Public Safety Crime Report

Columns

• DR_NO: A unique identifier for each crime report. This numerical code ranges from 817 to 252 million, with a mean value around 220 million and a standard deviation of 13.2 million.
• Date Rptd: The date when the crime was reported. This datetime field covers incidents from 1 January 2020 to 28 March 2025.
• DATE OCC: The actual date the crime occurred. This datetime field spans from 1 January 2020 to 27 March 2025.
• TIME OCC: The time the crime occurred, typically in 24-hour format. Values range from 1 to 2359, with a mean of 1.34k and a standard deviation of 651.
• AREA: A numeric code representing the geographical area of the crime. Codes range from 1 to 21, with a mean of 10.7.
• AREA NAME: The name of the geographical area. There are 21 unique area names, with "Central" being the most common, accounting for 7% of records, and "77th Street" for 6%.
• Rpt Dist No: The reporting district number for the incident. Numbers range from 101 to 2199, with a mean of 1.12k and a standard deviation of 611.
• Part 1-2: A classification of the crime, where "Part 1" indicates more serious crimes and "Part 2" less serious ones. Part 1 crimes account for approximately 60% of records, and Part 2 for approximately 40%.
• Crm Cd: A numeric code representing the type of crime. Codes range from 110 to 956, with a mean of 500.
• Crm Cd Desc: A description of the crime type. "VEHICLE - STOLEN" is the most frequent, making up 11% of records, followed by "BATTERY - SIMPLE ASSAULT" at 7%. There are 140 unique crime descriptions.
• Mocodes: Modus operandi codes describing the method used in the crime. This field has 15% missing values. Code "344" is the most common among non-null entries.
• Vict Age: The age of the victim. Ages range from -4 to 120, with a mean age of 28.9 and a standard deviation of 22. Note: The presence of negative ages may indicate data entry anomalies.
• Vict Sex: The gender of the victim. "M" (Male) accounts for 40% of records, "F" (Female) for 36%, and 24% are other or unspecified. This field has 14% missing values.
• Vict Descent: The ethnicity or descent of the victim. "H" (Hispanic) is the most common at 29%, followed by "W" (White) at 20%. This field has 14% missing values. There are 20 unique descent codes.
• Premis Cd: A numeric code for the type of premises where the crime occurred. Codes range from 101 to 976, with a mean of 306.
• Premis Desc: A description of the type of premises. "STREET" is the most common at 26%, followed by "SINGLE FAMILY DWELLING" at 16%. This field has a small number of missing values (588). There are 306 unique premise descriptions.
• Weapon Used Cd: A numeric code for the weapon used, if applicable. This field has a significant number of missing values (67%). Codes range from 101 to 516.
• Weapon Desc: A description of the weapon used. When a weapon is recorded, "STRONG-ARM (HANDS, FIST, FEET OR BODILY FORCE)" is the most common description, representing 17% of total records. This field also has 67% missing values.
• Status: A status code of the crime case. "IC" (Invest Cont) is the most common, making up 80% of records.
• Status Desc: A description of the case status. "Invest Cont" (Investigation Continues) is the primary status at 80%, followed by "Adult Other" at 11%.
• Crm Cd 1: An additional crime code for incidents involving multiple offenses. Values range from 110 to 956, with a mean of 500. Only 11 records are missing from this field.
• Crm Cd 2: A second additional crime code. This field has a large number of missing values (93%). When present, codes range from 210 to 999.
• Crm Cd 3: A third additional crime code. This field has 100% missing values (only 2314 valid records out of 1.01m). When present, codes range from 310 to 999.
• Crm Cd 4: A fourth additional crime code. This field has 100% missing values (only 64 valid records out of 1.01m). When present, codes range from 821 to 999.
• LOCATION: A text description of the crime location. There are over 66,000 unique location descriptions.
• Cross Street: A nearby cross street for the crime location. This field has a high percentage of missing values (85%). "BROADWAY" is a common cross street when specified.
• LAT: The latitude of the crime location. Values are primarily concentrated between 33.65 and 34.33, with some records at 0.00.
• LON: The longitude of the crime location. Values are primarily concentrated between -118.67 and -116.29, with some records at -2.37 to 0.00.

Distribution

Usage

• Trend Analysis: Identifying seasonal or yearly patterns in crime rates.
• Predictive Modelling: Developing machine learning models to forecast high-risk areas.
• Policy Planning: Supporting policymakers in designing targeted crime prevention strategies.
• Visualisation Projects: Creating heatmaps, dashboards, and visual reports for crime data.

Coverage

License

Who Can Use It

• Researchers: For academic studies on criminology and public safety.
• Data Analysts: To uncover patterns and generate insights into crime dynamics.
• Law Enforcement Agencies: For operational planning and resource allocation.
• Policymakers: To inform the development of evidence-based crime prevention strategies.
• General Public/Journalists: For understanding local crime landscapes and informing public discourse.

Dataset Name Suggestions

• Crime Records: 2020 Onwards
• Recent Crime Incidents Data
• Crime Trends 2020-Present
• Public Safety Crime Report

Attributes