Data Leveraging Partnerships to Improve Mass Notification Policies Lauren Stienstra MSc CEM Senior Manager Policy amp Research Arlington County OEM Agenda From Problem to Research to Results ID: 590009
Download Presentation The PPT/PDF document "Alerting +" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Alerting +Data = ?
Leveraging Partnerships to Improve Mass Notification Policies
Lauren Stienstra, M.Sc., CEM
Senior Manager, Policy & Research- Arlington County OEMSlide2
Agenda
From Problem to Research to ResultsResearch ProcessResultsSolution Development
Implementation & Lessons Learned
What’s Next?Slide3
Research Questions
HOW MANY ENROLLMENTS DO WE NEED?Who and Where are our people?
How do alerts affect behavior?
How do keywords/topics of the alerts affect enrollment?
How many people need to be enrolled?
Quality vs. quantity
Who shares information? Slide4
Where are our people?
Raw enrollments
Allotment by civic associationSlide5
Better Answer:
Heat map of counts within census tract (yellow borders)
Weighted by population density
Arlington National Cemetery and Reagan Airport tracts removedSlide6
By Location Name
Location name: a write-in option to describe location receiving alerts
Work, office,
ofc
,
trabajo
, travail
Most common: 2100 Clarendon Blvd. Slide7
Alerts: What are We Saying
Analyzed word frequency in all of the alerts sent out in the past two years
Most common words:
Warning
Watch
Severe
IssuedSlide8
Alerts: What do People React To
Analyzed word frequency of alerts where users replied “STOP”.
Most common words:
Traffic
Open/Reopened/Closed
Street
Lanes
FinalSlide9
How Should we Say it?
3 & 30 Principle (Chandler, 2010)Optimal messages relay key information in 3 short sentences with no more than 30 words
6 & 60 Principle (Chandler, 2010)
60 reading ease score & 6th grade comprehension level
Be honest, transparent, and direct
Include actions for people to take in response to the emergencySlide10
Who Shares Information
Intrinsic and extrinsic motivations influence member’s intention to share (Yeon et al., 2016)
If there’s motivation, people will share
People who enjoy helping others are more likely to share
Sharing is less likely if the information is widely available (Park, 2014)
If it’s easy, people will share information (Park, 2014)Slide11
WHAT ARE THEY LIKE?
Openness to experience: active imagination, aesthetic sensitivity, attentiveness to inner feelings, intellectual curiosity, and independence of judgment.
Agreeableness
: good-natured, forgiving, courteous, helpful, generous, cheerful and cooperative.
Conscientiousness
: dutiful, dependable, responsible, hardworking, and achievement-oriented.
Extroversion
Propensity to trust
Affective commitment
: emotional attachment to, identification with, and involvement in the organization and its goals.Affiliative tendency- preference for friends and attachments vs. independence and preference for group vs. individual activities. Public individuation, but only when there is high level of interpersonal interaction and they are given personal credit for their contribution. Slide12
Strong Interest Inventory Results
Teaching/Training (middle, high, elementary) Counseling (religious, medical, career, etc.) Restaurant/Arts/Entertainment Managers
Medical (Nursing, Allied Health, etc.)
Sales (especially insurance)Slide13
What did we do with this Information?
Based on per capita account enrollment distribution, direct outreach programming to the civic associations adjacent to Columbia Pike, as well as Hall’s Hill and Nauck
, for the next calendar year.
Based on the psychological profile of information sharing, target marketing towards the identified careers and community rolesSlide14
What did we do with this Information?
Based on the “STOP” message vocabulary assessment, evaluating how the number of traffic messages can be effectively reduced.
Based on the literature review of notification fatigue, review messaging policies and refrain from extraneous messaging (esp. around automatic weather alerts).Slide15
Lesson Learned:
Based on the data collection itself, improve the way data is collected, organized, and structured within the system. Slide16
Enrollment Campaigns
Putting Results to the TestSlide17
National Preparedness Month
Civic Association Contest“Check Your Prep” audits at Sugar ShackEmergency Kit Cook Off
ARLNow
posts
Preparedness Displays at 3 LibrariesSlide18Slide19Slide20
Distribution of New Accounts
Improvement!
Fairlington
Dominion Hills
Ballston-Virginia Square
STILL NOT WHERE WE WANT TO BESlide21
“Clicks for a Cause”
Pro-social behaviors can be leveraged when it’s easy to do:People like performing altruistic behaviors, especially when it’s easy/low cost
People engage in internet activity when it’s easy (Facebook likes, retweets, etc.)
Non-profit community has been able to leverage this for pro-social behaviors:
“Click this button to donate food to an animal shelter” on a page of ads
Page views/ad revenue fund the charitable work
Essentially, I got you to look at ads because you wanted to save the animals.
Can we use a similar system?
Translate pro-social impulses into Arlington Alert Enrollments?Slide22
Simplicity
Enroll in Arlington Alert, and we’ll make a donation on your behalf (enrollee gets to choose!):
Help yourself, help your community.
A Can of Beans
(AFAC)
A Pair of Socks
(ASPAN)
A Cup of Food
(AWLA)
A Box of Crayons
(APS/PTA)Slide23
Unexpectedness
Is this unique?Not in the non-profit world…
Greater Good suite of Websites
Heifer International
OXFAMSlide24
Concreteness
“Price Framing and Rewards” (Donor Psychology Term)Allows us to use the following persuasive forces:Visual Imagery (more motivating)
Stories
Value Attribution
Ownership BiasSlide25
Results
Only 250 new enrollments!Slide26
POSTmortem: Why did it fail?
Enrollment still too burdensomeLittle message amplification from partners (AFAC, etc.)
Competing messages/causes during “season of giving”
Privacy Concerns
“Preaching to the Choir” -> the people we target are already enrolled
Manpower/Short timelinesSlide27
But…
Average Monthly Growth (previous 7 months): 165Average County Population Growth: 179
Total population in Arlington 2016: 220,400
Total enrollments: 23,661
= 10.7%
Is this a ceiling?Slide28
What’s next?
Continued Research with Virginia TechCollaboration in the RegionDevelop new, more relevant strategic measures
Continue high impact, precisely targeted outreach