The Substitution test for learning from mistakes

Image from Wikimedia commons

“Learning from mistakes” is a goal of many lesson learning systems, and with good reason. If an organisation can learn from its mistakes, then it can avoid them in future, and so improve performance.  However many people do not like to learn from mistakes. They feel that examination of mistakes involves blame.

How do we resolve this?

In some ways, blame is important. If people do something deliberately wrong – something they know they should not do, like ignoring an instruction, or taking a shortcut – then they should  be blamed. On the other hand, if people make an honest mistake but are let down by the system they are tying to follow, then they should not be blamed. Instead the system should be improved so that the mistake does not re-occur.

Details of the substitution test can be found on the Skybrary site, as follows:

The Substitution Test helps to assess how a peer would have been likely to deal with the situation. Johnston (1995), a human factors specialist and an Aer Lingus training captain, has proposed the substitution test. When faced with an event in which the unsafe acts of a particular individual were clearly implicated, the judges should carry out the following thought experiment. 

Substitute for the person concerned someone coming from the same work area and possessing comparable qualifications and experience. Then ask: ‘In the light of how the events unfolded and were perceived by those involved in real time, is it likely that this new individual would have behaved any differently?’ If the answer is ‘probably not’ then, as Johnston (1996:34) put it, ‘apportioning blame has no material role to play, other than to obscure systemic deficiencies and to blame one of the victims’. 

A useful variant on the substitution test is to ask of the individual’s peers: ‘Given the circumstances that prevailed at the time, could you be sure that you would not have committed the same or a similar type of unsafe act?’ If the answer again is ‘probably not’, then blame and punishment are inappropriate.  

This simple test is very valuable for steering a safe course between Blame and No-Blame, when trying to learn from mistakes.