Not So Standard Deviations

Not So Standard Deviations

play.google.com/music/listen?u=0
Not So Standard Deviations: The Data Science Podc…
41 - Good to Two Decimal Places
Jun 22 • 56 min
This episode heralds the return of CEO Hilary. Also, Hilary and Roger discuss cybersecurity in the cloud, Theranos, residual variation in data analyses, and Airbnb’s Data University. Show notes: P-hacking or not?…
40 - It’s the CDs All Over Again
Jun 13 • 62 min
Hilary and Roger talk about the EARL conference, a data science tri-lemma, data science ethics, and Hilary’s CD collection. Show notes: P-hacking conversation: https://twitter.com/jakevdp/status/868564701239627776 Wrangle Conf: http://wrangleconf.com/…
39 - We Listen to Our Data Team
May 24 • 65 min
Roger and Hilary discuss the meaning of “data driven”, the value of Cloud Computing (again), and algorithmic transparency. Show notes: Artificial Intelligence Owes You an Explanation:…
38 - Banging on the Piano
May 14 • 59 min
Hilary and Roger revisit the desktop vs. cloud discussion, Hilary recaps csvconf, and Roger discusses the aesthetics of data analysis. Show Notes: Verena Haunschmid’s desktop: https://twitter.com/expectapatronum/status/857608745333006337 Roger’s desktop…
37 - It’s a Seller’s Market
Apr 27 • 70 min
Hilary and Roger talk about the EPA taking down data (or not?), whether statisticians are paternalistic, and how to evaluate machine learning consumer products. Show Notes: Mosh: the mobile shell - https://mosh.org Statisticians are paternalistic Twitter…
36 - Ripped From the Headlines
Apr 12 • 61 min
Hilary and Roger discuss Trump using an iPhone and implications for analyzing his tweets, computing setups for data science, and building simple versus complex tool chains to rely on. And the return of free advertising! Show notes: Trump is using an…
35 - Special Guest Sean Taylor
Mar 29 • 61 min
Hilary and Roger talk with Sean Taylor of Facebook about his new software package prophet, how to make tools usable and re-usable, and of course, Excel. Show notes: Sean Taylor http://seanjtaylor.com The Statistics Software Signal:…
34 - F Dist
Mar 12 • 58 min
Hilary and Roger discuss Stitch Fix’s massive Algorithms Tour, reproducibility from a physical scientist perspective, and interpretability in machine learning. Show notes F this: http://sgsa.stat.ucla.edu/f-this/ Interpretability is the new…
33 - Big Time Bubble Time
Feb 23 • 51 min
Hilary and Roger talk about the interview process for data science and data analyst jobs, opinionated analysis development, and what job interviews were like back in 1999. Show notes: Roger’s podcasting setup:…
32 - You Have to Reinvent the Wheel a Few Times
Feb 9 • 75 min
Hilary and Roger discuss training in PhD programs, estimating the variance vs. the standard deviation, the bias variance tradeoff, and explainable machine learning. Show notes: Support us through our Patreon page: https://www.patreon.com/NSSDeviations…
31 - The Word You’re Looking for is Magical
Jan 18 • 79 min
Hilary and Roger go on a podcast bender and discuss how they record the podcast, missing data methods, rstudio::conf, Uber Movement, diagnosis with prediction models, and Alexa/Siri. Whew! Show notes: Zencastr: https://zencastr.com Ecamm:…
30 - Philately and Numismatology
Jan 5 • 57 min
Hilary and Roger follow up on open data and data sharing in government. They also discuss artificial intelligence, self-driving cars, and doing your taxes in R. Show notes: Lucy D’Agostino McGowan (@LucyStats) made a great translation of Hill’s criteria…
29 -Standards are Like Toothbrushes
Dec 19, 2016 • 39 min
Roger travels to Washington, DC to talk with Daniel Morgan, Chief Data Officer for the U.S. Department of Transportation and Terah Lyons, Policy Advisor to the Chief Technology Officer of the U.S. The discussion covered open data, the process of making…
28 - Writing is a Lot Harder Than Just Talking
Dec 14, 2016 • 60 min
Hilary and Roger talk about building data science products that provide a good user experience while adhering to some kind of ground truth, whether it’s in medicine, education, news, or elsewhere. Also Gilmore Girls. Show notes: HIll’s criteria for…
27 - Special Guest Amelia McNamara
Nov 30, 2016 • 53 min
Roger sits down with Amelia McNamara, Visiting Assistant Professor of Statistical and Data Sciences at Smith College, to talk about data science, data journalism, visualization, the problems with R, and adult coloring books. Show Notes: Amelia McNamara’s…
26 - Our First Post-Election Episode
Nov 15, 2016 • 23 min
Hilary and Roger recover from last week’s U.S. election. There’s not really any data science (or much of anything else) in this episode, but if you’re just wondering how things are going, here it is. (Note: Patreon subscribers will not be charged for this…
25 - How Exactly Do You Pronounce SQL?
Oct 26, 2016 • 56 min
Hilary and Roger go through the overflowing mailbag to respond to listener questions. Topics include causal inference in trend modeling, regression model selection, using SQL, and data science certification. Show notes: Professor Kobre’s Lightscoop…
24 - 50 Minutes of Blathering
Oct 14, 2016 • 51 min
Hilary and Roger meet at a Jimmy John’s to talk data science, like you do. Topics covered include RStudio Conf, polling, millennials, Karl Broman, and more! Show notes: rstudio::conf https://www.rstudio.com/conference/ We Gave Four Good Pollsters the Same…
23 - Special Guest Walt Hickey
Sep 28, 2016 • 44 min
Hilary and Roger invite Walt Hickey of FiveThirtyEight.com on to the show to talk about polling, movies, and data analysis reproducibility (of course!). Show notes: FiveThirtyEight’s polling methodology:…
22 - #1 Side Project
Sep 14, 2016 • 54 min
Hilary and Roger celebrate their one year anniversary doing the podcast together by discussing whether there are cities that are good for data scientists, reproducible research, and professionalizing data science. Show Notes: Roger’s reproducible research…
21 - This Might Be the Future!
Aug 25, 2016 • 49 min
Hilary and Roger are apart again and this time they’re talking about political polling. Also, they discuss Trump’s tweets, and the fact that Hilary owns a bowling ball. Show notes: FiveThirtyEight election dashboard:…
20 - Face Time
Aug 3, 2016 • 63 min
Hilary and Roger are in Chicago (together!) for the annual Joint Statistical Meetings, the largest gathering of statisticians in the world. They discuss the session that Hilary organized, byzantine ASA bureaucracy, data scientists interacting with the…
19 - Divide by n-1, or n-2, or Whatever
Jul 18, 2016 • 54 min
Hilary and Roger talk about statistical software in fMRI analyses, the differences between software testing differences in proportions (a must listen!), and a preview of JSM 2016. Show Notes: fMRI bugs could upend years of research:…
18 - Back on Planet Earth
Jul 3, 2016 • 65 min
With Hilary fresh from Use R! 2016, Roger and Hilary discuss some of the highlights from the conference. Also, some followup about a previous Free Advertising and the NSSD drinking game. Show Notes: Theranos movie with Jennifer Lawrence and Adam McKay:…
17 - Diurnal High Variance
Jun 8, 2016 • 51 min
Hilary and Roger talk about Amazon Echo and Alexa as AI as a service, the COMPAS algorithm, criminal justice forecasts, and whether algorithms can introduce or remove bias (or both). Show Notes: In Two Moves, AlphaGo and Lee Sedol Redefined the Future…
16 - The Silicon Valley Episode
May 20, 2016 • 48 min
Roger and Hilary are back, with Hilary broadcasting from the west coast. Hilary and Roger discuss the possibility of scaling data analysis and how that may or may not work for companies like Palantir. Also, the latest on Theranos and the release of data…
15 - Spinning Up Logistics
Apr 28, 2016 • 45 min
This is Hilary’s and Roger’s last New York-Baltimore episode! In future episodes, Hilary will be broadcasting from California. In this episode Hilary and Roger discuss collaboration tools and workflow management for data science projects. To date, Roger…
14 - Sticker Unboxing
Apr 25, 2016 • 3 min
Hilary and Roger meet up face-to-face for the first time since the podcast started! This a short episode to document the unboxing of the official Not So Standard Deviations hex stickers that are available to our Patreon patrons who have signed up at the…
13 - It’s Good that Someone is Thinking About Us
Apr 7, 2016 • 62 min
Hilary and Roger talk about the difficulties of separating data analysis from its context, and Feather, a new file format for storing tabular data. Also, Hilary and Roger respond to some listener questions and Hilary announces her new job. If you have…
12 - The New Bayesian vs. Frequentist
Mar 25, 2016 • 49 min
In this episode, Hilary and Roger discuss the new direction for the journal Biostatistics, the recent fracas over ggplot2 and base graphics in R, and whether collecting more data is always better than collecting less (fewer?) data. Also, Hilary and Roger…
11 - Start and Stop
Mar 14, 2016 • 61 min
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…
10 - It’s All Counterexamples
Feb 29, 2016 • 57 min
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…
9 - Spreadsheet Drama
Feb 9, 2016 • 60 min
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…
8 - Snow Day
Jan 24, 2016 • 64 min
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,…
7 - Statistical Royalty
Jan 8, 2016 • 59 min
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…
6 - Google is the New Fisher
Dec 16, 2015 • 46 min
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…
5 - IRL Roger is Totally With It
Dec 1, 2015 • 54 min
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…
4 - A Gajillion Time Series
Nov 7, 2015 • 51 min
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…
3 - Gilmore Girls
Oct 24, 2015 • 50 min
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…
2 - We Got it Under 40 Minutes
Oct 1, 2015 • 39 min
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…
1 - RCatLadies Origin Story
Sep 16, 2015 • 47 min
Welcome to Episode 1! In this episode, Roger and Hilary talk about the origins of @RCatLadies, evidence-based data analysis, Roger’s new book, The Art of Data Science, and whole Python vs. R thing.
Naming The Podcast
Sep 16, 2015 • 1 min
Roger and Hilary figure out what to name the podcast. Also, Roger is really bad at making web sites.