Announcements, Reminders, & Upcoming Events
|
- Welcome back! Welcome to the Fall 2019 Semester and a new round of BRITE Blasts. As we start this new semester, if you have ideas for BRITE blasts or other BRITE initiatives, please be in touch with Justin.
|
|
- Rolling application procedure for BRITE Lab small grants this year (RFP). If you are interested in BRITE Lab small grants, please contact Lab Director Justin Sydnor with a short description of the study. Justin is willing to discuss preliminary ideas and he especially encourages graduate students or those newer to behavioral/experimental research to reach out at early stages of their study planning and conceptualization.
- Contact Felix to book time in the lab or indicate an intention to book. We currently have three studies scheduled to run in Fall 2019. Current bulk of dates open: the last two weeks of September and October, third week of November and December. Please check the calendar for full schedule.
- Send interesting things to Justin and/or Felix. If you see any papers, conferences, graduate student workshops, resources, etc... that might be of interest to the broader BRITE Lab community, please send along the info to Justin and Felix. Please also check below for our new thoughts into the IDEA section.
|
|
Michael Masterson,
Ph.D. Candidate of International Relations, Department of Political Science, recently ran an experiment in the lab.
Michael's study is on the effects of the emotional state of humiliation on international conflict preferences. He theorizes that humiliation increases individuals' preference for conflict by decreasing their sensitivity to the cost of conflict. In the study, he uses autobiographical essay tasks to treat humiliation. Respondents play an incentivized game where they must decide either to attack or not and the cost of war is randomly assigned to be either high or low.
|
|
The first Behavioral Operations Management Summer Institute for PhD students — co-organized by Ryan Buell (Harvard), Steve Leider (Michigan), and Jordan Tong (Wisconsin) — was held on June 10-14, 2019, at the University of Michigan. Generous support for the institute was provided by Michigan, Harvard, and Wisconsin (through the BRITE lab). The 1-week intensive program is designed to provide PhD students who are interested in behavioral research a solid foundation to conduct behavioral operations management research. Additionally, the summer institute encourages collaboration and the creation of a research community among the next generation of researchers in the field.
|
|
This inaugural year was a big success! We had 29 PhD students from most top US schools (Harvard, Stanford, Wharton, Columbia, MIT, Michigan…etc), as well as from top schools outside the US (Europe, Isreal, and China.) Thanks to the BRITE Lab for investing in advancing behavioral research for the next generation of scholars in the field. We plan to hold the summer institute every other summer rotating locations among Michigan, Harvard, and Wisconsin.
|
|
We will use this space each week in the news blast for Justin to share new ideas he thinks may be of interest for our behavioral & experimental community. If you ever have ideas for topics, please share your ideas with
Justin
.
This Week's BRITE Idea:
For the first BRITE Ideas post of the fall, I (Justin) wanted to highlight
one of
my
own studies
that was run both in the BRITE Lab and online with Mechanical Turk subjects.
Many studies attempt to estimate risk attitudes by asking people to make choices over lotteries. These choices can then be fit to structural models to estimate parameters like classic (utility curvature) risk aversion, probability weighting and loss aversion. My coauthors and I tested whether these models can predict decisions people make over insurance choices. In theory, a good model of risk preferences should be informative to how people make insurance choices. So we conducted an experiment where people chose over lotteries designed to allow us to calibrate a big number of common utility models (e.g., expected utility, prospect theory) and then also had to decide how much of a potential loss they wanted to insurance at different prices and probabilities of loss.
The punchline finding is that none of the structural models do a good job at predicting insurance choices. In fact, they typically perform worse that just random choice in predicting what people choose in the insurance task. There are a couple of key reasons for this, maybe most generally interesting that the models estimated from lottery choices predict that people should respond a lot more than they do to the price of insurance. Ultimately we think this study suggests a) that there are some deeper decision processes (e.g., heuristics) at play when people decide about insurance that are not captured by models of risk attitudes and b) the exercise of estimating models of risk preferences from experimental data may have extremely limited external validity. Of side interest for our group, we find more noise in M-Turk samples than BRITE Lab, but otherwise the patterns of choices and preference parameters look very similar.
|
|
Interested in booking research time in the Lab?
|
|
Volume 32 | September 7, 2019
|
|
Copyright © 2018 Justin Sydnor & Felix Zhan, All rights reserved.
|
|
|
|
|
|
|
