From a presentation by former Institute of Mathematical Statistics president Bin Yu:

More refined Venn Diagram by Drew Conway:

How does one answer a scientific question with data? Leek & Peng in Nature (2015) illustrate the data pipeline

1. Data collection
2. Data cleaning
3. Exploratory data analysis
4. Statistical modelling
5. Inference (form conclusions and communicate them)

Intro stats classes focus a bit on 3, 4, and 5 partially. This class will try to cover all aspects of the pipeline.

• Follow the complete statistical analysis cycle
• Real data: more interesting, not clean, violating statistical assumptions
• Data visualization: not just infographics, but as an analytical tool
• Use computational tools: R coding, R packages, scraping data from the web, building web apps
• Apply statistical methodologies: regression, correlated data, spatial statistics, text mining, etc.

• Teach in a language agnostic way. Transferable and generalizable ideas, not a class on R.
• Not learn a programming language, but learn how to learn a programming language
  • By doing
  • Google is your best friend
  • By suffering, like learning any other language

• Don't thrash. Really!
• Don’t be stuck for more than 20 minutes. This takes self-awareness and mindfulness.
• Seek expert advice; You’ll be on the other side soon enough
  • Your peers.
  • Me. Note I do prefer speaking in person than email.

For the first part of the class, we emphasize the two most important tools:

• Tools for manipulating/wrangling your data: dplyr package
• Tools for visualizing data: ggplot2 package, which is an implementation of the Grammar of Graphics

The beauty of these two R packages is there deep philosophy underlying their implementations.