Nov 17, 2020

By Marie Oldfield CStat CSci

Oldfield Consultancy and it’s working group have returned two calls for evidence from the Public Administration and Constitutional Affairs Committee (PACAC)

Oldfield Consultancy and its working group responded that the data coming from Government and numerous other bodies is not particulalry transparent. This investigation gave rise to our last few blogs on graphical analysis being discussed on linkedin. We also discussed the fact that, according to the Lords Enquiry, because the modelling was so poorly executed in the first place we were unsure as to what robust data was being referred to. So far, policy decisions do not seem to have been based on evidence or data and certainly, in the Lords Enquiry, modelling could not keep up with the pace of policy change. When PACAC asked who was responsible for the dissemination of data we stated that as the Government had signed a deal for ‘information dissemination’ with the media that it ought to make sure the data was understood. The problem with this being the data needs to be robust in the first place before dissemination occurs. We then demonstrated why the data currently being disseminated was not robust.

Since the beginning of covid, the resulting graphical analysis in the media has been, to one extent or another, unreadable by even a Statistician. Leaving the public to make up their own minds and to infer conclusions on graphs and data can be entirely misleading without good, contextual, explanation. This is why we then raised the issues of risk perception and dread fear because without understanding these principles the graphical output was not correctly targeted.

So far, robust graphical representation, contextual analysis and comparison are all seriously lacking in output concerning covid and as a result, we made recommendations to the committee. These recommendations concerned modelling best practise, data transparency and data dissemination.