WHAT CAN ARMCHAIR EPIDEMIOLOGY teach us about scientific expertise? Should scientists stay in their lane? Let's discuss ...

During the early days of the pandemic, we saw a dramatic rise in armchair epidemiologists. Many non-epidemiologists with quantitative backgrounds, who were stuck at home due to lockdowns, openly speculated on the incoming pandemic data.

Some of the more adventurous armchair epidemiologists started fitting models to that data and sharing those models with others on social media. Due largely to social influence coming from their non-epidemiologist lives, they often drowned out actual epidemiologists.

Even worse, their models were often in conflict with what actual epidemiologists had to say. The most adventurous non-epidemiologists started reading epidemiology papers, critiquing epidemiological models and challenging the epidemiological consensus.

Some armchair epidemiologists even speculated that they might be more qualified to understand pandemic data than the epidemiologists themselves. A very prominent economist publicly wondered about the types of standardized test scores required to enter the field of epidemiology.

The implication seemed to be that epidemiology as a field might not contain our most intelligent scientists. His comment was not well-received in the public health community.

At this point, I feel compelled to state that as far as I know, there's no scientifically-proven causal link between the average standardized test scores of a group and whether their collective efforts over decades can produce reliable mathematical models.

It's important to realize that this questioning of academic epidemiology came at one of the worst times possible, while most epidemiologists were scrambling to provide expertise to a world that was and is still struggling with an epidemiological disaster of unprecedented scale.

Epidemiologists were put in the position of choosing between defending their store of knowledge accumulated over more than a century to a largely uninformed and dismissively skeptical technical audience or providing their desperately needed expertise to a terrified world.

I spoke out publicly at the time because I felt that thousands of chattering armchair experts during a pandemic emergency was a dangerous distraction. However, others disagreed. They felt I was advocating excluding people from participating in science in a discriminatory way.

In science, we call this GATEKEEPING. Gatekeeping is often used to shut down good faith discussions about who should be considered an expert in a scientific field. Based on this view, we might decide that only epidemiologists should do epidemiology. End of discussion.

Many object to this. They want a world where we assume no expertise and only listen to the quality of the scientific arguments themselves. I think this only makes sense if the audience are experts as well.

For people who lack expertise, who don't understand the arguments, this can lead to going with the voices that seem the most confident or the most like themselves. It can also lead to people throwing up their hands and declaring that it's all confusing and there are no experts.

I see identification of experts as a service to the consumer of scientific expertise. It's like food labeling in the supermarket. Labeling food lets us know what we're putting in our bodies. Identifying experts lets us know what quality of arguments we're letting into our minds.

In this context, being overly concerned about gatekeeping can be bad because it shuts down conversations about which kinds of expertise are relevant for the question at hand.

I think the answer is somewhere in the middle. We need some gatekeeping but we also need to have real discussions about what constitutes expertise and be intentional about who gets labelled an expert depending on the context.

Within every scientific field, there's a hierarchy of expertise. People who study the field longer and have attained degrees, contributed scholarship, been awarded prizes and earned the respect of their peers are naturally considered more credible that those who have not.

Within any individual field, this seems like a reasonable and obvious way to think about credibility. The complications and conflict come when people cross fields. What bearing should credibility acquired in one field have on credibility within another field?

It seems naive to me to assume equal credibility in a scientific setting. Everyone is not equal in knowledge and skills. It's reasonable to request that armchair epidemiologists and others explain why their expertise is appropriate and to examine those explanations critically.

The idea that scientific credibility in general should transfer to credibility in epidemiology in particular is a logical fallacy. It's an appeal to authority. It is not the same as having credibility in epidemiology due to a track record of performance in epidemiology.

To logically demonstrate that credibility in some quantitative field should imply credibility in epidemiology we must establish that this field somehow speaks to epidemiology. To avoid the fallacy, this can't be generally assumed. It must be argued on the individual merits.