To install the refined eSTEeM: Measuring Qualification effects project’s framework create the suggested directory structure shown below and unzip the docs, scripts and taxonomies folders from The Open University’s online data store, DOI: 10.21954/ou.rd.7370444, and copy the README, LICENSE and requirements files.
Suggested directory structure:
|- LICENSE.txt |- README.md |- README.html |- requirements.txt | |-- docs | |-- mbz | |-- scripts | |-- taxonomies | |-- zipped-assignments
The directories are:
All other directories required by the framework are generated on demand when running each script.
The framework requires Python 3 to execute the scripts. Where possible built-in Python libraries have been used. However, there are two dependencies for specialist tasks:
draw_posts
script, andnltk.download('punkt')
.The dependencies have to be installed separately using Python’s supplied pip
installer.
You may find it more convenient to install a comprehensive Python bundle such as anaconda because it includes both matplotlib and nltk.
Note on versions used in developing version 1.0.0 of the framework: Python 3.6.6, matplotlib 3.0.0, nltk 3.3 and pip 18.1.
If you are looking to apply the framework, read usage.rst.
If you are a developer looking to maintain the framework, design decisions are documented in design.rst. For guidance on data suitable for testing, read testing.rst.
To address the potential breadth of audience for these documents, all source/\*.rst
files are available as build/\*.html
pages too.
Diagrams are developed in PowerPoint and saved as office/\*.pptx
files. They are exported in png format to images/\*.png
files.
The main report, office/Report.docx
, from the original Measuring qualification effects project has more information in it, especially relating to the output files and data tables. Examples of use of that original framework are in office/Cookbook.docx
.
There are four sub-folders: