My research agenda is founded on two broad
interests. First, I aim to understand decision-making under different
information settings and environments. Second, I seek to identify the
motivations for the economic choices that individuals make in different
contexts. I am an applied microeconomist and primarily use the
methodology of
experimental economics to answer research questions. I have training
and
experience in both laboratory and field experiment approaches, and
believe that
the two are complementary. I also believe in the value of
interdisciplinary
research for advancing the science of economics. I developed my research skills under the
supervision of Professor Tim
Cason as a Ph.D. student, and through my experience working with
Professor John
List as a Postdoctoral Scholar. I developed my knowledge in visual
analytics
through collaboration with Professor David S. Ebert.
My
research interests are in Experimental and Behavioral Economics, Public
Economics, Applied Economics, Health Economics and Visual Analytics. I believe that research that
is worth pursuing both advances the theory and practice of economics
and
provides actionable implications for organizations and policy makers.
I have had excellent opportunities to explore
decision-making
in many contexts, so my research projects are varied in scope.
Nevertheless, I
have a clear agenda for the direction of each of my ongoing areas of
interest:
Information
Settings.
I am interested in the effect of social
information
and reputation systems on behavior, and explore these effects in the
laboratory. In my job market paper (Savikhin and Sheremeta, 2011a) we
investigate the impact of recognizing group members on public good
provision,
and find that this social information plays a crucial role. This
research was
primarily my idea, and I also implemented the study (writing the Z-Tree
program, running the experiments) and played a leading role in writing
up the
paper. In a complementary paper, we study the effect of recognition on
contest
bids (Datta Mago, Savikhin and
Sheremeta, work in progress).Social
information also plays a role in a randomized e-mail message study we
conducted
on an online site (Klimeck and Savikhin, 2011). I have also studied the
effect
of reputation systems in exchanges with asymmetric information, and
found that
such systems are effective for high-valued, rather than low-valued,
goods
(Savikhin, 2011a).
In a series of papers, my co-authors and I
have
experimentally investigated the effect of participating in different
environments simultaneously in coordination games, public goods
environments,
and contests (Cason, Savikhin and Sheremeta, 2011; Savikhin and
Sheremeta,
2011b; McCarter, Savikhin and Sheremeta, work in progress; Krieg and
Savikhin,
work in progress). We find that participating in different environments
simultaneously or in sequence changes behavior relative to each
environment in
isolation – this is termed ‘behavioral spillover.’
I plan to continue my work in this area by
investigating additional ways of recognizing contributors to public
goods, and
want to discover the effect of differing beliefs about whether one is
being
observed. Together with Sheremeta, we plan to pursue a thorough
investigation
of the channel through which ‘behavioral spillovers’ occur in
simultaneous
decision-making settings.
[Job
Market Paper]Visibility of
Contributors and Cost of Information, with Roman Sheremeta, Under Review.
Motivation for
Contributions in an Online Scientific Community: Virtual Rewards,
Social Messages and 'Observation Cues,' with Gerhard Klimeck, Under review.
At the University of Chicago, I have been
spearheading the research program that is part of the Griffin Early
Childhood
Center (GECC). GECC is an innovative field experiment launched by John
List,
Steven Levitt and Roland Fryer, the goal of which is to test the effect
of
early childhood interventions on low-income households, either in the
form of a
full-day, free preschool or in the form of incentivized Parent Academy
classes.
570 families participated in GECC and in our experiments in Year 1, and
900
families are participating this year (Year 2).
In my (secondary) job market paper, I
designed and
implemented a field experiment with 4-5 year-olds to measure the gender
gap in
competitiveness (Savikhin, 2011b). I find that a gender gap between
girls and
boys exists even at this young age, but disappears with experience.
The bulk of the research at GECC is in
progress
since the program started in 2010. In one study, we are working to
disentangle
warm glow and pure altruism as giving motivators among 3-4 year old
children,
and to measure the level of racial discrimination (List and Savikhin,
work in
progress). I have spearheaded the design of the experiment, run pilots,
implemented
the study, and played a leading role in writing up the paper. We are
currently
conducting additional studies with older siblings. Other research that
we are
working on includes the propensity of parents to cheat in tasks when
children
are present or absent (Houser, List, Piovesan, Savikhin and Winters,
work in
progress), measuring and correlating the risk and time preferences of
young
children, older siblings and parents (Andreoni, Fryer, Levitt, List,
Savikhin
and Charlie Sprenger, work in progress), and measuring the effect of
social
information on children’s giving behavior (Ben-Ner, List, Putterman and
Savikhin, work in progress). In the studies discussed above, I have
managed a
team of RAs to implement the study, have fully developed or adjusted
the
instructions for the study, and have assisted with data analysis and
write-up.
I plan to continue these and other
experiments in
coming years. I am interested in how competitiveness evolves and the
problem of
developing interventions to reduce the gender gap.
[Secondary
Job
Market
Paper]Is there a Gender Gap in Preschoolers'
Competitiveness? An Experiment in the U.S.Working paper.
Another of my research areas is to
investigate the
effect of incentives on health behaviors. In a working paper (List and
Savikhin, 2011) we report on a large-scale field experiment testing
whether
gain or loss incentives, or educational messages, influenced the
child’s choice
of a cookie versus fruit. 1,978 children and adolescents ages 6-18
participated
across 17 different after-school programs. For this study, I worked
with the
Greater Chicago Food Depository (GCFD) to secure funding by developing
a
research idea and research plan that was acceptable both to GCFD and
the
funding agency ($115,000 secured through a collaboration with the
Griffin
Foundation). I managed a team of over 40 RAs to implement the study,
and played
a leading role in analyzing the results and writing the paper. We found
that
both gain and loss incentives significantly increase the take-up of the
healthy
choice (fruit) but that the educational message does not have a
significant
effect.
As part of the GCFD projects, we also
administered
several surveys to parents and children in the Chicago area, including
a
standard Food Frequency Questionnaire (FFQ) and Food Insecurity
questionnaires.
In addition, we conducted about 400 food recalls with children ages
6-18. The
results of the surveys were reported in a brief to the GCFD and we are
planning
to publish some of our results on the impact of food choice on (self
reported)
educational attainment.
List and I recently received a grant from the
Cornell
Center on Behavioral Economics in Child Nutrition Programs to
investigate my
research idea of how different verbal prompts in the lunch-line affect
child’s
choice to select and consume a healthy side item. These experiments
will be
conducted in Spring 2012. I am also at the beginning stages of working
with a
health marketing company to develop and evaluate a reward and reminder
system
to encourage patient compliance in taking medication as prescribed.
[Secondary/Health
Economics
Job
Market Paper]The Impact of Incentives and Educational
Messages on Child Food Choice, with John List. Working paper.
Not for
General Web Distribution due to Funder Preference - Please E-mail me
Applied
Visual Analytics for Economic Decision-Making.
The study of how applied visual analytic (VA)
tools
can be used for decision-support in economic and financial-decision
making
tasks was one of my first interests as a graduate student. Interactive
visual
interfaces are ideal for visualizing information in cases where
individuals are
limited in their processing capabilities. I work with experts in VA to
develop
tools – my roles are to provide expertise in identifying the problems
that
require decision support and to develop experiments in the laboratory
to
evaluate the effect of the tool on the decision-making process.
We have received grants from the National
Science
Foundation and the Social Security Administration that have allowed me
to
expand my research in this area over the past 3 years (I served as
Co-PI or
Co-I). To date, we have investigated the effect of different VA tools
for
financial portfolio selection (Rudolph, Savikhin & Ebert, 2009;
Savikhin,
Lam, Fisher & Ebert, 2011a; Savikhin and Ebert, 2011) and
information
search (Savikhin, Yi, Hur, & Kim, 2011). We find that appropriate
VA
increases the optimality of the decision, decreases the time spent
deliberating, increases exploration of the data space, and increases
confidence. The theoretical model I have begun to develop proposes that
the
effect on decision-making occurs through a decrease in the cost of
effort,
which in turn increases exploration of the data.
I am working with Annamaria Lusardi
(Dartmouth) and
her team of researchers to evaluate how VA tools, static brochures,
videos and
narratives compare in their ability to convey key financial concepts
(Lusardi,
Kapteyn, Glinert, Savikhin, Heinberg, Hung, Yoong, work in progress). I
am also
working on a field experiment that explores how new low- and moderate-
income
homebuyers in Ohio can benefit from a complete suite of visual
financial
planning tools and phone-based financial coaching sessions (Collins,
Loibl,
Moulton and Savikhin, work in progress).
In the next few years, together with Tim
Cason, we
plan to explore how visual analytics can proxy for experience in the
reduction
of asset market bubbles. I also plan to collaborate with organizations
to
measure the effect of different VA tools for assisting consumers in
practical
choices, and am currently in discussions with a financial-services
communication firm to develop a tool for 401(k) planning.
The Effect of Interactive Visual Displays
on Financial Decisions: An Experimental Study, with David S.
Ebert. Working paper.
An
Experimental Study of Information Search and Decision-Making with
Interactive Technology, with Ji Soo Yi, Sung-Hee Kim and
Inkyoung Hur.Working paper.
Not for
General Web Distribution - Please E-mail me
An
Experimental Study of Financial Portfolio Selection with Visual
Analytics for Decision Support, with Hon Cheong Lam, Brian
Fisher and David S. Ebert. Hawaii
International Conference on Systems Sciences.
Applied Visual
Analytics for Personal Financial Planning, with Stephen Rudolph
and David S. Ebert, IEEE Symposium
on Visual Analytics Science and Technology.
The
Application of Visual Analytics to Financial Decision-Making and Risk
Management: Notes from Behavioral Economics, Financial Records and Information
Management, forthcoming.
