You’ve probably heard a lot of statistics thrown around in the last year with Covid. But statistics can be super misleading. In fact, they’re notorious for being twisted around by whoever is putting them together.


Here’s a transcript of our conversation:

Emma: Hi, Brittany.

Brittany: Hi, Emma.

Emma: So, today we’re going to talk about statistics, and we’re not gonna be talking about the kind that you learn in math class.

Brittany: Thank goodness,

Emma: Not exactly. They’re, kinda related and they’re definitely sort of in the math world. But we’re gonna talk about the kind of statistics that you hear thrown around by people usually when they’re talking about politics or using it to back up their opinion. And I know that we’ve all heard a lot of statistics thrown around in the last year with COVID going around, and there’s something that have been used and weaponized by politicians and by people online and by the media. And something that, you know, I, think is important to talk about is that statistics can be very misleading based on how they’re put together. In fact, they’re actually pretty notorious for being twisted around. So before I get into that, I wanna just read a quick definition just so we can all get on the same page on what kind of statistics we’re talking about. So statistics are the practice or science of collecting and analyzing numerical data in large quantities, especially for the purpose of inferring proportions in a hole from those in a representative sample. So let me just explain that real quick. Yes.

Brittany: Unpack, if you will.

Emma: It’s a little wonky. So basically what that means is you take these large, these large numbers of data and you pull a small sample and you say this small sample. So I’m going to talk to people about their favorite political candidate or their favorite type of cheeseburger or whatever. You’re only talking to a few people, but those few people are supposed to represent like the whole country or, the whole Republican party or the whole of people who like cheeseburgers.

Brittany: So, which is me, It’s my favorite food.

Emma: Yes. Oh, yes. They’re my favorite as well. So they’re, notorious for being twisted around by whoever’s putting them together. one example of that that I think most of us have probably encountered at some point is Colgate, which is a brand of toothpaste. So you’ve probably seen those ads, Brittany, I’m sure you’ve seen them at one point, where, Colgate runs these ads saying 80% of dentists recommend our toothpaste. And the ads suggested that dentists actually preferred Colgate over any other brand of toothpaste. Right? When you hear that, you’re thinking, oh, well this must be the best toothpaste ever if all these dentists are recommending it. But the survey, the way that it asks these dentists was say, it listed several brands of toothpaste that these dentists would recommend. So maybe the dentist said Colgate in there, but they could have also listed three other different brands. So most people when they hear that, they’re like, wow, this must be really awesome. Like, I should go buy myself some Colgate. But the data actually did not really say that. it didn’t put their brand above other brands as being better. So it’s not quite the same claim. And that’s, just one example of how this stuff gets messed up. I know there are so many, Brittany, do you, have any thoughts on any of this? Have you ever encountered a weird statistic?

Brittany: Yeah. Well, I think one thing that really, you know, exemplifies this is the 2020, or no, sorry, yeah. 2016 election. Well, I forgot what year it was. So all these pollsters had Hillary Clinton winning. And pollsters that is statistics, right? Is who’s gonna win? So one thing that they did though is they were calling people on the phone and, surveying them. And that’s how we get these statistics. But Emma, what kind of people still have a house line? Not many people, right? Yeah. So a lot of the people that they were calling were in a certain demographic, so they knew or they didn’t know, but they were going to have a certain belief. I don’t mean to peg everybody as stereo, you know, stereotyping, but that’s just what happened. And so all these polls were like, okay, Hillary Clinton’s gonna win. Hillary Clinton’s gonna win. And then what happened in 2016, Trump won and everyone freaked out. So I bring that up to talk about sample size. You mentioned the sample size and who is in that sample. And a lot of times statistics are very, very, not tricky. Sneaky. So yes. So you can hand-pick who’s in your group, right? Or you cannot give all the information about who is in that group. And there’s the problem with correlation, not equaling. Wait, am I saying that right? Causation?

Emma: Yes. Correlation.

Brittany: Causation, yes. So the correlation is when something is like, oh, that’s kind of, those relate to each other, but causation is different. Cause causation is saying that because of A, B happened, right? So let’s say, I’m trying to think of a good example. So we know that smoking cigarettes can cause cancer. So we know that because of smoking, some people have a higher risk for cancer. I’m trying to think of a good correlation. What’s a good correlation thing that’s been used?

Emma: One that I have heard is like people with health problems that are being admitted to the hospital for COVID.

Brittany: Oh, that’s a great one. That’s an absolutely great one. Yes.

Emma: Yeah. Do you, wanna talk about that a bit? And sort of the different health factors that have gone into that?

Brittany: Yeah. So you know, all these numbers, and I don’t mean to be, I don’t know what the word is. I’m gonna be a little bit, snarky about this, but the survival percentage rate of COVID is very high. It’s not anything like, what’s the one? I can’t remember. not the measles. There was another plague I was just reading about it that came into America years ago. A typhus maybe, I don’t know, higher death rates, but they weren’t as spreadable. So you hear about these people dying and it’s, then you look at why they were dying and there was a lot of people that had preexisting health conditions. Yes. So things like diabetes, things like obesity, those were the people dying of COVID. Now that doesn’t mean they were the only people. That doesn’t mean there’s not exceptions to the rule. but, so that was, I think that’s a great way to put it. So there was causation for that too. So, we have to look really carefully at these statistics. I read an article that was like, here’s all the people under 30 dying. And it was meant to scare us. Right? And I, yeah. Scared me. I clicked on the article and then went through every single person and why they died and every single person had a pre-existing condition Exactly. Correlation is equal causation. So I think that’s a great example.

Emma: Yeah. And a lot of times too, because sometimes these statistics are being presented by the scientific community, which their job is to ask questions Yes. And find answers and be unbiased. But, when there’s a motive for them, which is like something that’s motivating them beyond just finding the truth. Maybe it’s who’s funding the studies or maybe it’s political pressure to prove a certain point. Maybe it’s trying to keep people more scared of COVID. I think that’s a really real one that we’ve seen. There could be a motive there that would actually prevent them from wanting to find the real truth, which is that most people who were admitted to the hospital for COVID had some sort of underlying condition. Yep. And some sort of health condition that made them more vulnerable. Now that doesn’t mean that they deserve to die of COVID, of course.

Brittany: No, of course not.

Emma: But what we’re saying is basically there are other factors at play here other than just did you contract COVID or did you not? So that’s kind of an interesting thing. And we heard a lot in the last year about cases versus deaths. And that’s something where a lot of times people will just be listening to the news and they’ll just hear a number thrown out, oh wow, we hit a record high of cases today and that doesn’t necessarily mean deaths. And there was that big problem of people being like, wow, well Florida is an open state and their cases are really bad. Yes. But then they would be comparing them to New York where they were actually talking about people dying. And dying is very different from contracting, you know, a virus that has a very high rate of survival. Those are two really different things. So you can end up with this problem of comparing apples to oranges if you’ve ever heard that phrase. It’s basically comparing two things that can’t be really compared. And that’s something we’re sort of that dishonesty and how things are presented can come into play. And there are all kinds of great quotes about, you know, statistics and why they’re not always super reliable. One of my favorites is from Mark Twain.

Brittany: Yeah. He’s a good one.

Emma: And he said that facts are stubborn things, but statistics are pliable. And I think that’s.

Brittany: Wait, does that not come from John Adams in the HBO O series? They, not the last part, but they are attributed to John Adams.

Emma: That, you know what? I actually believe that it was John Adams. I believe you not, because that’s, who I have seen it attributed to. Mark Twain was notorious for this. He would totally take people’s quotes and start, you know, passing them off as his zone really. So, yes. We’ll just say that one is John Adams because he was around long before Mark Twain.

Brittany: That is so funny.

Emma: But there are all sorts of different quotes like this. There’s one that’s kind of unknown where it came from, but Mark Twain, you know, was talking about this as well. Figures won’t lie, but liars will figure.

Brittany: I love that. That’s a great one.

Emma: Yeah, that’s a great one. Cuz it’s like, hey there’s this truth out there. The truth is, out there somewhere, scientists and statisticians, their job is to find what it is and, understand it. But a lot of the times because of, you know, we’ve talked about special interests a lot, special interests can sort of twist what is actually presented to the world. So you’ve talked about political polling earlier we talked about Hillary Clinton. An interesting thing that can happen when inaccurate polling starts coming out is it can actually motivate people to vote or not vote based on the polling that they’re seeing. And it can actually change the outcome of elections.

Brittany: So it’s manipulation.

Emma: Exactly. It’s manipulation. So a lot of people, if they see, oh, Hillary Clinton is leading by, by 20 points, which means she’s up like 20% above, above Trump, then they might say, oh, well I was gonna vote for Trump, but you know, I’m really tired. I don’t wanna go stand in line. If he’s gonna lose, then I’m just not gonna go. And that’s a huge problem that you see with some of these polls that can come out, is they can actually sway people and change the outcome of an election. And it brings up a really interesting question about the ethics of polling and how all of that should work. And it’s, you know, there are a lot of different things at play in the numbers that we are all presented to in our daily life. And it could be toothpaste ads, or it could be politicians and polling. But the best takeaway that I want you guys to take from this is that it’s good to be a critical thinker. Yes. And it’s good to be skeptical about the numbers that you hear thrown around by people. So maybe it’s on tv, maybe it’s online. even sometimes, you know, I remember growing up there would be crazy statistics in my history books that I would look into cuz I went to public school and they would be like completely off and not really have anything to back them up. So I’m not trying to get you guys to start doubting all of your school lessons.

Brittany: Questions, But always be questioning.

Emma: Yes. It’s good to be a critical thinker and to ask questions. And we’ve talked about this before, digging into the facts because the facts are what you ultimately want to find. And a lot of times statistics don’t exactly, don’t exactly represent the real facts.

Brittany: I think you’re right. And I think another thing, and we’ve kind of talked about this with special interest before, but follow the money as an expression, right? Yes. And a lot of these studies, and I say that with air quotes, are paid for by the people who want the outcomes to turn out a certain way. So they’re very biased. And one thing that you kind of touched on too that I want to say before we go is that, oh my goodness, I just forgot my thought. That’s so funny. I’m like, I have this great point to make, and oh my goodness. So never mind, we’re gone now.

Emma: You’re talking about special interests. It made me think about, we always talk about the dairy industry. I don’t know how we always end up on that but they have actually funded a lot of studies to show that drinking milk is like really necessary to have.

Brittany: That’s a good yes. Remember. And for some people, yes.

Emma: Well yeah, when we were growing up, Brittany and I don’t know if they’re still on tv, but the got milk ads were everywhere.

Brittany: Everywhere.

Emma: And they had these celebrities drinking milk and it was like this huge cultural thing. And those studies that they were citing were quite literally paid for by like Dairy Farmers of America. And like these huge, basically, associations paying to try to convince people to drink more milk cuz they wanted to sell more milk and make more money. So follow the money is a great takeaway. That’s, what I would advise you guys to do is follow the money, dig into the facts and think critically. Brittany, thank you so much for, jumping on here and sharing your thoughts. We will be sure to link some interesting videos. Oh, there’s a Ted talk video that I wanna link about statistics in here. I almost forgot. Definitely check that out. It’s got some cool, some cool examples of how this stuff can play out. And, that’ll be in show notes. So check those out and we will be sure to talk to you guys again soon. Thank you so much, Brittany.

Brittany: Talk to you later.