From Code to Concepts: Introduction to Data Science: Instructor Notes

(Hugo’s notes)

My idea is to have two objectives going in parallel:

• teach R syntax
• teach EDA concepts and principles

In that sense, R is being taught as a tool for data analysis. I would not aim to teach programming, thus the choice of focusing on tidyverse (although this choice is somewhat debatable, we could just as easily have chosen data.table or even base R).

I’m hoping that eventually the materials are rooted in a higher-level conceptual framework, inspired by:

• The Art of Data Science, by Roger D. Peng and Elizabeth Matsui
• Elements and Principles for Characterizing Variation between Data Analyses, by Stephanie C. Hicks and Roger D. Peng

Session 1:

This session is about getting started with working with data in R + RStudio:

• how to setup and organise an analysis project
• using functions and access their documentation
• import tabular data into R
• understand and distinguish between the fundamental data structures and data types in R
• discuss the importance of and apply basic quality control checks of data

Session 2:

This session is all about the visual display of information:

• recognise the necessary elements to build a plot using the ggplot2 package
• use ggplot2 to produce several kinds of visualisations (for continuous and/or discrete data)
• distinguish which types of visualisation are adequate for different types of data and questions
• discuss the importance of scales when analysing and/or visualising data

Session 3:

This session is about starting to manipulate tables and cleaning up data:

• use of the dplyr package to manipulate tabular data (add or modify variables, select and rename columns)
• understand and use “pipes” as a way to build a chain of operations on data
• discuss some common issues with data cleaning and use functions from the stringr package to help solve them
• discuss how to deal with missing data

Session 4:

This session is about subseting data based on specific questions or conditions:

• introduction to logical comparisons and logical data type
• remember and distinguish between different types of logical operators
• choose the right conditional operations to obtain specific observations from data using the filter() function
• apply conditional operations to clean data mistakes and/or highlight elements of interest in graphs using the ifelse() function
• how to save processed data into a file

Session 5:

This session is about working on sub-groups of data:

• recognise when to use grouped operations in data analysis
• differentiate between grouped summaries and other types of grouped operations
• apply grouped summaries using the group_by() + summarise() functions
• apply other grouped operations such as: group_by() + filter() and group_by() + mutate()
• recognise the importance of the ungroup() function

Session 6:

This session is about combining different datasets together and how to modify the “shape” of our data:

• understand and apply functions that reshape data from “long” to “wide” formats
• appreciate when each format is useful and which one to choose for the intended tasks
• join different datasets together
• distinguish between different ways to join data, and understand which one to choose depending on the desired outcome