« Swear and gossip: answer to question 2 | Main | Swear and gossip: answer to question 1 »


Feed You can follow this conversation by subscribing to the comment feed for this post.


You're arguing here for using absolute risk, rather than risk ratio, which in many cases is a better guide. It isn't for vaccines because the risk is quite variable, depending on country. Risk of having had covid ranges from less than 1 in 10,000 up to almost 1. For a country where everyone has covid the absolute risk reduction is 60-80% from being vaccinated. Without vaccination everyone would have covid.


Ken: Nothing in this post re-defines VE though. I'm just using the definition as used in the vaccine trials.

You raise two interesting issues about VE definitions.

1. "Without vaccination everyone would have covid"

This would imply we don't need a control/placebo group. But the statement is true only if we let time -> infinity, and not true within any study period.
Also think about VE for deaths. The analogy would be "without vaccination everyone would die from covid"

2. Relative risk ratio and absolute risk (difference) are both useful metrics, depending on the situation. With vaccines, the community has decided on risk ratios. With ratios, we are assuming that if the vaccine cuts infection by 50% on a base rate of 10% in one place, then the vaccine would cut infection rate from a base rate of 1% to 0.5% in another place. Yes, this allows us to apply the result to any regions and it's easy to forget that VE being the same everywhere regardless of base rate is an assumption. I don't think this assumption is obviously true.

The comments to this entry are closed.

Get new posts by email:
Kaiser Fung. Business analytics and data visualization expert. Author and Speaker.
Visit my website. Follow my Twitter. See my articles at Daily Beast, 538, HBR, Wired.

See my Youtube and Flickr.


  • only in Big Data
Numbers Rule Your World:
Amazon - Barnes&Noble

Amazon - Barnes&Noble

Junk Charts Blog

Link to junkcharts

Graphics design by Amanda Lee

Next Events

Jan: 10 NYPL Data Science Careers Talk, New York, NY

Past Events

Aug: 15 NYPL Analytics Resume Review Workshop, New York, NY

Apr: 2 Data Visualization Seminar, Pasadena, CA

Mar: 30 ASA DataFest, New York, NY

See more here

Principal Analytics Prep

Link to Principal Analytics Prep