COVID Statistical Errors

By Antonio Graceffo
Antonio Graceffo
Antonio Graceffo
Antonio Graceffo, Ph.D., is a China economic analyst who has spent more than 20 years in Asia. He is a graduate of the Shanghai University of Sport, holds a China-MBA from Shanghai Jiaotong University, and currently studies national defense at American Military University. He is the author of “Beyond the Belt and Road: China’s Global Economic Expansion.”
January 31, 2023Updated: February 2, 2023


As the pandemic rolls into its fourth year, the Biden administration renewed the COVID-19 emergency to keep the cash flowing. But the statistics justifying a public health emergency and vaccine mandates make very little sense.

In 1954, author Darrell Huff published the book “How to Lie with Statistics” about how statistics are misused, and errors in their interpretation can lead to wrong conclusions. Many of the articles and even official reports about COVID-19 and the vaccines are plagued by the “errors” Huff described in his book.

One of the first ways to lie with data is by not defining the time period. We’re told that COVID-19 is dangerous because it has killed 1.1 million Americans. But this is over a period of more than three years. We’re also told that deaths continue to climb, but this number is cumulative, so it never trends down.

Similarly, claims that most of the COVID-19 deaths have been among the unvaccinated may be accurate if we include 2020, when no one was vaccinated. In that case, the unvaccinated comprised 100 percent of the deaths. But if you track the comparison of deaths between vaccinated and unvaccinated now, the unvaccinated are steadily comprising a smaller percentage of total deaths.

And this leads to the next way to lie about statistics—definitions. Any report that wants to show that the unvaccinated are dying at a higher rate will use the strictest definition of vaccinated, which means having all available vaccine boosters. As booster uptake is steadily waning, fewer people are considered fully vaccinated. If you remove the one-shot, two-shot, or no-booster people from the unvaccinated group, you’ll find that the completely unvaccinated comprise the smallest percentage of people dying from COVID-19.

Claims of vaccine safety are similarly flawed. An article by Johns Hopkins Medicine purported to dispel the myth that the vaccine was negatively impacting women’s fertility. As proof, the article claimed, “during the Pfizer vaccine tests, 23 female volunteers involved in the study became pregnant, and the only one who suffered a pregnancy loss had not received the actual vaccine, but a placebo.”

The article doesn’t say how many women were in the two groups or how many of them attempted to get pregnant, however. It’s possible that 100 women tried to get pregnant and failed, for example. It also doesn’t tell us how many of those who became pregnant were in the vaccinated group. It states only that the women were “involved with the study.” It’s possible that all 23 of those who became pregnant were in the placebo group.

The same article claimed that “studies found that the two initial vaccines are both about 95 percent effective—and reported no serious or life-threatening side effects.” The statistical error here is not defining the term “effective.” This begs the question, effective at what? And how did they come up with that percentage? “How to Lie with Statistics” calls this error “an average of what.”

Epoch Times Photo
A sticker on the window of a pickup truck expresses disdain for vaccine mandates in Gainesville, Fla., on Jan. 21, 2023. City employees successfully challenged a vaccinate-or-terminate policy during the COVID-19 pandemic response. (Nanette Holt/The Epoch Times)

As for the claim about no serious or life-threatening side effects being reported, it’s unclear how these terms are defined. It’s also unclear if no serious or life-threatening side effects occurred or if they just weren’t reported. The article also doesn’t provide a timeframe. If someone became seriously ill three days later or a week later, was this taken into the statistics? When we see a statistic that seems shocking or unusual, such as this claim, it may be constructive to look at statistics for similar events.

According to the Mayo Clinic, the possible side effects of vitamin A supplements are bone thinning, liver damage, headache, diarrhea, nausea, skin irritation, pain in the joints, and bone and birth defects. Taking too much vitamin A can result in a coma or death, according to the National Institutes of Health. If even vitamins can be this dangerous, why would we assume that a vaccine has no side effects? All drugs and even vitamins are accompanied by disclosures of possible, life-threatening side effects. How can it be that only the COVID-19 vaccines have none?

Age stratification of risk is another mistake that everyone makes when they say the vaccine is effective for everyone. For those younger than age 65, the statistical likelihood of dying of COVID-19 is quite low. So when they claim that the vaccine is 95 percent effective, people misinterpret that number to mean that it reduces your chance of dying by 95 percent. But it doesn’t, it reduces your already small risk of dying of COVID-19 by 95 percent.

Another example of statistical chicanery is the Centers for Disease Control and Prevention’s published chart showing the chance of dying at any age compared to ages 18–29. If you’re 40–49 years old, you’re 10 times more likely to die than someone in the 18–29 age group. But if we chose the 85-plus age group as a reference, then you would be 0.02857 times as likely to die. That sounds a lot better.

The truth about COVID-19 numbers is that people older than 65 are more likely to die of COVID-19 than younger people. And this hasn’t changed since 2020. So the statistical risk of dying isn’t the same at all ages, and this is often omitted from articles or reports trying to convince you that you or your children are at risk. At the same time, we also know that vaccination rates are higher among people older than 65. And yet, across the whole population, the percentage of COVID-19 deaths that were above the age of 65 hasn’t changed dramatically. This raises a number of questions about the vaccine’s claim of being 95 percent effective.

In addition to official data, there’s a lot of anecdotal evidence being cited. People would go on social media and say, “I am a (doctor, nurse, or hospital janitor), and I can assure you that COVID-19 poses a threat to everyone because I see so many patients in my hospital.” If these stories are even true, then they’re guilty of selection bias. Because these people are in a hospital, they only see the people who were sick enough to be in a hospital instead of a broad cross-section of the population. In fact, 100 percent of the people they see are sick enough to go to the hospital, but this doesn’t mean that 100 percent of people are sick or will be sick enough to be in the hospital.

In conclusion, statistical data is very important for appropriate decision-making. When we decide how much life insurance or how much health insurance to buy, we conduct a risk analysis to see what the maximum damage of being uninsured would be. Then we compare that to the risk of a catastrophic event happening. We decide how much risk we can live with and how much we want to reduce it by buying insurance. Although the event of an earthquake in Chicago would be devastating, it’s very unlikely, so people don’t generally insure against earthquakes.

We’re permitted to make these decisions for every aspect of our lives, except when it comes to COVID-19. And to make an informed decision, you need accurate statistical data. But much of the data that has been presented is horribly misleading. Too many people apparently know how to lie with statistics.

Views expressed in this article are the opinions of the author and do not necessarily reflect the views of The Epoch Times.

Related Topics