Distinguising those who are competent from novices
Ask them to compile a series of .tex files. The first one should compile fine. The others have small problems to fix. See what they do. Problems might include:
- It has the beginning of a list but not the end
- It has a math equation out side of markers for an equation.
- It is set to compile to the wrong thing e.g. .bib
- Etc.
Give them a .pdf document and ask them to copy it in LaTeX given a style template. Parts might include:
- Add the date and the authors name
- Add a footnote
- Put text in bold and italics
- Insert an inline equation and a standalone equation.
Distinguising experts from those who are just competent
Ask them to shorten a document that is a page too long. Ask them to try different ways to get rid of the extra page without changing the text or removing figures/tables etc. They might for instance:
- Remove the spaces around boxes in various ways.
- Make equations into inline equations.
- Change the size of figures
- Reformat tables
- Resize the title or body fonts
- Change the margin sizes
- Move to a 2 column style
- Etc.
Have them compile a LaTeX project using a make file. Depending on how expert you expect them to be either have them write it or give them one to work with. One more step advanced is to ask them to take graphs produced using R and automatically pull them into the document and then compile.
Some notes:
- Given Greg’s point that you may not be able to evaluate competence until someone runs into a problem, some activities involve solving problems.
- In this case, these are most naturally used as formative assessments. There is no calibration to marks meaningful. Similar topics could be built into a quiz (maybe with an interactive worksheet), to provide a summative assessment.
- Good formative assessments might take smaller steps. The above exercises would be enough to give information to teacher to help move the student forward. It may not be enough for the student to move forward on their own.
**Could recording the screen help? **
**You would get significant insight from being able to sit with the student or record their screen. It would make it possible to distinguish between students who:
- Can do without looking it up quickly/easily, with some thought or rough work, with mistakes at first
- Must look up because they don’t remember the command, or because they don’t know the concept.
Some of the insight would come from recording just some features of the performance (e.g. how long it takes, are there earlier wrong attempts to compile, does the student look anything if so what are the search terms and what areas are visited?)
It could also save students time if it were possible to start with tasks that integrate multiple activities, but then to revert to simpler exercises to build up capacity if the student has trouble.