Personal Analytics Explorations to Support Youth Learning

TL;DR: In this article, personal analytics are used in the design of instruction for a high school student examining how sleep influences her mood and a sixth-grade class of students examining how deviations from typical walking behavior change distributional shape in plotted step data.

Abstract: While personalized learning environments often include systems that automatically adapt to inferred learner needs, other forms of personalized learning exist. One form involves the use of personal analytics in which the learner obtains and analyzes data about himself/herself. More known in informatics communities, there is potential for use of personal analytics for design of instruction. This chapter provides two cases of personal analytics learning explorations to demonstrate their range and potential. One case is of a high school student examining how sleep influences her mood. The other case is of a sixth-grade class of students examining how deviations from typical walking behavior change distributional shape in plotted step data. Both cases show how social support and direct experience with data correction are intimately involved in how youth can learn through personal analytics activities.

Summary

BACKGROUND

• Many who have documented self-tracking and personal analytics practices identify 2007 as the year in which a community and sociotechnical movement (the “Quantified Self”) emerged in the Silicon Valley, with Wired magazine editors Kevin Kelly and Gary Wolf being attributed as primary initiators (Choe et al., 2014; Lee, 2014).
• The two cases are also presented with an eye toward diversity to show the different kinds of data that could be obtained by youth and to demonstrate that personal analytics learning activities are usable with youth at different grade levels, with different levels of technology, and in different participation structures (i.e., working individually vs as a whole class).
• Immediately following, she began to consider a number of sleep variables that could impact her mood.
• While other girls participating in the project shared their own observations, experiences, and difficulties with the devices that they had tried, Melissa proceeded to pay partial attention to them and also entering the new values she obtained from the Up device into the spreadsheet she had started the week prior.

Commentary on the Playground Walking Case

• This in-class discussion about steps taken to get across the playground at school was not originally planned.
• The authors had expected that students would walk normally and produce normally distributed data through this activity.
• In the classroom, where there is a larger social setting in which multiple people are examining the data, gaining insight can come about by noticing what others can see and hearing how they explain the data.
• This represents an important consideration for how personal analytics activities could be sensibly implemented in schools and other learning environments.
• Thus, the authors suggest from this case that the combination of personal connection to the data, along with the software tool and the teacher’s facilitation of the discussion all jointly enabled the realization of a substantive learning opportunity.

CONCLUSION

• The overarching aim of this chapter was to suggest that personal analytics approaches represent an important area of personalized learning in today’s digital ecosystem.
• In contrast to other forms of personalized learning experience represented in this volume, the emphasis when using personal analytics for learning is to have the learner create data from their own activities and then be in the position of examining their own data.
• What is to be taught and learned by students is not automatically determined nor recommended to them; it is encountered in the process of making sense of what the data say about the students and their experiences.
• The first was of a high school student who participated in an afterschool program where she used a wearable device to explore possible relationships between her sleep and her mood.
• According to their records, she seemed to get there.

Figures

Figure 5. The horizontal line drawn by Melinda showing her prediction for how the distributional shape would look had all students walked normally

Figure 1. A screen capture of the Jawbone Up mobile device app that reports light and deep (sound) sleep and number of times woken up

Figure 2. The final plot Melissa and the others examined comparing mood against number of minutes of sleep on the preceding night, rendered in TinkerPlots

Figure 4. Histogram showing all student step records from the playground walk. Note the long tail distribution that extends to the right.

Figure 3. Students in Ms. Hayley’s class walking across the playground while counting steps. Note the two girls on the right doing intentionally abnormal forms of walking to change the number of steps that they are recording.

Full text

Digital Technologies and
Instructional Design for
Personalized Learning
Robert Zheng
University of Utah, USA
A volume in the Advances in Educational
Technologies and Instructional Design (AETID)
Book Series

Published in the United States of America by
IGI Global
Information Science Reference (an imprint of IGI Global)
701 E. Chocolate Avenue
Hershey PA, USA 17033
Tel: 717-533-8845
Fax: 717-533-8661
E-mail: cust@igi-global.com
Web site: http://www.igi-global.com
Copyright © 2018 by IGI Global. All rights reserved. No part of this publication may be reproduced, stored or distributed in
any form or by any means, electronic or mechanical, including photocopying, without written permission from the publisher.
Product or company names used in this set are for identification purposes only. Inclusion of the names of the products or
companies does not indicate a claim of ownership by IGI Global of the trademark or registered trademark.
Library of Congress Cataloging-in-Publication Data
British Cataloguing in Publication Data
A Cataloguing in Publication record for this book is available from the British Library.
All work contributed to this book is new, previously-unpublished material. The views expressed in this book are those of the
authors, but not necessarily of the publisher.
For electronic access to this publication, please contact: eresources@igi-global.com.
Names: Zheng, Robert, editor.
Title: Digital technologies and instructional design for personalized
learning / Robert Zheng, editor.
Description: Hershey, PA : Information Science Reference, [2018]
Identifiers: LCCN 2017024205| ISBN 9781522539407 (hardcover) | ISBN
9781522539414 (ebook)
Subjects: LCSH: Individualized instruction--Computer-assisted instruction. |
Instructional systems--Design.
Classification: LCC LB1031 .D54 2018 | DDC 371.39/4--dc23 LC record available at https://lccn.loc.gov/2017024205
This book is published in the IGI Global book series Advances in Educational Technologies and Instructional Design (AE-
TID) (ISSN: 2326-8905; eISSN: 2326-8913)

145
Copyright © 2018, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.
Chapter 7
DOI: 10.4018/978-1-5225-3940-7.ch007
ABSTRACT
While personalized learning environments often include systems that automatically adapt to inferred
learner needs, other forms of personalized learning exist. One form involves the use of personal analytics
in which the learner obtains and analyzes data about himself/herself. More known in informatics com-
munities, there is potential for use of personal analytics for design of instruction. This chapter provides
two cases of personal analytics learning explorations to demonstrate their range and potential. One case
is of a high school student examining how sleep influences her mood. The other case is of a sixth-grade
class of students examining how deviations from typical walking behavior change distributional shape
in plotted step data. Both cases show how social support and direct experience with data correction are
intimately involved in how youth can learn through personal analytics activities.
INTRODUCTION
One widely recognized model of personalized learning involves using computational tools to exogenously
recognize and respond to the immediate conceptual needs of a learner and automatically adjusting the
learning environment accordingly; for instance, an intelligent tutoring system can infer the knowledge
state of a user and then present appropriate tasks and instructional content to move them closer to a more
expert-like understanding (e.g., Graesser, Chipman, Haynes, & Olney, 2005). Yet there are still several
other ways in which personalization could take form in a technology-supported learning environment.
As examples, consider how using personalized story scenarios that map onto students’ out-of-school
interests can boost student learning in mathematics (Walkington, 2015; Walkington, Petrosino, & Sher-
man, 2013) or how personalization can also be achieved and supported by providing youth with genera-
tive and expressive computational media where they can design and engineer artifacts that are specific
to their own interests and prior expertise (Peppler & Kafai, 2007). In this chapter, my goal is to further
Personal Analytics Explorations
to Support Youth Learning
Victor R. Lee
Utah State University, USA

146
Personal Analytics Explorations to Support Youth Learning
expand the space of personalized learning by highlighting what opportunities exist when students are
able to explore “personal analytics”.
Personal analytics (Ruckenstein, 2014; Wolfram, 2012) refers to a set of data-related practices that
have been associated with the area of personal informatics (Li, Dey, Forlizzi, 2010) and the Quantified
Self (Nafus & Sherman, 2014). All typically involve collecting and analyzing aggregates of data obtained
through “self-tracking” (Lee, 2017). The use of the term has generally varied depending on the scholarly
community (i.e., personal informatics is more common parlance in the information sciences). Arguably,
there are nuances that distinguish the terms, but for current purposes, it is acceptable to think of personal
analytics as being comparable to the kinds of analytics that one might do with website data but instead
do so with data about one’s own self. At its core, personal analytics is self-inquiry using data. While
data collection and inquiry about one’s own self has been practiced for several years in a range of com-
munities (e.g., Kopp, 1988; Lee & Drake, 2013; Wallace, 1977, Wheeler & Reis, 1991), the widespread
availability of consumer-level mobile and wearable devices and automated data collection systems (such
as clickstream recording) has reduced some of the initial barriers associated with analyzing data about
one’s self (Lee, 2013) and popularized this approach. This has thus enabled some initial pioneering work
to integrate personal analytics routines and practices into the design of learning environments. Some
noteworthy examples include asking youth to use personal analytics data and gaming environments to
motivate healthier lifestyle behaviors (Ching & Schaefer, 2015) or to support interactive exhibit design
at settings such as zoos and museums (Lyons, 2015).
While some promising opportunities have been noted (Rivera-Pelayo, Zavharias, Müller, & Braun,
2012), much still remains to be understood about how educational designers can best support learning
that invites students to do the work of personal analytics. Part of this has to do with the disparity between
the most visible and noteworthy examples of expert-like learning through personal analytics and what
needs and challenges are encountered by novices. For example, noted polymath Stephen Wolfram (2013)
has presented some detailed cases and visualizations of his own personal analytics of email use, phone
calls, and meeting participation over the course of years. Power users who identify with the Quantified
Self movement (also known as QSelf-ers or QSers) regularly convene in major urban areas to share per-
sonal analytics projects that they have pursued and how that helped them to gain new insight and learn
about themselves (Choe, Lee, Lee, Pratt, & Kientz, 2014; Lee, 2014). Such examples are informative
and aspirational for the future, but they also presume proficiency with powerful visualization tools, flu-
ency with data and data representations, some formal understanding about correlational and potentially
experimental design, and instrumentation. Each of these could be, in a designed educational setting, a
set of learning goals on their own.
Moreover, personal analytics has been by and large dominated by an orientation toward using personal
data about one’s self to support planned behavior change, self-improvement, and performance optimi-
zation (Li et al., 2014). Empirical examination of adults who self-track and analyze their own personal
data actually suggests that the reasons for participation in self-tracking communities and activities are
actually more nuanced, with only a fraction of the broader population of self-trackers having aspirations
of behavior change in mind. Many are simply curious to see what possible stories come from their num-
bers or if their intuitions match an existing quantified scale (Epstein, Ping, Fogarty, & Munson, 2015).
Furthermore, the standards of scientific rigor are not always met nor understood by those who undertake
self-tracking and self-experimentation projects (Choe et al., 2014). With those observations in mind,
it is reasonable to expect that students and youth who are charged with performing personal analytics

147
Personal Analytics Explorations to Support Youth Learning
work to support their own learning may introduce other challenges or complications. Learning to read
graphs, for instance, is an ongoing challenge for many students (Leinhardt, Zaslavsky, & Stein, 1990).
Still, the idea of using data from and about students with those very same students holds appeal and
has been tantalizing enough for some to explore through educational research and design. The gap from
ideation to execution has only recently begun to be filled. This chapter is an effort to continue with that
gap-filling by presenting two cases of student learning activities with personal analytics obtained over
the course of a multi-year research program to explore the potential of personal analytics for youth
learning. The first case is of high-school student who participated in a multi-week afterschool program
where each participant was charged with embarking on her own personal analytics project. The goal
here was to modestly emulate what had been pursued in more advanced hobbyist practice (Choe et al.,
2014) and understand what supports were needed for youth to pursue similar endeavors. The second case
comes from a designed unit in a sixth-grade classroom where students worked continuously with data
from their own activities. In this second case, the data obtained were a count of steps required to walk
across the school playground and did not require sophisticated wearable devices. However, the case was
noteworthy in that the visualization of activity showed some abnormal walking behaviors and spurred
productive classroom discussion about distributions and how data were generated. Together, these two
cases illustrate some of the potentials for personal analytics to support learning, with different lessons
gleaned from each. The first hints at some of the supports that are needed and concerns that get raised
in the context of individual inquiry, but also provides the reader with a sense of what kinds of content
can be learned through personal analytics endeavors. The second demonstrates how the core commit-
ment of personal analytics – the examination of data about the self – can be sensibly integrated into a
classroom activity and how awareness of data creation can be conducive to classroom discussions about
visual representations of aggregated data.
BACKGROUND
Many who have documented self-tracking and personal analytics practices identify 2007 as the year in
which a community and sociotechnical movement (the “Quantified Self”) emerged in the Silicon Valley,
with Wired magazine editors Kevin Kelly and Gary Wolf being attributed as primary initiators (Choe et
al., 2014; Lee, 2014). Since that time, informal groups of self-trackers have appeared in over 100 cities
around the world, and an annual conference that brings together hobbyists, entrepreneurs, tool-makers,
medical professionals, and academic researchers boasts consistently strong attendance (i.e., a few hundred
attendees regularly). However, it is worth noting that like many activities popularized through the Silicon
Valley, self-tracking had actually existed in various forms for decades prior in academic communities
such as behavior analysis and organizational studies (Burns, 1954; Wallace, 1977; Wheeler & Reis, 1991).
Independent of the designated sociotechnical movement or prior behaviorist-oriented work, my re-
search group and I had been exploring student learning opportunities with wearable device-obtained data
about their own selves since 2008. From 2008 to 2017, we have been developing and pursuing integrated
and overlapping lines of educational research and development oriented toward personal analytics ap-
proaches to supporting teaching and learning. That has involved a number of design experiments (Cobb,
Confrey, diSessa, Lehrer, & Schauble, 2003) with youth working in small groups and collectively as full
classrooms (Lee & DuMont, 2010; Lee & Thomas, 2011). The two cases presented below come from
the broad data set obtained over those years in which different learning configurations and supports were

Frequently asked questions

Q1. What are the contributions mentioned in the paper "Digital technologies and instructional design for personalized learning" ?

Personal Analytics Explorations to Support Youth Learning this paper has been a hot topic in the field of personal analytics. 

Q2. What have the authors stated for future works in "Digital technologies and instructional design for personalized learning" ?

In looking toward the future, learning activities for individual novices and youth doing personal analytics work should incorporate supports to help students notice patterns and understand more about the phenomenon that they are quantifying through their projects. Still, they were able to use those records in productive ways, suggesting that while personal analytics is gaining in prominence because of increased availability of individual tracking devices, those are not absolutely essential for this sort of instructional approach to work. Thus, the prospects for personal analytics learning explorations to become an option for those who want to support personalized learning are promising. 

Q3. What was the commonly used data visualization software in their studies?

The most commonly used data visualization software in their studies was TinkerPlots, a novice-friendly data visualization tool developed to enable elementary students and above to use drag and drop interactions to produce dynamic data visualizations (Konold & Miller, 2005). 

Q4. What was the first case of a student who used a wearable device to explore possible?

While each student had a wearable device that would track their steps and the students examined data from those throughout the larger unit, the specific activity discussed here involved analytics on manually collected data. 

Q5. What are the prospects for personal analytics learning explorations?

There are questions that remain for us as a field to examine, such as what conditions promote an initial desire from students to look at their own data, how to support learning with different software tools, and what kinds of social configurations around a personal analytics learning activity enables broader participation and deep inspection of data. 

Q6. What did the girl think of the exceptions?

It took another girl’s comment that even though those exceptions existed, there did appear to be some upward152Personal Analytics Explorations to Support Youth Learningtrend with the “Dragging” bin having much lower values for minutes of sleep and the “Pumped Up” bin having the second highest number minutes of sleep.