Director’s Note : Iterative Learning and “Spark,” a New Series
You Pracademics are stellar “iterationists” (yes, I made up the word, but it helps to describe what you do, so I’m sticking with it). When you are involved with research you learn, irrespective of the outcome of the study, and often about more than the specific concern of focus. In this issue we highlight the value of iterative learning. Our partners are iterating in two ways, both important. 1. Iterating within your agency. That is, you are taking an idea and testing it again and again after making improvements. 2. Iterating across agencies. A great example of this is a series of BetaGov trials that test the impact of law-enforcement static cruiser lights at strategic locations to deter accidents and crime (we profile a snapshot from a recently completed trial in California). The first trial was a four-site RCT in Connecticut. We feature an interview with Jeremiah Johnson, the Pracademic who led the trial, for his perspective on his multi-site study and the iterative learning that followed when other jurisdictions replicated.
This month we are pleased to introduce another option for practitioners who have an innovative idea that may not be appropriate for a randomized controlled trial. Read more about this line of inquiry, the BetaGov Spark series, below.
Warm regards,
Angela Hawken
|
|
 Pracademia in Practice: Code-2 Lights
Lt. Jason Potts from the Vallejo (CA) Police Department identified promising methods to deter crime related to holiday shopping and tested them. He iterated on the initial four-site, four-month “cruiser light” experiment led by Jeremiah Johnson in Connecticut. Jason tested this approach in a large shopping center, after tailoring implementation and methods to his jurisdiction. He called his trial “Code-2 Lights” and randomly assigned patrol cars, by day of the week, to either “lights on” or “lights off” conditions. Over the month-long trial, significantly fewer motor-vehicle thefts (including auto thefts and thefts from vehicles) occurred in the lights-on condition. Thanks to all you iterationists, we are now accumulating a growing body of evidence about cruiser lights, having studied this approach in nine jurisdictions (we will soon share results of a freshly completed three-month study led by our Pracademic partner and NIJ LEADS scholar, Sgt. Matthew Barter, from the Manchester (NH) Police Department).
|
|
 Learning Corner: How to Learn Iteratively?
Iterative learning is our favorite sort of evidence-based decisionmaking. You start with an idea and you test it. What you learn from that test informs what comes next. Then you test that. And so on.
|
|
 News and Updates
We are proud to announce our new “Spark” series! BetaGov provides assistance to public agencies (and, in some cases, nonprofits) wishing to test an innovation, but some ideas don’t lend themselves to a rigorous experiment. Perhaps randomization isn’t possible, or the sample size is small. In these cases, we might consider a “Spark” project, the opportunity to explore an idea and look for a “signal” rather than statistical significance. It may also be that the innovative idea will include multiple steps or phases, starting with a small data dive or feasibility assessment or client survey and ending with a test of an intervention. So, while the first step or two might not be an experiment, they might serve to inform later RCTs. The Spark series enables us to offer assistance and acknowledge that sometimes it’s more practical and more responsible to dip our toes than to dive in headfirst. The Spark series allows us to share what we are learning during this innovation-incubation phase, with the caveat that these efforts are intended to be exploratory rather than rigorous field experiments. That said, in Spark we will be incubating some truly novel ideas and technologies, so stay tuned!
Here you can see our first Spark snapshot from Pracademics testing a program in a prison in Nigeria.
|
|
 Partner Spotlight: Jeremiah Johnson
This month we feature an interview with Jeremiah Johnson, a Sergeant with the Darien (Connecticut) Police Department and an NIJ LEADS Scholar. Jeremiah was the first Pracademic to test cruise/blue/code lights on patrol vehicles to deter crime. Since the first cruise-lights trial conducted with BetaGov, several others have been launched in the United States and Canada. Jeremiah’s experience and willingness to offer peer-to-peer consultation was instrumental in developing those trials. We recently talked with Jeremiah about his experience and the opportunities for iterative learning with each new cruise-lights trial.
|
|
|
Have an inspired Pracademic day!
|
|
|
|
BetaGov | The Marron Institute of Urban Management | New York University
60 Fifth Avenue, 2nd Floor | New York, NY 10011 | betagov.org
|
|
|
|
|
