top of page

Learning from Incidents to Create Safer Workplaces: Part 2 – Framework for Learning from Incidents

This is the second part of a two-part article. Click here to read the first part.

The second part of this article will introduce the framework (Figure 1) as conceptualised by Dane Lukic, Allison Littlejohn and Anoush Margaryan that different organisations can use to foster effective learning from safety incidents and other problem situations [1].

Figure 1

Framework for Learning from Incidents

This framework is characterised by five interconnected factors that are supplementary to one another:

1) Learning context

2) Type of knowledge

3) Learning participants

4) Type of incident

5) Learning process

The breadth of learning is represented by the first three factors, whereas the last four factors denote the depth of learning.

1) Learning Context

Learning context is essential for effective learning from incidents and entails understanding the extent of using formalised learning initiatives.

Most formal learning initiatives are embedded within safety procedures and performed through various means like safety meetings, formal reports and dissemination of safety alerts and bulletins by email. Although formal learning is often viewed as more valid as verification can be made by experts like workplace safety officers, it can likewise be implemented superficially.

In contrast, informal learning occurs when employees engage in everyday work practices and interact with other co-workers. As these activities are not deliberately created for learning, most remain unaware that learning has occurred [2]. One issue with informal learning is that the knowledge is usually not captured and codified; therefore, the knowledge cannot be reused, especially when these employees have changed roles or left the organisation.

In some cases, the distinction between formal and informal learning is not apparent, for instance, during toolbox talks and shift takeovers where employees discuss work and safety issues. These learning experiences are regarded as pedagogically rich as they allow employees’ knowledge to be extended and transformed, which may not otherwise be possible [3].

As a result, organisations should consider the degree to which the learning initiatives should be formalised when developing approaches for learning from incidents.

2) Type of Knowledge

Organisations should likewise consider the type of knowledge needed for addressing a problem or preventing future incidents.

Four types of knowledge are of importance to learning:

a) Conceptual knowledge

Conceptual knowledge, or 'knowing why’ and ‘knowing what’, is pertinent to facts, concepts, information and propositions. In the context of safety, conceptual knowledge may comprise understanding safety issues as well as the procedures and reasons that led to the incidents.

Solving complex problems would require deep conceptual knowledge to understand the nature of the issue and its relationship with other associated problems.

b) Procedural Knowledge

Acquiring procedural knowledge, or ‘knowing how’, is critical for expert performance at work. This type of knowledge entails the techniques, abilities and skills for performing conceptual knowledge and can be categorised into three hierarchical levels:

i) First-order procedural knowledge is procedures for achieving specific goals or tasks, such as tightening a screw.

ii) Second-order procedural knowledge is procedures for solving non-routine problems, for instance, figuring out how to replace casement windows in an old building in an aesthetic manner.

iii) Third-order procedural knowledge is procedures for managing first- and second-order procedures.

Procedural knowledge can be secured through participating in work activities, indirect guidance (e.g., observation and imitation) and obtaining direct and interpersonal (proximal) guidance from more experienced co-workers [4].

c) Dispositional Knowledge

Dispositions, which can be described as individuals’ tendencies to put their capabilities into action, are central to supporting the development of conceptual and procedural knowledge.

Dispositional knowledge comprises the values, attitudes and motivations associated with work (e.g., attitude towards wearing personal protective equipment) that can seriously affect workplace safety and health. Accordingly, having positive dispositions to secure new knowledge is vital for effective learning from incidents.

Dispositional knowledge can be developed through authentic work experiences that can afford indirect guidance and contribute to an employee’s micro-genetic development (i.e., moment-by-moment learning). Receiving proximal guidance from experts can also elucidate what may be hidden and reinforce the values and beliefs underpinning the ‘right’ actions. Nevertheless, developing dispositional knowledge can be an extensive and indeterminate process [5].

d) Locative Knowledge

Locative knowledge, or ‘knowing where’ and ‘knowing who’, is a type of meta-knowledge about the location and sources of pertinent knowledge (e.g., resources, people, and tools) that can be applied to develop conceptual and procedural knowledge. Securing such knowledge is central to the effective management of safety incidents.

Locative knowledge can be developed through training as well as networking and interacting with others within and outside the organisation.

3) Learning Participants

Understanding who should be involved and the extent of their participation in the learning process is essential for ensuring the effectiveness of learning from incidents. This would require managing the duality of inclusion and individual agency [6].

Safety literature has posited that effective learning from incidents should involve employees from different levels of the organisation, including the ‘shop-floor’ staff and top executives. Yet, some organisations tend to limit the opportunities for employees to participate fully in initiatives like safety meetings and investigations despite their desire for more significant involvement.

On the other hand, while most employees recognise that learning from incidents that are irrelevant to their work can bring important learning lessons, they do have concerns that this can result in information overload. Hence, they may only be able to pay attention to some learning initiatives due to time limitations.

Taken together, organisations must consider the extent to which stakeholders can be involved in shaping the learning process and questioning the status quo, as well as deliberate on the relevance of the learning initiatives to their roles before deciding who should participate.

4) Type of Incident

Organisations may sometimes oversimplify learning from incidents as they do not understand the nature and complexity of the incidents.

In order to develop an in-depth understanding of the complexity of incidents, organisations can refer to David Snowden’s ‘Cynefin’ (pronounced as ku-nev-in) framework (see Figure 2), which consists of four complexity domains.

Figure 2

Cynefin Framework

The orderly domains include the ‘simple’ and ‘complicated’ domains, whereas the disorderly domains comprise the ‘complex’ and ‘chaotic’ domains.

a) Simple Domain

Incidents belonging to the simple domain tend to have clear cause-and-effect relationships and straightforward solutions. Organisations may ascertain the facts, classify them, and apply ‘best practices’ to solve the specific issue.

b) Complicated Domain

The complicated domain is relevant for incidents that require in-depth analyses as the causal relationships are not apparent and cannot be readily determined. Effective solutions would encompass the ‘good practices’ usually identified after in-depth analyses.

c) Complex Domain

In the complex domain, urgent actions are required to cope with the incidents. Furthermore, as the causations are intertwined, the causes would only surface after the incident and following a comprehensive investigation.

d) Chaotic Domain

The chaotic domain deals with unpredictable incidents that require fast responses to alleviate the situation from chaotic to a complex and manageable state.

Such incidents would first necessitate what Donald Schön has termed as ‘reflection-in-action’, which entails reflection during practice to respond to problem situations, followed by in-depth analyses and the learnings involved in the other three domains.

It is important for organisations to apply appropriate solutions for incidents of different complexity and nature in order to ensure the effectiveness of learning from incidents.

5) Learning Process

Finally, effective learning from incidents should be supported by a robust understanding of the learning processes. The concepts of single- and double-loop learning postulated by Chris Argyris and Donald Schön would be a valuable basis for learning from incidents.

Single-loop learning approaches are characterised by corrective solutions for fixing direct and trivial causes of incidents, for example, technical corrections, sending employees for training or taking disciplinary actions. Indeed, research has shown that in most instances, organisations tend to embark on learning initiatives that constitute single-loop learning.

In contrast, double-loop learning approaches involve confronting, refining and replacing the current norms, assumptions, values and practices to resolve the problem and mitigate the risk of future incidents.

In order to foster double-loop learning, a blame-free attitude must be adopted across the organisation so that the underlying organisational issues can be examined and the solutions can be pursued. This, however, should also be supported by five shared values to promote learning [8]:

a) Transparency – The willingness to expose one’s thoughts and actions to others in order to receive feedback

b) Integrity – The willingness to seek and provide information regardless of its implications

c) Issue-orientation – Focusing on the relevance of the information to the issue under consideration regardless of the social standing or rank of the source or recipient

d) Inquiry – Persisting in the investigation until complete understanding is achieved

e) Accountability – The willingness to assume responsibility for learning and the implementation of lessons learned

In summary, the second part of this article has presented a framework that can serve as an evaluation and guidance tool for effective learning from safety incidents so that organisations can be empowered to create safer workplaces, which is vital not just for employees but also for the business.

Click here to read the first part of this article.


[1] Lukic, D., Littlejohn, A., & Margaryan, A. (2012). A framework for learning from incidents in the workplace. Safety Science, 50(4), 950-957.

[2] Simons, P. R.-J., & Ruijters, M. C. P. (2014). The real professional is a learning professional. In S. Billett, C. Harteis, & H. Gruber (Eds.), International handbook of research in professional and practice-based learning (pp. 955-985). Springer.

[3] Billett, S., & Noble, C. (2020). Utilizing pedagogically rich work activities to promote professional learning. Education et Didactique, 14(3), 137-150.

[4] Billett, S. (1994). Situated learning: A workplace experience. Australian Journal of Adult and Community Education, 34(2), 112-130.

[5] Billett, S. (1997). Dispositions, vocational knowledge and development: Sources and consequences. Australian and New Zealand Journal of Vocational Education Research, 5(1), 1-26.

[6] Billett, S., & Smith, R. (2007). Personal agency and epistemology at work. In S. Billett, T. Fenwick & M. Somerville (Eds.), Work, subjectivity and learning (pp. 141-156). Springer.

[7] Snowden, D. (2002). Complex acts of knowing: Paradox and descriptive self‐awareness. Journal of Knowledge Management, 6(2), 100-111.

[8] Lipshitz, R., Popper, M., & Friedman, V. (2002). A multi-facet model of organizational learning. Journal of Applied Behavioral Science, 38(1), 78-98.

bottom of page