Michael H. Birnbaum, PhD - US grants

This proposal outlines an ambitious investigation of behavioral theories of decision making with the goal of building a coherent method for measurement of utility and value. Previous studies of expected utility have produced two results: First, a variety of significant and reliable violations of utility theory have been reported. Second, several variations of the standard theory have been proposed to account for these violations. The proposed research is designed to begin the construction of a behaviorally- based coherent framework. A central idea behind the proposed research is the search for a common utility scale underlying behavior in a variety of tasks. The goal is to determine whether it is possible to provide a coherent system of utility equations to describe behavioral decision making. Procedures developed by the principal investigator in other domains will be applied to develop a framework for the research. The proposed approach has the potential to account for a wide range of phenomena and to resolve several controviersies in the literature. Even if the proposed effort does not work out, the experiments will yield interesting new findings which will enhance understanding of risky decisions. In all, the potential return from this project is substantial.

ABSTRACT Birnbaum, Michael H. California State University at Fullerton SBR-9410572 In collaboration with Ward Edwards (SBR-9409698) the PI proposes to study risk attitudes. The PI will investigate the behavior and attitudes of people who participate in "The Big Spin," which is the California State Lottery's TV show. The PI will be given access to the participants before, during, and after their participation in the televised portion of the lottery. The risk attitudes of these persons will be compared at various time periods, and their responses will also be compared to the responses of a control group of non-participants. A one-year follow-up of all participants is included in the design.

Abstract
***
9986436
Michael H. Birnbaum

The Internet provides a new domain of human experience. Estimates of
the growth rate of penetration of the Internet among the population
shows that its use is growing faster than previous innovations
(telephones, radio, and television) that have already become nearly
universal in industrialized nations. The majority of the U.S.
population will soon be using the Internet for communication, commercial
transactions, and entertainment. The Internet also provides an exciting
new medium for the conduct of behavioral research. By means of Internet
research, it is now possible to accomplish in a few weeks what used to
require six months or a year. Because the World Wide Web makes
available an extremely heterogeneous, international sample, it becomes
possible to examine the generality of laboratory research to demographic
groups that have rarely been studied in previous research. This allows
one to investigate if conclusions from laboratory research are unique to
college sophomores tested in university labs or if they have external
validity to a larger population tested under more naturalistic
conditions.
Are the cognitive processes used by the general population on the
Internet the same as those used by students in university laboratories?
There are two aspects to this question that need to be distinguished.
The first is that the people tested are different, so any difference in
results may be due to differences in education, age, wealth, or other
demographic differences among the groups tested. The second aspect is
that the Internet medium itself may itself induce different modes of
thought, reasoning, judgment, and decision-making. This research
explore these questions.
This is a proposal to continue and expand an on-going program of
research on the Internet. In addition to experiments on judgment and
decision-making on the Internet with 6,000 participants, I propose
to conduct surveys of behavioral scientists about the Internet and to
create tools that will make it easier for students and newcomers to
Internet research to adapt their studies to this new medium. By
comparing lab and Web studies, the validity of both procedures can be
tested. By comparing the results for different demographic groups, it
is possible to examine the linkage between such characteristics and
research findings. If behavioral research conducted via the Internet
lives up to its potential, then the process of facilitating the transfer
of such research to the Internet can result in acceleration of the
scientific enterprise.
***

This Advanced Training Institute in Social Psychology will provide training on the new methods and techniques by which social psychological research can be conducted via the internet. The internet allows for rapid collection of large samples of data with minimal costs associated with printing, mailing, testing, lab space, lab assistants, and data coding/entry. The Internet can also be used as a device for recruiting participants from different cultures or participants who have special characteristics that might be rare in the world's population. The training institutes will be held at California State University, Fullerton.

This proposal studies changes in opinion and judgment that follow as consequences of the ongoing "War on Terrorism," declared in response to the terrorist attacks of September 11, 2001. It is anticipated that military and diplomatic success and failure and renewed terrorist events will influence opinions regarding security versus liberty. The research will be done in the lab, by national telephone polling, and via the Internet using several of the new WWW methods for conducting research.

The primary goal of the research is to study models of how biasing phrases affect judged agreement with questions posed in polls. It has long been known that the results of polls depend strongly on how questions are worded. This research will attempt to separate the effects of bias from those of more stable "true" psychological values.

A secondary goal is to compare results obtained with inexpensive, efficient, new Internet methods to those obtained from more standard polling. This project will also attempt to determine which of several Internet methods of sampling is most accurate in matching the results of more standard polling.

SES - 0721126
Advanced Training Institutes: DRMS Research via the WWW
Birnbaum, Michael H.
California State University at Fullerton
This project will train scientists on how to use the World Wide Web (WWW) to conduct research in management, decision making, and economics. The Internet allows for rapid collection of large samples of research data with minimal costs associated with printing, mailing, testing, lab space, lab assistants, and data processing. One can also use the WWW to recruit participants from diverse demographic groups or to recruit people who have special characteristics that may be rare in the population. The methods facilitate a faster pace of scientific research and improved communication among scientists. The provision of advanced training is important because many scientists received their education before these modern computer techniques were developed; this grant will provide for advanced instruction in techniques that will help scientists to be more productive.
A board of six experts in this field will plan and present instruction at institutes with about 15 participants on three different occasions, approximately one year apart. The board will also plan some shorter sessions, to be held as one-day workshops to follow scientific meetings. These will have just two instructors and will be targeted to specific needs. We anticipate that trainees will not only be able to improve their own productivity but that many of our trainees will share these techniques with their students and colleagues. In addition, Web experiments help publicize scientific research to the general public, who are able to participate in research and communicate directly with scientists.