“Anyone who considers arithmetical methods of producing random digits is, of course, in a state of sin.”
― John von Neumann
It seems Avon & Somerset Human Remains Department just can not help themselves. To select further IODs for review they are going to use a random number generator.
This is a quote from local NARPO’s report on the IOD Liaison Group Meeting 17th December 2015
8 more claimants have been chosen using an EXCEL random picker programme and letters should be sent out in January 2016 asking for them to engage in the process. If you receive such a letter and are a member of NARPO Bristol Branch, we would like to know, please contact me through […] the branch secretary. If you require it we can offer advice or assistance with the process.
Let’s ignore the unsavoury and nonfactual word ‘claimants‘ used to describe those with an injury on duty award for a minute and focus on the words ‘random picker’.
No. It’s not ‘Lancelot’, the mechanical souped–up tombola that the National Lottery uses, but a computer program. A computer program cobbled together in a Heath Robinson manner using Microsoft Excel.
Does anyone spot the major flaw here?
Is their mistake that the numbers generated by Excel are not truly random, but pseudo-random? This is the output of program code that churns out numbers that appear to be random. Excel RAND’s output is only a simulation of a truly random process. Chance can not be programmed. That is why lottery providers use gravity pick or air mix mechanical machines. These 2 machine types have things in common. They are designed and proven using statistical analysis to produce random combinations of numbers.
Nope. Although correct, the above isn’t their ‘pig in the poke‘. They fail to grasp the mistake magnitude of their cunning ‘brainwave’.
The elephant in the room is that a suitable interval is not and can never be random. The Scoffield report dictates:
There should be a move away from automatic review for all cases at any fixed interval set in policy
A truly random selection would mean it is possible for a single former officer to reviewed consecutively. As Murphy’s Law wisely says, what-ever can happen will happen if there is enough trials.
Let’s give them the benefit of the doubt. Perhaps this supposed ‘random selection’ performed by A&S means that they have determined that everyone should be automatically reviewed and they are just randomising the order.
That doesn’t fit in with the interpretation of the Regulations either. An eminent Queen’s Counsel has determined an automatic review for all cases is not appropriate and reviewing everyone ‘randomly’ is just that – an automatic review based on ‘policy’.
A selection of any IOD for any review has to be made based on the individual. Some should never be reviewed. Time itself is not the determiner of a suitable interval – the circumstances of the individual and the detail recorded on the last final decision determines whether any interval is suitable.
But A&S has to wrongly rely on Excel and their fallacy of random selection because they have lost occupational health records and they can not read the handwriting of the doctor notes made in the records they do have. So they are unable to determine a true suitable interval. That is their problem – their faults should not and can not be passed on to a former officer.
IODPA will look forward to reading the future judicial review transcript where it is mentioned that the plaintiff was selected by a computer program and not on the medical merits on whether it was suitable to review.
Any letter sent by A&S to any individual has to be answered with a curt:
“Why me? Why Now?”.
If the answer from them is because ‘our Excel workbook’ says you’ve won the review lottery, then you can laugh your way to legal representation.
On a side note – IODPA wonders about the Data Protection Act and the use of names in a tool designed for pseudo-random selection. Perhaps Subject Access Requests should be made to determine if a person’s name exists on their list.
This DPA principle seems to suggest it is not right to use personal data in a list to unlawfully select someone for the wrong reason:
Personal data shall be obtained only for one or more specified and lawful purposes, and shall not be further processed in any manner incompatible with that purpose or those purposes.