Skip to content

Latest commit

 

History

History
73 lines (57 loc) · 3.42 KB

README.md

File metadata and controls

73 lines (57 loc) · 3.42 KB

StatTutor XBlock

This is the XBlock implementation of the StatTutor exercises originally developed for the Open Learning Initiative at Carnegie Mellon University using the Cognitive Tutor Authoring Tools.

StatTutor is a learning tool that provides students with a data analysis problem and supports them as they attempt to solve it, giving hints and feedback along the way. It assumes that the relevant data set is open in the background.

One of the main ideas behind StatTutor is to demonstrate that when solving data analysis problems we always go through the same set of steps. Those steps are shown on the left side of StatTutor, and students can click on them in sequence as they progress through the lab. Clicking on each step will cause a new worksheet to appear for students to complete.

Note that students must provide a correct answer to all questions in a given step or provide a written response before being able to proceed. The system will bounce them back to the step that does not have all the questions correct or is missing an answer.

Students can ask for hints by clicking on the hint button.

Installation

The steps to install all XBlocks can be found on edX's XBlocks integration page. Follow the instructions outlined in the Testing section if you're running the devstack or scroll down to those outlined under Deploying your XBlock if you're not.

How to Configure

To add one of the available StatTutor exercises, navigate to a unit in Studio, click Advanced and then click stattutor. This will add the first exercise. To change the exercise, click on Edit and use the drop down to select the desired exercise. A preview of the exercise will be displayed in Studio.

Configuring download types

The tutoring behavior is specified in a CTAT behavior recorder diagram (.brd) file. Also, students can use data files for various statistical packages to answer questions posed by StatTutor. However, OpenEdX does not recognize some of these file types by default. To register the various file types as valid mime types in OpenEdX, administrators must list them explicitly in the add_mimetypes() function defined in /edx/app/edxapp/edx-platform/[cms,lms]/startup.py by adding the following code:

    mimetypes.add_type('text/xml', '.brd')  # CTAT behavior recorder diagram
    mimetypes.add_type('application/octet-stream', '.8xg')  # TI Calculator
    mimetypes.add_type('application/octet-stream', '.mtw')  # Minitab workbook
    mimetypes.add_type('application/octet-stream', '.RData')  # R uses .RData

Adding new problems

To import new problems from OLI, place the problem specific files from content/webcontent/<problem directory> in the corresponding directory in stattutor/public/problem_files/<problem directory>. Once the XBlockStattutor is reinstalled in the OpenEdX server and it is restarted, the new problem should be available in the list of problems in the Edit panel for the xblock exercise in Studio. See the 'README.md' file in stattutor/public/problem_files/ for important notes and restrictions.

License

This is made available under a CC-BY-NC-SA 4.0 license by Carnegie Mellon University Open Learning Initiative.