After last week’s heavier stuff, I'd like to tell this time a little story about the stuff that sometimes - with the best of intentions, no doubt - is released onto the world in the name of safety. This is a relatively harmless example, but it’s from a rather successful and reputed European consultancy firm. An example that doesn't necessarily give confidence in the scientific soundness of safety work. As a profession, we have to shape up!

I found this rather silly example in a recent book about safety culture and (mostly) behaviour that I was asked to review by an associate. Lots to comment on that one, which I will do at a later point. Suffice for now to say ‘old wine, new bottles’.

One of the building blocks presented in the book is the ABC model of behaviours: Activators, Behaviour and Consequences. The rationale behind the model: People display certain (wanted, or in this case safe) Behaviour because of a trigger that came before (an Activator, or Antecedent) and that what happens afterwards (Consequences). As simple as ABC, indeed.

The book then tells us that Activators account less for behavioural change than Consequences. Activators get between 5 and 20% and Consequences between 80 and 95%. 

Let’s stop here for a minute.

Firstly, I think what they mean to say here is that Consequences do facilitate permanent or long-lasting behavioural change more than Activators. But, that might be a slip of the pen. Still, I wonder, how would you know, except that it has intuitive appeal? We do things because of the things that happen to us.

Secondly, let's assume that these numbers are actually based on something substantial (I mean, other than: “We’ve used 5 and 95 earlier in the book, people recognize that. Let’s also take the 80/20 rule - everybody knows that one.”). Just for the sake of the argument. They mention that the number “depends on the source”, but the sources are not clearly specified (hidden further on in the book they do point to some sources, but when I tried to verify some of them, the results were Zero).

Thirdly, let's also ignore the crudeness of the ABC model (article coming soon). And, fourthly, we must not forget to ignore that the book takes the liberty to redefine A and (to a lesser degree) C.

Having done that, we proceed to the next chapter where we are treated to a discussion of positive feedback (a form of Consequences). We are encouraged to do so because, as we saw, acting on Consequences is 80% more effective than acting on Activators and, by the way, positive feedback is 80% more effective than criticism and other forms of feedback.

I’ll leave it to you to spell out what’s wrong here, and then I’m not talking about the question where that final 80% came from, by then I had already disconnected (alas, dutifully I had to struggle with another 140 pages).

No sources, most likely fantasized numbers and shoddy basic maths… What did I say about shaping up?!

Ow, crap, I did it all wrong - this wasn’t positive feedback, after all… Well, the book looks great and is colourful, how’s that?


Also published on Linkedin.