Welcome/Intro

Welcome to SGSSS Decomposition Workshop!

Hello!

• Who we are
  • Ben; Josiah; Chloe; Trystan
• Who you are (Ben open https://www.menti.com/ala23cd37ta7)

Housekeeping

• Code of conduct
• Ask questions whenever 👍 You can do this through the Teams chat facility {demo this now}
• We’ve structured the sessions with regular breaks, but if you need to leave just leave!
• Materials will live online here so you can access them any time

Learning outcomes

By the end of the workshop you will:

1. Understand how standardization and decomposition can be used to answer social science research questions
2. Be familiar with different approaches to standardization and decomposition
3. Be able to apply standardization and decomposition to practical examples in R or Stata

Course schedule

• See the course website

Practical sessions

• You can follow along with the materials in a local installation of R and RStudio on your own computer
• You can find the course materials at https://github.com/josiahpjking/sgsss_std_decomp
• If you don’t have an R installation, you can follow along with the R practicals in your browser via posit.cloud (free account required) https://posit.cloud/content/12653236
• For Stata you’re relying on your own copy

Software

• R examples use the DasGuptR package. We (mainly Josiah) wrote this!
• Stata examples use the rdecompose command [@liRateDecompositionAggregate2017]. Seems neat!
• For obvious reasons we are more familiar with the R version, and so focus on that
• For some of the more complex examples we only provide R examples. You (or a friendly LLM) may be able to translate these to Stata

Or maybe not. I found that LLMs (the free ChatGPT one) was not at all good with rdecompose - perhaps because the documentation isn’t that extensive