commons-feedback

This repository contains two different but tigthly related applications:

• an Android mobile app, intended to collect student/attendee reactions to a synchronous lecture/event; its name, WE-COLLAB, is that of the omonimous Erasmus+ project (2022-2024) inside which it has been developed;
• a server-side Django application, commons-feedback, which extends the CommonSpaces collaborative learning platform to interface the mobile app and process its output.

Objectives

The objectives and the initial design ideas of both the mobile app and the Django application were set out in the blog article The design of a Student Feedback App (partially obsolete). Briefly, the main objectives pursued by commons-feedback are

• the growth of the student engagement and
• the improvement of the lecture quality, in synchronous events (remote or in presence).

Functionality

Two kinds of student feedback are supported:

• sending a verb, which codes the student reaction: an alert or a request to the lecturer, chosen from a small set of options; the choice is done with a 3x3 virtual keyboard;
• sending a free chat message.

The lecturer, and or her assistant, gets an aggregated view of the students' feedback through an Event Dashboard, which puts together:

• a raw log of the students' reaction messages;
• a Verb cloud, which shows the nine available options with font sizes related to their relative frequencies;
• the chat pane, which includes an input box and the message list; that is, also the lecturer can write to the chat.

Some context

Compared to Wooclap, which is a fairly well-known commercial product for a comparable "market", commons-feedback

• is free and open source;
• allows to send nine reaction types, while Wooclap has only "I'm confused";
• doesn't offer polls and question/answer functionality; but our feeling is that these are functions not really liked by teachers, since they imply a a non-trivial lesson design activity;
• instead, it comes with a flexible chat facility, which can support similar engagement objectives.

While commons-feedback doesn't support statistical processing directly useful for educational managers, we think that it can be quite useful for formative assessment:

• all feedback is saved by the native tracking functions ofCommonSpaces;
• most importantly, said feedback is forwarded to a Learning Record Store (LRS) as xAPI statements, for off-line processing.

Please, consider that:

• the WE-COLLAB project also includes original research and experimentation in the interpretation of neuro-physiological signals induced by learning situations, in terms of cognitive and emotional states;
• the availability of more explicit and more subjective reactions by the students, related to the same learning situations, should provide essential data for comparative evaluation.

Some technical aspects

The mobile (client-side) app has been entirely developed in Javascript for Android using the DroidScript on-device development environment.

The server-side web application has been developed as a Django app in Python, mostly as an extension of the CommonSpaces platform. The frontend uses also Vue.js.

The system architecture strongly relies on the WebSocket protocol, which was necessary for

• the real-time update of the Event Dashboard;
• the implementation of the chat on both sides.

The first version of the server-side software was generated following the channels tutorial referring to the design of a simple chat application.

As to the versions of the development and run-time environments and of many software libraries, please refer to the file requirements.txt of CommonSpaces.