In tor-gu/njoaguofdash: NJ OAG Use of Force data dashboard
NJ OAG Use of Force Data Explorer
This is a simple dashboard to illustrate the use of the njoaguof data package for R, which is a
repackaging of the the use of force data released by the New Jersey
Office of the Attorney General.
The NJ OAG has their own dashboard. If your
interest is in gathering summary data for this dataset, you should start there.
About the data
Current version
This site is based on njoaguof version r packageVersion("njoaguof"), which contains
data from r min(njoaguof::incident$incident_date_1) through r max(njoaguof::incident$incident_date_1).
About the data
incident and subject tables
There are two main tables: incident and subject
The incident table contains one entry for every use of force report filed by a law enforcement officer. Most incidents involve a single officer and a single subject. However:
- A single use of force report may refer to more than one subject of the use of force, and multiple types of force for each subject.
- Multiple officers may file separate reports for the same incident.
- Reports from different officers for the same incident may refer to exactly the same subjects, to different subjects, or to partially overlapping sets of subjects.
Because a single incident report may contain multiple subjects, the subjects are broken out into a separate table.
Unfortunately, the underlying dataset does not allow us to disentagle these potentially overlapping incidents and subjects. Thus, this application counts every incident and every subject separately. This means that the incident and subject counts are typically overcounts. (The NJ OAG dashboard makes the same decision.) However, see the special note on filtering multi-value subject fields, below.
Filtering on Single-Value and Multi-Value fields
Filtering may be done on either "single value" fields or on "multiple value" fields.
A "single value" field may contain only one value for each incident report. For example, there is only one value of officer_gender for each incident and only one value of arrested for each subject.
Multi-value fields are fields in which the law enforcement officer may select multiple values for a single field. For example, more than one location_type ("Street", "Business", "Residence", etc.) may be chosen for a single incident.
This means that when filtereing on multi-value incident fields, the "Relative" scale values may add up to more than 100% for a single region.
Multi-value subject fields {#subjectmulti}
There is an additional complexity with multi-value subject fields when there are multiple subjects in the same incident report.
In some cases, incident reports with multiple subjects include separate
selections for each subject -- for example, an incident report with multiple subjects may contain the the force_type value "Used pressure points on" twice, since it applies to two subjects in the incident report.
In other cases, the incident report will only include a value once, even if it applies to multiple subjects.
As a result, when filtering on multiple-value subject fields, the filtered count may be either an overcount (because the same subject my be included in multiple incident reports) or an undercount (because a single value applied to multiple subjects).
In particular, this means that when filtereing on multi-value incident fields, the "Relative" scale values may add up to either more or less than 100% for a single region.
About this site
This site is built with shiny.
I used the plotly package for the
interactive map.
The table uses the DT package.
The maps and the population data are from the US Census, obtained through the tigris and tidycensus packages, respectively.
The source code is available here.
tor-gu/njoaguofdash documentation built on July 27, 2022, 7:27 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
NJ OAG Use of Force Data Explorer
This is a simple dashboard to illustrate the use of the njoaguof data package for R, which is a
repackaging of the the use of force data released by the New Jersey
Office of the Attorney General.
The NJ OAG has their own dashboard. If your interest is in gathering summary data for this dataset, you should start there.
About the data
Current version
This site is based on njoaguof version r packageVersion("njoaguof"), which contains
data from r min(njoaguof::incident$incident_date_1) through r max(njoaguof::incident$incident_date_1).
About the data
incident and subject tables
There are two main tables: incident and subject
The incident table contains one entry for every use of force report filed by a law enforcement officer. Most incidents involve a single officer and a single subject. However:
- A single use of force report may refer to more than one subject of the use of force, and multiple types of force for each subject.
- Multiple officers may file separate reports for the same incident.
- Reports from different officers for the same incident may refer to exactly the same subjects, to different subjects, or to partially overlapping sets of subjects.
Because a single incident report may contain multiple subjects, the subjects are broken out into a separate table.
Unfortunately, the underlying dataset does not allow us to disentagle these potentially overlapping incidents and subjects. Thus, this application counts every incident and every subject separately. This means that the incident and subject counts are typically overcounts. (The NJ OAG dashboard makes the same decision.) However, see the special note on filtering multi-value subject fields, below.
Filtering on Single-Value and Multi-Value fields
Filtering may be done on either "single value" fields or on "multiple value" fields.
A "single value" field may contain only one value for each incident report. For example, there is only one value of officer_gender for each incident and only one value of arrested for each subject.
Multi-value fields are fields in which the law enforcement officer may select multiple values for a single field. For example, more than one location_type ("Street", "Business", "Residence", etc.) may be chosen for a single incident.
This means that when filtereing on multi-value incident fields, the "Relative" scale values may add up to more than 100% for a single region.
Multi-value subject fields {#subjectmulti}
There is an additional complexity with multi-value subject fields when there are multiple subjects in the same incident report.
In some cases, incident reports with multiple subjects include separate
selections for each subject -- for example, an incident report with multiple subjects may contain the the force_type value "Used pressure points on" twice, since it applies to two subjects in the incident report.
In other cases, the incident report will only include a value once, even if it applies to multiple subjects.
As a result, when filtering on multiple-value subject fields, the filtered count may be either an overcount (because the same subject my be included in multiple incident reports) or an undercount (because a single value applied to multiple subjects).
In particular, this means that when filtereing on multi-value incident fields, the "Relative" scale values may add up to either more or less than 100% for a single region.
About this site
This site is built with shiny.
I used the plotly package for the interactive map.
The table uses the DT package.
The maps and the population data are from the US Census, obtained through the tigris and tidycensus packages, respectively.
The source code is available here.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.