President's Column - August 2021

A Statisticians Role in the Pandemic of Misinformation

Warren Brettenny

20 August 2021

As we all adjust to a second year of “pandemic life”, I think it prudent that the president’s column should focus not only on what SASA is doing, but also the potential role that statisticians, as data experts, can play in our country during this increasingly difficult and trying time.

I was watching an interview with a medical correspondent, Dr Jon LaPook from NYU Langone Medical Center, who mentioned that we, as a global community, are in fact facing two pandemics. Firstly - and predominantly - the COVID-19 pandemic and secondly, with growing influence, a pandemic of misinformation and disinformation.

The former has had devastating effects on our lives, livelihoods and families. Outside of our own individual responsibilities to wear masks, stay socially distant and get vaccinated, the core responsibility to solve this pandemic fell (and continues to fall) into the hands of the virologists, microbiologists and chemists to create a safe and effective vaccine. Progress in this regard has been exemplary. With multiple vaccines developed and available, we are currently in a position we never dreamed possible even a decade ago. Statisticians, no doubt, played an important role in the progress that has been made in this area and I am proud that this is the case. The role of statisticians and data experts in this endeavour, however, is well defined and structured and their findings and results are subject to rigorous peer review.

The second pandemic (according to Dr LaPook) - as if it wasn’t bad enough that COVID-19 is wreaking havoc in our lives - is the “pandemic” of misinformation and disinformation. This “pandemic” complicates and frustrates the progress made by the scientists tasked with eradicating the pandemic and may indeed slow down the vaccine rollout and, in so doing, force us to live with COVID-19 for longer. In this “pandemic” of misinformation and disinformation, the role of statisticians and data experts is not well defined and often not even considered - but our role is nonetheless important. Misinformation and disinformation, while similar in name, carry somewhat different interpretations based on the context and intent of the information provided. Misinformation is information or interpretations of data which are inaccurate and/or are provided without the correct context - whether there is an intent to deceive or not. Disinformation, on the other hand, occurs when this information is presented with the direct intent to deceive. Either way, this information - which can come in the form of “half-truths” or outright lies - when proliferated, can have dire consequences on a personal, national and global level. A simple example of a “half-truth” is to compare only the COVID-19 case numbers of a large metropolitan city to those in a small town or village. Without the context of the population size and/or density of each of the compared cities, this only tells some of the whole (or true) story. It is in these cases that the role of statisticians - or indeed those who are simply data literate - is paramount.

At an educational level, our training of statistics students needs to emphasize the importance of context in the presentation of results. From the very foundations of the graphical and tabular representations of data to the most advanced modelling techniques, misrepresentations or misinterpretations of data and results need to be rooted out. We need to train our students to engage with a problem not only on the micro (solution) level but always with the bigger picture firmly in mind - so all conclusions that are drawn are accurate, relevant and unambiguous. This is particularly important to the research of postgraduate students - so instilling this way of thinking at an undergraduate level is exceedingly important.

Added to this, from a professional point of view, we need to call out these misrepresentations of data wherever they occur - from social media posts, to news broadcasts, to conversations with family and friends and even presentations by our senior colleagues. As trained experts in data analysis, the onus and responsibility - whether welcome or not - falls on us to set the record straight in times of confusion and the abuse of data. At the very least, our responsibility is to call into question any dubious conclusions from data by those with the influence to propagate misleading information. As a community we have a role to play and we should not be a shrinking violet in the face of such a misuse of our chosen discipline.

On a more administrative note, I am pleased to inform everyone that we are making great progress in our transition to a new and updated platform. Teething issues are a reality and I thank every one of you for your patience and understanding. We are aiming to host several online webinars in the coming weeks and months and I hope that you make the effort to engage with these. Furthermore, we are collaborating with the NGA-MaSS, the ICM, and NITheCS to bring additional potential benefits and opportunities to our members and society at large. Lastly, I would like to encourage you to attend our SASA 2021 hybrid conference, which will be hosted by the Department of Statistics and Actuarial Sciences at Stellenbosch University. Our return to hosting a conference will coincide with the 75th anniversary celebration of the department and is sure to be a very enjoyable and educational experience.

Keep well and stay safe