Streuspanne – Statistics and Its Curiosities

In the latest episode of our podcast »Streuspanne«, we take a critical look at the Harm to Benefit Ratio, a new key figure for assessing the Harm to Benefit Ratio of coronavirus vaccines. Based on a publication by Falk Mörl and colleagues, we discuss the reliability of this ratio, which provides particularly unfavorable results for the BioNTech vaccine.

The latest episode of the Streuspanne podcast is about election polls, forecasts and their uncertainties. Using Donald Trump's surprising re-election as an example, we shed light on how such forecasts are made, what sources of error there are and how we can assess their accuracy. We also discuss why polls are not forecasts, how to assess uncertainties mathematically and what Trump's election victory reveals about systematic errors.

In the third episode of our mini-series on »Artificial Intelligence (AI) and Statistics«, we look at »Shallow Learning«. In contrast to the deep and complex models of Deep Learning, these models require less data and computing power.

In the second episode of our mini-series on Artificial Intelligence (AI) and statistics, we look at Neural Networks. These models, inspired by the human brain, consist of several layers that recognize patterns and process information.

In the first episode of our mini-series on »Artificial intelligence (AI) and statistics«, we examine whether AI is just »glorified statistics«. We explain the terms AI and machine learning using concrete examples and offer an interactive task in which listeners consider how an AI can recognize a rectangle on a checkered sheet of paper.

What do Julius Caesar, Albert Einstein, Goethe, Marie Curie, Isaac Newton, Beethoven, Charlie Chaplin and Marilyn Monroe have in common? Not only were they all famous personalities, they were also left-handed. Today, August 13th, we are celebrating International Left-handed People's Day – with our new podcast episode and, of course, some curious statistics on the subject.

In the new episode of Streuspanne, we look at Benford's Law. This mathematical principle mysteriously reveals fraud in statistics. Our Streuspanne team investigates whether it puts financial investigators out of work and whether the law can disprove Bielefeld.

Today's short episode of the Streuspanne podcast is about the European Parliament elections. The special is under the bold statement: »There is no shift to the right. Unfortunately. – The AfD in the European elections since 2014«. The focus of the discussion is – following on from the last episode – the wording of the media in connection with statistics and why we need a real positive shift.

The new Streuspanne episode deals with a hot topic: the police crime statistics – PKS (Polizeiliche Kriminalstatistik acronym in German). These are published every year by the Federal Criminal Police Office and often attract a lot of media attention. The data on the origin of the perpetrators in particular is often used for striking headlines.

Our »Streuspanne Lexicon« is not a regular series, but a section to refer to small explanations without having to go on and on in a separate series. Today we continue with »F wie Freiheitsgrade« (English: D for Degrees of Freedom)

Today we continue the »Streuspanne Lexicon« with the entry »K wie kritische Werte« (English: C for Critical Values). Our lexicon is not a regular episode, but a section to refer to short explanations without having to go into detail again and again in a separate episode. The podcast team explains a term briefly and concisely in just a few minutes.

How much do you earn in rock music? Whether your answer will be »several million« or »that's breadless art« often has less to do with the question itself than with your own individual distortions and errors in thinking – the statistical noise.

Our »Streuspanne Lexicon« is not a regular series, but a section to refer to small explanations without having to go on and on in a separate series. Today we continue with »M for Mean« and »M for Median«. The Streuspanne team explains briefly why not all »means« are the same.

The new podcast episode is all about »p-values«, also known as empirical significance. The »Streuspanne« team uses statistics from the Wild West to explain what this is all about and how you can test whether a die is loaded, for example.

Today's »Streuspanne-Lexicon« entry is about what you need the binomial distribution for and what it actually is. As usual: explained briefly and clearly – in under five minutes.

In the new Streuspanne episode, representatives of the two largest schools of statistical thought face each other in a discussion. The team playfully introduces the audience to the two approaches to calculating probability: on the one hand, Jochen Fiedler slips into the role of Mr. Bayes for Bayesian statistics and on the other hand, Sascha Feth plays Mr. Frequentist, who speaks for the frequentist school of thought.

Our »Streuspanne Lexicon« is not a regular episode, but a section to refer to small explanations without having to go on and on in a separate episode. Here you will find terms such as »the estimate« or »the estimator« explained in less than five minutes.

Our German »Streuspanne Lexicon« is not a regular series, but a section to refer to small explanations without having to go on and on in a separate series. Today we continue with »C for Confidence Interval«. The Streuspanne team briefly and concisely explains the confidence interval.

In the first episode of 2024, the podcast continues with a direct paradox. What is the Simpson paradox? How can it influence meaningful data analysis? And how can you avoid falling for it?

As the year draws to a close, our Streuspanne team is getting into a fairytale mood. What exactly is the sleeping beauty paradox and what is actually paradoxical about it? The mathematical probability fairy tale is not about a kiss or a prince, but about a fairytale thought experiment – how could it be otherwise in probability theory – with a coin.

The »Streuspanne Lexicon« is a new rubric with the purpose of being able to refer to small explanations in the podcast without having to go into detail again and again in a separate episode. In this episode our Streuspanne team explains the »Monty Hall Problem« also known as the »Ziegenproblem« (Goat Problem) in German.

Probably everyone wonders how long they will actually live. You can't say exactly to the second, but statistics show very different aspects. Who lives the longest? What do the figures say?

This week, Dr Jochen Fiedler and Dr Sascha Feth, together with moderator Esther Packullat, take a closer look at the topic »end of the world«. They try to examine the end of the world with the help of the Doomsday Argument, which is based on probability theory and statistics. What conclusions do the three come to, what other considerations do they take into account and can the end of the world really be statistically predicted?

This week, our Streuspanne podcast is all about the topic of the speed limit in Germany. Questions about what is currently probably the most important topic in the transport sector are answered: Could the climate targets be achieved with the change in mobility? How, for example, can CO2 savings through a speed limit be determined empirically, i.e. with the help of statistics? And how can good arguments be made in favor of a speed limit?

In this episode, the two blogging statisticians Dr. Sascha Feth and Dr. Jochen Fiedler talk with host Esther Packullat and guest participant Nathalie Steil about audience ratings, their significance, and above all: the basics of their mathematical development.

What actually makes us scientists? We mean »real sciences« and not pseudo-sciences like parapsychology. Presumably our scientific methods, which also include empirical studies! Unfortunately, we pat ourselves on the back with this demarcation too early. Because especially the parapsychology manages again and again to hold significant studies under the nose of us »real scientists« – there is everything, from telepathy up to smart-talking rats. In this way, claims which do not fit into our (scientific) world view are supported with data. How can this be? Are these studies perhaps simply faked? Unfortunately, no. Worse, the detailed answer to this, might make us doubt whether empirical science really creates knowledge.

The Corona pandemic seems to be disappearing from our everyday lives and minds, and with it the masks from our faces. It feels like only half of the people still wear a mask in the supermarket. So today we ask ourselves the question: What good are masks that are hardly worn any more?

Today's episode is all about the topic of chance. How is the term »coincidence« defined mathematically and what conclusions arise from this for everyday situations that we interpret as coincidence? What distinguishes the world of quantum mechanics from the macrophysical one in this respect? What role do cats play, and what chances of success would an omniscient demon have in a game of dice? As always, we answer these and a few more questions for you from the perspective of statistics.

Last year, we addressed the topic of corona testing and its reliability in a series of blog posts. We received a question from a reader about this series: How much does the reliability of antigen tests increase if you test yourself more than once? Answers are provided by our blogging statisticians Sascha Feth and Jochen Fiedler.

Today's episode is all about so-called »embarrassing characteristics«. Alcohol problems, doping in sports and the partially still sensitive topic of corona vaccination – our blogging statisticians tell us what role social desirability plays and how science deals with this problem. In addition to methods from science, Dr. Sascha Feth and Dr. Jochen Fiedler also take a closer look at the unreported number of Corona infections and the start of the pandemic in Italy and Germany. As always, Esther Packullat guides the podcast with questions.