Jacob Kaplan's Concatenated Files: National Incident-Based Reporting System (NIBRS) Data, 1991-2021

Version 7 release notes:Adds 2021 data. Version 6 release notes:Adds batch header segments for all years. Version 5 release notes:Adds 2020 dataVersion 4 release notes:Fix bug where most years had arrestee and property were incorrectly window arrestee and window property segments.Changes R files from .rda to .rds. Version 3 release notes:Adds 2019 dataVersion 2 release notes:Changes release notes description, does not change data.These data are the FBI's National Incident-Based Reporting System (NIBRS) data for years 1991-2018. NIBRS data are incident-level data that have highly detailed information for each crime that is reported to the police agency. This data has 10 segments. Each segment has different data about the crime. AdministrativeBasic information about the crime incident - this is basically metadata about the other segments for this crime. This includes the date of the crime, the number of offense segments, the number of victim segments, the number of offender segments, the number of arrestee segments, if the crime was cleared exceptionally and (if it was) what date it was cleared. ArresteeArrestee-level information for those who are arrested. This includes demographics (age, sex, race, ethnicity), the date of the arrest (can be different than the date of the crime), what weapon (if any) was used, and the outcome of the case if the arrestee was a juvenile. Group B Arrest ReportsArrestee-level information for those who are arrested for Group B crimes. This includes the same variables as the arrestee segment. OffenderOffender-level information for each offender. Includes offender demographics (age, sex, race, ethnicity).OffenseDetailed information about each crime. Includes the weapon used (if any), the location of the crime, if the offender was intoxicated (including drugs and alcohol), and what their bias motivation (if any) was (if there is one, this would be considered a hate crime). PropertyInformation about property involved in the crime (i.e. drugs or stolen property). This includes the value of the property, what type of the property it was, when it was recovered. For drugs, this includes the drug and its quantity. VictimVictim-level information for each victim of a crime. Includes victim demographics (age, sex, race, ethnicity), injury, and relationship to the offender(s). Window ArresteeWindows segments have the same columns as their non-window counterparts and are incidents that occurred prior to the year of data or prior to when the agency started reporting to NIBRS. Window Exceptional ClearanceWindows segments have the same columns as their non-window counterparts and are incidents that occurred prior to the year of data or prior to when the agency started reporting to NIBRS.Window PropertyWindows segments have the same columns as their non-window counterparts and are incidents that occurred prior to the year of data or prior to when the agency started reporting to NIBRS. Due to the large file size, each year is its own file. All segment headers are available except for the batch headers. What I did here was read the data into R and save it as R and Stata files. No other changes to the data were made. The data was downloaded as NIBRS Master Files for each year from the FBI's Crime Data Explorer website - https://crime-data-explorer.fr.cloud.gov/downloads-and-docs.### 版本更新说明 V7版本更新说明：新增2021年数据集。 V6版本更新说明：为所有年份新增批次头段（batch header segments）。 V5版本更新说明：新增2020年数据集。 V4版本更新说明：修复了多数年份中逮捕人段与财产段被错误识别为窗口逮捕人段和窗口财产段的漏洞；同时将R语言格式文件从.rda格式转换为.rds格式。 V3版本更新说明：新增2019年数据集。 V2版本更新说明：仅更新更新说明文本，未对数据集本身进行任何修改。 ### 数据集概况 本数据集为美国联邦调查局（FBI）发布的1991-2018年国家事件报告系统（National Incident-Based Reporting System, NIBRS）数据集。NIBRS数据集属于事件级数据，收录了向警方报案的每一起刑事案件的高度细化信息。本数据集共包含10个数据段，每个数据段对应案件的不同维度信息： 1. **行政段（Administrative）**：收录案件的基础概况信息，本质为该案件其余数据段的元数据。涵盖案件发生日期、犯罪段数量、被害人段数量、作案人段数量、逮捕人段数量，以及案件是否以特殊方式告破、若告破则记录其告破日期。 2. **逮捕人段（Arrestee）**：收录被逮捕人员的个体级信息。包含人口统计学特征（年龄、性别、种族、族裔）、逮捕日期（可与案件发生日期不一致）、所使用的作案工具（若有），以及若被捕者为未成年人时的案件处置结果。 3. **B类逮捕报告段（Group B Arrest Reports）**：收录因B类刑事案件被捕人员的个体级信息，字段与逮捕人段完全一致。 4. **作案人段（Offender）**：收录每一名作案人的个体级信息，包含其人口统计学特征（年龄、性别、种族、族裔）。 5. **犯罪段（Offense）**：收录每一起刑事案件的详细信息。包含所使用的作案工具（若有）、案件发生地点、作案人是否处于中毒/受影响状态（包括毒品与酒精影响），以及是否存在偏见性作案动机（若有，此类案件将被认定为仇恨犯罪）。 6. **财产段（Property）**：收录案件涉及的财物信息（如毒品或被盗财物）。包含财物价值、财物类型、追回日期；若为毒品，则额外收录毒品种类与数量。 7. **被害人段（Victim）**：收录每一名刑事案件被害人的个体级信息，包含其人口统计学特征（年龄、性别、种族、族裔）、所受伤害情况，以及与作案人的关系。 8. **窗口逮捕人段（Window Arrestee）**：该类数据段的字段与对应非窗口数据段完全一致，收录数据年份之前或警方开始向NIBRS报送数据之前发生的案件相关信息。 9. **窗口特殊告破段（Window Exceptional Clearance）**：该类数据段的字段与对应非窗口数据段完全一致，收录数据年份之前或警方开始向NIBRS报送数据之前发生的案件相关信息。 10. **窗口财产段（Window Property）**：该类数据段的字段与对应非窗口数据段完全一致，收录数据年份之前或警方开始向NIBRS报送数据之前发生的案件相关信息。 ### 数据集存储与处理说明 由于数据集文件体积较大，故按年份分文件存储。除批次头段外，其余所有数据段的表头均已提供。本次处理仅将原始数据导入R语言环境后，分别保存为R格式与Stata格式文件，未对数据集本身进行任何其他修改。本数据集的原始数据为从美国联邦调查局犯罪数据探索者官网（https://crime-data-explorer.fr.cloud.gov/downloads-and-docs）下载的各年度NIBRS主文件。