What are you expecting to achieve, fixing workplace safety one behavior at a time or fix the context the worker is in that influences the behavior?

What do you mean by intervention, telling people that they are doing something wrong and how they should be doing it, or investigating the organizational factors that "cause" the behavior?

Are you looking at workers as a problem to be solved, or their behaviors they exhibit as an output from the systems they work in.

Not too sure where you are coming from, but you question seems pregnant with a bias towards seeing the worker as the source of problem behaviors

My achievement from this would be to generate conversation and sharing of learning from behaviour interventions. Behaviour is influenced by many many aspects of the workplace and psychological factors of an indivudual or their peers and other groups and I am in no way suggesting that people are the problem.

As an example problem behaviour - people not wearing their safety glasses in the workplace. This a simple behaviour on the surface that is not so simple when you look into it further. There are social influencors involved and people associating their behaviours with the norms of a group which need to be adressed. A behaviour intervention plan looks into all these areas in depth (which is what I should have said in the original post).

So I guess where I am coming from is how have people implemented behaviour intervention plans that positively influence people to display desired behaviours, whilst taking into consideration the psychological, social and oranisational aspects of a workplace (or as HSE's Human Factors put it Organisational, Job and Individual factors in HSG 48)

I just want to put I out there I am not a fan of BBS type programs

There is a lot a variability between organizations and industries, pulsing is not a bad plan as long as you expect to get a large variation in responses.  You alluded to an example in O&G in relation of observation programs, these are imagined (designed) with good intentions but, they are executed poorly and end up focusing on the individual.  I think this can be seen in construction and many other industries.

You mentioned norms of a group.  Group norms are a relative indication to the assumptions that are made between different hierarchical cohorts (we are really talking about the culture in each level of and organization, notice I didn’t say “safety” culture this is because a unicorn dies every time the term “safety culture” is used).  For example, an accident happened on a production line.  Workers misses a step in the process to try and get the job done quickly (at risk/problem behavior).  The worker is questioned and states that they were under pressure to get the job done quickly.  This is reported back as a cause.  The cause is challenged, management never talks about productivity to workers to make sure they don’t feel pressure to work fast and take short cuts.  Worker then asked to explain why they felt rushed.  Worker confirms management do not talk about productivity but states when your supervisor asks you how you are getting on, then the sales manager, then the project manager asks how you are getting on, this put pressure on you and is an indication that there is a rush (this is the assumption that the worker group makes about the other groups expectations based on those other groups behaviors).

In the above example you could have identified the problem behavior before the accident happened, corrected it, gathered the feedback in relation to the cause and reported it back and then progress to try and resolve the cause, but that response is backward looking because the intervention was still after the fact (after the behavior happened).  Also to catch and intervene requires someone with sufficient knowledge of the process to identify the “at risk”/“problem” behaviors and immediately intervene to correct it (resource intensive).  The problem however, is up stream and requires time to get leadership in the organization to understand how their actions and inactions become “antecedents” to the behaviors seen at the worker level.  This may well be the reason why many of these intervention programs don’t tend to work compared to the way they were imagined. 

So if you think about how much potential there is to make a mistake or exhibit a “problem” behavior per worker per shift, what kind of resources would you require to identify them, (would it be done continually or sampling etc.), what resources would you need to investigate the causes, identify effective corrective actions and then implement them monitor them etc.  These programs need to be very well resourced and managed to have the desired results.  In the above example a better time to intervene would have been when the Spv, Sales and project manager were asking “how they were getting on” stopping them from doing that would likely have stopped the problem behavior manifesting in the first place.

Here is a question, would it be easier to plan behavioral interventions and look at each problem behavior on and individual basis or to develop an organization that creates a context where workers do not exhibit at risk behaviors, where is makes sense and is easy for them to do what is expected and is difficult and nonsensical to exhibit at risk behaviors.  A good place to start might be to develop a sense of psychological safety ("being able to show and employ one's self without fear of negative consequences of self-image, status or career" (Kahn 1990)).  Pilots as an example can openly talk about problems without fear of retaliation.  If the organization above had this, the worker in this example could have said, “I need help” and he would have gotten it, resource required 1 extra person or more time to do the job safely……..

Think what might help you forcus the mind...and has been alluded to in previous post is that there are a number of organisation and cultural influences on how a person behaves...if you look the the wee model produced by the the Kiel centre a while back in thier investigation into behaviours offshore you will see that onlt 10% of the behaviours that you are talking about are the ones you observe day to day the rest are influenced from the top....which is why when I implemented a worldwid system it started with the board and then thier line etc etc...

So if you are now modifying that to bring in new teechniques like NLP in training and instruction that is worthy of debate...but generally we are now moving this into  a behavioural science debate rather than behavioural safety...

Ryan, I'm not sure you will get the sort of responses your question begs. Like many of my OHS colleagues I do not see behaviour as an isolated factor. There are far too many other factors which influence behaviours some of which have already been mentioned. Furthermore, by focusing on behaviours and interventions I believe the big picture is missed. As another person has once said, human failure is a symptom - not a cause.

Originally Posted by: Ryan.Donald

Good day all

As a person who has a keen interset in the psychology of behaviour and what mkes us "tick" I would like to open the subject up to the world of IOSH and its members to discuss behaviour intervention and planning. The aim is to share knowldge and experience in gathering data on problem behaviours, formally planing for behaviour intervention and successful interventions that have taken place (and advice for improving on the interventions that have not been such a success in hopes of future imporvement for future, similar interventions)

I would like to start by asking what we define behaviour as in a workplace context and get the discussion going. Here is my definition of behaviour;

"any interaction between a person and their environment, including the systems and peoiple in their environemnt that can be observed and measured"

looking forward to the discussion

Ryan

I think your investigation base is too broad and not necessarily in the remit of the jobbing H&S practitioner.  Yes, people's behaviours affect the way in which they work and the number of accidents, however, as I have said on many occasions, it cannot be introduced in isolation and seen as a panacea for all ills.  Behaviour based safety must be part of a robust management system and a good culture.

Also, I disagree with your definition - measuring behaviours?  I should be most interested in how you measure behaviours.  Do we take this enbloc or as individuals?  Do we take personal issues into account?  Are there underlying causes?  Did they not understand?  Is their behavioural issue our fault?  Yes, you can identify behavioural issues but you cannot measure these like counting beans - most behavioural issue will have a cause which is specific to that person at that time and may never occur again, even without intervention. 

Psychology is, of course, all about statistics and measurements and this can be applied to most fields, but not at the level at which the majority of H&S bods are working.  This quantification and statistical analysis is the basis for PhD's and is the standard in that field.  You can take a small part of H&S and focus the attention on that, complete a PhD and come up with a finding.  You can then look at applying that finding to other areas of H&S to see if the formulae and understanding you have proven in your PhD is transferrable.  However at the moment you are attempting to microanalyse a subject which is huge in its reach and breadth.

Behaviours are an integral part of a good H&S Management System and any practitioner will dig into root causes and corrective actions for failures, but you could have 50 different behavioural failures and 50 different root causes or groups of root causes.  How is this a measurable statistic?

The idea is to identify what behaviours are taking place and measure them yes, and behaviour is measureable (anything that a person does is measureable and quantifiable), and not consider subjective elements such as knowledge or ability, this will come out during review of the potential causes / influences. For example, if there are multiple instances of people not wearing PPE then that is a behaviour that is measureable and quantifiable through a number of dimensions;

The frequency of the behaviour - 10 out of 15 people are not wering PPE - observable and measureable

The intenstiy of the beahvour or the extent to which they demonstrate a behaviour - wearing some of the PPE not all of it - observable and measureable

The duration of the behaviour - do they wear it the whole shift or do they remove it at certain times - observable and measureable

What behaviour looks like - are people removing PPE or readjusting it because its uncomfortable? observable and measureable

Behaviour is observable and measureable, just like the beans

Measuring behaviours through observation groups or individuals. It can be carried out during specific tasks to identify behaviours during that task.

People seem to be attacking the fact that I am saying that the beahviour is the persons fault - this has been cleared up a number times now, and again in this reply. When we identify a behaviour that is not desired (irrespective of who pr what is at fault it is still undesired) an intervention plan is created (either formal or otherwise) and that plan would identify the tsteps that can be taken to positvley influence and reinforce desired behaviour

When we look at my example from above, we can sit in a team and brainstorm on the approaches on how we can tackle the behaviour and its influences. Included in the scope of this brainstorm are individual factors, job and organisational factors (HSG48) this can inlcude training, dealing with the group norm in some way (what is currently happening vs what is typically approved business wide)

Edited by user 27 November 2019 09:14:16(UTC)  | Reason: Remove original post text

I'm in an SME organisation so we don't have swathes of people all exhibiting the same behaviours that we can count. If we had 10 people out of 15 not wearing their PPE then we really would have a problem.

I'll be honest I am not sure what you are asking but I will list a problem behaviour within the various manufacturing firms I have work for (both in a H&S role and on the shop floor)

Broadly speaking the employees can split into two sub groups - too young to care and too old to believe that change is worth it. I know this is a mass simplification, but indulge me to illustrate my point. When it comes to H&S I find they both resist the same issues, but for different reasons. The classic example which I will use is injury vs health.

People on the factory floor want to know they are being protected. They want to see physical evidence of that protection and understand the benefits. Sharp blade? Put a guard on, now I can't cut myself. That sort of thing. This is fine, however when it comes to health issues they don't understand the benefits. Young people feel that they are immune and old people think it is too late anyway. In both instances they can't understand the benefits when compared to the hassle of wearing PPE (for example).

Is this the sort of behaviour problem you are wanting discussed?