MATH 241 Case Studies in Statistical Analysis

When I came up with the name for this class “MATH 241 Case Studies in Statistical Analysis” last year, I was still skeptical of the term “Data Science” as I felt it too buzzwordy.

I've since changed my stance. This class should be in fact “MATH 241 Introductory Data Science”

From the introduction to the OpenIntro Statistics textbook from MATH 141:

1. Formulate a scientific question
2. Collect data
3. Clean and manipulate data
4. Analyze data
5. Form conclusions & communicate them

In MATH 141, we tended to focus more on 4 and 5.

• 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, machine learning, etc.

For the first few lectures, we will work on developing our data toolbox. These tools are absolutely necessary before we can pursue any kind of meaningful analysis. Specifically

• Tools for visualizing data: the ggplot2 package
• Tools for manipulating/transforming your data: the dplyr package

The beauty of these two R packages is there is a deep philosophy underlying how to use them.

1. Tab: complete command and variable names
2. (From console) Up: scroll thru previous commands
3. (From editor) CMD+Enter or Control+Enter: execute in console either current line or highlighted code

To see them all, press alt+shift+k.