In case you haven’t realized it yet, TARTLE is a strong advocate for the importance of data collection and evaluation when it comes to making decisions. Part of that is recognizing that it is important that we have complete data, not just data that looks complete. We also have inherent biases to account for, biases that can color how we frame, process, and evaluate our data. Or to put it another way, we tend to see what we want in the data, or only look for the data that confirms what we already believe, not realizing that there are data that goes the other way as well.
The many controversies that have arisen in the wake of the COVID-19 virus illustrate all of this perfectly.
Here are just three examples:
Masks – People on the far right reject them entirely on the basis of a few studies that say their effectiveness is limited. People on the far left advocate for wearing them 24/7 based on the same studies. In fact, it may be that the right masks in the right situation is effective, and many situations where they don’t matter regardless of the mask.
Hospital Capacity – There are many reports that hospital capacity is being challenged. While the presence of a percentage certainly implies that some data has been analyzed, the fact is that data is very incomplete. To truly understand the situation, you need to compare the capacity with similar times over the last couple of years. Another question that needs to be asked is how many of these people need to be in a room with a bed? How long are those beds occupied by a given patient? Is it one person for a week or three people in a day? How is the counting being done? This matters because there is a perverse incentive on the part of hospitals to over-treat people. The more services a patient uses and the more time is spent using them, the more the hospital can charge to the insurance company.
Case Numbers: Cases are indeed increasing but you’re not hearing a whole lot about the death rates. That’s because the numbers there are remarkably low and the average of those dying from it is within a couple years of life expectancy when looked at nationally. In some places the average age of death from COVID is over 80. Also, nearly everyone has some sort of underlying condition that made them less able to deal with the virus.
Another important example has taken a more prominent role in the headlines over the last month. That would be the recently announced COVID vaccines. There are a number of competing ones at this point, each touting an effectiveness of 90% or greater. Now, if that’s proves to be accurate that is amazing. But let’s dig just a bit into that data, at least the data we have as of this writing.
The Moderna vaccine is advertising an effectiveness of 94.5%. To get there, they had an extremely large sample size, a total of 30,000 participants, evenly split between a control group that received a placebo and a group that received the vaccine. What’s interesting is that only 90 people in the control group got COVID; 90 out of 15,000 or 0.6%. Among that small group only eleven got significantly sick. The study made no mention of any deaths.
Of the group that got the vaccine, five contracted COVID and none got significantly sick. Now based on that, we can surmise that the vaccine is effective. However, there are important points missing in the data. For example, we have no idea of the sex, age, health, and daily habits of any of the participants. These are all important factors in determining the effectiveness of the vaccine. If all of the people who contracted COVID in each groups were healthy and in their twenties, that tells us something different than if they were all in their seventies. For example, whether or not they were wearing masks would also be useful. Were they largely staying at home or working as normal? What climate were the participants in? Again, these are important variables.
All of this goes to show that having data, all of the data available in a truly open and transparent forum that anyone can look at is incredibly important. Without that, we are prey to biases, those of others and our own. We need as much unvarnished data as possible to make better decisions and counteract those biases so we can get at the truth of things.
What’s your data worth? Sign up and join the TARTLE Marketplace with this link here.
