The Wonder of Reality

We've talked about science and probability, so it's time to bring them together in this, our third episode of the miniseries  "What is Science and How Does It Work?". This episode we're joined by Alexandr Diaz-Papkovich, a Master's Student of Statistics to talk about how we can apply statistics to science. First we model some data, analyze the results, discuss error and bias, then get into advanced statistics. How can analyzing data differently result in completely different answers? What should we do when we find data contrary to our theory? Find all answers in this episode, +/- 1 answer, 19 times out of 20. Corrections/Clarifications: When Alex says that it's impossible to get the value .6156 meters, he's referring to a continuous range of infinite size; there are an infinite number of points between 0 and 1 (Numberphile has a good explanation of infinity). In reality though, our measurement devices are only so accurate, so the same number can come up twice, as it did with our data. Links: Our Show Notes for the episode (Modelling, Confidence Intervals, Error and Bias, Simpson's Paradox, and Bayes' Theorem) The Additional Information from the episode (Modelling, Confidence Intervals, Simpson's Paradox, and Bayes' Law). We recommend reading through the Simpson's Paradox section as you listen; seeing what we're talking about is very helpful. Probably the easiest place to learn about the Central Limit Theorem is Wikipedia A simple explanation of Bias vs Error The Dewey Defeats Truman polling issue Nate Silver's 2012 election predictions and LISPOP's website Wikipedia's information on Simpson's Paradox; we recommend checking out our Additional Notes on the topic as you listen to this part Bayes' Law "in Plain English" as well as our Additional Notes on the subject

After looking at how to conduct good science in our first episode, it's time to deal with the basis of probability in this, the second episode of the miniseries  "What is Science and How Does It Work?". The most important thing about probability is the difference between dependent and independent events. Why do hot and cold streaks apply to only some casino games and not others? If the odds are 1 in 6 of rolling a 3, does rolling a die 6 times guarantee you'll get exactly one 3? We also discuss the famous Monty Hall Problem and if firemen cause fires. Dive into the episode below; just don't buy ice cream as it may cause drowning. Corrections/Clarifications: In the show we mention that the Monty Hall Problem is from Let's Make a Deal, a game show hosted by Monty Hall. In actuality, the Monty Hall Problem is only loosely based on the game show and is best known from Marilyn vos Savant's column in 1990. Links: Our raw Show Notes for this episode (Expected Value Fallacy, Gambler's Fallacy, Law of Large Numbers, Monty Hall Problem, and Correlation vs Causation) Our polished Additional Information for those looking for more on why rolling 6 dice doesn't guarantee a 3 coming up and the Law of Large Numbers The Expected Value Fallacy is better known as the so called Law of Averages The Gambler's Fallacy The Wikipedia page on the Law of Large Numbers (as mentioned in the show, this is a Master's level topic, so don't feel bad if you don't understand it) If you enjoyed the Monty Hall Problem, you'll also like the Gender Paradox which is another simple problem but has a strange outcome Wikipedia has more information on correlation versus causation; it's also where we got our examples We'd be remiss if we didn't include this great comic by xkcd on Correlation and Causation

Welcome to the first episode of The Wonder of Reality! Since we're excited about science, we first need to understand what science is and how to conduct good science. This episode is the first of the miniseries "What is Science and How Does It Work?" and we're dealing with the philosophy of science. How do you prove things in science and is that even possible? How do you remove sources of error? And do white shoes really show us that ravens are black? All this and more await! Corrections/Clarifications: The "White Shoe" argument against proof by induction is attributed to Hume (not "Howe") and an in-depth explanation can be found here When we originally recorded the episode, we only knew of the Bayesian approach to probabilistic science. Since then we've talked with Alexandr Diaz-Papkovich, a friend of ours and current Masters' student in Statistics (who you'll meet in episode 3), who has informed us of the Frequentist approach. We'll have more on this before the miniseries is over. Links: Our raw Show Notes for this episode Our polished Additional Information for those looking for more (covers "All Ravens are Black" being the same as "All Non-Black Things are Non-Ravens" and the consequences) An amazing YouTube video about science called "The Poetry of Reality" by "Symphony of Science"; it really captures our feelings towards science A simple illustration on how theories are refined (reddit discussion here) The distinction between deduction and induction An excerpt from Karl Popper on Falsification and Science Want to see us rip phone books? The video will be up soon on our YouTube channel NOVA explains the history of the search for Nessie up to 1999 and the BBC disproves Nessie with 600 separate sonar beams in 2003 A brief explanation of double-blind studies from Wikipedia; they even mention triple-blind studies on the same page How Stuff Works explains Occam's Razor (and yes, the spaceship and fallen tree example is from Eureka's "Alienated") NASA on Dark Matter and Dark Energy and how they're at the fringe of science Discovery News has a piece on magnetic monopoles and a bit of their history. A lot of the article covers the season 2 finale of The Big Bang Theory which deals with magnetic monopoles; you can read more on the science behind the episode here