# Parachutes & Evidence

*Tom Geraghty · Models & Critique, Measurement, Safety & Human Error, Politics, Diversity & Equity*

Two newsletter pieces published together, the second correcting the first, on what counts as evidence for psychological safety. The first uses Smith and Pell's 2003 BMJ review of parachute trials to answer a recurring demand: that psychological safety prove itself by randomised controlled trial before it is taken seriously. The counterfactual cannot ethically be run, so the mechanism is instead made visible by its failures, and the catastrophe record is the evidence. Tenerife, Chernobyl, Deepwater Horizon, Challenger and the run-up to 2008 are each cases where a concern existed at the sharp end and did not travel to where it was needed. Successes leave no trace, because an averted disaster is structurally invisible, and that asymmetry is the parachute's epistemic signature. The second piece is the more interesting half, and is a public correction: an RCT does exist. Yeh et al (2018) randomised twenty-three participants to jump wearing either a parachute or an empty backpack, and neither group came to harm, because everyone approached at altitude declined to enrol and the trial ran instead from a parked biplane at a mean altitude of 0.6 metres. The randomisation and analysis were sound and the result was still absurd, because the conditions that made the trial possible were precisely the conditions that removed the effect. It measured its sample rather than its intervention. Glasziou et al (2007) on the dramatic effect supplies the honest alternative standard, and the observational record clears it for parachutes and for organisational disasters alike. The critical edge runs through both halves: the catastrophe record establishes the mechanism, not the survey, the stage model or the metric on the dashboard, and nobody should be allowed to sell parachutes that do not open. It closes on the critic who says people should simply speak up, and observes that such critics are usually right about themselves, because they are standing on the tarmac.

- **This page:** https://explore.psychsafety.com/n/parachutes-and-evidence-based-thinking/
- **Read the full article on psychsafety.com:** https://psychsafety.com/parachutes-and-evidence-based-thinking/
- **Interactive map:** https://explore.psychsafety.com/?node=parachutes

## Neighbours (11)

- [Who Gets to Decide if PS Matters?](https://explore.psychsafety.com/n/who-gets-to-decide-if-psychological-safety-matters.md)
- [Normalisation of Deviance (Challenger)](https://explore.psychsafety.com/n/normalisation-of-deviance.md)
- [The Tenerife Disaster](https://explore.psychsafety.com/n/the-tenerife-disaster-of-1977-were-going.md)
- [When are randomised trials unnecessary? Picking signal from noise](https://explore.psychsafety.com/n/glasziou-et-al-2007.md) (paper)
- [Amagasaki Disaster](https://explore.psychsafety.com/n/the-amagasaki-disaster.md)
- [Beyond Metrics](https://explore.psychsafety.com/n/beyond-metrics.md)
- [Chernobyl](https://explore.psychsafety.com/n/chernobyl.md)
- [PS Index Critique](https://explore.psychsafety.com/n/the-psychological-safety-index-a-critical-look.md)
- [Too Much PS?](https://explore.psychsafety.com/n/can-workplaces-have-too-much-psychological-safety.md)
- [PS & Science](https://explore.psychsafety.com/n/psychological-safety-68-science.md)
- [Social Science vs Epidemiology](https://explore.psychsafety.com/n/social_science_epidemiological.md)