People Are Statistically Illiterate

Added: 2026-05-12

[00:00:00]Numbers do not speak for themselves.

[00:00:02]Numbers never have spoken for themselves. Everyone serious in the world of statistics knows this. And yet an entire population has still been soped into taking dogmatic legitimacy from vague studies, scientists or statistics that they've never read. They do not understand and they do not even understand how is done. The reason this is an issue is the following. It encourages dogmatic thinking. People think about kind of religions in this sense that there is a one higher power and you cannot question anything that for example a holy book says. But I'm here to tell you that statistics and science today work exactly in the same way. It has become a pseudo religious dogma. It makes people easy to manipulate and control. If you want to take quite a cynical view of this, then this may actually be the intended purpose, but I'm going to talk about that a bit more later. and it confuses you and me and the general population into procrastination and paralysis and inaction. The purpose of this video then is to explain why each of these is a problem in more detail as well as explain a bit of the fundamental basics of how statistics and data science actually works and therefore how to avoid being manipulated and teach you how to make your own conclusions about things. Okay, let's just specify first exactly what it is we're talking about.

[00:01:16]We have for example uh the kind of person that comes onto the news outlet and they will be the so-called expert and they will be sounding very smart and showing you these kinds of lines and graphs and things and this of course when it comes to news you probably know this has some politically aligned agenda but this is not the only thing we're talking about.

[00:01:36]It could be, for example, someone simply trying to make content on a platform and they are just um showing you all these graphs and evidence of things they have never actually read and things that you won't read so they know they can use it and perhaps I don't know sell some kind of health supplement or some [ __ ] Okay.

[00:01:51]The first thing we need to understand a little bit in more detail is how statistics and uh mathematical modeling actually works. We don't we don't need to go into highle technical detail about whatever distribution to use at any given time. No, no, that's not what I'm going to explain here. But I am going to tell you the underlying philosophy of how statistical modeling works. The first thing you need to understand is that nothing is objective. Objectivity does not exist. That is the golden rule of statistics. You cannot say the numbers show this. The evidence shows this. Numbers do not speak for themselves. They never have. Like I explained earlier, all that you are actually doing as the mathematician, as the modeler, you are looking at a system, a real system that actually exists and you are deciding subjectively, yes, but usually with some real consu some real reasons as to why you're selecting what is the hypothesis that I'm trying to um do a test on. So for example, am I trying to see how a population um behaves if X or Y happens?

[00:02:47]That's the first thing. Then they pick an objective and constraints and variables that they can actually test and collect data on. So models very crucially rely on assumptions and actually one huge issue with something like academia is that academics are very out of touch with reality. I made a separate video going into full detail on that specifically so you can watch that after you finished this one. But the point is that a model is only as good as its assumptions. You can get numbers to do or say almost anything. But real intuition about a system is where actual usable insights will will come from, not from how good you can make your mathematics look. This is why you can get um models or studies that show so many things but are very useless to you and me. You've probably experienced it before yourself that there is some kind of study that is not very relevant to you. And the reason for that is because by the nature of doing scientific experiments, you need to be quite precise and quite um sterile with what you're talking about. And that loses the intuition of it being actually practically useful. So hopefully we should understand now that there is nothing mystical, there is nothing vague or complex or far away about the idea of carrying out um scientific testing or mathematical modeling. Anyone that tells you otherwise is trying to confuse you.

[00:04:06]what they are actually doing is far simpler than you realize it. You do this yourself every day. You create mental models or mental tests of things within your day-to-day life. When you go to a new restaurant, for example, you're testing whether that restaurant is good.

[00:04:19]If you have some kind of feedback, you have actual data on that restaurant, you you eat some dish and it's it's like dodgy and you get a dodgy stomach, that is real feedback and you've made a conclusion about the restaurant afterwards. for example, I'm not going to eat there again because last time I ate there, the feedback, the dish I ate was kind of dodgy. This is exactly the same thing that they are doing. The only difference is that they're applying some layers of like technicalities on top of that. So things you see as the general population about what like mathematicians do. It looks super complex. You see all these equations and graphs and stuff. These are literally like the 1%. These are like the final layer to just compute whatever it is you're trying to do. Now we also need to try and understand um where the incentives as to manipulate data come from. I think we all know that anyone with some kind of personal agenda will be pushing a certain statistical model that favors that. So when it comes to uh politics perhaps you have a certain political agenda that you want to push.

[00:05:19]Look, I want to explain something about mathematics, right? mathematics and pure logic is actually far more um rigorous and far more rigid than you think.

[00:05:31]When you see someone doing statistics, it's not objective because if two models are both equally objective, they must reach the same conclusion. Let me explain that concept. If two people start with the same axioms, the same um like basic rules in mathematics, there is only one conclusion they can ever reach. That is just a law of how logic works. When you see two people debating, for example, on TV some kind of political agenda and they both are talking about the same thing, but they're coming to different conclusions.

[00:06:03]This can only happen because the models are not objective. And that's fine.

[00:06:07]That's not an issue. By the way, models in reality cannot be objective. Logic does not apply to the real world. That's that's a misconception that people have.

[00:06:17]Logic is a really specific branch of mathematics that's super rigid and you can barely do anything useful with it.

[00:06:22]So what that means is that we as a population need to actually take some accountability. And the reason that we need to take accountability is because we [clears throat] have given these people the power uh and legitimacy that they have. Right? So legitimacy just comes from collective agreement. If a hundred people all agree that someone's an expert, if all of you watching this agree that I'm an expert, then I literally am an expert. If as a society, we all agree that something like like science and uh data in a really vague sense that nobody really understands because nobody does it is so important and is like literally objective truth and fact. Then then it then it becomes that then it becomes that. And what happens is that we allow that to be the manipulation tactic of someone that wants to misinform you. And the negatives this has on you personally, why this is actually relevant and why I'm even making a video about it.

[00:07:15]Because I don't care about, you know, widescale societal philosophies, but this becomes relevant because it essentially confuses you as the average person because actions we take in our own lives, we want to at least support them with some kind of legitimacy. So it it's kind of taboo to do something based on intuition or emotion. It shouldn't be and I again I have a separate video on that. Uh check out my channel if you want to see that. But the point is we want to support our conclusions with legitimacy. And the uh model of legitimacy that we currently have is something that most people don't understand. It's science and statistics and data and things like this. And this means that instead of carrying out your own uh say experiments, if you want to go to the gym and there's a kind of diet that you want to follow, you've been told by this supposed expert that keto diet is the best and you've been told by this one that only eating red meat and you've been told by this one that only eating vegetables, you become confused and you end up doing nothing. You end up having analysis paralysis. Now, relating that back to the objective of this video, hopefully by demystifying what it actually is that um scientists and statisticians, for example, are doing, you should gain a new found kind of um confidence in yourself to be able to carry out your own experiments for anything. What this means is that you can you can literally test. You do not need someone to tell you. Grow a pair of [ __ ] balls. You do not need someone to tell you what you can and can't do, what is effective and not effective for you. you are able through some basic training and understanding how these systems actually work to do that yourself. The main push back that someone will have if I say that is for example there are going to be some things that you can't test yourself. You don't have the money or the equipment or something like that. And to that I say well anything that falls into that category is not relevant for you to test anyway. So let's say for example like physicists at CERN like the large hadron collider they have billions of dollars worth of equipment and they're coming to conclusions on things that you are never going to be able to test in your day-to-day life. Fine, fair enough. I concede that. However, this is not even relevant to you. At most anything that falls into this category will be something that you read and find interesting. It's not going to be something that [clears throat] has direct impact on something you do in your actual life. The other thing is that you do not need to stay at the base level you are right now. You can learn and develop how to explicitly model things properly and how to do your own tests. Remember, academics don't gatekeep knowledge. You can take the knowledge of something that exists within academia, within something like mathematics, and you can apply it to your own life to carry out your own tests on things that you deem relevant.

[00:09:55]And you can learn and develop in that skill too. You don't need to be constrained by right now you don't know for example formal mathematics. Perhaps you never will need to learn formal mathematics. The problem with academia is that you learn a bunch of technical skills and then you try and apply them.

[00:10:09]In reality the most effective thing to do is to try and naturally discover whatever is relevant to you at any given point and then learn whatever technicalities are required at that point. If that message resonates with you and you want to learn more about how you can apply uh mathematical thinking in daily life as well as seeing more analysis like this exposing things like established systems, then make sure to subscribe to this video and interact by liking and commenting your opinions down below so that the discourse can continue to grow. That's all I have to say for now. Thank you for watching and have a nice day.