General review points

[PBBlomquist]

No description provided.

[PBBlomquist]

Data needed:

• Initial clean linelist
• Inject 1
• Inject 2 for Quarto
• COVID19 data

positive RDT results get tested (that's why none of the PCR results get lost), we filter the data to create df_cases after the first case_def (in module 5). We then filter the data again after producing df_joined to remove PCR negative.

in module 5 show epicurve for all areas including unclear? just for more panels

Case study

• intror10-covid

Word doc sitrep

Folder

• intror_course
  • install_packages.R (add appliedepidata?)
  • cholera
    • scripts
      • backup
        • 3
        • 4
        • 5
        • 6
        • 7
        • 8 - XXX
        • 9 - quarto - XXX
        • word doc example sitrep
    • data
      • backup (note - rename all to cholera)
        • Week 1 data (1 clean file)
        • All inject 1 data (5 files)
        • inject 2 data (5 files)
  • covid19
    • backup
    • data

Slide standardisation - CONTENT

• Structure:
  1) Title slide
  2) scenario/learning objectives depending on best fit. Learning objectives should address what has been done in part B
  3) Add 'functions to learn today' consistently?
  4) Teaching content (a) concepts explained if needed, (b) functions slide on purpose etc, then examples
  5) Exercises
  6) Recap key concepts and key exercise solutions, including any new details that came up in exercises
  7) For part B only: what code should look like
  7) For part B only: Review of learning objectives (saying what we have done)
  8) For part A only: A title slide saying 'Break!'

• Title slide is formatted right and has title/subtitle/author and blue formatting (see slide 1A)
  - [ ] Subtitle - should match the exercise description?

• Learning objectives are phrased correctly

• Explanations of functions generally (purpose, why, example etc)

• Make sure all pipes are new pipe |>

• Exercises/scenario stuff with light blue background

• Review column names in examples (e.g. for group_by i had written group_by(grouping_column) - is that clear enough?)

Slide standardisation - CODE/APPEARANCE

• - Add .fragment. Remove unnecessary multiple slides
• - Standardise how code is presented. Currently long code uses a proper chunk which means font colour changes if functions/string etc, while initial very simple short code uses the code-box or code-box-small container (misleading names as code-box-small is the wider one)
• Play with code/text boxes so that they fit
• Thing to check:
  • Should brackets go to the next line?
  • use gt to convert or just present as the actual code as if in cnsole?
  • are example slides called example or the function name?
• Colours are consistent (as recorded in css):
  • 

Bigger content edits needed - SLIDES

• - Some slidesets seem way too long and others very short
• 8A - wide table a bit hard to interpret?
• Explanations of unite and coalesce - they seem repeated

Additions still needed - SLIDES

• Date formatting
• To mutate: add 7 days
• Mention that organising script in different ways is ok
• scales??
• missingness/NAs
• - explain c() somewhere (it comes up in 4B)
• - explain different types of missing values (comes up in 4B in case_match())
• Still need to add split

Bigger content edits needed - Exercises

• - Transfer everything to correct tech
• - Some seem way too long and others very short
• Add for RStudio/import session to look at missingness
• - Add to level I

Data

• Change raw data to have actual NAs not blanks?? Or how to deal with it as so ugly?
• Change to not be about Nepal
• Multiple lab files
• Change report dates so one is ymd and they all have years (report currently just day and year)
• Add stuff to deduplication so can partially deduplicate
• lol "dry mouf"? for symptoms
• Change period of time for datastage 2 so there is more data
• Change the onset date to have MDY at the start