Feedback Authoring for Exploratory Activities: The Case of a Logo-Based 3D Microworld

Abstract

This paper presents AuthELO an authoring environment that can be used for the configuration of logging and authoring of automated feedback for exploratory learning objects (ELOs). ELOs are web components (widgets) that can be integrated with learning platforms to synthesise highly interactive learning environments. AuthELO has been developed in the context of the MCSquared project that is developing a platform for authoring interactive educational e-books. This platform comprises an extendable set of diverse widgets that can be used to generate instances of exploratory activities that can be employed in various learning scenarios. AuthELO was designed and developed to provide a simple, common and efficient authoring interface that can normalise the diversity of these widgets and give the ability to non-experts to easily develop or customise the feedback that is provided to students using a data-driven approach. In this paper we describe the architecture and design characteristics of AuthELO and a small-scale evaluation with activities in a logo-based 3D microworld called Malt+. We reflect on both the challenges of the authoring process and the pedagogical potential of the feedback when these activities are used by students.

A review of automated feedback systems for learners: Classification framework, challenges and opportunities

“…The second popular approach is a control group experiment employed in 56% reported evaluations, e.g. in de Vries et al (2015); Cheng (2017); Karkalas et al (2016). Moreover, the results of pre-and post-tests are compared in 36% cases.…”

“…The most common full name is by far Intelligent Tutoring System, used by 35% of the authors. Next frequent keyword combination was "learning environment" (15%), which was used together with some adjectives and prefixes such as: virtual (Kennedy et al, 2013), interactive (Karkalas et al, 2016), digital (Lukáč and Sekerák, 2015), intelligent (Sun et al, 2018), training (Hülsmann et al, 2018), web (Correia et al, 2017), e-(Marcus et al, (Naismith and Lajoie, 2017), open-ended (Basu et al, 2017) and example-based (Watson et al, 2011). The most frequent single keywords are "system" (55%), "intelligent" (42%), "learning" (27%), "feedback" (11%) and "tool" (10%).…”