DIY Ecological
Momentary Assessment
CHI 2023 - 18th April 2023 (online)
Session 1 Time: 13:30 - 14:45 (CET)
Session 2 Time: 15:00 - 16:15 (CET)
Session 3 Time: 16:30 - 17:45 (CET)
Brought to you by:
What is Ecological Momentary Assessment?
Ecological Momentary Assessment (EMA) is a methodology which aims to reduce biases connected to the recall of past experiences. EMA is able to achieve this by collecting data on the participant throughout their usual day, at multiple time points – asking them about their experiences in that moment rather than using recall. In healthcare this has been used to track fluctuations in different health conditions to reduce the bias associated with recall and with altering their daily routine to visit a clinician.
Ecological Momentary Assessment
Shiffman, S., Stone, A., and Hufford M. R.
Annu. Rev. Clin. Psychol. 2008. 4:1-32
| Annual Review of CLinical Psychology
What can EMA do for you?
The benefits of EMA include: reducing recall bias with ‘in-the-moment’ measurement; improving ecological validity with ‘in-the-context’ measurement and enabling exploration of the dynamic interplay of variables from repeated measurement.
What will you do in this course?
The figure shows the session task breakdown across four directed blocks representing the progression between the four tasks covered in the course. The first block is shorter indicating that the first task is completed more quickly than the others. Brackets around the blocks show that task 1 and task 2 make up teaching session 1. Task 3 is covered in teaching session 2 and task 4 is covered in teaching session 3.
What will you get out of this course?
You will create the output shown in the image below which you are free to adapt for your project needs.
The flow diagram displays the connections between the different programs involved in the EMA system. There are two flows displayed. The first flow involves sending a message. The flow starts with the scheduling of messages via Outlook. Next, Power Automate is used to request a message to be sent from Twilio. Twilio is then able to send messages and dynamically respond to a participant, represented by a phone icon. The second flow involves storing a message. Power Automate is used to request messages from Twilio, which returns a list of messages. Power Automate is then able to store the messages in Excel. The flow diagram also shows a double-headed arrow of communication between Power Automate and Azure, which is used to ensure messages are sent securely.
Why should you take this course?
This course will teach participants about EMA by providing hands-on practical skill development in how to use off-the-shelf software to develop their own EMA system. At the end of the course participants will be able to rapid prototype and run custom EMA studies securely and at scale without the usual associated costs.
Who are we?
Sydney Charitos is a PhD student in Digital Health and Care at the University of Bristol and she is supported by an EPSRC doctoral training grant. Sydney created the original teaching materials and has used them to teach clinicians, researchers and students how to build EMA systems using this approach. With Jon Bird she has supervised Masters's students who have used the approach to rapidly build EMA systems to carry out research.
Dr Amberly Brigden is a Lecturer in Digital Health in the computer science department at the University of Bristol. She leads the University of Bristol’s Ecological Momentary Assessment (EMA) Interest group, which spans the Faculty of Engineering and the Faculty of Health Sciences. She holds Research Grants focusing on EMA and is supervising EPSRC-funded PhD students to develop EMA applications. Amberly teaches within the Faculty of Engineering and has been recognised for her teaching as a 2022 nominee for the Advanced HE Award for Teaching Excellence (CATE Award).
Dr Jon Bird is a Senior Lecturer in Digital health in the computer science department at the University of Bristol. He has taught university students for over 30 years and also has extensive experience in developing and running workshops and courses to teach technical skills, such as mobile app design and physical computing, to participants with non-technical backgrounds. Teaching technical skills is also a research interest and he has explored how to support troubleshooting when people are building systems that combine software and hardware. He proposed the initial idea of using Twilio to develop EMA systems to Sydney who then fully developed the idea and found ways to use Power Automate, Excel and Azure to automate message scheduling and secure data storage. One of his current research interests is using EMA to better understand self-harm.
Designing Troubleshooting Support Cards for Novice End-User
Developers of Physical Computing Prototypes
Booth, T., Bird, J., and Stumpf, S.
IS-EUD 2019
| City Research Online
Crossed Wires: Investigating the Problems of End-User
Developers in a Physical Computing Task
Booth, T., Stumpf, S., Bird, J., and Jones, S.
CHI '16
| CHI Conference on Human Factors in Computing Systems
Learning Outcomes
Learn how to do the following over three - 1 hour and 15 min - classes:
Learn More About EMA
EMA is a methodology that is being increasingly used in HCI
to measure a wide range of subjective experiences, for
example, mood, substance usage and binge eating.
Understand the Benefits of EMA
Understand the benefits of EMA for participant data
collection, including: reducing recall bias, increasing
ecological validity and improving participant
adherence.
Understand the Benefits of Rapid Prototyping
Understand the value of rapid prototyping in the context of
EMA both as an educational tool and for the customisation of
EMA protocols.
Create an EMA System
Create an EMA system using off-the-shelf tools, that is
able to dynamically respond to study participants.
Automate an EMA System
Automate the scheduling of questions to participants while
ensuring the secure transfer and storage of their responses
using Microsoft technologies (Power Automate, Outlook
calendar, Excel, Azure Key Vault). The automation of the
system reduces long term maintenance costs.