possible new chapter: "Starting the Analysis"

[wibeasley]

@oneilsh as I said in a review (of a chapter that I'm suggesting "Introducing the Enclave Tools"), I think the book needs a chapter that starts the reader with code for graphs & models. I think it could resemble @jerrodanzalone's and my day of the 2022 short course. And probably start with a section that includes some sql code.

[wibeasley]

@oneilsh recapping yesterday's ET-DT meeting, people agreed the code goes into another chapter and suggested it go in the "Start-to-Finish Examples" Special Topics Chapter. In our book outline, all it says is "High-level overviews of research projects along with individual unique challenges and paths (but more detailed and realistic than ch 2)"

What do you think of this content as a starter? If you like, we can introduce at next week's meeting.

Getting Starting with an N3C Analysis

Workbooks

1. manipulation-1: reads from the (OMOP & LL) source tables to produce a dataset that's ready for analysis

2. descriptives-1: Table 1 for the manuscript (headed for a DRR)

3. descriptives-2: basic demographics & descriptives

4. propensity-1: propensity matching to balance participants, including diagnostic graphs

5. analysis-1: regression models & graphs in R

6. analysis-2: regression models & graphs in Python

Details

For each workbook:

1. Describe its purpose

2. Provide starter code that can be pasted directly into transform. Including the Global Code panel.

3. Detail the input object for each transform (eg, Spark DataFrame, R Transform Object, R data.frame, Pandas data frame)

4. Recommended Enclave environment (eg, driver count, executor count, RAM, Spark/Python/R packages)