reSee.it Video Transcript AI Summary
The speaker explains a distinction between absolute risk and relative risk and shows how both can be technically true but convey different impressions. Absolute risk is described as the real chance of something happening, while relative risk compares two small numbers to emphasize one as much larger than the other, often making the risk appear much bigger or smaller than it truly is.
Using a jar example, absolute risk is the chance of drawing a red marble from a jar with 10,000 marbles and only one red marble—a one in ten thousand chance. Relative risk, by contrast, compares two jars: if another jar has two red marbles, the statement would be that you have a 100% greater chance of drawing a red marble in this jar than in the first jar. Although the numbers are both small, the relative risk has doubled. The speaker argues that relative risk is a favorite tool of fearmongers because it makes tiny numbers sound large, whereas absolute risk shows the real-world odds.
The speaker then applies this to headlines. A headline claims you are 800 times more likely to get sick from raw milk than from pasteurized milk, labeling this as a relative risk number. It is technically true, but the absolute risk of illness from raw milk is about 1 in 13,000 for the people who drink it, which is less than one one-hundredth of a percent.
A similar framing is discussed with COVID-19 vaccines. The Pfizer vaccine is described as 95% effective in headlines, which the speaker notes is the relative risk reduction. In the trial data, the absolute risk reduction—the actual difference in risk between vaccinated and unvaccinated groups—was about 0.8%, less than one percent. The speaker asserts that this shot lowers actual risk by less than one percent, but the media emphasized the 95% figure. While not called a lie, this framing is characterized as incredibly misleading and capable of influencing public decisions.
The overarching message is that statistics can be technically true yet used to manipulate public opinion through framing. The speaker urges readers to compare whether a number refers to relative risk or absolute risk whenever confronted with alarming or astonishing headlines. Relative risk is said to sell headlines; absolute risk is said to tell the truth.