Module: Writing Observer

The Writing Observer module is designed to analyze and visualize students' writing data. It provides various aggregators, reducers, and other features to process incoming data and serves information to dashboards through APIs. In this document, we will discuss the features provided by the Writing Observer module and its components.

Course Aggregators

Course aggregators are responsible for aggregating data from different reducers and generating a summary of statistics for each student. The Writing Observer module currently defines:

This area is a work in progress while we determine the communication stream.

Reducers

Reducers process incoming data and generate relevant information. The Writing Observer module defines the following reducers:

1. time_on_task: Calculates the time spent by students on writing tasks.
2. reconstruct: Reconstructs the text written by students.
3. event_count: Counts the number of events for each student.
4. document_list: Provides a list of documents for each student.
5. last_document: Fetches the latest document for each student.

Extra Views

Extra views are additional views that can be added to the module. The Writing Observer module currently defines:

• NLP Options: This view serves the INDICATOR_JSONS static JSON data, which is related to natural language processing (NLP) options.

System configuration (creds.yaml)

The following settings are available to set on a per-system basis in your creds.yaml file.

• use_nlp (boolean): whether to use AWE_Components to compute NLP data (True) or generate fake data (False)
• use_google_documents (boolean): whether to automatically update the reconstruct data with Google API (True) or not (False)

Example:

modules:
  writing_observer:
    use_nlp: true
    use_google_documents: false