|
|
|
|
Bias, Bigotry, and Availability |
|
183 |
|
A separate experiment sought to determine how extreme traits in a small portion of a population might influence availability and thus stereotyping. Participants were told the height of 50 men, one at a time, and later asked to guess the number of them who were taller than 6 feet. In both treatments, exactly 10 of the men were taller than 6 feet, and the mean height was 5 feet 10 inches. In the extreme condition, the tallest man was 6 feet 11 inches, and in the control condition the tallest man was 6 feet 4 inches. In the control, people believed there had been about 10 men taller than 6 feet, and in the extreme condition they believed it was 15 men. Thus, because the same number were much taller than 6 feet, people thought more of them were taller than 6 feet. The extremeness of the height led to greater availability of the trait when evaluating the group. Similarly, examples of extreme behavior by even a small number in a group can lead to unfair stereotyping of the whole. Such biases would be difficult to eliminate when making subjective judgments of résumés with very little information about individual character or ability. Firms may be able to overcome such biases by using more-objective measures to make a first pass at résumés (e.g., highest degree completed, years of experience). If subjective decisions are held off until the number of candidates is small, and the decision maker has been able to interview each candidate, the effects of such stereotyping may be minimized.
History and Notes
Psychologist Ward Edwards began to examine Bayes’ rule as a behavioral model in the late 1950s. His interest was primarily to discover the cognitive processes behind information processing and the factors that could affect that process. His early work on conservatism was highly influential in the later work of David Grether, Daniel Kahneman, Amos Tversky, and others. Related work by Edwards examined how people misperceive probabilities in general, finding generally that people overestimate small probabilities and underestimate large probabilities. As you will see in later chapters, this finding became foundational in the study of behavioral decision under uncertainty. Edwards advised and mentored Amos Tversky. Edwards’s work fed directly into Tversky and Kahneman’s first study introducing the representativeness heuristic and later related work developing the concept of the law of small numbers. Ongoing work by economists has sought to incorporate these concepts into mathematical models and explore applications to economic decision making. The availability heuristic is a related concept whereby the small representative samples are the events that are most easily recalled. Although it is clearly an important determinant of economic behavior (see the earthquake example), less work has been done to formalize the availability heuristic in economic modeling.
|
|
|
|
|
184 |
|
REPRESENTATIVENESS AND AVAILABILITY |
|
|
|
|
|
|
|
|
|
|
|
|
Biographical Note
Ziv Koren/Polaris/Newscom
Daniel Kahneman (1934–)
B.A., Hebrew University of Jerusalem, 1954; Ph.D., University of California at Berkeley, 1961; held faculty positions at Hebrew University in Jerusalem, University of British Columbia, University of California at Berkeley, and Princeton University
Daniel Kahneman received his undergraduate training in psychology and mathematics and his Ph.D. in psychology. Following this undergraduate training, he served in the Israeli military, where he was assigned to evaluate the character of new recruits to determine whether they were
fit for officer training. He noted after some experience that their methods had very little predictive ability, yet that they were still used to determine admission to the officer ranks. He dubbed this puzzle the illusion of validity, and he later used this concept in his research. He credits many of the tasks he performed in the military as leading directly to the lines of research for which he has become so well known. He describes his early career as “humdrum,” consisting of rather tame research. When Amos Tversky by chance discussed the conservatism work of Edwards in one of Kahneman’s classes, a debate ensued regarding whether people jump to conclusions or fail to react to new information. From this debate, the collaborative effort of the two was born. Kahneman’s contributions to the psychology of judgment and decision making, primarily coauthored with Tversky, provided the foundation for much of behavioral economics. His later collaborations with Richard Thaler formally introduced heuristics and biases into economic models of decision making. More-recent work has focused on what makes an experience pleasant or unpleasant and how we recall experiences. This work has the potential to redefine our notions of utility and enjoyment. For his contributions to the development of behavioral economics, Kahneman won the 2002 Nobel Prize for economics. In addition he has won numerous prizes and honorary degrees. Kahneman attributes much of his curiosity about psychological phenomena to his upbringing. Though born in Palestine before the founding of Israel, he lived much of his boyhood in France. During World War II, as Jews in occupied France, his family moved often to avoid internment in prison camps. His father was at one point interned for six weeks, though he was later released through action by his employer. In this climate, his mother’s gossip—short stories about human behavior—served as his primary entertainment and curiosity.
|
|
|
|
References |
|
185 |
|
T H O U G H T Q U E S T I O N S
1.In this chapter, brief mention was made of the false consensus as a form of the availability heuristic. Consider an entrepreneur who has developed a product that she finds very useful in her own life. What might the false consensus have to say regarding her beliefs that the product is marketable to a more general audience? How might these beliefs affect her decision to invest in a new business venture distributing the product, and what impact will this have on the riskiness of her investment? Suppose we were to examine a large sample of entrepreneurs who each had developed products around their own needs. Given the false consensus, what types of entrepreneurs are most likely to succeed?
2.In 2003 Andy Pettitte pitched for the New York Yankees baseball team, a team that won the American League pennant and qualified for the World Series. In the postseason, the Yankees played a series of games with each of three teams: Minnesota, Boston, and Florida. In each series, Pettitte pitched the second game and won. A prominent sportswriter noticed this and wrote an article touting this notable streak of wins when pitching the second game of a series. Over the season, Pettitte had pitched in 29 games and won 21 of them. Is this a streak? Why might the sportswriter believe this is a streak? How could you profit from this perception? Model the sportswriter’s beliefs supposing that the individual has two mental urns. One urn (average) has three balls, with two marked “win” and one marked “lose.” Suppose the other urn (streak) also has three balls but all are marked “win.” Suppose that the urns are never refreshed. What is the lowest probability of a streak that would lead the sportswriter to interpret this series of wins as a streak? Suppose
R E F E R E N C E S
Bertrand, M., and S. Mullainathan. “Are Emily and Greg More Employable than Lakisha and Jamal? A Field Experiment on Labor Market Discrimination.” American Economic Review 94 (2004): 991–1013.
Clotfelter, C.T., and P. J. Cook. “The ‘Gambler’s Fallacy’ in Lottery Play.” Management Science 39(1993): 1521–1525.
Croson, R., and J. Sundali. “The Gambler’s Fallacy and the Hot Hand: Empirical Data from Casinos.” Journal of Risk and Uncertainty 30(2005): 195–209.
instead that the urns are refreshed after every two games. Now what must the unconditional probability of a streak be before one would believe one was observing a streak?
3.Suppose there is an unconditional probability of a bull market of 0.8, and a 0.2 probability of a bear market. In a bull market, there is a 0.7 probability of a rise in stock prices over a one-week period and 0.3 probability of a fall in stock prices over the same period. Alternatively, in a bear market there is a 0.4 probability of a rise in stock prices in a one-week period and a 0.6 probability of a decline in stock prices in a one-week period. Suppose, for simplicity, that stock price movements over a week are independent draws. In the last 10 weeks, we have observed four weeks with rising prices and six weeks with declining prices. What is the probability that you are observing a bear market? Suppose a cable news analyst behaves according to
Grether’s generalized Bayes’ model of belief updating, with βP = 1.82 and βL = 2.25. What probability would the news analyst assign to a bear market? Finally,
suppose a competing news analyst behaves according to Rabin’s mental urn model, refreshing after every two weeks of data. Suppose further that this analyst has 10 balls in each urn with distributions of balls labeled “rise” and “fall” corresponding to the true probabilities. What probability will he assign to a bear market? What if the analyst had 100 balls in each urn?
4.Many lotteries divide the winnings evenly among all those selecting the winning number. Knowing this, how could one use the gambler’s fallacy to increase the expected earnings from playing the lottery? Under what conditions would it be profitable to do so?
Edwards, W. “Conservatism in Human Information Processing.” In D. Kahneman, P. Slovic, and A. Tversky (eds.). Judgment under Uncertainty: Heuristics and Biases. New York: Cambridge University Press, 1982, pp. 359–369.
Folkes, V.S. “The Availability Heuristic and Perceived Risk.” Journal of Consumer Research 15(1988): 13–23.
Gilovich, T., R. Vallone, and A. Tversky. “The Hot Hand in Basketball: On the Misperception of Random Sequences.” Cognitive Psychology 17(1985): 295–314.
|
|
|
|
|
186 |
|
REPRESENTATIVENESS AND AVAILABILITY |
Grether, D.M. “Bayes Rule as a Descriptive Model: The Representativeness Heuristic.” Quarterly Journal of Economics 95(1980): 537–557.
Hogarth, R.M., and H.J. Einhorn. “Order Effects in Belief Updating: The Belief-Adjustment Model.” Cognitive Psychology
24(1992): 1–55.
Kahneman, D., and A. Tversky. “On Prediction and Judgment.”
Oregon Research Institute Research Bulletin 12(1972).
Kahneman, D., and A. Tversky. “On the Psychology of Prediction.”
Psychological Review 80(1973): 237–251.
Langer, E.J. “The Illusion of Control.” In D. Kahneman, P. Slovic, and A. Tversky (eds.) Judgment under Uncertainty: Heuristics and Biases. New York: Cambridge University Press, 1982, pp. 231–238.
Lichtenstein, S., P. Slovic, B. Fischoff, M. Layman, and B. Combs.
“Judged Frequency of Lethal Events.” Journal of Experimental Psychology: Human Learning and Memory 4(1978): 551–578.
Rabin, M. “Inference by Believers in the Law of Small Numbers.”
Quarterly Journal of Economics 117(2002): 775–816.
Rothbart, M., S. Fulero, C. Jensen, J. Howard, and P. Birrel. “From Individual to Group Impressions: Availability Heuristics in Stereotype Formation.” Journal of Experimental Social Psychology
14(1978): 237–255.
Shelor, R.M., D.C. Anderson, and M.L. Cross. “Gaining from Loss: Property-Liability Insurer Stock Values in the Aftermath of the 1989 California Earthquake.” Journal of Risk and Insurance
59(1992): 476–488.
Tversky, A., and D. Kahneman. “Belief in the Law of Small Numbers.” Psychological Bulletin 76(1971): 105–110.
Tversky, A., and D. Kahneman. “Availability: A Heuristic for Judging Frequency and Probability.” Cognitive Psychology
4(1973): 207–232.
Tversky, A., and D. Kahneman. “Judgments of and by Representativeness.” In D. Kahneman, P. Slovic, and A. Tversky (eds.)
Judgment under Uncertainty: Heuristics and Biases. New York: Cambridge University Press, 1982, pp. 84–100.
|
|
|
Confirmation and Overconfidence |
|
|
8 |
|
|
|
|
|
|
|
|
Throughout the last several years, the cry of political bias has been directed at the news media from nearly all quarters. Interestingly, though, one hears very different accusations from each quarter. In particular, when asked, a majority of political conservatives accuse the network news and major newspapers of a deep liberal bias. However, when one questions the most politically liberal, they claim that the same media display a conservative bias. If we consider both groups of people to be sincere, it appears that they have come to exactly opposite conclusions while viewing the exact same information. In fact, because of the supposed bias, many have chosen to obtain their news primarily from cable news outlets that seem much more open about their bent. Thus, more-conservative people tend to view more-conservative networks, and more-liberal people tend to view more-liberal networks. Here they may view news that generally confirms their already-held beliefs and are unlikely to encounter news that would cause them to question or abandon these beliefs. What drives people to these havens of safe information?
Other interesting behavior can be seen on the part of entrepreneurs. Entrepreneurs take great financial risks on the bet that their business idea will produce a successful venture. More often than not, however, these ventures fail, leaving in their wake lost dreams, lost money, and often lost marriages. Even those founding successful businesses tend to put in greater amounts of work for less money than they would make in alternative employment. Given the overwhelming prevalence of failure, why would any rational person take such risks? Moreover, among well-established firms, we often see waves of mergers. One firm buys another, hoping that the two pieces together will provide a greater profit than they do separately. If the firms perform some set of overlapping functions, it is possible to eliminate the redundant portions, thus reducing costs, and obtain the same revenues. However, this is not how it generally works out. In fact, the overwhelming majority of mergers lead to lower profits for the purchasing firm. What would lead firms to systematically misjudge the benefits of such mergers?
This chapter builds on the previous chapter in exploring the ways people seek new information and how this systematically affects beliefs and subsequent actions. In searching for new information, we often have a choice as to what type of information we will see. For example, we can choose to consume news with a conservative spin or with a liberal spin. One might have a greater tendency to confirm our currently held beliefs, and others might tend to disconfirm our currently held beliefs. In making decisions under uncertainty, we often must
187