We’ve started a Patreon page! Now you can support the podcast directly by going to http://patreon.com/NSSDeviations and making a pledge. This will help Hilary and me build the podcast, add new features, and get some better equipment. This is an all craft episode of Not So Standard Deviations, where Hilary and Roger discuss starting and ending a data analysis. What do you do at the very beginning of an analysis? Hilary and Roger talk about some of the things that seem to come up all the time. Also up for discussion is the American Statistical Association's statement on p values, famous statisticians on Twitter, and evil data scientists on TV. Plus two new things for free advertising. Show notes: NSSD Patreon Page (http://patreon.com/NSSDeviations) Jan de Leeuw (https://twitter.com/deleeuw_jan) Douglas Bates (https://twitter.com/BatesDmbates) Sports Night (https://en.wikipedia.org/wiki/Sports_Night) ASA’s statement on p values (http://goo.gl/JFz7ic) Basic and Applied Psychology Editorial banning p values (http://goo.gl/O8kL60) J. Kenji Alt’s Vegan Experience (http://www.seriouseats.com/vegan-experience) #fieldworkfail (http://fieldworkfail.com/)

Hilary and Roger talk about the motivation behind the explainr package and the usefulness of automated reporting and interpretation of statistical tests. They also struggle to come up with a quick and easy way to explain why statistics is useful when it sometimes doesn't produce any different results. Show notes: The explainr package: https://github.com/hilaryparker/explainr Google's CausalImpact package: https://google.github.io/CausalImpact/CausalImpact.html Software is Eating the World: http://www.wsj.com/articles/SB10001424053111903480904576512250915629460 Many Rules of Statistics are Wrong: http://allendowney.blogspot.com/2015/12/many-rules-of-statistics-are-wrong.html Jeff Leek is mentioned at 5:31

Special guest Jenny Bryan (@jennybryan) joins us from the University of British Columbia! Jenny, Hilary, and Roger talk about spreadsheets and why some people love them and some people despise them. They also discuss blogging as part of scientific discourse. Show notes: Jenny's Stat 545: http://stat545-ubc.github.io/ Coding is not the new literacy: http://goo.gl/VvFyXz Goldman Sachs spreadsheet error: https://goo.gl/mC0Qz9 Jingmai O'Connor episode: https://goo.gl/hNloVr De-weaponizing reproducibility: http://goo.gl/IYDwn1 Vintage Space: https://goo.gl/n02EGP Tabby Cats: https://goo.gl/H3YgV6

Hilary and Roger were snowed in over the weekend, so they recorded Episode 8 of Not So Standard Deviations. In this episode, Hilary and Roger talk about how to get your foot in the door with data science, the New England Journal's view on data sharing, Google's "Cohort Analysis", and the possibility of predicting a movie's box office returns based on the movie's script. Show notes: Remembrances of Peter Hall (http://goo.gl/eUU2AK) Research Parasites (NEJM editorial by Dan Longo and Jeffrey Drazen) (http://goo.gl/HbMu87) Amazon review/data analysis of Fifty Shades of Grey (http://goo.gl/83DvvO) Time-lapse cats (https://youtu.be/55psWVYSbrI) Pocket (https://getpocket.com/)

Hilary and Roger ring in the new year with a discussion of Theranos and the pitfalls of diagnostic testing, Spotify's Discover Weekly playlist generation algorithm (and the need for human product managers), and of course, a little Star Wars. Also, Hilary and Roger each give some "free advertising" to something interesting that they think other people should know about. Show Notes: Gosset Icterometer (http://goo.gl/JDk6ni) Entertainment medicine Spotify's Discover Weekly (http://goo.gl/enzFeR) Variance Explained (http://varianceexplained.org) What3Words (http://what3words.com)

In this episode Roger and Hilary discuss analyses that seem easy but are actually hard, like power/sample size calculations and A/B testing. Notes Roger's books on Leanpub (https://goo.gl/X0TFt9) KPIs (https://goo.gl/VO0ckP) Reply All (http://replyall.soy), a great podcast Use R! 2016 conference (http://user2016.org) where Don Knuth is an invited speaker! Liz Stuart's directory of propensity score software (http://goo.gl/wUcTBT) A/B testing (https://goo.gl/CibhJ0) iid (https://goo.gl/qMyksb) R 3.2.3 release notes (https://goo.gl/qHVzWQ) pqR(http://www.pqr-project.org/) John Myles White's tweet (https://goo.gl/pFOVkx)

Roger and Hilary open the mailbag for the first time and go over some feedback they've gotten on previous episodes. The rest of the time is spent discussing the importance of reproducibility in data analysis, both in an academic research setting and an industry setting. Notes Hilary's talk on reproducible analysis in production: https://youtu.be/7B3n-5atLxM Duke Scandal Starter Set: http://goo.gl/rEO5QD Keith Baggerly's talk on Duke Scandal: https://youtu.be/7gYIs7uYbMo The Web of Trust: https://goo.gl/RtpBZa testdat R package: https://goo.gl/MlM0gu

Hilary and Roger take on the statistical challenges in DevOps and the inherent difficulties of detecting rare anomalies in real time. But first Hilary schools Roger on what DevOps is. Also Ben and Jerry's. And the t-test. Notes: Nobody Loves Graphite Anymore (https://goo.gl/259VKI) A response: http://goo.gl/zB7wM9 Why Gosset is awesome: https://goo.gl/7PgLKY Gosset's grandson: http://goo.gl/fhaqHx

At the suggestion of Roger's brother, this episode is all about Hilary and Roger. Well, really it's about their jobs. Hilary and Roger talk about the life of the data scientist in academia and in the tech industry and how it doesn't seem to be all that different....

In this episode, Hilary and Roger talk about user testing for statistical methods, navigating the Hadleyverse, the crucial significance of rename(), and the secret reason for creating the podcast (hint: it rhymes with "bee"). Also, Roger erroneously claims that Bill Cleveland is way older than he actually is. Sorry Bill. In other news, we are finally on iTunes so you can subscribe from there directly if you want (just search for "Not So Standard Deviations"). Notes: Bill Cleveland's paper in Science, on graphical perception, published in 1985 (http://goo.gl/xtkSJy) TomFest (https://goo.gl/Sbtz0o)