What the Data Told Us: Fear, Trust, and the Stories That Never Get Told. Part 2 of 3.
Blog 2 of 3: The Research and What It Found
This is the second in a three-part series drawn from Storytelling to Learn: What Happens Underwater, Stays Underwater — an MSc thesis from Lund University. Blog 1 covered the problem and the literature. This one covers the research design and findings. Blog 3 addresses what needs to change.
Research about human behaviour in high-risk environments is only as good as the willingness of people in those environments to speak honestly. That willingness is itself one of the central problems the research set out to investigate. So before diving into what the data showed, it is worth reflecting briefly on how it was gathered — and what that process revealed in its own right.
How the Research Was Conducted
The study used a mixed-methods approach, deliberately designed to capture both breadth and depth. Three distinct datasets were collected between May and July 2023.
An online survey ran for two weeks and gathered 676 completed responses from across the diving community internationally, covering recreational divers, technical divers, cave divers, and instructors. A subset of the 55 questions focused directly on the research question: how respondents defined an incident, what risk meant to them, whether they understood just culture, and what single change they believed would most improve context-rich storytelling in diving.
Five focus groups were planned with recreational divers, technical divers, cave divers, diving instructors, and a planned but ultimately absent group of training agency staff. A sixth contribution came as an unsolicited letter from a recreational diver who, following the focus group, was directly involved as a bystander in a fatal diving event and wrote about how the reporting and discussion of that event had unfolded. Two interviews with practising lawyers, one UK-based, one US-based, provided a legal perspective on what influences what can and cannot be said following an adverse event.
It is worth noting that the planned agency staff focus group never happened. After a presentation to the British Diving Safety Group explaining the research, four additional dates were offered for an online session. There was no response from any of the nine agencies represented. That silence is itself a data point, and a significant one given what the rest of the research found about organisational factors.
The qualitative data from focus groups and interviews was coded using a two-stage process, open coding followed by focus coding, before Pareto analysis was applied to identify the themes that carried the most weight across the different communities of practice. Where the volume of survey responses made manual coding impractical, ChatGPT v4.0 was used to generate and test themes across multiple passes, with variability between sessions assessed as part of the reliability process.
What the Data Found: Eight Themes in Two Clusters
The analysis produced eight sub-themes that sit under two supra-themes. The first cluster concerns organisational and cultural factors and tensions — the higher-level conditions that make it easier or harder for any story to be told. The second cluster concerns defining the event and how to tell a learning-focused story — the lower-level, practical factors that shape what gets told, and how.
These are not independent. They interact, reinforce, and constrain each other in ways that reflect the complexity of the system in which diving takes place.
Cluster One: Organisational and Cultural Factors
Social, cultural, and organisational factors. Across every community of practice, the data pointed to the role of the wider social and organisational environment in shaping what people would or would not say. The instructor group was particularly revealing. One instructor described being taught that, if something goes wrong on a course, the training agency "will cut you and you're on your own." Another described being a PADI centre where instructors wrote incident reports internally but felt no confidence those reports served any learning purpose: "This report was going nowhere. Unless they themselves wanted it to go." A lawyer in the data confirmed this directly, stating that what incident reporting means in practice for one major agency is fundamentally about litigation protection — and that this orientation ended around 2010.
Fear of. Fear appeared across all six datasets, in multiple forms. Fear of litigation. Fear of peer criticism. Fear of being blacklisted. Fear of losing certification. Fear of vulnerability. One instructor described waiting two to three years — including a public webinar — before finally being willing to share a story about running low on gas. Another noted that sharing a story often means "showing vulnerability" and "having done something wrong that they admitted to," and that the social consequences of this are real and unpredictable. The fear is not always rational, and it rarely needs to be substantiated by an actual adverse outcome — its presence is sufficient to suppress disclosure.
Trust between and across the system. The data showed that trust enables storytelling within tight, familiar groups — and that trust breaks down almost completely in public settings. One cave diver was explicit: "I've had some things that I would get on the phone and talk with a buddy, but I'm not putting it in a text message." An instructor described a deep hunger for "some kind of place or conference where we can talk about these things where it's open to everyone, but closed and controlled and safe." This tension between the need for a wider audience and the vulnerability required to speak to one is at the heart of why stories stay local or stay silent.
The tension between learning, litigation, and speculation. The lawyers in the study offered a perspective that complicated the expected narrative. Both were broadly supportive of context-rich storytelling, arguing that it reduces rather than increases litigation in the long run — because it helps families understand that accidents are rarely as simple as "someone messed up." But the instructor community saw this very differently. The concern was not just litigation itself but the threat of speculation: once a story enters public circulation, control over its interpretation is lost. Lawyers might read a nuanced account and find a standards violation embedded in it. This keeps many instructors silent even when the event itself has no legal dimension.
Cluster Two: Defining the Event and How to Tell It
Understanding what an incident is. The survey asked 662 respondents to define a "diving incident" in their own words. The composite themes that emerged across all communities of practice were: deviation from plan, potential for harm or actual harm, near-misses, equipment or protocol issues, and negative outcomes. On the surface, that looks reasonable. But nearly every dive deviates from plan to some degree, and diving always involves potential for harm because the environment does not sustain life. So at what point does normal diving become an incident?
More strikingly, a keyword search across all 662 responses for the word "learn" in any form — learn, learned, learning — returned just seven instances. If divers do not conceptually connect incidents with learning opportunities, it is difficult to see why they would go to the effort of sharing them.
What does risk mean? The responses to this question highlighted a significant split between instructors and other communities of practice. Where recreational, technical, and cave divers focused on understanding, assessing, and mitigating risk, themes that matched the composite picture across all respondents, instructors focused primarily on controls and compliance. Their top themes did not match the overall pattern at all. The hypothesis is that professional development in diving, with its heavy emphasis on compliance and liability management, has produced a particular understanding of risk that is more about protecting oneself from legal exposure than making sense of what is actually happening at the sharp end.
What does just culture mean? Respondents were asked to define just culture in the context of diving. The results were sobering. Only 21% of respondents gave a definition aligned with the established meaning (from Reason, 1997, or EASA's Regulation 376/2014). Thirty percent gave definitions that were counter to just culture, including, in some cases, responses that described it as an excuse for poor performance, a culture of complacency, or a "cop out." Twenty-five percent simply did not know what it meant. Given that the presence of a just culture came second in the survey's "one thing to improve" rankings, behind only an organisational commitment to capturing near-miss stories, the gap between what people want and what they understand is a significant practical problem.
How to tell a learning-focused story. The focus groups repeatedly showed that more context changed the response to an account. Once people understood the conditions surrounding an event, the tendency to judge and the tendency to share both shifted — they became less likely to blame and more likely to engage. One technical diver put it plainly: "It's only stupid when you have the extra information. It doesn't become stupid until you have more information to go: actually, that was a really bad idea." But divers are rarely taught what context matters or how to describe it. Training materials, to the extent they address incidents at all, focus on proximal causes. The systemic factors are not on the curriculum.
How to create learning through sharing. The instructor group offered some of the most practically hopeful data. Multiple instructors described how incorporating context-rich stories into their own teaching, including stories about their own mistakes, changed the learning environment for students. One described a culture shift across 50 instructors at a single dive centre, which then rippled outward into the wider community of roughly 150 instructors in the city. Another argued that if the industry stopped selling the lifestyle and started talking honestly about the risks, people would actually dive more, not less.
The Survey's Single Most Important Question
When asked to choose the one thing that would most improve context-rich storytelling in diving, 25.8% of respondents chose "an organisational commitment to improved diving safety by capturing, analysing, and sharing near-miss and non-fatal incident stories." The second choice, at 19.4%, was "the presence of a just culture in the industry."
These two responses are deeply interdependent. An organisational commitment to capturing stories is meaningless without a just culture to make it safe to share them. A just culture that people do not understand cannot function as intended. The data suggests that the diving community intuitively recognises what is needed but lacks both the vocabulary and the structural conditions to make it real.
What Is Not Said Is Often More Important Than What Is Said
The near-total absence of training agency voices from the research data is significant. One agency member attended a preliminary session; none participated in the research. This matters because the organisational factors that the data identified as the most powerful influence on storytelling in diving — trust, fear, litigation culture, and what incident reporting actually means in practice — are substantially shaped by what training agencies do and do not do. Their absence from the conversation about improving learning is not neutral. It is a choice that has consequences.
Part 3: The Discussion. What Does It Mean? What Needs to Change?
References
Chan, W. T.-K., & Li, W.-C. (2023). Development of effective human factors interventions for aviation safety management. Frontiers in Public Health, 11, 1144921. https://doi.org/10.3389/fpubh.2023.1144921
EC. (2014). Regulation (EU) No 376/2014 of the European Parliament and of the Council of 3 April 2014. European Commission.
Reason, J. (2016). Managing the risks of organizational accidents. Routledge. https://doi.org/10.4324/9781315543543
