Skip to main content

Evidence gathering and analysis

In any inquiry, evidence falls into four broad categories: people, machine, environment, mission. Investigators initially focus on gathering evidence that could disappear or change. The Commission has extensive powers to protect and gather evidence. Evidence gathering continues as investigators follow different lines of inquiry.

Gathering Evidence

There are four broad categories for the evidence we collect: 

  • People – who was involved, relevant professional and personal background, their training, what they knew, thought, experienced, and did.     
  • Machine – the vehicle itself, including individual and type history, performance, maintenance, components and design.
  • Environment – what was going on around the vehicle, including geography, weather, operating company safety system and culture, traffic control, regulatory.
  • Mission – where the vehicle came from and where it was going, the purpose of the journey, the vehicle operator’s relevant processes and their use in this case.

The accident / incident scene

TAIC investigators work with emergency services, the vehicle operator, and the relevant transport regulator to ensure evidence is protected. After emergency services have rescued survivors and stopped further physical and environmental damage, the scene is frozen until our investigators arrive or we give permission for other activity. The Commission may formally seize or order protections of the vehicle, scene and other evidence. TAIC has ultimate control of the scene, and co-ordinates the access required by other authorities.

Typically, two of the investigation team travel to the scene. They may be supported by TAIC specialists in human factors, vehicle recorders and data, a communications officer or logistics support. External support may come from Search and Rescue, Coastguard, Fire and Emergency NZ, Police, or similar specialist agencies. Health and safety considerations are particularly important for all of the people at the scene. Risks and hazards can include weather and terrain, accident vehicle dangers such as sharp, flammable or unstable wreckage, pollutants and disease.

The evidence

Evidence collection is broad at first to support the many routes that an investigation could follow. The initial focus is on gathering evidence that could disappear or change. This includes recording the accident scene, recovering wreckage, and interviewing witnesses. Investigators also gather maintenance and vehicle records and documents related to the vehicle, its operating company (operator) and key personnel. Investigators may also seek information from related services such as traffic control. 

If there are fatalities, TAIC's medical consultant will liaise with the forensic pathologist and Coroner. 

Investigators have authority to use the Commission’s legal powers to protect and gather evidence, and sometimes the Commissioners may decide to interview witnesses themselves.

All the evidence gathered has extensive legal protection from disclosure. This prohibits us from talking in detail about what we have seen and heard. If another agency is investigating the same event, we may give them access to physical evidence (but not our interpretation of it) under strict conditions.

Early assessment

Early in the inquiry, Investigators brief the Commissioners on what is happening, including the scene examination and associated health and safety considerations. They discuss emerging lines of inquiry and potential focus areas to be considered as evidence gathering continues and analysis begins. They may consider the need for urgent recommendations or whether to produce an interim report setting out the key facts established so far.

Interviews and collection of physical and documentary evidence can continue over the first few months as different lines of inquiry are followed. Work in later stages, particularly during analysis, may identify the need for more evidence collection.

 

 

Swiss Cheese Model of safety incident causation

Diagram of the Swiss cheese model of accident causation. Multiple yellow slices with holes represent defensive layers or controls. Red arrows labelled 'Hazards' pass through holes in each layer. Some arrows are blocked, showing 'Losses prevented'. One arrow passes through all the layers, labelled 'Loss not prevented (Incident)'. Other arrows that nearly align but stop are labelled 'Near miss'.
Figure 5.1: Swiss Cheese Model of safety incident causation (after Reason, 1997a)
Source: “Swiss Cheese Model” by Ben Aveling, Wikimedia Commons, CC BY-SA 4.0

Analysing evidence

Analysis involves sorting, corroborating, and linking evidence and facts to prove, disprove and weigh competing theories.

The events leading to an incident or accident can appear to be clear, particularly those facts and factors closest to the occurrence. However, incidents and accidents rarely have a single cause; contributing factors are often complex and reach beyond the accident vehicle and its operation to wider systemic issues. For instance, a pilot may activate the wrong control, but contributing factors revealed in analysis could include the pilot’s training, the airline’s training system and its operation, the control design, and relevant regulations covering any of these aspects.

Testing the analysis

While TAIC’s investigation team come from a range of professional disciplines, we often call on external experts for their specialist knowledge, skills and expertise. These can include metallurgists, engineers, psychologists, medical specialists, vehicle type experts, forensic data specialists, and so on. 

Commission staff who have not worked on the case are called on for the benefit of their fresh minds, enabling the larger group to consider the potential relevance of previous similar occurrences, including some from overseas. 

Sometimes we may gather further evidence.

Endorsing the analysis

The investigation team presents its completed analysis to Commissioners for testing and to decide whether it's necessary to do further work. Their considerations include whether all necessary facts have been established, robustness of evidence and analysis, and whether the safety issues have been properly identified. Analysis may also lead to consideration of whether it is necessary to issue urgent recommendations. 

Verbal probability expressions

It is rare that evidence allows for absolute findings; usually the Commissioners make findings with a stated degree of probability.

For clarity, the Commission's reports use standardised terminology to describe the degree of probability (or likelihood) that an event happened, or a condition existed in support of a hypothesis. TAIC has adopted this terminology from the Intergovernmental Panel on Climate Change and Australian Transport Safety Bureau models.

Terminology  Likelihood  Equivalent terms
Virtually certain > 99% probability of occurrence  Almost certain
Very likely  > 90% probability  Highly likely, very probable
Likely  > 66% probability  Probable
About as likely as not  33% to 66% probability More or less likely
Unlikely  < 33% probability  Improbable
Very unlikely   < 10% probability  Highly unlikely
Exceptionally unlikely  < 1% probability  

 

Last updated: Tuesday, 26 August 2025 - 12:37