Um ... Hello, What Do I Always Say?

I got a question in my inbox last night that I thought was of sufficient import that it needed a blog. In a nutshell, the question was:

"what is the difference between evidence-based and evidence informed approaches, and, given the practical realities of policing, is an evidence informed approach more realistic than trying to urge an organization to become evidence based?"

I cannot express how much I love this question and how grateful I am to my Saturday night correspondent! Any subsequent snark is definitely not directed at him.

Let's start with some basic definitions:

Evidence-based

In the context of policing, we're talking about a perspective in which practitioners believe that policing strategies, policies and practices should be developed based on careful targeting of identified problems, and, once implemented (preferably as controlled pilot projects), tested and tracked to ensure they work effectively and efficiently (Sherman 2015). The key here is:

1. to combine scientific and police knowledge to effectively identify and develop treatments for crime and other operational issues;

2. to rigorously test those treatments to ensure they actually meet their objectives

3. to track those interventions over time to ensure they continue to 'work' and/or to course correct when required, testing those corrections as required.

As Larry Sherman has extensively argued - as well as a whole host of other EBP advocates, including myself - the field of policing is littered with a dizzying array of policing initiatives that have been nothing less than dismal to spectacular failures*. As a small example, take a look at our Square 1 assessments and you will find the majority of the programs reviewed come with significant cautions and/or statements that no research evidence supports the program.

The other major element of what it means to be 'evidence based' rather than 'evidence informed' lies in the idea of a 'base.' Some have mistakenly assumed 'evidence based' simply means you base your decision, your policy or your practice on a reading of the relevant studies you could find.

The key here is that base should mean the entirety - that is, the cumulative sum of the effects and/or findings - of all of the decent quality research available. Not just the stuff you cherry picked. And, yes, guy from Alberta who once presented at a conference on how he cherry-picked his literature review so that 'his study' magically came to his pre-determined conclusion, I am talking to you!

Why emphasize this notion of a base? Because when a researcher, such as myself, tells you that you can reduce burglaries in an area by targeting prevention efforts at select homes near a place that's just been hit, what would you prefer:

a. that I could point to hundreds of credible studies that have been independently conducted by hundreds of different researchers and police organizations in cities and towns across the globe, or

*b.I googled a couple of things and found some interesting stuff on some websites that looked pretty good**?

Evidence informed

This is a sort of catch-all term that has been picking up speed lately in response to the growth of evidence based practice. Whereas some practitioners use the two terms interchangeably, within the health and management fields the term has come to be used to highlight practitioner concerns that requiring rigorous research and evaluation of proposed interventions sets the bar too high. Whereas EBP advocates near-universally privilege the marriage of high quality scientific research with police expertise, some proponents of evidence informed approaches welcome less rigorous approaches to data, including what is collected, how it is collected, analyzed and used. For EBP, clinical judgements based on an systematically gathered sample of 5 practitioners would never fly, nor would case studies of a n of 1 (n=1 in this case refers to one person)***. Based on my readings in the health sciences, it is not at all clear to me that EIP advocates would feel similarly. Further, it also unclear whether EIP would demand testing of possible interventions, being satisfied they were simply 'evidence informed.' It is this setting of the bar -- at a much lower level -- that worries people like me, being all too familiar with how significant elements of policing as a whole has taken similar models of the past - such as problem oriented and community policing - and used them as little more than public relations exercises on websites and social media. Or worse yet, stood up at conferences and self-extolled their use of 'CP', 'POP' and lately 'EBP' while describing practices that bear zero resemblance (and, yes, I haven't forgotten you, Mr. Cherry Picked Lit Review).

I can hear it now, "But, Laura, EBP is HAAAAAAARRRRRDDDDD! Mwah wah wah wah wah wah."

Oh, sorry, what were you saying? Oh, that's right. Unh ... no one said you had to become EBP over night. What myself, Sherman and others are advocating for is an ideal to which we feel policing should aspire, hopefully enacting steps towards this ideal over time in a series of thoughtful, careful increments. What no one is suggesting is something akin to one of those crazy 2 year organizational reshaping/restructuring/modernizing plans of which the same overwhelmed, under-resourced and over-burdened police services seem rather fond and inclined to enact on a roughly 5 to 7 year basis.

Remember: it's about evolution, not revolution

* Renee Mitchell and Stuart Lewis raise an interesting question: given the enormous burden - in terms of operational budgets, demands on staff, training costs, potential burdens to local communities - that untested policing initiatives can generate, do police leaders have a moral imperative to rigorously test such strategies before approving their implementation? Mitchell, among others, clearly believes they do.

** Option 3: Or, I have a pet theory based on what happened to my neighbor's cousin and I'm a criminologist, so who else would you take crime prevention advice from?

*** Yes, I've seen case studies in the Health Sciences that actually had a n of 1. The argument is that this case is so interesting, other people need to know about it. That generally wouldn't, and shouldn't fly, outside of texts of medical oddities.