U.S. flag

An official website of the United States government, Department of Justice.

Data Corrections

wutzkohphoto/Shutterstock.com (see reuse policy).

Corrections to BJS data are posted on this page. Links to the revised publications are provided with a short description of the correction. All related documents affected by the corrections have been revised.

TitleDate of RevisionDescription
Contacts Between Police and the Public, 2020       3/1/2024On page 8 in table 5, the value for the percent of persons ages 45-64 who were given a warning in 2020 has been corrected from 42.4% to 43.2% with no other impact on text in the body of the report.
Federal Law Enforcement Officers, 2020 – Statistical Tables       9/29/2023During a subsequent verification of the Federal Law Enforcement Officers, 2020 — Statistical Tables report, analysts discovered that the 2020 Census of Federal Law Enforcement Officers collected data from 45 (versus 44) Offices of Inspectors General (OIGs). This error appeared on the cover page but has been corrected with no other impact in the body of the report.
Federal Justice Statistics, 2020       7/18/2023

BJS was made aware of errors caused by duplicate records introduced into the FY 2020 analysis files. The following report text, tables, and data have been adjusted and corrected to account for these errors:

  • Report text, tables 3 & 4, and appendix tables 1 & 2
  • Federal Criminal Case Processing Statistics (FCCPS) data tool:
    • Suspects in matters investigated (FY 2020)
    • Suspects in matters concluded (FY 2020)
Criminal Victimization, 2021       7/5/2023During verification of the Criminal Victimization, 2022 report, analysts discovered that the alignment had shifted for the "Threatened rape" measure within the table on page 22, resulting in an incorrect classification of the elements of sexual violence. As a result, formatting has been corrected.
National Crime Victimization Survey Redesign Field Test Topline Report: Comparing Condition 1 and Condition 2 by Interleaving Treatment (Third-party report)       3/28/2023

A programming error impacted the Condition 2, non-interleaved (NIL) Attack screener series. The NIL estimates of violent crime incident rates in the following figures and tables have been adjusted to account for this error:

  • Figures 3-4, 3-5, 3-6, 4-1, 4-2, 4-5, 4-8, and 4-10
  • Table 4-2
  • Appendix Tables A3.3, A4.1, A4.1a, A4.3, A4.7, A4.9, A4.11, and A4.13
Contacts Between Police and the Public, 2018 - Statistical Tables       2/3/2023

During verification of the Contacts Between Police and the Public, 2020 report, analysts discovered  differences in calculation methods between the 2020 and 2018 reports. Since the 2020 report displays data from both reports, 2018 estimates were recalculated with similar methods as the 2020 report. As a result, the following errors in the 2018 report have been corrected:

  • Table 4 and appendix table 4 - most of the estimates under 2018 in the number column
  • Table 5 - the total percent for Handcuffing
  • Appendix table 5 - the percent for Pushing/grabbing/hitting/kicking for whites
Impact of COVID-19 on State and Federal Prisons, March 2020–February 2021       12/20/2022BJS was notified by the Hawaii Department of Corrections of an error in the 2020 admissions data. As a result, state and U.S. totals in tables 2, 3, appendix table 1, and their corresponding .csv files were affected, as well as associated text on page 5 of the report. These corrections have been made.
Facility Characteristics of Sexual Victimization of Youth in Juvenile Facilities, 2018 – Statistical Tables       12/9/2022The second bullet of the front page Highlights states an incorrect incident type. Estimates are for reported staff sexual misconduct, not reported sexual victimization. This error has been corrected with no other impact in the body of the report.
Local Police Departments, 2016: Personnel       11/17/2022

BJS was made aware of an error in the population served value for Kansas City (MO) PD. Further investigation found errors in population served values for many law enforcement agencies that served populations across multiple counties (which included Kansas City PD). Recalculation across all responding agencies led to approximately 120 agencies with changed population served values. Following analysis of the corrected data, numerous reported estimates have small differences as compared to published estimates. These program errors affected the following figures and tables:

  • Figure 2
  • Tables 4-6, 8-14
  • Appendix tables 1, 5-7, 9, and 12-17

Edits were also made to the text referencing these tables.

Indicators for Crime Estimates Using NIBRS Data9/14/2022On the cover page, corrections were made to the contributing authors of the report.
Jails in Indian Country, 2019–2020 and the Impact of COVID-19 on the Tribal Jail Population3/31/2022During the 2021 Survey of Jails in Indian Country data quality follow-up with the Navajo Department of Corrections in Crownpoint, the respondent informed BJS that the 5 expedited releases reported in the 2020 survey was incorrect, and that the actual number was 768. As a result, in the page 1 highlights, the fourth bullet’s text changed from “about 7% (1,510)” to “about 11% (2,300).” This and other related changes also affected text and table 5 on page 6.
Prisoners in 2020 – Statistical Tables12/14/2021On page 46 in appendix table 2, the Texas cell values were erroneously shifted left. This has been corrected with no impact on text within the body of the report
Background Checks for Firearm Transfers, 201810/29/2021On page 19, the jurisdiction note for Oregon indicated that the state was a non-point of contact state when it should have been classified as full point of contact. This has been corrected.
Jails in Indian Country, 2019–2020 and the Impact of COVID-19 on the Tribal Jail Population10/21/2021On page 14 in appendix table 1, the 2019 total for juveniles has been corrected from 47 to 123. This correction has no impact on the text within the body of the report.
Recidivism of Prisoners Released in 34 States in 2012: A 5-Year Follow-Up Period (2012–2017)9/8/2021On page 21, the text “… lower among those released in 2005 than among those released in 2012 …” was corrected to read “… lower among those released in 2012 than among those released in 2005.” 

On pages 3 and 7 and in the footnotes in table 3, figure 3, and table 6, the text describing offenses included in the analysis was revised. “Juvenile offenses may be excluded from this analysis” was changed to “some juvenile offenses may not be accounted for in the analysis.”
State and Local Law Enforcement Training Academies, 20137/7/2021The Bureau of Justice Statistics is reissuing this report using a different weighting methodology for the data collected in the 2013 Census of Law Enforcement Training Academies. The original report weighted unit nonresponse based on the number of recruits in each type of academy, which required imputing values for the number of recruits for both item and unit nonresponse. These imputed values could not be replicated using the archived datafile. The methodology used in this revised report weights respondents by academy type (e.g., State POST or local police department; see table 17) and applies that weight to all values provided by responding agencies. This resulted in different estimates from those previously released. For example, the overall estimates of recruits changed from 44,891 to 45,149. Similarly, there were small changes in tables and text throughout the revised report.
Immigration, Citizenship, and the Federal Justice System, 1998-20181/27/2021Updated data resulting from late postings on suspects in matters concluded during fiscal year 2018 were made available after this report had been published. This affected tables 12, 15, 16, 17, and 18. An associated bullet on page 2 was revised as well.
Local Police Departments, 2016: Personnel1/27/2021Shortly before publishing, the population served by the Las Vegas Metropolitan Police Department was updated to 1,048,717 in appendix table 1. This change shifted the classification of population served from 500,000 to 999,999 to 1 million or more, but this change was not reflected in other tables that describe local police departments by population served. Tables, figures, and the summary have been updated accordingly.
Sheriffs' Offices, 2016: Personnel1/27/2021The note in appendix table 1 was revised to better reflect which agencies were included in the table, and also added additional non-respondents to the 2016 LEMAS survey who were in the top 50 largest offices in 2012.
Prisoners in 201910/23/2020In table 10, the incorrect number of U.S. Hispanic female residents and other female residents (denominator) was used in calculating the imprisonment rates of Hispanic and other females. These rates were revised. An associated bullet on page 16 was revised as follows: "The imprisonment rate of Hispanic females (63 per 100,000 Hispanic female U.S. residents) was 1.3 times the rate of white females in 2019, and was higher than all age groups except white females ages 45 to 49."
Criminal Victimization, 20189/14/2020A coding error assigned victimizations with an attempted (but not completed) theft as trespassing in household-level victimization rates. These victimizations are now coded as burglary. This resulted in a decrease in trespassing estimates and an increase burglary estimates. It affected victimization rates for all years; however, the total burglary/trespassing estimate was not affected. See Criminal Victimization, 2019 for updated estimates.
Drug Use, Dependence, and Abuse Among State Prisoners and Jail Inmates, 2007-20098/10/2020As originally written, table 14 included estimates of those who participated in drug or alcohol treatment, not those who participated in drug treatment, and table 15 included estimates of those who met the drug dependence and abuse criteria, not those who participated in drug treatment. These program errors affected tables 14 and 15 (appendix tables 19 and 20). Edits were made to the text referencing these tables as well.
Mortality in Local Jails 2000-2014 Statistical Tables1/27/2020In tables 8 and 9, the data in the convicted and unconvicted rows are transposed.
Full-Time Employees in Law Enforcement Agencies, 1997-201610/25/2019One of the reporting agencies in the 2016 LEMAS dataset was ineligible, and three local police departments misreported the number of full-time sworn officers and non-sworn employees. The dataset was updated with corrected counts of full-time sworn officers and civilians.
National Crime Victimization Survey (NCVS) API2/5/2019When requesting the NCVS household population data for a given year in Comma Separated Values (CSV) format only, the API returned incident numbers instead of population numbers for households.
Prisoners in 20168/7/2018Oklahoma updated their count of prisoners (both sentenced and total).
Criminal Victimization, 20154/3/2018Issues were identified with several standard error calculations. This affected tables 1, 7, and 8 (appendix tables 2, 11, and 12). Edits were made to the text for table 1 to reflect these changes as well.
Federal Prosecution of Commercial Sexual Exploitation of Children Cases, 2004-201311/21/2017Updated “possession of child pornography” category in tables 1-8 and related text to “possession, receipt, or distribution of child pornography.”
State-Administered Indigent Defense Systems, 20135/3/2017The number and types of cases closed, number of full-time litigating attorneys, and caseload calculations were updated for Connecticut, Kentucky, North Carolina, and Virginia. These changes were made based on feedback from indigent defense experts in those states. The Methodology was updated to include more detail about universe identification and calculation of full-time equivalent workloads.
Federal Justice Statistics, 2014 - Statistical Tables3/30/2017In table 1.3, the values for statistics on race were presented in the incorrect order. Data have been reordered accordingly.
Probation and Parole in the United States, 20152/3/2017In Appendix table 1, Georgia 2015 percent change should be -10.0 and not -1.0.
Arrest-Related Deaths Program Redesign Study, 2015-16: Preliminary Findings12/22/2016In table 1, the order of the data sources was revised to (1) ARD Program, (2) Fatal Encounters, (3) Killed by Police, (4) The Counted, (5) Gun Violence Archive, and (6) National Police Shootings Database.
Police Behavior during Traffic and Street Stops, 201110/27/2016The total percentage of persons identified as American Indian or Alaska Native who were involved in street stops in 2011 has been revised to 0.4%. Data in table 1 and appendix table 3 have been updated accordingly.
Census of Problem-Solving Courts, 201210/12/2016Highlight bullet #7 was revised as follows: "Overall, 57% of all problem-solving courts reported that more than half of the exits were successful program completions."
Key Statistics: Corrections8/8/2016Incorrect data were presented for the total correctional population in 2007 and the rate of correctional supervision from 1980 to 2014. Tables and figures have been updated accordingly.
Survey of State Criminal History Information Systems, 20146/16/2016An earlier version of this report presented incorrect data for Michigan and Missouri on table 14. These totals have been corrected, and counts presented in applicable descriptions on page 7 have been updated accordingly.
Female Victims of Sexual Assault, 1994–20105/23/2016Figure 3 was calculated from a base of both male and female victims rather than only female victims. Because males account for a small portion of rape and sexual assault victims, the overall trend does not change but the annual percentage of victimizations reported to the police changed slightly.
Correctional Populations in the United States, 20141/26/2016A number was changed on page 1 in the first paragraph." After peaking at 7,399,600 person in 2007..." was changed to "After peaking at 7,339,600 in 2007."
Arrest Data Analysis Tool11/17/2015Two issues with agency-level counts have been identified and corrected. Arrest counts for the 55 to 59 age group and 60 to 64 age group incorrectly displayed the same numbers for several agencies in different years. Also, the “all offense” totals being displayed were incorrect for some years and agencies.
Criminal Victimization, 201410/8/2015On page 4 (Prevalence of crime), 2nd column, 3rd paragraph, it states “The prevalence of serious violence committed by strangers declined slightly from 2013 (0.19%) to 2014 (0.23%).” "Decline" was changed to increase.
Prisoners in 201410/8/2015"State” and “U.S. total” columns in Figure 1 of the .csv files were adjusted to reflect the correct column.
Intimate Partner Violence, 1993–201010/8/2015On page 4, the last sentence in the last paragraph, “2000” was changed to 2010.
HIV in Prisons, 2001-20103/24/2015In table 5, the incorrect denominator was used when calculating the 2001-2010 rate of AIDS-related deaths among black and Hispanic inmates. These rates were revised.
Federal Criminal Case Processing Statistics (FCCPS) data analysis tool2/23/2015The trends module in the Federal Criminal Case Processing Statistics (FCCP) web query tool was generating two types of errors for queries that involved more than one year of data and federal judicial district. For “Persons arrested and booked (trends),” an error message occurred when the user queries more than one year of data and federal judicial district. Also, the tool displayed incorrect data for queries on “Defendants charged in criminal cases” (AO), “Defendants in criminal cases closed” (AO), and “Offenders sentenced” (USSC that included more than one year of data and federal judicial district.
Indigent Defense Services in the United States, FY 2008–20122/17/2015The statute numbers and code references in the Texas section under local administration were incorrect. An official from Harris County, Texas, emailed the correct references, and the document was updated.
Probation and Parole in the United States, 20131/29/2015The footnotes in Appendix Tables 1 and 2 mistakenly state that the rates were computed using the estimated U.S. adult population in each jurisdiction on January 1, 2013. The estimates were based on the population for January 1, 2014.
Probation and Parole in the United States, 20121/29/2015The footnotes in Appendix Tables 1 and 2 mistakenly state that the rates were computed using the estimated U.S. adult population in each jurisdiction on January 1, 2012. The estimates were based on the population for January 1, 2013.
Also, on page 6 the report incorrectly states that 54% of probationers were supervised for a felony offense in 2012. The correct percentage is 52%, as shown in Appendix table 3.
Corrections Statistical Analysis Tool (CSAT) - Parole1/13/2015In the parole data analysis tool released in October 2014, the labels for "Incarceration sentence of more than one year" and "Incarceration sentence of a year or less" were reversed. The labels have been corrected.
Prisoners in 201310/22/2014Appendix table 1 has been updated with new capacity estimates for Ohio.
Prisoners in 2012: Trends in Admissions and Releases, 1991-201209/18/2014Unlike previous years, the 2011 federal data included felons from the District of Columbia. For consistency, these felons have been excluded from the data and all related items in the pdf, csv files, and CSAT web tool have been updated.
Jail Inmates at Midyear, 2013 - Statistical Tables09/04/20142013 Annual Survey of Jails - Rated capacity error for one jail jurisdiction in the final dataset delivered from the Census Bureau. As a result, all data related to rated capacity for 2013 has been revised. The Census Bureau is investigating the error as a potential technical issue in their data system.
Capital Punishment, 2012 - Statistical Tables09/04/2014Table 4 on page 17 - The name of two states have been transposed: The row currently labelled Massachusetts should be Delaware, and the subsequent row labelled Delaware should be Massachusetts.
Immigration Offenders in the Federal Justice System, 201010/31/2013For 2002 and 2003 data, age was computed incorrectly as age at commitment instead of age at fiscal yearend. After 2004, age was computed correctly as age at fiscal yearend. To ensure consistency for all years, table 13 and associated text in the report were corrected to display all years as age at fiscal yearend.
State Corrections Expenditures, FY 1982-201010/31/2013In table 4, Florida's 2001 medical expenditures were not originally calculated in 2010 real dollars. The changes, percentages, and per capita medical expenditures for Florida were recalculated to reflect inflation-adjusted 2010 real dollars.
Stalking Victims in the United States - Revised9/28/2012Since the original January 2009 release of the report Stalking Victimization in the United States (NCJ 224527), which was based on the Supplemental Victimization Survey (SVS), two errors have been identified:The estimates were calculated using an incorrect sample selection procedure.
Incorrect populations were used to generate rates in tables 1 and 3.
Justice Expenditure and Employment Extracts, 2009 - Preliminary9/19/2012The expenditure part of table 8 was based on state expenditures only. It has been revised to use state and local expenditures.
Prisoners in 20102/9/2012After releasing Prisoners in 2010, the Georgia Department of Corrections resubmitted its previously verified 2009 and 2010 custody, jurisdiction, and local facility prisoner counts. The text discussing state imprisonment rates incorrectly cited national imprisonment rates. Also, table 5 had incorrect growth-adjusted release rates.
Sexual Victimization in Juvenile Facilities Reported by Youth, 2008-092/2/2012Appendix tables 2, 4, and 5 have been corrected because of an error reported by the Juvenile Correctional Center in Lewiston, ID. Due to an error in the administration of the survey, only 4 of 29 interviewed youth received questions on both forced and non-forced sexual contact. Of these 4, only 1 reported victimization. Due to the small number of completed interviews, this rate has been suppressed.
Federal Justice Statistics, 2009 - Statistical Tables1/26/2012Table 4.4 in the pdf document displayed 2008 data. It has been corrected to show 2009 data. The .csv spreadsheet file was not affected by this error.
Crime Against Persons with Disabilities, 2008-2010 - Statistical Tables12/22/2011Aged-adjusted rates of violent crime against persons with disabilities by sex, race, and Hispanic origin have been recalculated.
Sexual Victimization in Local Jails Reported by Inmates, 200712/22/2011Estimates in the report described the National Inmate Survey's limited reporting of sexual victimization incidents to "the last 6 months, or since admission to the facility, if less than 6 months." After the report was published, it was discovered that the reporting period programmed in the audio computer-assisted self interview (ACASI) computer instrument was based on incidents occurring in "the last 12 months, or since admission, if less than 12 months." The estimated numbers in the report were not impacted by this amendment.
Prisoners in 200912/13/2011Table 4, footnote c should read …”and other unconditional releases.”
P. 12 Definitions: conditional releases should read “Includes releases to probation, supervised mandatory releases, and other unspecified conditional releases.”
p. 13. Definitions: unconditional releases should read “Includes expirations of sentence, commutations, and other unconditional releases.”
p. 14 Jurisdiction notes: Add Alabama – Operational capacity represents physical capacity for inmates but is not based on staffing, programs, and services.
p. 14 Jurisdiction notes: Add North Carolina – Standard operational capacity is equal to one inmate per cell or 50 square feet per inmate in a dormitory setting.
Appendix table 9, foot note a should read …”See Jurisdiction notes.” (not Methodology).
Appendix table 23, North Carolina change to data (state respondent updated capacity data in 2011). Rated capacity is “…”; Operational is 40,014; Design is 34,364; Highest capacity is 100 and Lowest capacity is 117.
Jail Inmates at Midyear, 2010 - Statistical Tables6/28/2011Miami-Dade County, FL, revised its rated capacity from 6,035 to 5,845. Los Angeles County, CA, noted that its 2008 and 2009 rated capacity was its operating capacity. In addition, the number of inmates in 2008 was revised to 785,533 from 785,536. These changes affected data in several tables and have been highlighted in the report. Percentage change in average daily population for 2010 in table 1 was misreported.
Criminal Victimization, 20056/23/2011Estimates were calculated using incorrect sampling weights. Corrected weights result in relatively small revisions to estimates of victimization counts and rates. For example, the violent crime rate dropped from 21.2 per 1, 000 person age 12 or older to 21.1 per person. The variance data for the revised estimates show that all conclusions and findings presented in the bulletin remain unchanged. See the bulletin for additional details.
Local Police Departments, 20076/23/2011On page 33, the partial form weight of 20.04 should read 2.21.
Indicators of School Crime and Safety, 20101/12/2011Executive Summary - Key Findings: In 2008, among students ages 12 -18, there were about 1.2 million victims of nonfatal crimes at school, including 619,000 thefts and 629,800 violent crimes (simple assault and serious violent crime).
Mortality In Local Jails, 200712/7/2010This report, originally released on 10/7/10, has been revised to exclude duplicate death records identified in the data and to include updates reported to the Bureau of Justice Statistics (BJS) after release. Thirteen death records were determined to be duplicates, reducing the total number of deaths in local jails during 2000 through 2007 to 8,097 from 8,110. In addition, 2 jails updated cause of death information on a total of 12 deaths. An errata sheet explaining the purpose of these revisions is also available for review.
Public Defender Offices, 2007 - Statistical Tables6/17/2010The Bureau of Justice Statistics is re-issuing the report, Public Defender Offices, 2007 - Statistical Tables, originally published on November 19, 2009, to correct errors recently identified in the data. The revised report corrects caseload data for one office, excludes caseload data that was determined to be unreliable (out-of-range) for another office, and excludes all data for seven additional offices that were found to be ineligible (out-of-scope). These changes affected data in the text and tables. An errata sheet explaining the purpose of these revisions is also available for review.
Prisoners in 20084/1/2010Selected characteristics of the prison population under state and federal jurisdiction. Tables 1 and 2.
Table 1. Estimated prisoners by race.
Table 2. Imprisonment rate per 100,000 persons in the U.S. resident population, by race, Hispanic
origin, and gender, 2008.
HIV in Prisons, 2007-083/28/2010Missouri found an error in their original submission for 2008. New data were submitted and pages 1-3, including tables 1-4, were revised.
The total number of inmates held in state or federal prison as of December 31, 2008, who were HIV positive or had confirmed AIDS, was revised to 21,987.
Female Victims of Violence10/23/2009Added footnote on "Fatal intimate partner violence," page 2.
Asterisk was added to explain bullet on page 2 - Homicide data are voluntarily reported to the FBI by law enforcement agencies active in the Uniform Crime Reporting Program. Offender information (and, therefore, information on the victim-offender relationship) is missing for about 1 in 3 murders reported. This information is missing because either no offender was identified or information on the identified offender was not sent to the FBI. For this report, missing victim-offender relationships were estimated by assuming that the distribution of relationships in murders, for which the relationship was known, was the same as in murders for which the relationship information was missing.
Table 2. Revised the age at offense for percent of intimate partner violence by victims and defendants.
State and Local Law Enforcement Training Academies, 20064/10/2009Page 2: Approximately half (52%) of academies employed the fulltime equivalent of 25 instructors or more. These academies employed more than four-fifths of both full-time (86%) and part-time (81%) instructors.
Page 10: More than three-fifths of academies operated by county police (89%), state police (75%), sheriffs' offices (71%), or municipal police (66%) had training environments they described as either predominantly stress or more stress than non-stress.
Prison Inmates at Midyear 2008 - Statistical Tables4/8/2009Table 6. Percent of all sentenced prisoners was revised to 93.1%.


Date Created: May 13, 2021