecutive years. Average annual change--average (mean) annual change in a population across a specific time period. Community supervision population--estimated number of persons living in the community while supervised on probation or parole. Community supervision rate--estimated number of persons supervised in the community on probation or parole per 100,000 U.S. residents of all ages (i.e., total community supervision rate) or U.S. residents age 18 or older (i.e., adult community supervision rate). Correctional population--estimated number of persons living in the community while supervised on probation or parole and inmates under the jurisdiction of state or federal prisons or held in local jails. Correctional supervision rate--estimated number of persons supervised in the community on probation or parole and inmates under the jurisdiction of state or federal prisons or held in local jails per 100,000 U.S. residents of all ages (i.e., total correctional supervision rate) or U.S. residents age 18 or older (i.e., adult correctional supervision rate). Imprisonment rate--estimated number of prisoners under state or federal jurisdiction sentenced to more than 1 year per 100,000 U.S. residents of all ages (i.e., total imprisonment rate) or U.S. residents age 18 or older (i.e., adult imprisonment rate). (The imprisonment rate is presented and discussed in The difference between measures of the incarceration rate and imprisonment rate text box.) Incarcerated population--estimated number of inmates under the jurisdiction of state or federal prisons or held in local jails. Incarceration rate--estimated number of inmates under the jurisdiction of state or federal prisons or held in local jails per 100,000 U.S. residents of all ages (i.e., total incarceration rate) or U.S. residents age 18 or older (i.e., adult incarceration rate). Indian country jail population--estimated number of inmates held in correctional facilities operated by tribal authorities or the Bureau of Indian Affairs (BIA), U.S. Department of the Interior. These facilities include confinement facilities, detention centers, jails, and other facilities operated by tribal authorities or the BIA. (This estimate is presented in appendix table 4.) Local jail population--estimated number of inmates held in a confinement facility usually administered by a local law enforcement agency that is intended for adults, but sometimes holds juveniles, for confinement before and after adjudication. These facilities include jails and city or county correctional centers; special jail facilities, such as medical treatment or release centers; halfway houses; work farms; and temporary holding or lockup facilities that are part of the jail’s combined function. Inmates sentenced to jail facilities usually have a sentence of 1 year or less. Military prison population--estimated number of service personnel incarcerated under the jurisdiction of U.S. military correctional authorities. (This estimate is presented in appendix table 4.) Parole population--estimated number of persons who are on conditional release in the community following a prison term while under the control, supervision, or care of a correctional agency. Violations of the conditions of supervision during this period may result in a new sentence to confinement or a return to confinement for a technical violation. This population includes parolees released through discretionary (i.e., parole board decision) or mandatory (i.e., provisions of a statute) supervised release from prison, those released through other types of post- custody conditional supervision, and those sentenced to a term of supervised release. Prison population--estimated number of inmates incarcerated in a long-term confinement facility, run by a state or the federal government, that typically holds felons and offenders with sentences of more than 1 year, although sentence length may vary by jurisdiction. **Prison jurisdiction population--estimated number of prisoners under the jurisdiction or legal authority of state or federal correctional officials, regardless of where the prisoner is held. This population represents BJS’s official measure of the prison population and includes prisoners held in prisons, penitentiaries, correctional facilities, halfway houses, boot camps, farms, training or treatment centers, and hospitals. Counts also include prisoners who were temporarily absent (less than 30 days), in court, or on work release; housed in privately operated facilities, local jails, or other state or federal facilities; and serving concurrent sentences for more than one correctional authority. **Prison custody population--estimated number of prisoners held in the physical custody of state or federal prisons regardless of sentence length or the authority having jurisdiction. This population includes prisoners housed for other correctional facilities but excludes those in the custody of local jails, inmates held in other jurisdictions, inmates out to court, and those in transit from one jurisdiction of legal authority to the custody of a confinement facility outside that jurisdiction. (This estimate is presented in appendix table 5.) Probation population--estimated number of persons who are on a court-ordered period of supervision in the community while under the control, supervision, or care of a correctional agency. The probation conditions form a contract with the court by which the person must abide in order to remain in the community, generally in lieu of incarceration. In some cases, probation can be a combined sentence of incarceration followed by a period of community supervision. Often, probation entails monitoring or surveillance by a correctional agency. In some instances, probation may not involve any reporting requirements. Territorial prison population--estimated number of prisoners in the custody of correctional facilities operated by departments of corrections in U.S. territories (American Samoa, Guam, and the U.S. Virgin Islands) and U.S. commonwealths (Northern Mariana Islands and Puerto Rico). (This estimate is presented in appendix table 4.) ************** Methodology ************** Data sources **************** The statistics presented in this report include data from various Bureau of Justice Statistics (BJS) data collections, each relying on the voluntary participation of federal, state, and local respondents. For more information about any of the following data collections, see the Data Collections webpage at www.bjs.gov. Annual Surveys of Probation and Parole. The Annual Surveys of Probation and Parole (ASPP) began in 1980. The reference date for the surveys is December 31, and they collect data from probation and parole agencies in the United States that supervise adults. Both surveys cover the 50 states, the District of Columbia, and the federal system. In these data, adults are persons who are subject to the jurisdiction of an adult criminal court or correctional agency. Persons age 17 or younger who were prosecuted in criminal court as if they were adults are considered adults, but persons age 17 or younger who were under the jurisdiction of a juvenile court or agency are excluded. Annual Survey of Jails. The Annual Survey of Jails (ASJ) has collected data from a nationally representative sample of local jails each year since 1982, except in 1983, 1988, 1993, 1999, and 2005, when a complete census of U.S. local jails was conducted. Jails are confinement facilities, usually administered by a local law enforcement agency, that are intended to hold adults, but they may also hold youth age 17 or younger before or after they are adjudicated. The ASJ data used in this report include inmates age 17 or younger who were held either before or after they were adjudicated (about 4,200 persons in 2014). To maintain the jail series in this report, all tables and figures that include national estimates of the local jail population as of the last weekday in June were provided through the ASJ, except in 1999 and 2005 when a jail census was completed (see Census of Jails). Because the ASJ is designed to produce only national estimates, tables and figures in this report that include jurisdiction-level counts of the incarcerated population and the total correctional population were based on jail data collected through two other BJS sources, specifically the Census of Jails and the Deaths in Custody Reporting Program. (See Census of Jails and Deaths in Custody Reporting Program.) Census of Jails. The Census of Jails began in 1970 and was conducted in 1972, 1978, 1983, 1988, 1993, 1999, 2005, and 2006. In 2013, BJS expanded the 2013 Deaths in Custody Reporting Program—Annual Summary on Inmates under Jail Jurisdiction to act as the 2013 Census of Jails. (See Deaths in Custody Reporting Program.) The census is designed to produce a complete enumeration of jail facilities in the United States. It is part of a series of data collection efforts, including the Census of Jail Inmates and the Census of Jail Facilities, aimed at studying the nation’s jails and their inmate populations. The reference date of the 2013 census was December 31, while the reference date for prior iterations was the last weekday in June within the reference year. BJS relied on local jail counts provided through the census in 1999, 2005, and 2013 to generate jurisdiction-level estimates of the total incarcerated population and total correctional population that appear in appendix tables 2 and 3. Because they include the 2013 local jail estimates as of December 31, the national totals of the correctional and incarcerated populations reported in appendix tables 2 and 3 are not consistent with the national totals of the populations reported in the other tables and figures of this report, which include BJS’s official estimates of the total correctional and incarcerated populations. Deaths in Custody Reporting Program. The Deaths in Custody Reporting Program (DCRP) is an annual collection that provides national, state, and incident-level data on persons who died while in the physical custody of the 50 state departments of corrections or the approximately 2,900 local adult jail jurisdictions nationwide. To reduce respondent burden for the 2013 iteration, BJS combined the 2013 DCRP collection with the 2013 Census of Jails. For more information, see Census of Jails and Census of Jails: Population Changes, 1999–2013, NCJ 248627, BJS web, December 2015. The DCRP began in 2000 under the Death in Custody Reporting Act of 2000 (P.L. 106–297), and it is the only national statistical collection to obtain comprehensive information about deaths in adult correctional facilities. In addition to the death count, BJS requests that jails provide summary statistics about their population and admissions. All jails, including those with no deaths to report (which includes about 80% of jails in any given year), are asked to complete the annual summary survey form. BJS relied on the local jail counts provided through the DCRP in 2014 to generate jurisdiction-level estimates of the total incarcerated population and total correctional population that appear in figures 2, 3, and 4 and appendix tables 1 and 2. Because they include the 2014 local jail estimates as of December 31, the national totals of the correctional and incarcerated populations reported in appendix tables 1 and 2 are not consistent with the national totals of the populations reported in the other tables and figures of this report, which include BJS’s official estimates of the total correctional and incarcerated populations. National Prisoner Statistics program. The National Prisoner Statistics (NPS) program began in 1926 under a mandate from Congress and has been conducted annually. It collects data from the nation’s state departments of corrections and the Federal Bureau of Prisons (BOP). The NPS distinguishes between inmates in custody and prisoners under the jurisdiction of correctional authorities. To have custody of a prisoner, a state or the BOP must hold that inmate in one of its facilities. To have jurisdiction over a prisoner, the state or BOP must have legal authority over that prisoner, regardless of where the prisoner is incarcerated or supervised. Some states were unable to provide counts that distinguish between custody and jurisdiction. See Jurisdiction notes in Prisoners in 2014 (NCJ 248955, BJS web, September 2015) to determine which states did not distinguish between custody and jurisdiction counts. With the exception of appendix table 5, the NPS prisoner counts in all tables and figures of this report are consistent with the jurisdiction counts and findings reported in Prisoners in 2014. The jurisdiction counts represent BJS’s official measure of the prison population. The NPS prisoner custody counts are presented in appendix table 5 and include all inmates age 17 or younger who were serving time in a state or federal correctional facility after being sentenced in criminal court as if they were adults (about 1,000 persons in 2014), and inmates in the six states in which prisons and jails form one integrated system, including inmates age 17 or younger who may have been held before or after adjudication. Through the annual NPS collection, BJS has obtained yearend counts of prisoners in the custody of U.S. military authorities from the Department of Defense Corrections Council since 1999. In 1994, the council, comprising representatives from each branch of military service, adopted a standardized report (DD Form 2720) that obtains data on persons held in U.S. military confinement facilities inside and outside of the continental United States. These data are presented in appendix table 4 of this report. See Prisoners in 2014 for more statistics and information. Since 1995, through the annual NPS collection, BJS has collected yearend counts of inmates from the departments of corrections in U.S. territories (American Samoa, Guam, and the U.S. Virgin Islands) and U.S. commonwealths (Northern Mariana Islands and Puerto Rico). These data are presented in appendix table 4 of this report and represent all inmates in the custody of prison facilities in U.S. territories or commonwealths. See Prisoners in 2014 for more information, including nonresponse. Survey of Jails in Indian Country. The Annual Survey of Jails in Indian Country (SJIC) has been conducted annually since 1998 with the exception of 2005 and 2006. The SJIC collects detailed information on all adult and juvenile confinement facilities, detention centers, jails, and other facilities operated by tribal authorities or the U.S. Department of the Interior’s Bureau of Indian Affairs. The SJIC data in this report includes inmates age 17 or younger who are in the custody of Indian country jails (about 200 persons in 2014). These data are presented in appendix table 4. See Jails in Indian Country, 2014 (NCJ 248974, BJS web, October 2015) for more information. ************************************* Counts adjusted for offenders with multiple correctional statuses ************************************* Offenders under correctional supervision may have multiple correctional statuses for several reasons. For example, probation and parole agencies may not always be notified immediately of new arrests, jail admissions, or prison admissions; absconders included in a probation or parole agency’s population in one jurisdiction may actually be incarcerated in another jurisdiction; persons may be admitted to jail or prison before formal revocation hearings and potential discharge by a probation or parole agency; and persons may be serving separate probation and parole sentences concurrently. In addition, state and federal prisons may hold inmates in county facilities or local jails to reduce crowding in their prisons. Through the ASPP, BJS began collecting data on the number of probationers and parolees with multiple correctional statuses in 1998 and has since expanded on the information collected. Through the NPS, BJS began collecting data in 1999 on the number of prisoners under the jurisdiction of state or federal prisons who were held in county facilities or local jails. Table 6 includes adjustments that were made to the total correctional population, total community supervision population, and total incarcerated population estimates presented in this report to exclude offenders with multiple correctional statuses to avoid double counting offenders. The estimates from the ASPP are based on data reported by the probation and parole agencies that were able to provide the information within the specific reporting year. Because some probation and parole agencies did not provide these data each year, the numbers may underestimate the total number of offenders who had multiple correctional statuses between 2000 and 2014. Due to these adjustments, the sum of correctional statuses in tables 1, 2, 4, and 5; figure 1; and appendix tables 1, 2, and 3 will not equal the total correctional population. In addition, the sum of the probation and parole populations for 2008 through 2014 will not yield the total community supervision population because the total was adjusted for parolees who were also on probation. Also, the sum of the prison and local jail populations for 2000 through 2014 will not equal the total incarcerated population because prisoners held in local jails were excluded from the total. Decomposing the decline in the correctional supervision rate ********************************* To decompose the decline in the correctional supervision rate, the following formula was used: ?R = [P1 * (1/GP1)] – [P0 * (1/GP0)] = [P1 * ((1/GP1) - (1/GP0))] + [(1/GP0) * (P1 – P0)] = [(1/GP1) * (P1 – P0)] + [P0 * ((1/GP1) – (1/GP0))] In this formula, ?R is the change in the correctional supervision rate, P1 is the total correctional population for the most recent year, P0 is the total correctional population for the earlier year, GP1 is the U.S. adult resident population for the most recent year, and GP0 is the U.S. adult resident population for the earlier year. The components [(1/GP0) * (P1 – P0)] and [(1/GP1) * (P1 – P0)] provide the change in the correctional supervision rate due to the change in the total correctional population. These two components were summed, and the average was used to estimate the amount of change in the correctional supervision rate attributed to the change in the total correctional population during that period. The components [P1 * ((1/GP1) – (1/GP0))] and [P0 * ((1/GP1) – (1/GP0))] provide the change due to the U.S. adult resident population. These two components were summed, and the average was used to estimate the amount of change in the correctional supervision rate attributed to the change in the U.S. adult resident population during the period. Nonresponse adjustments to estimate population counts ************************************ *********************************************** Probation, parole, jail, and prison populations *********************************************** Probation, parole, jail, and prison population counts were adjusted to account for nonresponse across the data collections. The methods varied and depended on the type of collection, type of respondent, and availability of information. For more information on the nonresponse adjustments implemented to generate national and jurisdiction-level estimates of the probation, parole, and prison populations, see Prisoners in 2014 (NCJ 248955, BJS web, September 2015) and Probation and Parole in the United States, 2014 (NCJ 249057, BJS web, November 2015). For more information on the nonresponse adjustments implemented to generate national counts of the jail population that are included in the tables and figures of this report that include only national estimates, see Jail Inmates at Midyear 2014 (NCJ 248629, BJS web, June 2015). *********************************************** Jail population—jurisdiction-level estimates *********************************************** The response rate to the 1999 Census of Jails was 99.8%. Six jail jurisdictions did not respond to the census. Data for critical items, including the population count on the last weekday in June, were imputed based on previous survey and census reports. For more information, see Census of Jails, 1999 (NCJ 186633, BJS web, August 2001). Considering that the response rate to the 2005 Census of Jail Inmates was 100%, no nonresponse adjustments were implemented. For more information, see Prison and Jail Inmates at Midyear 2005 (NCJ 213133, BJS web, May 2006). Nonresponse in the 2013 Census of Jails and 2014 DCRP was minimal. The unit response rate to the 2013 Census of Jails was 92.4% and the 2014 DCRP was 95.8% at the time of this report. The item response rate for the December 31 confined jail population total was 99.3% in the 2013 Census of Jails and 99.0% in the 2014 DCRP. For jails that did not participate in the 2013 Census of Jails or 2014 DCRP or were not able to provide the yearend confined population count, a sequential hot-deck imputation procedure was used to impute values. This procedure used respondent (donor) data as a substitute for missing values. The donor for each nonrespondent was randomly selected from within a set of similar jails, which was sorted by the previous-year population value. The resulting imputed values are generally similar to previous- year reported values, but are not identical due to differences between each donor and nonrespondent pair and the year-to-year fluctuation in donor population values. Because the 2013 Census of Jails and 2014 DCRP data collections used a census design (no sampling), each jail was initially self- representing and had a design weight of 1. To reduce nonresponse bias, responding jails had their weight adjusted via post- stratification to allow their responses to represent jails that did not respond. The description of the weighting used in the 2014 DCRP is described next. The method used for the 2013 Census of Jails was similar. For more information, see Census of Jails: Population Changes, 1999–2013 (NCJ 248627, BJS web, November 2015). Control totals for the 2014 confined jail population from the DCRP were estimated at the state level as follows: * The year-to-year change in confined jail population among respondents to both the 2013 and 2014 DCRP was computed within the state. * Plausible values for the 2013 confined population were imputed for jails that did not report to the DCRP in 2013 using a hot- deck procedure that randomly selected a donor for each nonrespondent from within a set of jails that reported similar confined jail populations in the prior year. * Estimated 2014 values were calculated by multiplying the yearly change rate and the 2013 DCRP estimate of confined population for jails that did not respond to the 2014 DCRP. * The sum of reported, item-imputed, and DCRP-estimated values for the 2014 confined jail population for each state served as the control totals for the post-stratification procedure. The post-stratification weight adjustment factor was identical for all jails within a state and was computed as the ratio of the control total for state i to the sum of the reported and item-imputed 2014 DCRP confined jail population values for state i: The final analysis weight is the product of the design weight and the post-stratification adjustment factor. Because the design weight was 1 for all jails, the analysis weight is equal to the adjustment factor. *********************************************** Nonresponse adjustments to estimate males and females under correctional supervision *********************************************** The number of males and the number of females on probation or parole in 2013 and 2014 were adjusted to account for nonresponse using a ratio adjustment method. For jurisdictions that did not provide data on sex for a single year, the sex distribution reported the prior or subsequent year was used. For jurisdictions that did not provide data on sex for a portion of their population, the sex distribution of the known portion of the population was used to impute for the unknown portion because it was assumed that the distributions were the same. For jurisdictions that were unable to provide any data on sex for more than 1 year, the state national average was used to impute the number of males and females supervised in those states. Adjusted jurisdiction totals were then aggregated to produce national estimates of the number of males and females on probation and parole. The counts of prisoners by sex in 2013 and 2014 were adjusted to account for nonresponse using either external sources or a ratio adjustment method. When possible, BJS used information available on state department of corrections’ websites to impute the number of males and females under the jurisdiction of that state’s prison system within the reference year. Otherwise, the sex distribution reported by the state in a recent, prior year was used to impute the number of males and females in the reference year. For more information, see Prisoners in 2014 (NCJ 248955, BJS web, September 2015). For jails that were unable to report the number of males and females confined at yearend 2013 or 2014, the same sequential hot-deck imputation procedure described in Jail population— jurisdiction-level estimates to impute for the confined jail population was also used to impute for the number of males and females confined in jail. Control totals for the 2014 confined jail population by sex were estimated at the state level as described in the section about weighting under the heading Jail population—jurisdiction-level estimates. A similar method was used to estimate 2013 control totals by sex. More information can be found in Census of Jails: Population Changes, 1999–2013 (NCJ 248627, BJS web, November 2015). To generate estimates of the total correctional population in 2013 and 2014 by sex and jurisdiction, ratio estimation was used to account for male and female offenders with multiple correctional statuses in each jurisdiction. These adjustments were made by correctional status and were based on reported counts, by jurisdiction, of the number of offenders by sex and the number of offenders with multiple correctional statuses: * To estimate the number of male and female prisoners held in local jails, the distribution of the prison population by sex within the reference year was applied to the number of prisoners in local jails by jurisdiction. The estimated number of male prisoners held in local jails was then subtracted from the total number of males under correctional supervision by jurisdiction. This same method was used to adjust the number of females under correctional supervision by jurisdiction * The correctional population estimates in each jurisdiction were also adjusted to account for the number of males and females on probation who were held in prisons or local jails. The distribution of the local jail population by sex was applied to the number of probationers in local jails by jurisdiction within the reference year to estimate the number of males and females with both correctional statuses. In addition, the distribution of the prison population by sex was applied to the total number of probationers in prison within the reference year to estimate the number of males and females with both correctional statuses. The estimated number of male probationers in prisons and local jails was then subtracted from the number of males under correctional supervision within the reference year by jurisdiction, and this same method was used to adjust the number of females under correctional supervision by jurisdiction. This method was also employed to account for parolees held in prisons or local jails and the totals, by sex, were excluded from the number of males and females under correctional supervision in each jurisdiction. * To estimate the number of males and females on parole who were also on probation in 2013 and 2014, the distribution of the parole population by sex within the reference year was applied to the number of parolees on probation in each jurisdiction. The estimated number of males with dual community supervision statuses was then subtracted from the number of males under correctional supervision by jurisdiction. This same method was used to adjust the number of females under correctional supervision. *********************************************** Comparability of jurisdiction-level estimates over time *********************************************** All jurisdiction-level estimates included in this report are based on data reported within the reference year. Some jurisdictions update their population counts for different reasons after submitting their data to BJS. Updated population counts usually include data that were not entered into the information system before the survey was submitted or data that were not fully processed by yearend. Also, some jurisdictions have experienced reporting changes for one or more correctional population collections over time. These changes may result because of administrative changes, such as consolidating databases or implementing new information systems, resulting in data review and cleanup; reconciling offender records; reclassifying offenders, including those on probation to parole and offenders on dual community supervision statuses; and including certain subpopulations that were not previously reported. For these reasons, comparisons between jurisdictions and comparisons between years for the same jurisdiction over time may not be valid. More detailed information about updates and reporting changes that impact the ability to make jurisdiction- level comparisons over time can be found in the source reports for each of the four correctional populations, such as the Probation and Parole in the United States series or Prisoners series, within the particular reference year. ************************************************************** The Bureau of Justice Statistics of the U.S. Department of Justice is the principal federal agency responsible for measuring crime, criminal victimization, criminal offenders, victims of crime, correlates of crime, and the operation of criminal and civil justice systems at the federal, state, tribal, and local levels. BJS collects, analyzes, and disseminates reliable and valid statistics on crime and justice systems in the United States, supports improvements to state and local criminal justice information systems, and participates with national and international organizations to develop and recommend national standards for justice statistics. William J. Sabol is director. This report was written by Danielle Kaeble, Lauren Glaze, Anastasios Tsoutis, and Todd Minton. Lauren Glaze, E. Ann Carson, and Todd Minton verified the report. Lynne McConnell and Jill Thomas edited the report. Tina Dorsey produced the report. December 2015, NCJ 249513 ************************************************************** ************************************************** Office of Justice Programs Innovation * Partnerships * Safer Neighborhoods www.ojp.usdoj.gov ************************************************** ************************ 12/16/2015/JER/11:10am ************************