It has been widely reported that a data scientist in Florida was fired for refusing to manipulate the data being presented on the COVID-19 dashboard she had created. This incident raises many issues and it is appropriate that the Data Science Foundation expresses an opinion.
Before the opinion, a few observations. Firstly, the fact that the story has received so much global coverage, suggests not only interest, but perhaps shock and outrage. Some reporters are concerned about health and well-being, others about an individual’s right to professional integrity and others still about a political agenda favouring a small section of society. A few thankfully, are concerned about the role of the data scientist and the fact that accurate unbiased data, and how they are analysed and presented, are the foundation of the decision-making process. Secondly, are we surprised that those in positions of authority want to visualize results in a way that favours their interests? And finally, that this action exposes what some have come to believe, that the data upon which a visualization is based are manipulated prior to the visualisation. In this case, the dashboard did not support a policy decision.
Should we be concerned? As data scientists we should be distressed, as citizens we should be alarmed. There is more at stake here than professional integrity or the ability of a profession to do what is right. Data scientists work to enable data to tell a story, one that is accurate, fair, and unbiased. But they can only influence decisions, they do not have the authority to make them. Data scientists typically do not own the data they work with, so cannot dictate how it or the visualisations they create are used.
In this era of fake news, elected officials who are selective with the truth or habitually lie, how can the public be expected to participate the democratic process? Maybe they are not expected to participate, but to believe what they are told with data visualisations becoming the new statistics. Public institutions should have a duty to present an accurate and balanced account of the truth, including the uncomfortable truth which is not to their liking. Unfortunately, this often is not the case.
The answer is to be found in education. If more people had the ability to understand how data is collected, cleaned, and analysed. There would be less opportunity for the unscrupulous to manipulate it. Educated people do not accept what they are told at face value, they question it. They ask for supporting information and references. It is important that journalists, as gate keepers of the news, are the first to become data aware. Journalists have a responsibility to question the news for validity, not just to report on what they are told. This responsibility could be extended to include information that is being released directly to the public.
If anything, this story has told us that the big news is not the numbers presented on the dashboard, but the manipulation of the numbers. If more journalists had the ability to interrogate data, those in authority would be less likely to pressure data scientists in to being selective with the truth. This would result in fewer data scientists losing their jobs for doing what they believe is right and would prevent further corrosion of the democratic process.
This opinion relates to the scenario reported and will be updated in the light of new information.
If you found this Article interesting, why not review the other Articles in our archive.