Decision Science News

THE PSYCHOLOGY OF SUNK COSTS

The sunk cost effect is sometimes called the Concorde fallacy

Hal Arkes and Catherine Blumer asked a group of 48 people the following scenario involving sunk costs:

As the president of an airline company, you have invested 10 million dollars of the company’s money into a research project. The purpose was to build a plane that would not be detected by conventional radar, in other words, a radar-blank plane. When the project is 90% completed, another firm begins marketing a plane that cannot be detected by radar. Also, it is apparent that their plane is much faster and far more economical than the plane your company is building. The question is: should you invest the last 10% of the research funds to finish your radar-blank plane, yes or no?

Then asked another group of 60 people this very similar scenario, which one will notice has no sunk costs:

As president of an airline company, you have received a suggestion from one of your employees. The suggestion is to use the last 1 million dollars of your research funds to develop a plane that would not be detected by conventional radar, in other words, a radar-blank plane. However, another firm has just begun marketing a plane that cannot be detected by radar. Also, it is apparent that their plane is much faster and far more economical than the plane your company could build. The question is: should you invest the last million dollars of your research funds to build the radar-blank plane proposed by your employee?

Which did people choose?

Researchers refer to this phenomenon as the “sunk cost effect … manifested in a greater tendency to continue an endeavor once an investment in money, effort, or time has been made.” One could argue that the decision to complete the project shouldn’t be influenced by the information which differs between the sunk cost and non-sunk cost scenarios, and yet the data show that the framing matters very much to people.

To see 9 (!) other studies on the sunk cost effect, some involving real-world decisions such as to use pre-paid season tickets, see Arkes and Blumer, below.

Reference:
Arkes, H. R. & Blumer C. (1985). The Psychology of Sunk Cost. Organizational Behavior and Human Decision Processes, 35, 124-140.

HARVARD CENTER FOR RISK ANALYSIS QUIZ ON CAUSES OF DEATH

The folks at Risk World let me know about this mini-quiz on odds of various causes of death at Harvard’s Center for Risk Analysis.

It’s a lot more fun if you take the time before going to that page to complete the following quiz.

The annual risk of dying from the following causes is 1 in ___:
heart disease ___
cancer ___
stroke ___
accidents of all kinds ___
motor vehicle accident ___
Alzheimer’s disease ___
alcohol ___
suicide ___
homicide ___
food poisoning ___
drowning ___
fire ___
bicycle accident ___
lightning ___
bioterrorism ___

Now go take the quiz.

Risks of death are a popular topic among judgment and decision making crowd, the most common reference being to:

Slovic, P., B. Fischhoff, and S. Lichtenstein (1982), Facts Versus Fears: Understanding Perceived Risk, in D. Kahneman, P. Slovic, and A. Tversky (eds.), Judgment Under Uncertainty: Heuristics and Biases. Cambridge: Cambridge University Press.

WHEN TO STOP INFORMATION SEARCH?

Today’s highlight article comes from Dan Gilbert’s mention of it in his recent New York Times Op-Ed piece:

“Two psychologists, Peter Ditto and David Lopez, told subjects that they were being tested for a dangerous enzyme deficiency. Subjects placed a drop of saliva on a test strip and waited to see if it turned green. Some subjects were told that the strip would turn green if they had the deficiency, and others were told that the strip would turn green if they did not. In fact, the strip was just an ordinary piece of paper that never changed color.

So how long did subjects stare at the strip before accepting its conclusion? Those who were hoping to see the strip turn green waited a lot longer than those who were hoping not to. Good news may travel slowly, but people are willing to wait for it to arrive.

The same researchers asked subjects to evaluate a student’s intelligence by examining information about him one piece at a time. The information was quite damning, and subjects were told they could stop examining it as soon as they’d reached a firm conclusion. Results showed that when subjects liked the student they were evaluating, they turned over one card after another, searching for the one piece of information that might allow them to say something nice about him. But when they disliked the student, they turned over a few cards, shrugged and called it a day.”

For over a quarter century we’ve known that people use heuristics to make decisions. The challenge of the next quarter century is to figure out what governs which heuristics are pulled from the toolbox when. When do we choose between a heuristic that cuts search short and one that digs through information? These kinds of considerations, and not fat (as in overweighted) linear models, are a start.

Reference: Motivated Skepticism: Use of Differential Decision Criteria for Preferred and Nonpreferred Conclusions. Peter H. Ditto and David F. Lopez. Journal of Personality and Social Psychology 1992, Vol. 63, No. 4, 568-584.

MOTIVATION AND AFFECT IN DECISION MAKING CONFERENCE, DEAD SEA, ISRAEL, DECEMBER 12-15 2006

Keynote speaker: Nobel Laureate Daniel Kahneman
Venue: Crowne Plaza Hotel, Ein Boqeq, The Dead Sea
Submission deadline: July 10, 2006

The Decision Making and Economic Psychology Center at Ben-Gurion University of the Negev and the Behavioral Decision Making group at the UCLA Anderson School of Management invite submissions of papers for the above conference. The conference will include mostly one-track talks by invited participants. Submitted papers, if accepted, will be assigned to either a parallel session section or a poster session. While most of the invited talks focus on the relationship between affect, motivation, and decision making, submission of papers in other areas of decision making is also possible.

For more information about the program, invited participants, registration and other relevant issues, please consult the conference website, at http://www.bgu.ac.il/~dmep

DECISION SCIENCES TENURE TRACK JOB AT IESE

Spain has some of the finest decision scientists anywhere. Consider living the good life and applying for the following post.

The Decision Sciences department at IESE Business School (Barcelona) invites applications for a tenure track position (assistant professor level). Applicants should have a research interest in managerial decision making, broadly defined, and show ability do mathematical modelling as well as experimental research work. IESE is an international business school with top-ranked MBA and executive programs. The school has campuses in Barcelona and Madrid, and offers regular programs in Munich and Sao Paulo.

Interested candidates should send a cv, research statement, and representative paper/s by E-mail to mbaucells at iese.edu. Candidates should also arrange for two or three letters of recommendation to be sent separately. Review of materials will begin on May, 5th, 2006, and will continue until the position is filled.

ASSESSING RISK TOLERANCE WITH A SIMPLE GAMBLE

Want to estimate risk tolerance with just one question? Thanks to Bob Clemen who has written a nice piece about it in the Decision Analysis Newsletter, and to the SJDM mailing list, here’s the one question:

“Suppose you face a gamble where you can win $x with probability 0.5 or lose $x/2 with probability 0.5. What is the largest x for which you would be just willing to take this gamble? I’m looking for the largest x that makes you just indifferent between taking the gamble or not.”

In Clemen’s book Making Hard Decisions, in the solutions manual, he gives a derivation of how to get from the respose to that question into an estimate of the risk tolerance parameter R in the exponential utility function U(x) = 1-exp(-x/R).

Another way to estimate risk tolerance is using the Distribution Builder

WHO TALKS TO WHOM: INTRA- AND INTERDISCIPLINARY COMMUNICATION OF ECONOMICS JOURNALS

A talk by Jeffrey Pfeffer currently touring the world cites some stats from Who Talks to Whom? Intra- and Interdisciplinary Communication of Economics Journals, a 2002 paper by Rik Pieters and Hans Baumgartner:

” * 90% of citations in economics is intradisciplinary
* The 5 economics journals studied made no citations to management, marketing, anthropology, or psychology journals
* Economics is cited 4333 times by its sister disciplines between 1995 & 1997, while economics cites other disciplines (mostly finance at 79%) 601 times”

ABSTRACT of Pieters and Baumgartner:
Citation patterns between 42 journals in economics from 1995 to 1997 are examined, plus between economics and anthropology, political science, psychology, sociology and five business disciplines. Building on social network theory, we identify a hierarchical organization of journals in economics and seven journal clusters. Major citation flows are found from all areas of economics to the general interest and theory and method clusters, but not the other way around. Economics emerges as a significant source of interdisciplinary knowledge for the other social sciences and business. However, no area of economics appears to build substantially on insights from its sister disciplines.

Also note this recent piece in Foreign Policy by Moisés Naím suggest economics is overly inward gazing given its limited ability to preform its bread and butter task of forecasting the future(a point also made in a forthcoming book by Nassim Taleb).

AUTHORS:
Jeffrey Pfeffer
Rik Pieters
Hans Baumgartner

R STATISTICAL LANGUAGE MULTI-SITE SEARCH

Image (C) R Foundation, from http://www.r-project.org

Decision researchers and fellow bloggers Jon Baron and Andrew Gelman are big fans and supporters of the R project for statistical computing.

Searching for information on R can be difficult (though Baron’s R search tool is a great help), so DSN has put together a search widget that only queries R sites:

R Statistical Language Multi-Site Search

Simply click the image above to search and see the results from many sites. Sites searched include:

finzi.psych.upenn.edu
cran.r-project.org
wiki.r-project.org
lib.stat.cmu.edu/S
lib.stat.cmu.edu/R
tolstoy.newcastle.edu.au/R

and DSN welcomes suggestions for additions and deletions.

INFORMATION SCIENCE

When DSN was visiting Stanford’s Department of Management Science and Engineering in the late 90s, it was called “Engineering-Economic Systems and Operations Research”. A change for the better, no? One thing that hasn’t changed is the excellence of the people there.

We just got our copy David Luenberger’s new Information Science. Not only is it a handsome book (big but not heavy, with cottony paper), it’s like an entire college education on a field you never knew existed, looking at everything from file compression to marketing to microeconomics through one beautiful framework set forth by Claude Shannon in 1949. It includes a nice Shannon quote, from 1953:

The first [method] I might speak about is simplification. Suppose that you are given a problem to solve, I don’t care what kind of problem—a machine to design, or a physical theory to develop, or a mathematical theorem to prove or something of that kind—probably a very powerful approach to this is to attempt to eliminate everything from the problem except the essentials; that is, cut it down to size. Almost every problem that you come across is befuddled with all kinds of extraneous data of one sort or another; and if you can bring this problem down into the main issues, you can see more clearly what you are trying to do an perhaps find a solution. Now in so doing you may have stripped away the problem you’re after. You may have simplified it to the point that it doesn’t even resemble the problem that you started with; but very often if you can solve this simple problem, you can add refinements to the solution of this until you get back to the solution of the one you started with.

Luenberger comments “Shannon’s approach of abstraction to an essence should become clear as we study his contributions throughout this text. His work is a testament to the power of the method.”

REFERENCE:
Shannon, Claude E. Creative Thinking. Mathematical Sciences Research Center, AT&T, 1993.

WEB EXPERIMENTS MUST BE GOOD WEB SITES

To do a good web-based experiment, a researcher needs to keep the drop-out rate as low as possible. Good Web design is key to getting people to take experiments seriously and to finish what they start. By encouraging good coding practices, it also improves cross-platform and cross-browser performance.

DSN likes the simple heuristics-based approach of the book Defensive Design for the Web by 37signals, Matthew Linderman, and Jason Fried, which contains 40 practical guidelines for how to improve a site.