Welcome to our new website!
Feb. 8, 2022

The Enemy of Good is Perfect: Except when it comes to data and precision policing

The Enemy of Good is Perfect: Except when it comes to data and precision policing

I am sure he could feel the adrenaline still surging through his body as he finished his run. Once in his car, he probably looked at himself in the rear-view mirror focusing squarely on the red hooded sweatshirt he was wearing that was now soaking up his sweat. His probable glance down at the passenger seat where he had tossed the fruits of his labor - the stolen credit cards he had just robbed – reminded him of what he just did. Perhaps this was added to his repeated pensive scans of the rearview mirror, which telegraphed what he was thinking.  Did anybody see me - Could they identify me - Are there cops on my tail?

Like so many criminals, his false sense of bravado would have kicked in - and as he calmed himself, his thoughts shifted to turning plastic into cash.

What Terrence Rhue of Plainfield, New Jersey, did not know, was that following his heinous act of allegedly breaking into the home of a Westfield woman, sexually assaulting her at knifepoint after her early morning walk - then stealing her credit cards, was that within two days he would be in custody thanks to the collective efforts of local, county, and state investigators who were able to leverage technology assets to identify him.

While Rhue wore a blue surgical mask and medical gloves to hide his identity, what he could not conceal was his past criminal history, and those bits of digitalized data identifiers that captured his movements and documented his actions would be central to his identification. There were the various home surveillance videos from the neighbors’ doorbells and porch mounted cameras, the strategically placed license plate reader technology and registration records that connected him to the car he used which he borrowed from his father.

For Rhue, the coup de grace would be the search warrant executed at his house that turned up his red sweat-stained hoodie and the credit cards he had stolen. Innocent till proven guilty, he currently faces first-degree aggravated sexual assault, first-degree robbery, second-degree burglary, and weapons offenses.[1]

The technology used to identify Rhue is not something new to law enforcement. In fact, surveillance cameras and license plate readers have been around for some time and are proven measures for placing perpetrators at the scene of the crime, linking criminals to those crimes, and most importantly providing investigative leads. Why? Because the data that these technologies provide to law enforcement is indispensable for narrowing criminal suspects from the public at large – the right suspects. When these technologies are aggregated with other law enforcement data sets, they enable investigators to become much more precise in who they are targeting for crimes.

But what happens if law enforcement takes data for granted?

Human slippage is no different than Murphy’s Law in that it is always lurking in the shadows and when not accounted for can spoil your day. Take for instance a few years back when we were looking for suspects involved in a robbery in Atlantic City and we could not figure out why we were unable to pick up on technical surveillance data leads on suspects leaving the city. To our chagrin we found out that the person responsible for checking on the portable data readers had not done so after a Nor’Easter had blown through.

Not sure who was more shocked, us or him when the reader was found overturned in the marsh and had been there for approximately two weeks. From a data collection perspective, it was no different when we found that our remote readers that were powered by solar were left powerless when the effect of less sun and shorter days was worsened by snow- and ice-covered sensors - rendering the technology temporarily useless.

Then there is plain old human error. Maybe a spelling mistake on a ticket or using the wrong code on a report. When data is critical - every bit and piece counts. Whether it is for identifying suspects, steering operations, or guiding strategic initiatives like precision policing, data is crucial!

In a recent blog post I discussed the two core principles of precision policing. The first being crime-and-disorder enforcement and the second being community policing. Both of these principles equally require the collection, analysis, and evaluation of data sets that can empower law enforcement agencies to focus their collective efforts in order to ensure the right balance of enforcement and neighborhood policing that can and should take place.

In the past, if the systems were down or even non-existent, or if the data had errors, the traditional policing methods – reactive by nature – really did not suffer all that much or did it. Maybe, people only expected that which could be done at the time – less solutions available – less expectations. However, in this new era of precision policing data is key.

In 2017, The Economist published an article entitled, The world’s most valuable resource is no longer oil, but data. The article went on to describe that 97% of all businesses utilize data to power business opportunities. With that as a backdrop, it is no wonder that data can be a law enforcement organization’s most valuable asset when that data can be trusted. When law enforcement agencies work with incomplete or untrustworthy data for any reason the outcomes can result in incorrect insights, skewed analysis, or faulty recommendations. With the underpinning of precision policing being robust data, it is critical that sustainable measures are always in place to ensure that trustworthy data can be accessed upon demand.

Yet, the reality is that in most law enforcement organizations, to include fusion centers and real time crime centers, the folks that manage the data are buried deep within the organization’s chain of command, and hardly gain face time with leadership unless there is a crisis moment. The mindset that “IT” folks are only there to ensure that applications and systems are running will have to extend into the strategic realm. This will ensure that data sets like the ones used to identify Terrence Rhue are constantly being pursued and integrated into law enforcement operations. In 2022 and beyond, this requires police executives to have a fundamental understanding of the value of data as it relates to carrying out their individual policing missions whether at the tactical, operational, or strategic levels.

One area that police leaders can begin with to better understand data is to acknowledge the two terms that best describe the condition of data: data quality and data integrity. While these two terms have been often used interchangeably, there are important distinctions that must be recognized. Data quality is a subset of data integrity and is assessed against the five elements below:

Is the data complete?

Is the data unique?

Is the data valid?

Is the data timely?

Is the data consistent?

Quality data must meet all of the above criteria. If it does not, then data-driven initiatives, like precision policing will suffer. For example, if an agency has a list of known offenders that is accurate and valid, but there is no supporting data that can provide context about those criminals, the types of crimes they commit, and the association to a particular jurisdiction than the list of these offenders quickly loses its value.

            Data integrity on the other hand refers to whether data is complete, accurate, consistent, and has context. It is data integrity that makes data useful to owners. While data quality is a component of data integrity it is not the only pillar. The four main pillars of data integrity are:

a)    Data integration – data must be seamlessly integrated regardless of its original source in order to provide visibility

b)    Data quality – data must be complete, unique, valid, timely, and consistent

c)     Location intelligence – data can be much more actionable when a geocoding provides location insight and analytics

d)    Data enrichment – when data can be enriched by outside sources it can provide greater context to users and more robust analyses

For law enforcement, data can be leveraged as a tactical, operational, or a strategic asset. In the case of Terrence Rhue above it was the aggregation of disparate data quality sets that led to his identification and arrest. Yet, strategic data-driven initiatives such as precision policing will have to move past just data quality, and on to measures that increase data integrity measures that can drive better planning, decisions, and ultimate outcomes.

This begins with law enforcement leaders and practitioners understanding that perfect data can enable and enhance good public safety.

Of course, at the end of the day, even with the most perfect of data sets, we must still depend upon the dedicated men and women of law enforcement to turn that data into the proper actions required to protect and serve the public.


[1] Note: The details described in the case above came from NJ.com article entitled, Surveillance footage led to arrest of man accused in Westfield home invasion sex assault, cops say, accessed from https://www.nj.com/union/2021/10/surveillance-footage-led-to-arrest-of-man-accused-in-westfield-home-invasion-sex-assault-cops-say.html, on February 3, 2022.