People Provided 28 Fascinating Examples Of Popular Statistics That Are Misunderstood The Most

A popular saying is that numbers don’t lie. But apparently, people do. It might seem like statistics is a branch of science that should not be argued with. But as it turns out, it is possible for people to exaggerate or misuse certain facts. According to Datapine, “Misleading statistics refers to the misuse of numerical data either intentionally or by error. The results provide deceiving information that creates false narratives around a topic. Misuse of statistics often happens in advertisements, politics, news, media, and others.”

Not only do regular people tell half-truths with stats, but it’s also professionals who sometimes muddy the waters. A shocking survey of scientists found that 33.7% of them had engaged in questionable research practices like changing the results of studies to improve outcomes, subjectively interpreting data, and holding back on certain analytical details.

Bored Panda interviewed Professor Jennifer Visser-Rogers, the Vice President for Statistical Research and Consultancy, PHASTAR, to understand a bit more about statistics and why people misuse them. Here’s what she shared: “A statistic on its own is just a number. It’s the context that tells the story, but unfortunately, this can result in a misleading story if the context is inaccurate.”

People also tend to fall prey to the issue of sample sizes. For example, let’s say there’s a survey wherein 60% of the respondents preferred Version A over Version B. Obviously, you’d imagine that Version A was the superior option. If the number of people surveyed were 10,000, that would be a pretty significant sample size to back up the research. But if you found out that the number of people surveyed was only 20, then it would mean that just 12 people preferred Version A. That’s why the context behind the statistic matters.

Jennifer provided an example to explain how people should interpret statistical data. She said: “So, for example, if I tell you that eating bacon every day increases your risk of pancreatic cancer by 20%, that’s quite a shocking figure. But what does that 20% increase even mean? It’s a relative risk, which tells me what the risk is in one group compared to another. But it doesn't tell me anything about what my risk actually is.” 

“If I were to tell you that we have a 1 in 80 lifetime risk of pancreatic cancer, that means that out of 400 people who don't eat bacon, we would expect 5 of them to get pancreatic cancer anyway. A 20% increase in this means that if you took 400 people who ate bacon every day, we would expect 6 of them to get pancreatic cancer. So only an extra 1 person in every 400. Which doesn't sound anywhere near as shocking as a 20% increase,” she added.

Another fascinating example Jennifer shared illustrated how easily information could be manipulated to make a statistic seem more appealing. In fact, many brands and companies might use this tactic, which is why people should start paying more attention.

She stated, “One of my favorite stories is that Colgate came under fire for one of its advertising campaigns that stated that ‘80% of dentists recommend Colgate.’ It was reported for being misleading and suggesting that this meant dentists recommend Colgate over other brands. But that wasn’t how the data was collected, dentists were told to recommend several toothpaste brands, so it wasn’t that 80% recommended Colgate and 20% recommended other brands,” she added.

An interesting phenomenon in the world of statistics is the Simpsons Paradox (not based on the show). It shows how oversimplification of data can cause people to come to dangerous conclusions and why a healthy bit of skepticism is important. A popular example of this is when UC Berkley feared they would face a lawsuit for gender bias because they admitted fewer female applicants in one year. However, a statistician found that upon dividing the school into departments, there was gender bias in favor of women for 4 out of the 6 departments and no significant gender bias in the remaining 2.

Apart from sample sizes and people cherry-picking stats, Jennifer also shared that “Another thing that people often get wrong is the difference between percentages and percentage points. If, say, inflation goes from 4% to 9%, this doesn't mean that is has increased by 5%, it has gone up 5 percentage points. It’s a really subtle difference but with completely different meanings. But I get that concepts like this can be really difficult for people to get their heads around.”

“We are bombarded with data and statistics now on a daily basis, so I think it’s really important for us all to have a certain level of critical thought to challenge what we see. If we’re expected to use statistics to make decisions about our day-to-day lives, it’s really important to understand what questions we need to ask to be able to fully understand the context of what we’re being presented with,” she added.