Physical Distancing: 3 Feet vs 6 Feet

7 min readMar 16, 2021

The news is buzzing with news of a new research study showing that “3 feet of social distancing is safe in schools”, which of course many people are adjusting to “3 feet of social distancing means I can’t get COVID”. Let’s talk about that.

You can find the actual study here.

To really understand what this study means, we need to understand exactly what the study was actually about.

A Brief Overview

This study compared rates of COVID in school districts that implemented 3-foot physical distancing policies and those that implemented 6-foot physical distancing policies, and controlling for community spread found that there were not significant differences between them. This is important research — it reflects empirical testing of policy implementation, something we should be doing more often!

The data generally support the results. They found no statistically significant difference between school districts with 3-foot policies and 6-foot policies. They even managed to produce one of those miraculous graphs that really doesn’t need statistics to be interpreted!

Limitations of the Study

Despite providing very clear analysis, there are also important limitations we should consider when deciding how this study informs our behavior. There are four main critiques of the study that I’ve seen floating around, and I would like to take a moment to discuss each one.

1. Policy vs Behavior

A key limitation of this study is that it focuses on written policies rather than observed behavior. As anyone who has left their house in the past 12 months knows, these are very different things. As we’ve seen, schools with 6-foot distancing policies and universal masking requirements are often not consistently or meaningfully enforcing them (1 2 3) . As documented in recent research from the CDC (Gold et al., 2021), even in schools where 6 foot policies were in place it was not always possible to seat students more than 3 feet apart due to classroom layouts and the need for small group activities. The found that while mask compliance was high during observations, they were removed entirely during in-classroom lunches that “might have facilitated spread”. There were also specific instances of “lack of or inadequate mask use” were identifiable in five clustered outbreaks.

This presents a significant concern. If the schools who had 6-foot policies on paper didn’t enact them in real life, then it would be unsurprising to find no differences between them. The authors of the new physical distancing study do recognize that the possibility of differences between policy implementation and reality might be significant, but strangely they only seem to consider it in the opposite direction — that perhaps schools that allowed 3 feet were able to attain greater distance.

2. Reported vs Actual Cases

An issue that’s recurrent in COVID research is that it’s often difficult to get accurate measurements of cases in the community. Because some people (particularly children and teens) may only have mild symptoms or be asymptomatic, they may never seek out testing; in other cases people with clear and pronounced symptoms may avoid testing for financial, political, or social stigma related reasons and thus not be represented in the data. Asymptomatic and symptom disregarded cases are of particular interest when considering children, as they are more likely to have milder symptoms and families may feel economic pressure to avoid having kids stay home sick. In this study, only cases which were reported to schools and subsequently to the state were included, and surveillance testing was not widely used. The authors do acknowledge this limitation:

Practically speaking, because this source of error is likely evenly distributed between the two policy conditions (though one might hope a school with 3-foot distancing would be even more vigilant), it is unlikely that this contributes significantly to the results.

3. Community Spread vs School Spread

The goal of school-based mitigation policies is to reduce or prevent the spread of COVID in the school environment, where the group education setting provides a higher risk environment for transmission of respiratory diseases. Ideally, studies would be able to focus solely on the infections that occur in this space, but in reality it is very difficult to separate community-based spread from school-based spread. This is because many infections are not reported, and because school staff and student’s membership in the community creates a potential bidirectional relationship between community and school infection rates. As a result, studies often rely on tracking clusters of cases with known contacts.

In the 3-foot vs 6-foot study, researchers found a strong correlation between community spread and cases in students and staff, which is to be expected.

The authors did statistically control for community spread, and found that it didn’t make a significant difference. Again, this is not surprising, as community spread is likely to impact schools similarly.

4. Safe vs Safer

An area of disconnect between media reporting and the study itself is on the issue of “safety”. Safety is inherently difficulty to assess, as it ultimately requires a subjective evaluation of the degree of risk that’s acceptable relative to the potential benefits. Perfect safety — zero risk — isn’t possible short of closing schools indefinitely.

As a result, the physical distancing study focuses on whether there’s a difference in risk of infection (as reflected in incidence) between the two conditions — in essence seeing if one is safer than the other rather than declaring one or the other “safe”. This is a reasonable approach for a research paper to take.

Unfortunately, in media reporting has taken “safer” and translated it to “safe”. This can be misleading. As we’ve seen with previously with state regulations around indoor dining making something “safer” or meeting minimum safety requirements does not eliminate all risk, and if the risk assessor places greater value on economic or symbolic benefits than on the health risks then significantly risky activities may be endorsed as “safe”.

When considering relative safety, we need to take a broad lens. As this and other data suggest, the relative incidence of COVID in school children is lower than for adults, and their symptoms tend to be more mild. This is good news for them! But opening schools also means considering the risk for adults working in those settings (which are notably greater) and the resources we provide them to offset those risks. As the recent fight over forcing teachers to return to classrooms without vaccination shows, this is a real issue; unfortunately we still often fail to consider other contacts like janitorial staff, bus drivers, administrative staff, and paraprofessionals who may have significant contact with students while schools open but may not have similar health benefits to the teachers and higher administrators who are in the current media focus.

Summary & Conclusion

Ultimately, the study is an important piece of information — it shows us that the policies don’t differ significantly. What it doesn’t tell us is why. It could be that we’ve been enforcing more distance than required. It could be that we’ve been failing to enforce distancing, making the policy difference moot. It might even be that 6-feet was always inadequate for a classroom setting, and 9 feet would be necessary to reduce transmission. It could be that other failures of mitigation, like maskless classroom dining and poor ventilation, provide enough risk that any benefits from distancing were erased. And regardless of the cause, we don’t know if any of this is generalizable to adults, or outside of a school setting yet.

In short, this is promising research, but it leaves many questions unanswered. As discussed in the summer, there are differing opinions about how much distancing is necessary, but you’ll never err by having too much distancing to reduce your risk.

As always, I am not trained in epidemiology, and defer to recognized experts in the field on all issues. These analyses and commentary are solely designed to help lay persons approach the publicly available data and larger public health conversations.

Stay Home. Mind Your Distance. Wear a Mask.