How to use systems thinking to carry out a learning needs assessment

Many practitioner guides to carrying out an LNA draw on the idea of the O-T-P principle, first outlined by McGehee and Thayer in the 1960s. This stands for Organisation, Task and Person, and recommends that an LNA starts by looking at the organisation’s goals or objectives, then working out what tasks need to be carried out to achieve this, and finally analysing the persons who will carry out the tasks. This sounds quite logical for making strategic change but has some significant limitations.

First, it is a top-down and very unitarist approach, with an unquestioned assumption that what is good for the organisation is good for the people employed. Of itself, it does not encourage much employee participation in developing objectives at any level and hence working towards employee commitment.

Second, it is a reductionist approach, meaning that we start off with one or two high level objectives and then break these down into multiple lower level objectives, each relevant for particular tasks or people. The assumption here is that by adding together these objectives we will achieve the high level objective, but this is a dubious assumption. This is because it is incredibly difficult to break down complex ideas that a high level objective embodies into meaningful sub-tasks. The reductionist process ignores the complex sets of factors that link individual activities together, and we often end up with different people doing different tasks which overall do not have the desired effect.

A third weakness in this approach is that it is less well suited to situations where a particular local problem has been recognised and where we are not really dealing with a strategic change.

In such a case an alternative approach is to focus on the local problem and use what is called something like performance problem analysis or performance engineering. One influential model in this approach is Tom Gilbert’s Behaviour Engineering Model. This says that when analysing a performance that we think needs improving that we look at two areas; the supporting environment and the performer’s ‘repertory of behaviour’.

In the supporting environment we have to consider whether or not adequate information is provided, whether the performer has suitable equipment and what incentives are provided for desired levels of performance. For the performer we have to consider whether they understand information available, that they can use necessary equipment, and that they have the necessary motivation to perform adequately. The BEM helps us to develop a more holistic understanding of what affects local performance, but it can still be difficult to work out what particular changes will be most useful. Gilbert himself suggested that when trying to improve performance we should work through the supporting environment factors in turn to see what works and then move on to the repertory of behaviour, but that approach can take a lot of time. It is also another reductionist strategy and ignores any systemic connections between the different factors.

A development of the BEM is the Mager and Pipe Performance Analysis Flowchart. This provides a useful checklist which can take the analyst through a series of questions to help them identify what potential causes of inadequate performance exist and to suggest general ideas that may help.

Again, this is a useful tool for helping us to consider a performance problem in a more holistic way, but it still relies on the analyst being able to being able to understand what, for example, rewards poor performance or what other obstacles might exist. This means that the analyst needs to be able to understand how an organisation is functioning and what could be improved.

It is at this point that a systems thinking perspective can be very useful. A systems thinking tool that is ideally suited to using in an LNA is called the Viable System Model, or VSM. VSM was developed by thinking about how biological organisms remain viable, through processing information from their environment and regulating internal processes so that they adapt to the world around them.

As such, VSM is interested in how information flows around an organisation and pays little attention to structure. As a result, we can use a VSM analysis at any level in the organisation, from looking at the organisation as a whole down to seeing how an individual carries out their day-to-day work. For an LNA we would probably be looking at how a department or team functions, but can use the same principles to see how that departmental functioning integrates with the organisation as a whole.

In the VSM model there are five systems of information flow that we need to think about. Starting from the highest level, System 5 is the Policy system which is where the department defines overall departmental goals and operational ethos. System 4 is called Development, and this gathers information from the operational environment to understand how this is changing. This interacts with System 5 so that the goals and ethos can remain relevant to the outside world. System 4 also provides information to System 3, Delivery. This is the system that sends information to operational teams to help them carry out the customer-facing work of the department. System 3 also has a feedback system, usually called System 3*(3 star). This sends information back about what is happening at an operational level. System 1, or Operations, is the work that the department does that the outside world sees: in a software company this would be coding or in an engineering company machining or casting, and in all companies includes a sales function. Organisations will generally have multiple System 1s. System 2 is a system which coordinates the activities of the various System 1 activities, for example, making sure that coding activities are aligned with sales of software.

In a viable organisation all of these systems need to operate effectively at all levels, so, for example, a departmental level System 1 will have its own five systems, including a System 4 which gathers information from its own operational environment and feeds information upwards through the organisation.

This might sound a bit complicated but it is all logical and once you have grasped the essentials, it can make carrying out an LNA much easier. We can draw on ideas from other models such as the BEM and the performance flowchart and use a VSM analysis to understand how well information is flowing through all the different systems relevant to the level of our analysis. That will then give us an idea about weaknesses in the system so that we can decide whether there are knowledge and skill issues to be addressed or structural matters to be remedied.