Relationship and Sexual Violence Prevention
Peter Bode ke Saat a Asher,
Katie Chew, Sarah Pritchard, Shih-Ving Cheng, Jill Kuhlberg, & Patrick Fowler A
Washington University in St. Louis . oe :
ian ec er creas
ugust Jn ’ :
eee ST: a i ana ES
Acknowledgements
¢ Provost’s Office
* Holden Thorpe
* Sharon Stahl
¢ Adrienne Davis
¢ Institute of Public Health
¢ Center for Violence and Injury
Prevention
* Relationship and Sexual Violence
Prevention Center
¢ Social System Design Lab
= Washington
University in St.Louis
INSTITUTE FOR PuBLIC HEALTH
Dd
Center for Violence and Injury Prevention
9 Relationship & Sexual
Violence Prevention Center
STUDENT AFFAIRS AT WASHINGTON UNIVERSITY
ff
Al
Social System Design Lab
Overview
1. Background to the problem of relationship and sexual violence
prevention on university campuses and Washington University’s
response
2. Structural violence and then need for new methods
3. Conceptual individual level model of resilience in response to
insults
4. Next steps and future work
Timeline
Release of AAU
campus climate
survey
(Sep ’15)
\
/Launch of
a ~—~ Relationship and
( Sexual Violence
Dear Colleague \_ Sexual Misconduct Assessment
letter pserrcam Initiative
(April ‘11)
\
\_ Sexual Assault and
Relationships
Violence Task Force
«Policies and Processes
«Prevention and Education
Cy «Support and Advocacy
‘~~ Center for “Assessment
Violence and
Injury Prevention
*VTB project
(Matthieu, Co-P1)
*Scriptapedia
2009 2014 2015 2016
Lifetime prevalence of sexual assault by age and gender for persons who have attended college (N=9,079)
from analysis of National Violence Against Women (NVAW) survey
Cumutative
Lifetime prevalence of partner physical assault by age and gender for persons who have attended college
(N=9,079) from analysis of National Violence Against Women (NVAW) survey
Cumutative
Rationale
¢ With close to 50 percent of the US population attending four-year
institutions, prevention systems that show a demonstrated reduction in
sexual assault and relationship violence could have significant
population health impact.
* Universities have an innovative role in prevention of sexual assault and
relationship violence in other communities
* Data on population and services
* Dynamic population
* University as a “testbed” for designing and demonstrating an adaptive
prevention system
Goal: To develop a comprehensive assessment system for the
prevention and response to campus sexual assault and
relationship violence.
Specific aims:
1. Form transdisciplinary research teams to develop innovative solutions to
prevention and response to campus sexual assault and relationship violence;
2. Develop scalable methods for a comprehensive campus sexual assault and
relationship violence public health surveillance and evaluation of prevention
and response programs and policies;
3. Train the next generation of public health prevention specialists, direct service
providers (e.g., counselors, doctors), advocates and civic leaders to create
community systems that prevent and respond more effectively to sexual
assault and relationship violence at the community level.
Structural violence as systemic patterns
When one husband beats his wife
there is a clear sense of personal
violence, but when one million
husbands keep one million wives in
ignorance there is structural violence.
Johan Galtung (1969). Violence, peace, |.
Violence
and peace research. Journal of Peace level
Research, 6(4), p.171
Violence as systemic, distributional
versus structural injustice, and
concept of thrownness of social
groups.
Iris Young (1990), Five faces of
oppression. In Justice and the politics
of difference. Princeton, New Jersey:
Princeton University Press
Personal violence or event
Pattern or
structural violence
Ci
Time
(Redrawn from Galtung)
Need for methodology (methodology = study of methods)
Events Populations and outcomes
Practice innovations
Patterns over time (applications)
Structural insights
Structure (methods)
Methodological
innovations
(methodology)
Values, attitudes, and norms
Two major methodological problems in studying
relationship and sexual violence
* Time delays = right censoring of data and biases in
underreporting
* Dynamics of identity labels = biases in reporting and assessing
risk of marginalized populations
* Constructs tied to vulnerability and risk changing quickly in a dynamic
population
¢ Hence, missing data and not missing at random
Time delays in recognizing and self-reporting
victimization experiences (i.e., right censoring)
Number of Respondents in Physically Abusive Relationships by Year
arnt nnn nnn nnn nnn nnn nn nn nn nnn nnn ene grcsee 5 Decrease,
431 however, is
probably due
to censoring
as a result of
being in an
abusive
relationship,
ie, under-
reporting
Increase probably due to
84 distribution of
a bi
respondents’ ages
Number
1930 1940 1950 1960 1970 1980 1990
Year
Data from NVAW survey
Dynamics of identity and labels (i.e., not missing
at random)
shi: ‘SOCIOLOGY AND SYSTEM DYNAMICS
<<< °° * Scientific discourse relies on
N .
7 oi stata RES understanding labels as
= = Saas immutable
¢ Understanding how labels change
(“looping effect”)
¢ Changing social norms, process of
crescive legitimation
Cowan T, A LeBlanc. 2018. Feelings under dynamic description: the asexual spectrum and new ways of being. Journal of
Theoretical and Philosophical Psychology 38(29-41); Jacobsen C, H Law-Yone. Sociology and system dynamics. Dynamica
10(1): 2-8. (originally presented at the first ISDC at Chestnut, MA in 1983)
Ways to think about mathematical modeling
Like an engineer As a basic natural scientist
How do we solve a problem? How do we explain natural ph ?
E.g., Petroski (2011); Simon (1996) E.g., Newtown (1686); Lakatos (1970); Meehl (1990)
Two types of propositions in mathematical
modeling in a progressive program of research
1. Conjectures
Statements about what is logically entailed by the assumptions of the model
of a theory (what does the model “say” ?)
* Explored and verified through computer simulation
* Testing the dynamic hypothesis in system dynamics
2. Hypotheses
Statements logically implied by the model that can be empirically tested
* Comparing statements entailed by a model against empirical reality
Black M. 1962. Models and metaphors: Studies in the language and philosophy. Cornell University Press, Ithaca, NY.;
Lakatos |. 1970. Falisfication and the methodology of scientific research programmes. In Lakatos I., A. Musgrave (eds.),
Criticism and the Growth of Knowledge. Cambridge University Press, New York, NY, pp. 91-196; Bunge M. 1967. Scientific
research Il: The search for truth. Springer-Verlag, New York, NY.; Meehl PE. 1990. Appraising and amending theories: The
strategy of Lakatosian defense and two principles that warrant it. Psychological Inquiry 1(2): 108-141.; Ostrom E. 2005.
Understanding institutional diversity. Princeton University Press, Princeton, NJ.
System dynamics simulation modeling
1. Macrosystem view of population, risk, 2. Microsystem view of individual
prevention, and response trajectories
https://tinyurl.com/y75d7gsn https://tinyurl.com/y9f6jaua
Different responses to insults
Ee
==? Resilient;
Wellness
Event, Time
shock or
system
insult
Resilient,
Recovery
Incomplete
recovery
Delayed
* Chronic
Adapted from Bonanno, G. A., & Diminich, E. D. (2013). Annual Research Review: Positive adjustment to adi ity
resilience. J Child Psychology Psychiatry, 54(4), 378-401.
Resiliency model
Goal
Insults
on
)
¢
‘
B2
Developing
Coping
Skills
Wellness
Developing
Fractional Coping
Growth: Skills
Rate.
Effect of
Wellness on
Growth
Treatment
B3 Developing Learning Skills
Resilience Fraction Development
INIT Fraction
Fractional
Growth
Rate
Developing
Resilience
Example of an individual factual-counterfactual
comparison
No insults + treatment
Wellness
coe =
N No insults zt
an Microaggression + shock with treatment
ge aa 88
a | Man,
= oe WV A ADPOPOPONNINIAAAN AAA
Wea Microaggressions
| an
Microaggressions — cone
wo | + check = = 7
© Prenary + Secon
Prenary + Saconary mst
a
© T T T T T
ti) 12 24 36 48
As frequency of microaggressions increases, perceived impact
decreases while cumulative impact increases
Wellness
Se
N No insults
AA
© _| Vt ES = 5% and T = 2 months
= Va
2 [one ES = 5% and T = 2 weeks
=
ES = 5% and T = 1 week
2 J
o
ES = 5% and T =3.5 days
e4
°
Using the model to generate synthetic data for developing and
testing innovative resource allocation algorithms
Sarah Busmann, Neeharika Kotte, and Carley Maupin. (2018). Intelligently Segmenting the
Long Tail. Research mentor: Brendan Juba
Sarah Busmann Neeharika Kotte Carley Maupin Brendan Juba, PhD
Assistant Professor
of Computer Science
and Engineering
Next steps and future directions
* Using model to design/test research evaluation designs
* Brown School Evaluation Center leading effort to develop RSVP program
evaluation plan for prevention and response
¢ Educational supports for P-12
¢ Addressing capability traps in Tier 1, 2, and 3 needs and services
¢ AAU Campus Climate Survey
* 27 institutions
¢ Sampling size of 779,170 with 196,984 responses
¢ Extend to design of a more general diversity and inclusion model
Your invited!
#2 Washington University in St.Louis
Keynote speaker:
Ixsrrrue For Pusuic HEauTHt
e
indtiative Jody O'Sullivan
Professor & Dean of the
UMSL/Wash U Joint
Undergraduate Engineering
Program and The Samuel C.
Sachs Professor of Electrical
Innovations in Evaluation:
Expanding the Boundaries of
Privacy and Security through
Technology
Engineering
Agenda:
1-2 PM Keynote
2-3 PM Developing a
Comprehensive
Evaluation Plan
3- 4 PM Poster Session
For more information about RSV-AI: contact Peter Hovmand, PHD, MSW (phovman d@wustl.edu) or Sarah Pritchard,
rpritchard@wustl.edu) or visit https://publichealth.wustl.edu/r
and-sexual-violence-
