Current Global - We need to rediscover the stories in statistics

We need to rediscover the stories in statistics – our humanity depends on it.

“Data are just summaries of thousands of stories – tell a few of those stories to help make the data meaningful.” — Chip and Dan Heath

According to Alan Bleakly, a professor of medical education in the UK, there are two ways of knowing: science and stories. Conventional analytical methods are descriptive, while the value of a story is in its emotional impact. In other words, a narrative adds soul to statistics.

Of course, ‘big data’ and machine learning are already revolutionising medicine. Thanks to massive data sets, medical researchers can analyse patients’ records around the world to see which treatments work more effectively and which do not. In addition, utilisation of analytics in medicine has the potential to reduce costs of treatment, predict outbreaks of epidemics, avoid preventable diseases, and improve the quality of life in general.

The problem is this. To protect people’s privacy, medical data are often deidentified, which removes any sensitive information that can be used to potentially identify a patient.

On one hand, this makes total sense. The experience of COVID-19 has demonstrated that personal data can be potentially used as a way to restrict movement, civil liberties or shared with law enforcement.

But deidentification also can amount to dehumanisation. It means we lose crucial details about people’s experiences. When no trace of a person’s identity is left, they become nothing but points in a data set.

This is where storytelling comes in. It’s a critical tool when you need to gather insights into human behaviour.

Take cancer screening: while data can tell us there are low rates of breast cancer screening among African American women, storytelling through focus groups helps us to understand why. Researchers in North Carolina, for example, found a widespread perception of low risk stemming from conflicting information from family, media, and health providers, as well as a belief that cancer is a “White disease.” Storytelling was further used to combat misinformation and engage women with compelling information about the importance of screening. None of this would have been possible from deidentified data points alone.

It is imperative we connect science and storytelling together. Although every data set tells a story, not all of us can engage with it. By allowing for effective communication of insights from any data set using narratives or visualisations, data-led storytelling offers a compelling way to bridge the divide. It turns it into perception that resonates with us emotionally, one that we have capacity to appreciate.

In the age of misinformation, it is now essential that scientists take an active role in educating the public about what they do, why they do what they do, and why it matters.

Storytelling is the tool that allows listeners to engage with the essence of complex ideas and concepts in a more meaningful way, as well as making science more accessible and inclusive to communities that are often excluded from it.

At the end of the day, you can have as much data as possible, but it means nothing if you don’t know the humans behind these numbers.