Jacob Kaplan's Concatenated Files: Uniform Crime Reporting Program Data: Offenses Known and Clearances by Arrest, 1960-2019
- Kaplan, Jacob (University of Pennsylvania)
AbstractFor any questions about this data please email me at firstname.lastname@example.org. If you use this data, cite it.
V15 release notes:
- Adds data for 2019.
- Please note that in 2019 the card_actual_pt
variable always returns that the month was reported. This causes 2019 to
report that all months are reported for all agencies because I use the card_actual_pt variable to measure how many months were reported. This variable is almost certainly incorrect
since it is extremely unlikely that all agencies suddenly always
report. However, I am keeping this incorrect value to maintain a
consistent definition of how many months are missing (measuring missing
months through card_actual_type, for example, gives different results for previous years so I don't want to change this).
- Adds arson data from the UCR's Arson dataset. This adds just the arson variables about the number of arson incidents, not the complete set of variables in that dataset (which include damages from arson and whether structures were occupied or not during the arson.
- As arson is an index crime, both the total index and the index property columns now include arson offenses. The "all_crimes" variables also now include arson.
- Adds a arson_number_of_months_missing column indicating how many months were not reporting (i.e. missing from the annual data) in the arson data. In most cases, this is the same as the normal number_of_months_missing but not always so please check if you intend to use arson data.
- Please note that in 2018 the card_actual_pt variable always returns that the month was reported. This causes 2018 to report that all months are reported for all agencies because I use the card_actual_pt variable to measure how many months were reported. This variable is almost certainly incorrect since it is extremely unlikely that all agencies suddenly always report. However, I am keeping this incorrect value to maintain a consistent definition of how many months are missing (measuring missing months through card_actual_type, for example, gives different results for previous years so I don't want to change this).
- For some reason, a small number of agencies (primarily federal agencies) had the same ORI number in 2018 and I removed these duplicate agencies.
- Adds 2018 data
- New Orleans (ORI = LANPD00) data had more unfounded crimes than actual crimes in 2018 so unfounded columns for 2018 are all NA.
- Adds population 1-3 columns - if an agency is in multiple counties, these variables show the population in the county with the most people in that agency in it (population_1), second largest county (population_2), and third largest county (population_3). Also adds county 1-3 columns which identify which counties the agency is in. The population column is the sum of the three population columns. Thanks to Mike Maltz for the suggestion!
- Fixes bug in the crosswalk data that is merged to this file that had the incorrect FIPS code for Clinton, Tennessee (ORI = TN00101). Thanks for Brooke Watson for catching this bug!
- Adds a last_month_reported column which says which month was reported last. This is actually how the FBI defines number_of_months_reported so is a more accurate representation of that. Removes the number_of_months_reported variable as the name is misleading. You should use the last_month_reported or the number_of_months_missing (see below) variable instead.
- Adds a number_of_months_missing in the annual data which is the sum of the number of times that the agency reports "missing" data (i.e. did not report that month) that month in the card_actual_pt variable or reports NA in that variable. Please not that this variable is not perfect and sometimes an agency does not report data but this variable does not say it is missing. Therefore, this variable will not be perfectly accurate.
V11 release notes:
- Adds data in the following formats: SPSS and Excel.
- Changes project name to avoid confusing this data for the ones done by NACJD.
V10 release notes:
- Adds card column to the monthly data. The "pt" cards can be used to indicate if no return was submitted that month. Thank you to Tanaya Devi for the suggestion.
- Adds the mailing address columns to both monthly and yearly data.
- In 2017 the agency "New Orleans" in Louisiana (ORI = LANPD00) reported hundreds of times more unfounded attempted burglaries than actual attempted burglaries for every month. These were all changed to NA (unfounded total burglaries was also changed to NA.
- Adds column for index crimes. Index Violent is the sum of murder, total robbery, total rape, and total aggravated assault. Index Property is the sum of burglary, total motor vehicle theft, and total theft (arson is an index crime but is not available in this data set). Index Total is the sum of Index Violent and Index Property. When looking at these columns please consider that some agencies don't always report all crimes (e.g. Chicago didn't report rape between 1986-2012) so these variables may be undercounts.
- All data now from FBI, not NACJD. See here for the R code I used to read in the files and clean data, and the setup files made to read them in. https://github.com/jacobkap/crime_data
- Changes some column names so all columns are <=32 characters to be usable in Stata.
- Adds 2017 data.
- Removes SPSS (.sav) and Excel (.csv) files.
- Changes column names for clearances to "tot_clr_..." to make explicit that this is all clearances, not just adult clearances.
- The formatting of the monthly data has changed from wide to long. This means that each agency-month has a single row. The old data had each agency being a single row with each month-crime (e.g. jan_act_murder) being a column. Now there will just be a single column for each crime (e.g. act_murder) and the month can be identified in the month column.
- Adds a month column and a date column. This date column is always set to the first of the month. It is NOT the date that a crime occurred or was reported. It is only there to make it easier to create time-series graphs that require a date input.
- Removes all card columns. This was done to reduce file size.
- Reorders crime columns to the order of assaults/deaths of officers, actual crimes, total clearance, clearance under age 18, unfounded. Within each category the columns are alphabetized.
- Monthly data and yearly data are now in different zip folders to download.
- Fix bug where Philadelphia Police Department had incorrect FIPS county code.
- Changes the word "larceny" to "theft" in column names - eg. from "act_larceny" to "act_theft."
- Fixes bug where state abbrebation was NA for Washington D.C., Puerto Rico, Guam, and the Canal Zone.
- Fixes bug where officers_killed_by_accident was not appearing in yearly data. Note that 1979 does not have any officers killed (by felony or accident) or officers assaulted data.
- Adds aggravated assault columns to the monthly data. Aggravated assault is the sum of all assaults other than simple assault (assaults using gun, knife, hand/feet, and other weapon). Note that summing all crime columns to get a total crime count will double count aggravated assault as it is already the sum of existing columns.
- Reorder columns to put all month descriptors (e.g. "jan_month_included", "jan_card_1_type") before any crime data.
- Due to extremely irregular data in the unfounded columns for New Orleans (ORI = LANPD00) for years 2014-2016, I have change all unfounded column data for New Orleans for these years to NA. As an example, New Orleans reported about 45,000 unfounded total burglaries in 2016 (the 3rd highest they ever reported). This is 18 times largest than the number of actual total burglaries they reported that year (2,561) and nearly 8 times larger than the next largest reported unfounded total burglaries in any agency or year. Prior to 2014 there were no more than 10 unfounded total burglaries reported in New Orleans in any year.
- There were 10 obvious data entry errors in officers killed by felony/accident that I changed to NA.
- In 1974 the agency "Boston" (ORI = MA01301) reported 23 officers killed by accident during November.
- In 1978 the agency "Pittsburgh" (ORI = PAPPD00) reported 576 officers killed by accident during March.
- In 1978 the agency "Bronx Transit Authority" (ORI = NY06240) reported 56 officers killed by accident during April.
- In 1978 the agency "Bronx Transit Authority" (ORI = NY06240) reported 56 officers killed by accident during June.
- In 1978 the agency "Bronx Transit Authority" (ORI = NY06240) reported 56 officers killed by felony during April.
- In 1978 the agency "Bronx Transit Authority" (ORI = NY06240) reported 56 officers killed by felony during June.
- In 1978 the agency "Queens Transit Authority" (ORI = NY04040) reported 56 officers killed by accident during May.
- In 1978 the agency "Queens Transit Authority" (ORI = NY04040) reported 56 officers killed by felony during May.
- In 1996 the agency "Ruston" in Louisiana (ORI = LA03102) reported 30 officers killed by felony during September.
- In 1997 the agency "Washington University" in Missouri (ORI = MO0950E) reported 26 officers killed by felony during March.
- Merges data with LEAIC data to add FIPS codes, census codes, agency type variables, and ORI9 variable.
- Makes all column names lowercase.
- Change some variable names
- Makes values in character columns lowercase.
- Adds months_reported variable to yearly data.
- Combines monthly and yearly files into a single zip file (per data type).
- fixes a bug in Version 2 where 1993 data did not properly deal with missing values, leading to enormous counts of crime being reported.
All the data is from the FBI and read into R using the package asciiSetupReader. All work to clean the data and save it in various file formats was also done in R. For the R code used to clean this data, see here. https://github.com/jacobkap/crime_data.
As a description of what UCR Offenses Known and Clearances By Arrest data contains, the following is copied from ICPSR's 2015 page for the data.
The Uniform Crime Reporting Program Data: Offenses Known and Clearances By Arrest data set is a compilation of offenses reported to law enforcement agencies in the United States. Due to the vast number of categories of crime committed in the United States, the FBI has limited the type of crimes included in this compilation to those crimes which people are most likely to report to police and those crimes which occur frequently enough to be analyzed across time. Crimes included are criminal homicide, forcible rape, robbery, aggravated assault, burglary, larceny-theft, and motor vehicle theft. Much information about these crimes is provided in this data set. The number of times an offense has been reported, the number of reported offenses that have been cleared by arrests, and the number of cleared offenses which involved offenders under the age of 18 are the major items of information collected.
1960-01-01 / 2019-12-31Time Period: Fri Jan 01 00:00:00 EST 1960--Tue Dec 31 00:00:00 EST 2019
Is version of
Dave, Dhaval, Monica Deza, and Brady Horn. “Prescription Drug Monitoring Programs, Opioid Abuse, and Crime.” Cambridge, MA: National Bureau of Economic Research, August 2018. https://doi.org/10.3386/w24975.
- ID: 10.3386/w24975 (DOI)
Fischer, Matthew. “Impartiality, Social Network Effects and Collective Memory: Three Essays on Trust in Police.” University of Louisville, n.d. https://doi.org/10.18297/etd/3258.
- ID: 10.18297/etd/3258 (DOI)
Garner, Maryah, Anna Harvey, and Hunter Johnson. “Estimating Effects of Affirmative Action in Policing: A Replication and Extension.” International Review of Law and Economics 62 (June 2020): 105881. https://doi.org/10.1016/j.irle.2019.105881.
- ID: 10.1016/j.irle.2019.105881 (DOI)
Ryo, Emily, and Ian Peacock. “Jailing Immigrant Detainees: A National Study of County Participation in Immigration Detention, 1983–2013.” Law & Society Review 54, no. 1 (March 2020): 66–101. https://doi.org/10.1111/lasr.12459.
- ID: 10.1111/lasr.12459 (DOI)
Valasik, Matthew, Elizabeth E. Brault, and Stephen M. Martinez. “Forecasting Homicide in the Red Stick: Risk Terrain Modeling and the Spatial Influence of Urban Blight on Lethal Violence in Baton Rouge, Louisiana.” Social Science Research 80 (May 2019): 186–201. https://doi.org/10.1016/j.ssresearch.2018.12.023.
- ID: 10.1016/j.ssresearch.2018.12.023 (DOI)
Update Metadata: 2020-10-22 | Issue Number: 1 | Registration Date: 2020-10-22