Albany Data Stories

Albany Data StoriesAlbany Data StoriesAlbany Data Stories

Albany Data Stories

Albany Data StoriesAlbany Data StoriesAlbany Data Stories
  • Home
  • About Albany Data Stories
  • Albany's AIM Funding
  • Albany's Budget 2017-2025
  • Albany's Vendors
  • Albany's Population
  • Albany's Poverty
  • Albany's Taxable Property
  • Albany's Developable Land
  • Albany's Vacant Buildings
  • Albany's Housing
  • Albany's LowInc Housing
  • Albany's APD Complaints
  • Albany Crime Reports Pt 1
  • Albany Crime Reports Pt 2
  • Albany's PILOT program
  • Albany's Financial State
  • Albany's Open Data
  • What's Next

What insight is in the Albany Police DEPT's complaints data?

The Albany Police Department (APD) publishes many statistics and visualizations on openAlbany, more than half of all listings originate from the APD and they represent the most current data found on openAlbany.  


There are several datasets related to citizen complaints and we wanted to explore those datasets to understand what we can learn.  The APD posts a data file of all citizen complaints dating back to 2009 (there is some additional data from prior years but it is not significant).  In addition, there is a month-by-month visualization here.   

 We downloaded the citizen complaint data file and began exploring it.  The data is very straightforward.  There are 5 columns of data - a case number, a type of complaint, a date, and allegation and a count (which is always=1).  There are two values for “Type”: Citizen Complaints are initiated by the public and Administrative Investigations are generated within the Police Department.   

Analysis by year

 We did a small amount of cleaning, deleting any records prior to 2009 and some other invalid data.  


This left 7,223 complaints recorded between 2009-2024.  We can view complaints by year, including both citizen complaints and internal administrative investigations.   2010 had the highest number of complaints (845) and 2018 is the full year with the lowest number of complaints (198).  Note that 2024 includes only 9 months of complaints data.  

all complaints by quarter

 We started to graph out the data for 2009-2024 regardless of type of complaint.  We wanted to avoid the lumpiness of month-to-month analysis so we chose to analyze by year & quarter.  For example, in the graph below for Q3 2009 (the months July-September 2009) there were 225 citizen complaints (the third bar from the left).  


Between the graph and the yearly statistics table above, the data suggests a drop in overall complaints, with a dramatic drop between 2012 to 2018.  

administrative investigations by quarter

We wanted to pull apart the two types of complaints - Citizen Complaints and Administrative Investigations.  We started graphing Administrative Investigations which you will recall are generated internal to the Police Department.  Graphing the number of Administrative Investigations by quarter shows a trend similar to that of all complaints. 

citizen complaints by quarter

 Lastly, we graphed Citizen Complaints (issues initiated by the public).  We generally see the same trends as the previous graphs, however there are more spikes or surges in citizen reports.  For example, in 2014 the reports by quarter are:

  • Q1 (January to March) - 24 citizen complaints
  • Q2 (April to June) - 69 citizen complaints
  • Q3 (July to September) - 16 citizen complaints
  • Q4 (October to December) - 45 citizen complaints 

observations on complaints data

 Looking at the trends there are a number of things that could be going on.  Three possibilities are 1) complaints are decreasing, 2) police force training is having an impact, and/or 3) processes to self-report data are not being utilized as frequently since the system’s inception.  However there is nothing in this data that suggests what the root cause is (or causes are) for a decrease in overall complaints; is it a mix of those three possibilities or is something else going on? 

analyzing allegations coding

Administrative investigation by type of allegation

 

In an effort to better understand the data and identify additional trends we analyzed the “Allegations”.  The “Allegations” field provides additional detail on the complaint or issue.  


We only looked at the Allegations data between 2020-2024; prior Allegations data was frequently freeform (e.g. “CO alleges that the officer put handcuffs on him and tightened them for over a period of 45 minutes”) and not specifically categorized.  



Citizen complaint by type of allegation



  Looking at just the Administrative investigations between 2020-2024 there were 979 Administrative investigations.  64 of the Administrative investigations (7%) were coded with an Allegation; the remaining 93% of Administrative investigations were blank (not coded).  We can review the coding of the Administrative investigation Allegations in the table below.  “Off Duty Conduct” was the most frequent Allegation value. 


Reviewing the Citizen Complaints between 2020-2024 there were 187 Citizen Complaints.  50 of the Citizen Complaints (27%) were coded with an Allegation; the remaining 73% of Citizen Complaints were blank (not coded). 

Recommendations

The Albany Police Department is the most active department on openAlbany and should be recognized for the frequency of their open data posts.  Unfortunately, there are some limitations in the usefulness of the Complaints data that APD is posting.  We have several recommendations that will increase the utility of the Complaints data:

  1. Allegation coding definitions - the data should be published with a glossary that defines the Allegation codes, i.e. when is “Rudeness” used, what is “Call Handling” and how is that different from “Arrest Authority and Procedures”
  2. Allegation coding comprehensiveness - 100% of all Citizen Complaints and Administrative Investigations should be coded with an Allegation, or reasons should be provided when an issue is not coded.  In addition, the APD should be clear about how multiple complaint issues are handled, e.g. if an internal investigation reveals an issue that transcends Use of Force and Vehicle Operations how is that handled?
  3. The APD should construct a narrative - ideally quarterly, at the least yearly - that comments on the trends.  This narrative should describe any cause and effect that could be driving trends (e.g. “we instituted ABC training which resulted in XY showing up in the data”)
  4. The APD should review and specifically comment on the efficacy and comprehensiveness of self-reporting.  Importantly, when we see a drop in issues (which should be a good thing), we want to ensure that the drop is not an artifact of gaps in self-reporting.


Questions or comments?  email us at AlbanyDataStories@gmail.com

Copyright © 2025 Albany Data Stories - All Rights Reserved.

This website uses cookies.

We use cookies to analyze website traffic and optimize your website experience. By accepting our use of cookies, your data will be aggregated with all other user data.

Accept