The previous two units (Confirmatory Data Analysis and Inferential Statistics and Predictive Analytics) have focused on teaching both theory and practice in ideal data scenarios, so that our more academic quests can be divorced from outside concerns about the veracity or format of the data. To this end, we deliberately stayed away from datasets not already built into R or available from add-on packages. But very few people I know get by in their careers by using R and not importing any data from sources outside of R packages. Well, we very briefly touched upon how to load data into R (the read.* commands) in the very first chapter of this book, did we not? So we should be all set, right?
Here's the rub: I know a few people who can get by using simple CSVs and tab-delimited text locally with the primary read.* commands and can get by not using outside sources of data at all! The unfortunate fact is that many introductory analytics texts largely disregard this reality...