Create gradable exams as interactive Shiny documents from R markdown documents. Learners can attempt an exam one or more times, and each attempt uses a (potentially) different random seed to generate the question text and answers.
For a full list of features and to learn how to create your own exams, visit the documentation at https://dakep.github.io/examinr.
The package is still in development and not yet available on CRAN. You can install it via
devtools::install_github("dakep/examinr")- Questions and text are rendered on-demand for each attempt:
- Allows for user- and attempt-specific content (e.g., randomized values in questions and answers)
- Questions are not sent to the user’s browser until they are allowed to see them
- Support for different question types:
- Simple text answers
- R code exercises
- Numeric answers (with auto-grading support)
- Multiple-choice (with auto-grading support)
- Users are authenticated via configurable authentication providers
- Exams can be configured to show only one section at a time
- Fine-grained control over attempts:
- Control how many attempts each user gets
- Configure exams to be only accessible after the opening time and until the closing time
- Timed attempts where users are required to submit the exam within the specific time limits
- Adjust the configuration for specific users
- Support for message localization
- Interfaces for grading (for instructors) and for accessing feedback (for users).
- High-contrast theme and screen-reader friendly output
- Users’ inputs are also stored in their browser to allow restoring input values after a reload or in case of server issues.
The package is still in a development phase and some features are not yet available, most notably the following:
- Download an exam as RStudio project to give learners the option to write the code on their local computer.
- Storing the evaluation environment to allow results from one code box being used in other code boxes.
- File upload question.
- Multiple text inputs per text question.
- Improved rendering of help pages.
- Auto-grading of code exercises.
If you are looking for specific features which are not yet supported by examinr and not yet requested, please create an issue or, even better, a pull request.
Several features in examinr are inspired by the learnr package, in particular the exercises.