Hi everyone, I'm Will, a PhD student at Stanford doing research in programming and cognitive science. I'm currently working on an "Auto TA" project to help new R users (particularly those with no prior programming experience) learn to overcome basic R errors. Here's a screenshot of a prototype my team has built where we diagnose a syntax error:
Currently, our primary learning goals for the tool are:
- Learn how to read error messages (e.g. locate most relevant information to understanding error)
- Learn how to systematically diagnose potential causes of errors
- Learn how to write effective search queries for Google/StackOverflow
We're starting by addressing the
object not found,
could not find function,
unexpected *, and
cannot open the connection errors.
My questions for you:
- Are there particular errors that you found hard to solve, especially as a novice?
- Are there particular concepts in R you found hard to learn, or kinds of syntax you found confusing?
- Do you have any other ideas for ways a tool could provide automated feedback for common R errors?
You can follow our progress on Github if you're interested: https://github.com/willcrichton/r-autota