Slides and links

Slides available on slideshare

Selected links from the slides

• The Big Data Brain Drain: Why Science is in Trouble
• The data science Venn diagram
• Slides from Hadley Wickham’s talk in the Simply Statistics Unconference
• 2014-08-17 NYT article For Big-Data Scientists:, ‘Janitor Work’ Is Key Hurdle to Insights by Steve Lohr
• Data carpentry (response to NYT article re: janitor work)
• The Forgotten Job of a Data Scientist: Editing
• 10 things statistics taught us about big data analysis
• The 80/20 rule of statistical methods development
• Don’t use Hadoop - your data isn’t that big
• Most data isn’t “big,” and businesses are wasting money pretending it is
• Big data is like teenage sex
• Edward Tufte book chapter 5 on the Challenger Disaster
• “Let’s Practice What We Preach: Turning Tables into Graphs”. The American Statistician, Volume 56, Number 2, 1 May 2002, pp. 121-130(10) by Gelman A, Pasarica C, Dodhia. via JSTOR
• Data Sharing and Management Snafu in 3 Short Acts, a video
• The R Project
• RStudio IDE
• R Markdown
• knitr
• “FINAL”.doc from PHDCOMICS
• Git
• GitHub
• The Best Map Ever Made of America’s Racial Segregation
  • The Racial Dot Map
  • unorthodox123/RacialDotMap on GitHub
• Foodborne Chicago finds dodgy restaurants with tweets, and R
  • corynissen/foodborne_classifier on GitHub
• Roger Peng
• GNU Make for Reproducible Data Analysis
• stackoverflow

Register a GitHub account

Do this first!

• Register for a free individual GitHub account and request the student discount. This should give you some free private repositories for a couple of years. Do this ASAP! I think a human vets these requests, so it could take a couple of days.
  • As with Twitter (see below), think before you set up your public profile. I suggest you use a real name and begin to build a scholarly identity in social media and coding, but it’s a decision you should make with intention.
• We will elicit your GitHub username soon so we can add you to the STAT 545 GitHub Organization.

Git(Hub) and Git client

• Install Git
• Optional but recommended: Install a Git client

R and RStudio

Follow these instructions. Stick with it to the bitter end, where you try to get Git and RStudio talking to each other.

If you have a pre-existing installation of R and/or RStudio, we highly recommend that you reinstall and upgrade to the most recent version. It is very easy and RStudio, specifically, is changing rapidly and positively (written 2014-09). If you upgrade R, you will need to also update any packages you have installed.

Twitter

https://twitter.com

I will use the @STAT545 Twitter account to make micro-announcements, share interesting links, and facilitate a conversation amongst ourselves in public.

In class, we’ll talk about Twitter, its scholarly use, and privacy. Some relevant links:

• Disposable Twitter Accounts for Classroom Use from ProfHacker on the Chronicle of Higher Education
  • Good quote: “I strongly encourage you to create a disposable account if for any reason you prefer not to share your personal account for classroom activities.”
• Resources for exploring digital identity, privacy and authenticity from Catherine Cronin
• Some advice to young graduate students from C. Titus Brown
• How to be the grad student your advisor brags about from the blog of Impactstory
• 10 Commandments of Twitter for Academics from Do Your Job Better on the Chronicle of Higher Education
• Using Twitter in university research, teaching and impact activities: A guide for academics and researchers (PDF)

If you choose not to get a Twitter account, you will want to visit the @STAT545 profile page regularly to see what’s up.

If you set up (or already have a Twitter account), you should follow @STAT545. We will elicit your Twitter handle soon so we can follow you back.

You could also follow me as an individual, an account I use very differently (and sparingly), at @JennyBryan. Ditto for the course TAs. You can learn some good people to follow re: data science and R by looking at who we follow and who we have Twitter conversations with.

RPubs

http://rpubs.com

This is a free hosting service provided by RStudio. It is the fastest way to get a report up on the web, based on an R script or an R Markdown file. But I think its days are numbered. Still, we may use this a little before we move on to other techniques. Registering for a free account is a “nice to do”, not a “must do”.