Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Measuring Stress in Doctoral Students - Stanford

by

Russell Butson

on 9 August 2017

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Measuring Stress in Doctoral Students - Stanford

the study
Stress levels in doctoral students have long been of concern (Kearns, Gardiner, & Marshall, 2008; Offstein, Larson, McNeill, & Mwale, 2004). Not only do they affect doctoral completion rates and retention, they have been found to have a detrimental impact on students overall wellbeing.
People Data
Generally, the interest in
analytics
within Higher Education is centred on the use of
Big Data
for
institutional decision making
.

Whereas this study explores the use of
personal

analytics
to add
value to people's lives
.

Higher Education Development Centre University of Otago
New Zealand
Participants (partners)
Measuring Stress in Doctoral Research
Russell Butson
A case for reality mining across
psychological, physiological and environmental dimensions.
Dunedin
Austr
alia
New
Zealand
Traditionally, studies of student stress in higher education settings have been based on
student self-reports
of stress through various punctuated collection methods.
Perception Data
Research suggests that student stress and burnout are not indicators associated with the individual but reflect an inconsistency in the relationship between the individual and the environment (Meriläinen & Kuittinen, 2014).
Digital devices allow data to be captured continuously - without intruding on the personal daily life.
Naturally Occuring Data
All full-time - 3 year PhD
Physical Education
2nd year of his PhD
Fieldwork
Data Analysis

Self-Reported Stress Level:
high
<- Medium
- Low
Stressors
Workload
Multitasking
Time pressures
Financial pressures
Archaeology
2nd year of her PhD
Data Analysis

Self-Reported Stress Level:
high
<- Medium
- Low
Stressors:
Data issues
Supervisor meetings
Presentations
Pharmacology & Toxicology
3rd year of her PhD
Data Analysis
Final Write-up

Self-Reported Stress Level:
high
<- Medium
- Low
Stressors:
Supervisors
Experiment Failures
Stressors
Workload
Family
Deadlines
Presentations
Physiotherapy
1st year of her PhD
Literature Review
Research Design

Self-Reported Stress Level:
high
<- Medium
- Low
Devices
Biometric Data - Empatica E4
Context Data - Narrative Clip Auto-Camera
Electrodermol activity
Blood Volume Pulse [BVP]
Accelerometer
Heart Rate
Skin Temperature
Time of day
Measures
Detail

Scale
scale: 5 minutes
Data Streams
EDA - this was the core measure
BVP
- too little known of its relevance/measure
ACC – used to mitigate against perspiration from physical exertion
HR
– was irregular and therefore not reliable
TEM – used to mitigate against perspiration in conjunction with ACC
Measures
Process
Training
x4 weeks
Bands worn
time awake
not-time asleep
study
Cameras
time awake
not-time asleep
Discussion
-each week-
review data
+capture
perceptions
Social Gathering
Meals together
Mon-Fri
recording times
10+ hours per day

Mon-Fri
per person generating
over 100,000 photos
Challenges Analysis
Area of inquiry
Data Types
Data Collection
Data Management
Data Cleaning
Signal Processing
Data Science
Infomatics
Statistics
Phase-1: Exploring EDA Load
of
We are interested in the individual as opposed to the crowd. In this way our work is idiographic rather than nomothetic
The Individual
Personal Capture
Devices
Access to Live
Data Feeds
Self management of accumulating data.
Autonomous Learners
capable of
quantifying & qualifying
their learning
Specifically...
I know I know what I know -

not because the institution tells me so
...leading to learner
autonomy is the
greatest agent
for change in
Higher Education
its all about people
Raw data
Detecting Peaks
Detecting Artifacts
Taylor, S., Jaques, N., Chen, W., Fedor, S., Sano, A., and Picard, R. "Automatic Identification of Artifacts in Electrodermal Activity Data" In EMBC, August 2015.
Taylor, S., Jaques, N., Chen, W., Fedor, S., Sano, A., and Picard, R. "Automatic Identification of Artifacts in Electrodermal Activity Data" In EMBC, August 2015.
Signal Processing
Social Scientist
Data Scientist
Research Design
Phase-2: Detecting Stressors
Rationale
There is a hidden cost to learning that is
insidious and pervasive

One night during the third year of my PhD program, I sat on my bed with a
packet of tranquilizers
and a
bottle of vodka.

I
popped a few pills
in my mouth and swigged out of the bottle, feeling them burn down my throat.

Moments later, I realized I was making a terrible
mistake
. I stopped, trembling as I realized what I’d
nearly done
(Walker, 2015).

The aim of this study was to explore psychological, physiological and
environmental factors associated with stress through the use of devices that afforded the capture of continuous naturally occurring data.
#the aim
The
oretical
FRAMEWORK
BIG DATA
In Particular...
Personal Analytics
Access to live
processed data
Each person generated
100,000+ images
Each person generated
approximately 15 hours of
biometric data per day over
a period of 12-16 weeks:
1,600 - 2,000 hours of data
STRESS
Average daily EDA for student 1
Average daily EDA for student 2
Average daily EDA for student 3
Average daily EDA for student 4
Pre-meeting
Post-meeting
Capturing both data streams
EDA - 5mins
Pics - 15mins

Location
Rationale
Method
Participants
Devices


Process
Challenges

Signal Processing
Prelim. Findings

Study Spaces
Social Work Spaces
Teaching Spaces
Movement
Social Spaces
Acknowledgements
Assistant Professor Dr. Bert Arnrich
Bogazici University
Department of Computer Engineering
Istanbul, Turkey
Professor Rachel Spronken-Smith
Dean (Graduate Research School)
University of Otago
New Zealand
As a result, measures of stress are typically guided by
post-event recollections
as opposed to measures situated at the time of the stressful event.
long-term goals
Earthquake

Researcher
Perspective

In

these

early stages of

exploration

it was helpful
to couple the
psychological,
physiological &

environmental

dimensions.



photo every 30sec
Its always about

people [data]


Student

Perspective


Physiological

measures
had a profound
impact on

their understanding
of STRESS
Feild of Higher Education

Academic Developer

40|40|20

NZ Funding for research
in this area?
Undergraduates | Postgraduates | Faculty
background
study...
aWear 2016 - Stanford University
Privacy: Photo removed
Privacy: Photo removed
Privacy: Photo removed
Privacy: Photo removed
essential
Tension
between
Psychological
Physiological
Full transcript