In my course on Statistical Communication and Graphics, each class had a special guest star who would answer questions on his or her area of expertise. These were not “guest lectures”—there were specific things I wanted the students to learn in this course, it wasn’t the kind of seminar where they just kick back each week and listen—rather, they were discussions, typically around 20 minutes long, facilitated by the outside expert.
One thing that struck me about these discussions was how fun they were, and how various interesting and unexpected things came up in our conversations.
And that made me think—Hey, we should do a podcast! I can be the host and have conversations with these guests, one at a time, and then release these as (free) 15-minute podcasts. How awesome! The only challenge is to keep them lively. Without a roomful of students, a recorded conversation between two people could get stilted.
Also we need a title for the series. “Statistics Podcast” is pretty boring. “Statcast”? The topics we’ve had so far have been focused on statistical communication, but once we go through that, we could cover other statistical areas as well.
And then there’s the technical details: how to actually set up a podcast, also maybe it needs to be edited a bit?
So here’s what I’m needing from you:
– A title for the podcast series.
– Advice on production and distribution.
Our starting lineup
Here are some of the visitors we’ve had in our course so far. I’d plan to start with them, since I’ve already had good conversations with them.
I list the topic corresponding to each visitor, but the actual conversations ranged widely.
-
Thomas Basbøll, Writing Consultant, Copenhagen Business School (topic: Telling a story)
Howard Wainer, Distinguished Research Scientist, National Board of Medical Examiners (topic: Principles of statistical graphics, although the actual discussion ended up being all about educational testing, because that’s what the students’ questions were about)
Deborah Nolan, Professor of Statistics, University of California (topic: Student activities and projects)
Jessica Watkins, Department of Education, Tufts University (topic: Facilitating class participation)
Justin Phillips, Professor of Political Science, Columbia University (topic: Classroom teaching)
Beth Chance, Professor of Statistics, California Polytechnic State University (topic: Preparing and evaluating a class)
Amanda Cox, Graphics Editor, New York Times (topic: Graphing data: what to do)
Jessica Hullman, Assistant Professor of Information Visualization, University of Washington (topic: Graphing data: what works)
Kaiser Fung, Senior Data Advisor, Vimeo (topic: Statistical reporting)
Elke Weber, Professor of Psychology and Management, Columbia University (topic: Communicating variation and uncertainty)
Eric Johnson, Professor of Psychology and Management, Columbia University (topic: Communicating variation and uncertainty)
Cynthia Rudin, Associate Professor of Statistics, MIT (topic: Understanding fitted models)
Kenny Shirley, Principal Inventive Scientist, Statistics Research Department, AT&T Laboratories (topic: Understanding fitted models)
Tom Wood, Assistant Professor of Political Science, Ohio State University (topic: Displaying fitted models)
Elizabeth Tipton, Assistant Professor of Applied Statistics, Teachers College, Columbia University (topic: Displaying fitted models)
Brad Paley, Principal, Digital Image Design Incorporated (topic: Giving a presentation)
Jared Lander, statistical consultant and author of R for Everyone (topic: Teaching in a non-academic environment)
Jonah Gabry, Researcher, Department of Statistics, Department of Political Science, and Population Research Center, Columbia University (topic: Dynamic graphics)
Martin Wattenberg, Data Visualization, Google (topic: Dynamic graphics)
Hadley Wickham, Chief Scientist, RStudio (topic: Dynamic graphics)
David Rindskopf, Professor of Educational Psychology, City University of New York (topic: Consulting)
Shira Mitchell, Postdoctoral Researcher, Earth Institute, Columbia University (topic: Collaboration)
Katherine Button, Lecturer, Department of Psychology, University of Bath (topic: Communication and its impact on science)
Jenny Davidson, Professor of English, Columbia University (topic: Writing for a technical audience)
Rachel Schutt, Senior Vice President of Data Science, News Corporation (topic: Communication with a non-technical audience)
Leslie McCall, Professor of Sociology, Northwestern University (topic: Social research and policy)
Yair Ghitza, Senior Scientist, Catalist (topic: Data processing)
Bob Carpenter, Research Scientist, Department of Statistics, Columbia University (topic: Programming)
P.S. Lots of suggested titles in comments. My favorite title so far: Learning from Numbers.
P.P.S. I asked Sharad if he could come up with any names for the podcast and he sent me these:
White Noise
In the Noise
The Signal
Random Samples
I’ll have to nix the first suggestion as it’s a bit too accurate a description of the ethnic composition of myself and our guest stars. The third suggestion is pretty good but it’s almost a bit too slick. After all, we’re not the signal, we’re just a signal. I’m still leaning toward Learning from Numbers.
Looks excellent! Can’t wait to listen.
Statchat?
Too similar to the name of the popular blog ‘stats chat’.
Sounds great! I’ll offer that since it’s a conversational approach how about “Two statisticians walk into a bar” Or perhaps something punnier “Two statisticians walk into an X-Bar”? Of course, that’s a lot better (maybe? It’s still a bad pun) if you could actually use a proper math symbol.
Some name ideas:
– Sufficient Statistics
– Standard Deviations
Ending with -cast, in my own order of preference:
– Heteroscedasticast
– Stochasticast
– Statisticast
FYI, there’s already a Not So Standard Deviations podcast with Hilary Parker and Roger Peng.
How about Stats in a Pod?
Nice trans-disciplinary cast (a comment, not a suggested title.)
Title: Variation and Uncertainty Uncast Cast.
A title so ugly – it will be safe from kidnappers ;-)
For advice on production see if this stuff helps: “How to Start a Podcast” by the guy making The British History Podcast.
http://www.thebritishhistorypodcast.com/how-to-start-a-podcast
Stat-Squawk
Stats-Quack
S(t)asquatch
“Gelman on the Street”
Not replicable
Garden of forking casts
This sounds awesome.
Your Columbia colleague Vincent Racaniello (whose blog has been a hub for posts on the PACE trial) is a prolific podcaster: http://www.microbe.tv/about/
He and his cohosts were just complaining about null hypothesis significance testing on the last episode of his flagship podcast, ‘this week in virology’ (#374).
Professor Racaniello frequently emphasizes the importance of good audio. Since you’ll have multiple people in the room you’ll probably want someone to mix the audio.
Various people offer free advice and consulting on getting started, e.g. http://thepodcastersstudio.com/consulting/
I should have written “…free advice and *paid* consulting”.
Statcast?? No, no, no. You need something snappy, like The Statistical Modeling, Causal Inference, and Social Science Podcast.
If you want an example of a successful podcast with excellent, highly intellectual, one-on-one discussion and excellent production values, check out…
http://www.econtalk.org/
I don’t know the folks there, but I suspect they would be willing to provide good advice on the technical aspects.
I like “StatChat” as a name…good idea from above.
I was clicking to comment this same thing. This podcast is very popular with econ students and econ-interested laymen. It’s been going for ten years and I listen religiously.
I totally agree with the recommendation of EconTalk. I wish every interviewer did such a fine job. I’ve long since lost track of the number of books that I’ve read after discussions with their authors on EconTalk.
My suggestion is don’t use stats or any portmanteaus in the title. Sounds cheap.
Though mine isn’t much better:
-Within the Margins
How about ‘Standard Deviations’? Or just ‘Deviations’? Something playing off the fact that most conversations will deviate from their originally intended focus.
Ooooh, this is so exciting. I’ve been looking for that type of podcast for a while, I’ll definitely listen.
Hopefully there are some good title ideas – it’s hard to avoid a boring and/or punny title.
Some naming ideas:
The Null
Sampling From My Posterior
BayesWatch
Evidently We’re Talking
“Sampling From My Posterior” got the biggest laugh from me. Too bad it doesn’t match Andrew’s personality.
OK, then how about just “Samples.”
Though in this day and age, people should want to know where samples come from.
Call it “Stat Talk” and model the conversation on “Car Talk” … since it’s a podcast you can edit the calls into the format, and give the commentators a little time to think before they provide commentary or advice. I also recommend a weekly “Puzzler” feature. With best regards – Reid Hastie (University of Chicago)
I like dsparks idea of “Sufficient Statistics” or perhaps even “Efficient Statistics?” Here are some other potential titles I like: “Latent Variables,” “Regression towards the Mean,” “Central Limits,” “Robust Estimators,” “Random Coefficients,” and “Normal Distributions.” I appreciate your blog and look forward to your podcast.
Boring Statisticians?
If you like statistics, what could be more fun than telling people, “I’m so excited–there’s a new Boring Statisticians podcast!”? If the podcasts are any good, the word of mouth campaign writes itself: “It’s called Boring Statisticians, but it’s actually really interesting!”
If you don’t like statistics, that’s what you’ll be thinking anyway–no matter what it’s called.
Cool.
Statistical practice from practicing statisticians
Stimulating Statisticians
Statistical practice from purposeful statisticians
Stimulating Purposeful Statisticians
How about ‘Deviations’ or ‘Standard Deviations’ because the conversations will deviate from their original focus?
I’ve used https://auphonic.com/ before for automatic post-production. Pretty easily.
Residual Deviants, surely?
Random talks?
How about…
estimable podcast
I see your p value
Alt hypothesis
Two for t-test
Maybe pushing it a bit, but perhaps
“A man, A stan, A plan, A podcast!”
or something to that effect.
I think you should host a web poll where your readers vote on which questions they want the next guest to answer. And maybe get a grad student to voice these questions if you want to keep the audio more “lively”.
Plus, you could always run an experiment on us to evaluate if we vote differently when we can or cannot see the distribution of previous votes in the poll.
Title suggestion: The Causeway
Which comment is suggesting “Learning from Numbers”? I don’t see it.
“Two Statisticians Walked Into a Bar”
Since you may have participants from far away, I could offer the suggestion of setting up an Asterisk server and having people call in over the internet via SIP to a conference channel on the Asterisk server, maybe using the Opus codec in 16 bit format, and then record the conference call to disk in say signed linear 16 bit. Post editing in Audacity. Encode final thing to Ogg vorbis.
Just as a suggestion of something to look at.
It’s a different story if you’ve got people in the same room.
Suggested title: Static.
Love this idea and podcasts generally. Based on listening to a lot of podcasts, I’d say the key thing you should do is get a good editor. There’s too many podcasts that just hit record and upload the conversation with no editing and it can make them drag a bit.
Suggested title: High Probability Intervals
Great idea! A few suggestions:
* Try to speak clearly, and to be expressive. You can win hearts with appropriate laughter :D
* If you don’t have good ears yourself, get someone involved who can give you the required reality-check. I recently came across a video tutorial series that was recorded in a really echoey room. Oops!
* If you end up doing any voice-overs, or have to read anything (e.g. discussing programming, results, graphics, etc), be extra-careful to enunciate very clearly and avoid monotone. Some people tend to mumble.
* Do a practice session or two with friends, and get critical feedback, before going live with your A-listers.
I love the Philosophy Bites Podcasts ( http://philosophybites.com/ ). They are short (max. 15 mins), crisp and to the point. The interviewers do their homework so they ask precise, relevant questions. The pace is fast. Not much of ummm ahhh -ing.
I am often amazed how they do this. Perhaps it employs smart post-interview editing?
At the other extreme there’s Russ Robert’s EconTalk. Not that I don’t like it but it has a very leisurely format. Most episodes exceed 60 minutes. There’s lots of rambling and tangential off shoots of discussion etc. with not much of editing.
You could get some ideas from Ryan Adams’ Talking Machines podcast (http://www.thetalkingmachines.com). In their case, there’s a full-time employee who works strictly on managing recordings, editing, sound tracks, and so on (they’re funded by a Kickstarter project). Recording in-person discussion at key events such as JSM, and then editing and releasing them over periods of time, is also a great technique to take advantage of.
Gelmanifesto
Gelmania
+2
Rahul, I was just about to suggest Gelmania. We should meet someday ;)
Uncanny! And, yes, would love to meet someday! :)
Signal from the Noise
Some less boring ideas ;)
Gelman’s Pod-Pourri
Statatouille
Gelman’s marvellous numbers
Gelman’s world of uncertainty
Not so sure
Ma’am, you have a flat prior
Statfinders
The Bayes Players
Might come up with more later.
Andrew:
Did you consider inviting Donald Green (of the Bruno LaCour fame) to your class? That could be an interesting session. He’s right next door at Columbia, right?
OK, another silly one: Aching Desire (echoing Tukey).
Can’t wait to see this in action.
I’m kind of surprised that nobody’s suggested “The Scatterplot Thickens” especially since the main topic for the first guest on the list involves telling a story.
And you could append it with a more explanatory subtitle if you wanted to.
I asked a friend that has been producing a decently popular podcast (trcpodcast.com) for the past few years, and here are his tips:
– Levelator (conversationsnetwork.org/levelator) is your friend. Use it.
– 64Kb is probably enough for any spoken word podcast. Don’t make the file huge.
– If you’re serious, get a host. Don’t try to feed it off a website. Libsyn and podbean are the big ones.
– Set your RSS feed through something like feed burner rather than directly from a host. That way if you change things up, in terms of hosting, it’s as simple as repointing.
Also, if you encounter the scenario of interviewing someone remotely over skype (or similar software):
– Both interviewer and interviewee should record their audio using an app like audacity. Interviewee needs to send resulting wav to interviewer, likely via a media drop site (or a direct connection via justbeamit.com) as such recordings will likely be too large to email.
– Both interviewer and interviewee should wear headphones so that the audio they receive from their counterpart does not contaminate the recording of their own audio
– Interviewer should record the audio received via skype just in case something goes awry with the recording being made on the interviewee’s end.
Does Skype have an option to record audio received via Skype? Or do you need another tool / trick to do that?
No, skype doesn’t provide any built-in method of recording audio, but if you’re on a mac, it’s fairly easy to do using soundflower, a free/open source audio re-routing app (here’s an example tutorial: http://www.addictivetips.com/mac-os/record-audio-from-skype-calls-high-quality-mac/). I haven’t tried it, but for windows & linux the app “skype call recorder” (http://atdot.ch/scr/) seems to do the same.
How about The Statisphere? Check in the mail? :)
Lots of expert advice at http://5by5.tv/podcastmethod
Andrew, what’s the intended audience for this podcast?
I only listen to two podcast-type things these days, Slate Money and SciFri. In Slate Money it’s just three people, no audience, and it doesn’t lack in liveliness. Maybe you need a third person on this show. SciFri is just one person interviewing an expert, but people can call and talk directly to the interviewer, which generally works well; it certainly leads to more engagement from the audience.
Maybe you should make it the Daily Show of statistics; that would be a huge hit, but you would have to develop the talents of Jon Stewart.
I’ve had a great deal of fun doing a food safety themed podcast. Email me if you’d like tips on production and distribution.
If your issue with “The Signal” as a title is that it seems to singular, just drop the “the”. I think “Signal” would be a great name.
Excited to hear it!
I’ve enjoyed Talking Machines, and think it could be a great template for you. It’s probably longer than you want, and too focused on the guest’s bios rather than a specific substantive topic, but they do a good job.
Statpocalypse!