Week 1 Session: The Incident Review Committee

A fictional collision, five conflicting accounts, and fifty minutes to turn "someone should do something" into a question a dataset could actually answer.

22 February 2027

Before the session#

Read the week 1 lecture. Bring one popular-media claim about kangaroo–vehicle collisions you’ve encountered — a news story, a road sign, a rental-car pamphlet — that you suspect oversimplifies.

In the session#

You’re convened as the Incident Review Committee for a single collision: the Yarralong Road case. This case, and everyone in it, is fictional teaching material invented for this session — not a real incident, real person, or real location.

The incident briefing#

At approximately 6:40pm on a weekday in late autumn, a sedan travelling the Yarralong Road — a regional two-lane road with an 80 km/h limit, no dedicated wildlife fencing, and intermittent tree cover along both verges — struck an adult kangaroo that entered the road from the eastern verge. The driver braked and swerved but did not avoid the animal. The driver stopped, called it in, and waited for a wildlife carer to attend. No other vehicles were involved. The nearest wildlife-crossing sign is approximately 400 metres further along the road, past the collision site.

Driver’s statement#

“I was doing about 80, maybe a bit under — it’s hard to say exactly. I didn’t have my eyes off the road. It came out of the trees on the left, and there just wasn’t time. I’ve driven this road every week for two years and never seen one that close to the road before, but I don’t really look for them, if that makes sense.”

(The driver’s speed estimate is not backed by any independent record — there is no fixed camera or data logger on this stretch. Treat it as a recollection, not a measurement.)

Wildlife carer’s notes#

Attended 7:05pm. Adult female eastern grey kangaroo, no pouch young observed. Injuries consistent with vehicle impact; deceased on arrival. Site is unfenced verge with moderate tree cover extending roughly 15m back from the road. No tag or prior record for this individual. Unable to determine how long the animal had been in the immediate area before the collision, or whether this stretch has a prior collision history — that would need to come from the road authority’s own records, which I don’t have access to from the field.

Five role cards#

• Researcher — Your stake: whether any general claim can be drawn from this one case. You know how to ask what would need to be true for a pattern (not just an instance) to exist. You do not know the collision history of this specific road, or how representative this case is of anything.
• Road authority — Your stake: whether the existing signage and any future spending decision is defensible. You know the sign placement (400m past the site) and the general design of the road. You do not have, in front of you today, the road’s collision history — the carer’s notes flag this gap deliberately.
• Driver — Your stake: whether your account is treated as fact or as one input among several. You can report what you remember seeing and doing. You do not have any way to independently verify your own speed or reaction time.
• Wildlife carer — Your stake: the animal’s welfare and an accurate clinical record. You know the animal’s condition, species, and that its injuries are consistent with vehicle impact. You do not know anything about traffic volume, signage history, or the driver’s account beyond what’s in this briefing.
• Kangaroo representative — Your stake: that the animal’s perspective (habitat, movement, why this verge, why this time of day) isn’t skipped in favour of blaming the driver. You hold the session’s one standing objection: “Before we discuss my decision-making, could we discuss why the road is here?” — raise it before the group settles on blame.

1. 10 minutes — sort the briefing, statement, and notes into established fact, someone’s interpretation, and genuinely unknown.
2. 15 minutes — each role states the question they most want answered.
3. 15 minutes — the group agrees on one research question, one falsifiable hypothesis derived from it, and three pieces of missing data that would help test it.
4. 10 minutes — report back: check whether the media claim you brought would survive the committee’s own standard of evidence, and name one conclusion the group still can’t draw from this case alone.

Afterwards#

Submit a one-page Incident-to-Question Brief. It’s formative practice for turning an account into a testable claim, not a graded artefact. Fill in:

Incident-to-Question Brief

• Research question the case raises: ____
• Hypothesis derived from that question (must be specific enough that evidence could contradict it): ____
• Three established facts from the case material: 1. ____ 2. ____ 3. ____
• Three pieces of missing information that would help test the hypothesis: 1. ____ 2. ____ 3. ____
• One conclusion this case does not yet support: ____
• One source of evidence that could move it forward: ____

This session doesn’t require the shared collision dataset — that’s introduced properly in week 2, where the Road Ecology Research Group’s synthetic collision log replaces this one fictional case with a much larger set of cases. You’ll leave today with a partner for the week 3 mapping exercise.