Page 2 - R_EdQuire White Paper Nov 2017 v3.4
P. 2
EdQuire White Paper: Computer learning behaviour in K-12
Nov 2017 V3.4
Page 2 of 15
1. Introduction
Students and teachers in Australian classrooms use a wide range of computer resources to facilitate the
st
learning process (Crook and Sharma 2013) and instil in students the ICT and higher order 21 Century skills.
In-class usage those resources actively shapes the nation's educational outcomes. However, we have a
limited understanding of how computers are used in classrooms, of the categories of software and internet
resources that are utilized to facilitate learning and of how both relate to learning outcomes (Crook and
Sharma 2013).
The body of K-12 ICT literature suggests the emergence of a second digital divide between those with
advanced ICT skills and those without. However, the research has invariably employed surveys, manual logs
and interviews. This has prompted calls for more objective data on classroom computer use from many
educators faced with stagnant or even falling educational outcomes and ICT skills despite the ever increasing
expenditure on K-12 ICT (NAP-ICT 2014).
The purpose of this paper is to report on objective data from high school student computer use during trials
of our EdQuire real-time Learning Analytics classroom tool in four schools. edQuire is a private 4 year
philanthropic project in collaboration with a number of teachers and academics located in Sydney, Australia.
Its aim is to use objective data and artificial intelligence to make computer learning processes in classrooms
visible to teachers and educators, in order to forge a better understanding of how ICT is used in schools and
how it can be made to benefit all students in all schools.
This report is an observational report of computer use by 549 Year 7-12 students from four independent high
schools over one year, during which time teachers were given real-time Learning Analytics data of each
student’s computer use and engagement in a single glanceable colour coded web page. We used an edQuire
background agent on students' computers to continuously send application and internal activity data used by
students during lessons to our cloud-based data warehouse. Using an AI-based learning analytics algorithm
we categorized and analysed the educational relevance of student computer usage data in real time. We
also continuously displayed this data to the teachers on a web console their students’ classroom engagement
in the form of a glanceable browser screen with student icons colour coded according to their engagement.
Students on an assigned resource (On-Task) were coded green, students on a self-discovered resource
(OwnTask) blue and students off task (Off-Task) red. We analysed this data and provided reports to teachers
on engagement, distractibility, task switching and search engine use.
We report here on the collected data, with early analysis, including by gender and by year grade. We also
report on a small prospective controlled substudy examining students’ learning behaviour after giving them
access to their own analytics data.
We present a preliminary analysis describing computer use in classrooms: characterising the extent of active
computer use in classrooms, the nature and cognitive level of activities undertaken, the proportions of on-
task and off-task computer use and an overview of search engine use by students. Finally, we show that
giving students feedback of their own data can have a dramatic beneficial effect on their self-regulation.
© Copyright 2017 | All Rights Reserved by FIC Technology