95 – Statistics Bias, Fake News, Question Authority

Life Unsettled with Thomas O'Grady

This may start out sounding a little bland, but I want to blow your socks off. It is also not about politics except as an example. You need to be aware because biased statistics has you following bad medicine; eating the wrong things and making bad decisions for your business or life. I may be talking about biased statistics, statistical bias, etc., but what are we really talking about? We’re talking about the information you get and hear about every single day from the media, from teachers, from other people. If you’ve ever seen some of those movies or listen to some of the songs from the 60s, one of the big themes was: Question Authority. Question Authority Question authority is important for lots of different reasons, but what causes bias? Some of it is biased in the sample the way they choose people. Who do they choose to ask about a certain question? Some of it is actually that they don’t know any better, but many times what people are doing is they’re trying to influence you and give you some reason to believe in what they believe. Now, they may be thinking that they’re doing well, but they’re giving you bad information to base your decision on and trying to convince you or tell you what is the case. I’m going to give you a lot of different examples of this, a lot of background, and a lot of things for you to think about, some of which you probably believe – but they’ve been proven wrong or certainly have not been proven that you should listen to them. It’s one of my favorite topics anyway, but it came up because I was sitting there watching some things on TV, and they were talking about the popularity of Trump. I don’t really care whether you like him or dislike him. I’m trying to talk about the statistics. Both sides might do it, all kinds of people, I don’t really care about that. What I care about is you getting good information. Politics only as a Statistics Bias Example The reason I’m talking about it is not because of Trump or politics, but because if you’re doing a business, starting up a business, or trying to grow your business. You have to be able to discern and listen to true authority (how to establish authority), and understand a little bit about what may happen in statistics. People might be doing to convince you, convince you of their product, their company, or whatever. For example, what should you be investing in? What should you be joining or starting up? There are different reasons that people have. Some of it is just bad information; they don’t know any better, they’ve just never really understood statistics enough to know what’s good or bad. I’ll give you an example of that, too, when I go into some of this. First of all, what about that political example? Just recently Rasmussen and John McLaughlin of the McLaughlin Group came out, and they both have Trump’s favorability ratings at about 50%, and John McLaughlin was being interviewed because it differs so much from some of the results that you’re hearing about from the others. They are the Quinnipiac, Gallup, and Pew polls, which all had Trump down in the 30s someplace, low 30s to low 40. What was the difference? This has got to be not ignorance but actually deliberate because it is so bad—in the Quinnipiac poll only 23% of the sample were Republicans, Gallup – 26%, and then in the Pew poll they sampled 51% Democrat. Now, in the last election only a few months ago the people that identified and voted as Republicans were 33%. They should have been polling something that was a reasonable sample of people who had voted or likely voters. You can only wonder whether this was deliberate in order to tell a story, and to sell their samples and to sell their polling. Again, forget about Trump and all that other stuff. First thing, though, this was pointed out and then in the case of Fox News they continued to talk about the polls as if they were legitimate and used their numbers. Why use numbers that you’ve just discovered were bad?