<%BANNER%>

Do People Brace Sensibly? Risk Judgments, Outcome Importance, and Risk Prevalence


PAGE 1

DO PEOPLE BRACE SENSIBLY? RISK JUDGMENTS, OUTCOME IMPORTANCE, AND RISK PREVALENCE By KATHARINE DOCKERY A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2003

PAGE 2

Copyright 2003 by Kate Dockery

PAGE 3

ACKNOWLEDGMENTS I thank Dr. James Shepperd for his constant guidance and encouragement and for never allowing me to do less than my best work. I thank my husband and my parents for their unconditional love and support, without which this process would have been difficult and painful. Finally, I thank my God, who is able to do immeasurably more than all we ask or imagine. iii

PAGE 4

TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iii LIST OF TABLES.............................................................................................................vi LIST OF FIGURES..........................................................................................................vii ABSTRACT..................................................................................................................... viii CHAPTER 1 INTRODUCTION........................................................................................................1 Judging Likelihood.......................................................................................................3 Avoiding Disappointment as a Source of Judgment Errors..........................................6 Outcome Likelihood and Bracing.................................................................................9 Overview and Hypotheses..........................................................................................12 2 METHODS.................................................................................................................14 Overview.....................................................................................................................14 Participants.................................................................................................................14 Procedure....................................................................................................................14 3 RESULTS...................................................................................................................17 Risk Estimates............................................................................................................17 Are the Results Due to Outliers? ................................................................................21 Mediation Analyses ....................................................................................................23 4 DISCUSSION.............................................................................................................27 Possible Explanations.................................................................................................28 The Role of Disappointment.......................................................................................30 Limitations and Implications......................................................................................33 Conclusion..................................................................................................................34 iv

PAGE 5

APPENDIX A QUESTIONNAIRE A................................................................................................35 B QUESTIONNAIRE B................................................................................................37 LIST OF REFERENCES...................................................................................................39 BIOGRAPHICAL SKETCH.............................................................................................42 v

PAGE 6

LIST OF TABLES Table page 1 Frequency analyses of optimistic, realistic, and pessimistic risk estimates.............25 2 Frequency analyses of risk estimates.......................................................................26 vi

PAGE 7

LIST OF FIGURES Figure page 1 Risk estimates as a function of risk condition and need..........................................18 2 Mean risk estimates as a function of risk condition and need..................................20 vii

PAGE 8

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science DO PEOPLE BRACE SENSIBLY? RISK JUDGMENTS, OUTCOME IMPORTANCE, AND RISK PREVALENCE By Katharine Dockery December 2003 Chair: James A. Shepperd, PhD Major Department: Psychology Previous research shows that people become pessimistic about impending bad news to brace for the worst. The current study examined whether the commonality and importance of an event moderates bracing. Students learned about a billing error that would result in an unexpected bill for either 20% (rare event) or 80% (common event) of the students at their university. Students in the common event condition made higher personal risk estimates than did students in the rare event condition. Financially needy students also made higher risk estimates than did non-needy students. Comparing risk estimates to the base rates provided to participants revealed that students in the rare event condition were pessimistic about their risk of receiving a bill, with the financially needy students making the most pessimistic estimates. In contrast, students in the common event condition were optimistic about their risk, with non-needy students making the most optimistic estimates. The discussion explores several possible explanations for these findings. viii

PAGE 9

CHAPTER 1 INTRODUCTION During the fall of 2002, a series of sniper attacks terrorized life in the greater Washington, DC, region. People living in the area felt unable to leave their houses or send their children to school for fear that they might be the next victims. By the time the assailants were apprehended on October 25, 2002, they had left in their wake ten fatalities and three injuriesout of approximately 4.4 million people living in the counties where the attacks occurred. The chances of any given person being attacked by the snipers were less than 0.00003%, yet activity in Virginia and Maryland was paralyzed until the snipers were caught. Although the objective likelihood that any given person would fall prey to the snipers bullet was quite low, the behavior of people living in the area (e.g., withholding their children from school, postponing trips to gas stations) suggests that people felt otherwise, perceiving their risk as quite high. What made people so drastically overestimate their risk? Numerous studies find that people do not fully account for base rate information (Tversky & Kahneman, 1982), give undue weight to salient examples (MacLeod & Campbell, 1992; Slovic, Fischoff, & Lichtenstein, 1982), have a poor understanding of small numbers (Tversky & Kahneman, 1974), and rely on stereotypes (Kahneman & Tversky, 1972) when estimating their risk. Undoubtedly, these errors in thinking contributed to an overestimation of personal risk of being attacked. In this thesis I propose an additional reason people might overestimate their risk. Specifically, I propose that in some instances people may overestimate their risk to avoid negative feelings such 1

PAGE 10

2 as disappointment associated with having bad events come as a surprise. That is, people may overestimate their risk by making unduly pessimistic predictions in an attempt to brace for the worst (Carroll, Dockery, & Shepperd, 2003). The pessimism of residents of the Washington DC area about the risk of a sniper attack may have been motivated in part by an attempt to brace for the worst possible scenario. In fact, people who are bracing themselves may consistently overestimate the risk of rare events in their effort to prepare themselves for the blow of a negative outcome. This research examines whether the commonality of an event moderates bracing. I specifically address whether people brace more for rare events or for common events. On the one hand, people may brace more in anticipation of negative events that are common because of the greater objective likelihood of experiencing common event, which prompts more worry and fear of disappointment. For example, during an epidemic when large numbers of people are testing positive for a disease, people objectively face a greater risk of catching the disease and thus should be more preoccupied with the disease. In contrast, people should be less concerned about a rare disease for which their objective risk is lower. Thus, one might expect that people are more pessimistic, relative to the base rate, about common negative events. On the other hand, it is also possible that people brace more in anticipation of negative events that are rare. Research shows that the intensity of disappointment depends on the unexpectedness of a negative event. The more unexpected the event, the greater the disappointment people experience (van Dijk, Zeelenberg, & van der Pligt, 1999). Because bracing serves to reduce or avoid negative feelings such as disappointment, people may be more inclined to brace for rare events, which have greater capacity for producing disappointment, than for common events.

PAGE 11

3 Indeed, people may not only brace more for rare events, they may also prepare more in other ways, a finding that would have important implications regarding risk-taking behaviors. For example, travelers who mistakenly believe that they are likely to crash in a plane may take to the far more dangerous highways. The current study tests the competing possibilities regarding the influence of commonality on bracing. Judging Likelihood Although people often attempt to predict future outcomes (e.g., the likelihood that rain will spoil a picnic, that stock prices will rise, that traffic will be lighter on one route or another), try as they might they often predict inaccurately (Tversky & Kahneman, 1974). In fact, people often fail to adequately consider information about the actual frequency of an event, or its base rate. The base rate for an event, or more accurately, peoples perception of the base rate, can influence estimates of an events likelihood. However, people are often unaware of the base rate for an event, and even available base rate information does not guarantee accurate estimates. People often neglect such information, attending instead to less informative details of a situation (Tversky & Kahneman, 1982). This phenomenon is related to the process of anchoring and adjustment, another example of base rate use in probability judgments. When people employ the anchoring and adjustment heuristic, they make estimations by accounting for present information and then adjusting for differentiating characteristics (Tversky & Kahneman, 1974). Although anchoring and adjustment can lead to better predictions when people objectively consider relevant information, people often do a poor job of using such information or rely on biased perceptions of their capacity to avoid negative outcomes. For instance, participants in one study read scenarios in which the base rates for several

PAGE 12

4 events (car accident, burglary, pet illness, parachute failure, and raffle win) were either high or low and the target was the participant or another person. For every scenario, participants estimated a greater likelihood when the base rate was high than when the base rate was low. However, the estimations for the negative events were consistently lower when the participant was the target of the estimate than when another person was the target (Greening & Chandler, 1997). On the whole, people inadequately consider base rate information when estimating personal likelihood. Researchers have identified several cognitive biases that can prompt errors in judgments about base rates. First, people sometimes misjudge probability by overusing the availability heuristic, a shortcut in decision making in which people base judgments of likelihood on how easily they can generate examples of the event. For instance, people often overestimate the likelihood of a fatal plane crash or homicide because they can easily bring to mind poignant instances of these events from the media (Slovic, Fischoff, & Lichtenstein, 1982). The availability heuristic results in overestimation of probabilities for very salient events (often due to recency, personal experience, or media coverage) and underestimation for less salient events, regardless of the actual rate of occurrence. One study demonstrated use of the availability heuristic by objectively measuring availability and relating it to perceived probability. The results showed that quicker speed of retrieval of an occurrence of the event corresponded to higher probability judgments for the event (MacLeod & Campbell, 1992). Second, people sometimes misjudge probability because they overuse the representativeness heuristic, a shortcut in decision making that involves prototype matching. Specifically, people have prototypes for entities such as events, things, and

PAGE 13

5 people, and base judgments of likelihood that a particular example falls within a category on how closely the example matches the prototype. In other words, people perceive their risk for an event as higher when they closely match their stereotypes of someone who would experience that event. For example, peoples judgments of their likelihood of being mugged are influenced by how similar they think they are to their stereotype or prototype of the typical mugging victim. The more similar people perceive they are to the stereotype, the more likely they think they will be mugged. People also hold stereotypes about the likelihood of certain kinds of events. For example, if people believe a negative event is serious, they will also perceive it as rare. The converse is also true if people believe an event is rare, they will perceive it as more serious (Jemmott, Ditto, & Croyle, 1986). Although these beliefs may often be true (e.g., diseases that cause death are usually relatively rare), it is not always the case. For example, the human papilloma virus (HPV) can have serious consequences including infertility and cervical cancer, yet over 40% of college-age women are infected with the virus (Ho, 1998). Clearly, assumptions about the correlation between severity and risk can be misleading and even dangerous. Third, people sometimes err in judgments of likelihood based on biased perceptions of control, perceiving that desirable items are more likely to occur when they are controllable than when they are uncontrollable. More specifically, people underestimate their risk for certain events when they believe that they can control the outcome. In addition, people tend to overestimate their personal control (McKenna, 1993; Klein & Kunda, 1993), leading them to be overly optimistic about their ability to avoid certain negative events. In a study by Greening and Chandler (1997) described

PAGE 14

6 earlier, people perceived their likelihood of experiencing a variety of negative events as lower than another targets likelihood. For example, most people believe they are better than average drivers, and estimate their chances of getting in a car accident as less than average (McKenna, 1993). However, by definition most people are average drivers, and such flawed logic is likely a result of an overestimation of personal control. Thus, probability judgments based on perceptions of personal skill or control are likely to be underestimations. In addition, people may believe that they can control not only their actions, but also their outcomes. For example, smokers may believe they are less likely to get smoking related illnesses than other smokers because, unlike other smokers, they will not be smokers in the future (McKenna, Warburton, & Winwood, 1993; Lee, 1989). In summary, a variety of factors can cause people to make errors in estimates of an events base rate, or probability. First, people overuse the availability heuristic, over relying on salient examples to give them clues as to the likelihood of an event. Second, people overuse the representativeness heuristic, judging likelihood based on how closely an event or person matches their prototype for the event or person. Third, people underestimate their risk for negative events and overestimate the likelihood of desirable events when they overestimate how much control they have over their actions or the outcomes. Avoiding Disappointment as a Source of Judgment Errors The prior sources of error in likelihood judgment represent cognitive errors that arise from discounting or ignoring information that is hard to comprehend (underusing the base rate), or from overusing common shortcuts in making judgments (the availability and representativeness heuristics), or from attempts to calibrate judgments based on personal information (misperceptions of control). I propose another source of error that

PAGE 15

7 is more motivational in flavor arising from a desire to reduce or avoid negative emotions such as disappointment. Consistent with previous researchers, I view disappointment as the experience of outcomes falling short of expectations ( Van Dijk, Zeelenberg, & van der Pligt, 1999; van Dijk & van der Pligt, 1997; Zeelenberg et al., 2000). People feel disappointed about shattered hopes or expectations, not simply about negative outcomes. In this way, disappointment can be distinguished from the similar emotions of sadness, anger, frustration, and regret (Zeelenberg et al., 2000). Disappointment is specifically associated with absence of a positive, hoped for outcome, whereas sadness, frustration, and anger are generally associated with the presence of a negative outcome (van Dijk, Zeelenberg, & van der Pligt, 1999). Disappointment can be further distinguished from its close relative, regret, in that people feel regret over actions and disappointment over outcomes, regardless of the precipitating actions (Zeelenberg, van Dijk, & Manstead, 1998). The intensity of disappointment depends primarily how peoples outcomes compare with their expectations. The more expectations exceed outcomes, the more intense the disappointment (van Dijk & van der Pligt, 1997). If people set their sights very high in anticipation of feedback, they are likely to be disappointed by almost any outcome. For example, a student who expects a perfect score on an exam will almost certainly be disappointed. On the other hand, very poor outcomes also make disappointment likely. A student receiving a failing grade is also likely to experience disappointment. Decision affect theory (DAT; Mellers et al., 1997) discusses the emotional consequences of the relationship between expectations and outcomes, noting

PAGE 16

8 that how people feel about an outcome depends on the value of the unobtained outcome (was it better or worse than the outcome obtained?) and the perceived likelihood of receiving the outcome. For example, in a study presenting participants with gambling tasks with varying possible outcomes, subsequent affect depended not only on absolute loss or gain, but also on the possible alternative outcome and the probability of winning (Mellers et al., 1997, Study 1). Given that the experience of disappointment depends on the relationship between outcomes and expectations, how can people avoid feeling disappointment? Even at the moment of truth, when all control over outcomes is gone, people can still moderate their disappointment over a negative outcome by lowering their expectations (Carroll, Dockery, & Shepperd, 2003; Shepperd, Oullette, & Fernandez, 1996, experiments 2 & 3; see also Gilovich, Kerr, & Medveck, 1993; Sackett, 2002; Sanna, 1999). This process of bracing for the worst results in pessimistic likelihood estimates at the point of feedback. For example, students in one study estimated their scores on an exam at four points in time: one month before the exam (Time 1), immediately after the exam (Time 2), one hour before receiving their exam grades (Time 3), and immediately before receiving the grades (Time 4). Students shifted, first from optimism before the exam to realism immediately after the exam was completed. The realism persisted at Time 3, when the students did not expect feedback for another hour. However, when the graded exams were being distributed, the students became pessimistic in their predictions (Shepperd et al., 1996). Furthermore, not all outcomes are of equal importance. For example, a test for HIV could have far greater consequences than a test for Strep throat. Do people brace

PAGE 17

9 equally for all events, regardless of importance? Research shows that people, in fact, become more pessimistic about an outcome with severe consequence than they do for one with only mild consequences (Taylor & Shepperd, 1998). Importance can also be in the eye of the beholder. What seems life-threatening to a high school student, for instance, is usually insignificant to a mature adult. If people see an outcome as unimportant or inconsequential for any reason, they will likely maintain realism or even optimism. In summary, people experience disappointment when outcomes fall short of expectations. People can avoid disappointment by either changing their outcomes or change their expectations. At the moment of truth, however, people can no longer change the outcome. They can, however, change their expectations, avoiding disappointment by making a pessimistic prediction. However, people will only embrace pessimism to avoid disappointment over outcomes they view as important. Outcome Likelihood and Bracing Do people brace more for common events or for rare events? On first blush it might appear that reliance on the availability heuristic would lead people to brace more for common events. As discussed earlier, judgments of frequency are based on how easily examples come to mind. The pervasiveness of common events is likely to facilitate access to examples of common events, prompting greater estimates of frequency reliance on the availability heuristic leads to concern over salient events. For example, people who lived in the days of the influenza virus were certainly aware of the high numbers of fatalities and would likely have rated their chances of contracting the illness as high. People would have heard stories the influenza victims, accounts of how the virus devastated families, and what could happen to them if they did not take the appropriate precautions. However, although this argument is intuitively appealing, in reality the

PAGE 18

10 commonality and availability of an event are distinct. As stated earlier, people frequently underestimate the likelihood of very common events. Thus, although some common events are also highly available in memory and the consequences easily imagined, equally often availability and commonality are unrelated. Availability aside, people may brace more for common events because they have difficulty understanding probabilities. Instead of thinking of likelihood in terms of percentages or probabilities, people may think about likelihood in binary terms -an event will either happen or it wont. Accordingly, for low probability events, people may conclude that because the event is rare, it will not happen. For high probability events, however, people may conclude that the event is inevitable even though there is some chance that it will not happen. Binary thinking would thus lead people to see common events as inevitable and rare events as entirely avoidable. While some evidence suggests that people will brace more for common events than for rare events, other evidence suggests that people will brace more for rare events than for common events. First, a great deal of research shows that people underestimate the likelihood of common events and overestimate the likelihood of rare events (Fischoff, 1981; Johnson & Tversky, 1983; Lichtenstein, Slovic, Fischoff, Layman, & Combs, 1978; Pulford & Colman, 1996; Slovic, 1987; Weinstein, & Lyon, 1999; Rothman, Klein, & Weinstein, 1996; Brandsttter, Khberger, & Schneider, 2002). For example, people underestimated their risk for common diseases, such as the human papilloma virus and chlamydia, and overestimated their risk for rare diseases such as chronic liver disease and cirrhosis (Rothman et al., 1996, see also Lichtenstein et al., 1978). The availability heuristic may account in part for these errors in estimation in that media attention to rare

PAGE 19

11 diseases may prompt people to overestimate the base rate and, as a consequence, overestimate their risk (Slovic et al., 1982). Recent events make this point clear. Following the events of September 11, 2001, people became afraid to travel by air. Although, the risk of dying in a plane crash is extremely low, the vivid plane crashes appeared to have created the perception that travel by plane is highly dangerous and that fatal plane crashes are widespread. Second, the disappointment literature reveals that people feel negative outcomes are particularly aversive when unexpected (Mellers et al., 1997; Shepperd & McNulty, 2002). The intensity of disappointment depends not only on the negativity of the outcome, but also on the expectations regarding the outcome (van Dijk & van der Pligt, 1997), and people would naturally have positive expectations when a negative event is unlikely to occur. In other words, if a negative event has a known base rate of 1%, people will probably be fairly certain that the event will not happen to them. On the other hand, if the negative outcome were to occur, it would be extremely disappointing because it is so rare and unexpected. In sum, the question of how the commonality of an event moderates bracing has two possible answers. The first, and most intuitive prediction is that people brace more for common negative events. This possibility is supported by the possibility that people think in binary terms rather than considering probabilities and statistics. The second, opposing prediction is less intuitive and proposes that people brace more for rare negative events, and is supported by research on disappointment and probability judgments of common and rare events.

PAGE 20

12 Overview and Hypotheses The present study examines whether people brace more for common vs. uncommon negative outcomes by exploring responses to news of a possible financial setback. Specifically, participants believed that either 20% or 80% of students at their university would be affected by an error in the registrars office that would result in those students receiving an unexpected bill. Hypothesis 1: I hypothesized that participants in the common risk (80%) condition would estimate their likelihood of receiving a bill as greater than would participants in the rare risk condition (20%). Consequently, I expected to find a main effect of risk condition on risk estimates, with the mean of the common risk condition greater than the mean of the rare risk condition. Hypothesis 2: In line with the finding that people will be pessimistic about outcomes that are important to them, I hypothesized that, when averaging across the common and rare events, participants who were financially needy would estimate their likelihood of receiving a bill as greater than would participants who were not needy. Consequently, I expected to find a main effect of need on risk estimates, with the mean of the high need participants greater than the mean of the low need participants. Hypothesis 3: Among participants receiving news of a rare billing error, I predicted that high need participants would be more likely than low need participants to estimate that they would receive a bill. Among participants receiving news of a common billing error, I predicted that high need and low need participants would not differ in their risk judgments. These sets of means would manifest themselves as an interaction of need and risk level.

PAGE 21

13 Hypothesis 4: Based on the literature on disappointment, I hypothesized that high need participants would be pessimistic in the rare risk condition but would be realistic in the common risk condition. Consequently, I predicted that a dependent t-test comparing the predictions of low need participants would reveal that predictions in the rare risk condition would differ significantly from the 20% baseline, but that predictions in the common risk condition would not differ significantly from the 80% baseline. Hypothesis 5: I hypothesized that low need participants would be realistic in both the common risk and rare risk conditions. Consequently, I predicted that dependent t-tests comparing the predictions of low need participants in the rare and common risk conditions with their respective baselines (either 20% or 80%) would reveal no differences from the baselines.

PAGE 22

CHAPTER 2 METHODS Overview The methods were adapted from those used Shepperd et al. (2000). This experiment examined students reactions to the prospect of a financial loss after learning that a high or low percentage of students would be experience the loss. The greater disappointment experienced when a loss is unexpected led me to predict that people would be more pessimistic about the prospect of a loss, relative to the baseline, when they were told that few students would be affected than when they knew most students would be affected. Moreover, I expected that the effect would be due to greater pessimism on the part of financially needy students. Participants Introductory psychology students (N = 234) participated voluntarily as part of three class sections and were randomly assigned to either the high likelihood (80%) of receiving a bill condition or the low likelihood (20%) of receiving a bill condition. Procedure Participants in three classes received a description of a recently discovered billing error that would affect either 20% or 80% of the student body (see appendix A). Participants in each situation learned that students affected by the error would receive a $178 bill in three to four weeks and that failure to pay the bill would result in their records being flagged. 14

PAGE 23

15 Participants then completed a questionnaire regarding their reactions to the news of the billing error. First, immediately after learning of the billing error, participants responded to five adjectives assessing state anxiety (calm, nervous, anxious, relaxed, worried). Participants responded to each item with how they felt right now, at this moment, using a four-step scale (1 = not at all; 4 = very much so). These items were summed (after reverse coding) and divided by five to produce a measure of anxiety, range = 1 to 4, M = 2.06, SD = 0.76, Cronbachs = .85. Second, participants completed the primary dependent measure, estimating their likelihood of receiving the bill. This item asked participants to use a 0 to 100% scale to estimate the probability that they would receive a bill. Third, previous findings reveal that only financially needy students braced at the prospect of a financial blow (Shepperd et al., 2000). As a consequence, in the present study I assessed financial need by having participants complete the same six items assessing financial need used by Shepperd et al. (2000). Specifically, participants indicated (a) the extent to which they were on a tight financial budget (1 = not on a tight budget; 11 = extremely tight budget), (b) how much difficulty they had making ends meet (1 = extreme difficulty; 11 = no difficulty), (c) how much the bill would impact their lives (1 = little impact; 11 = great impact), (d) what effect the bill would have on their finances (1 = little impact; 11 = great impact), (e) how dependent they were on financial aid (1 = not at all dependent; 11 = very dependent), and (f) the extent a bill would affect their budget (1 = not at all; 11 = a great deal). Three of these items (a, b and e) assessed the extent to which participants faced financial challenges and the remaining two items assessed the financial consequences of receiving the bill. I combined the five items to

PAGE 24

16 form a single index of need because collectively the five items provided a more complete picture of financial neediness. These five items were summed, after a reverse coding item b, and divided by five to form a single index with a potential range of 1 to 11, M = 5.63, SD = 2.70, Cronbachs = .93. Of note, the first two items were assessed prior to the description of the billing error, and the last three items were assessed after the description of the error. It is possible that the risk manipulation influenced responses to the last three financial need items. However, an independent t-test revealed no difference between participants in the rare and common event conditions on their responses to these three items, t(178) = 1.43, p > .15. Participants were also asked if they currently were receiving a Bright Futures Scholarship (unique to public universities in Florida), if their tuition was pre-paid, and whether they were paying in-state or out-of-state tuition. Fourth, participants reported the extent to which they were thinking about the financial difficulties that they would experience in the immediate future as a result of receiving a bill (1 = not at all; 11 = a great deal) and how disappointed they would be if they received a bill (1 = not disappointed; 11 = extremely disappointed). These two items were included to address possible processes mediating any effect of financial need on personal estimates. Finally, participants indicated if they estimated their personal risk to be more or less than the 20% or 80% given in the scenario, why they gave that estimate. They were given several options to choose from, including the option to write a reason other than the ones provided. Of note, one version of the questionnaire included several additional items. However, these items did not yield significant results and are irrelevant to the purposes of this study.

PAGE 25

17 CHAPTER 3 RESULTS From the initial pool of 234 participants I omitted from analyses data from 54 participants because they doubted the authentic ity of the billing error, a fact that was evident from their responses to the item, I estimated that my probability of receiving a bill wasbecause. More specifically, I omitted data from 28 participants because they indicated that they knew their bill well and would have noticed the error, 5 participants because they indicated that scho larships would cover all expenses, 8 because they did not believe that a tuition error had occurred, and 13 due to a clerical error that rendered their responses unusable. Of note, in cluding the responses of these participants did not change my basic findings. As noted earlier, data were collected from three classes. Preliminary analyses including class as a variable in the model revealed no ma in effects or interactions involving class. Consequently, I collapsed ac ross class in all subsequent analyses. Risk Estimates Did risk judgments vary as a function of n eed and risk level? Figure 1 displays the probability estimates as a function of need (h igh or low) and risk condition (rare or common). My first three hypotheses were 1) participants in the common risk (80%) condition would estimate their likelihood of receiving a bill as greater than would participants in the rare risk condition (20%), 2) participan ts who were financially needy would estimate their like lihood of receiving a bill as grea ter than would participants who were not needy, across risk conditions, and 3) among participants receiving news of a

PAGE 26

18 rare billing error, high need pa rticipants would be more likel y than low need participants to estimate that they would receive a bill, but among participants receiving news of a common billing error, high need and low need pa rticipants would not differ in their risk judgments. Consistent with Hypothesis 1, participants in the rare risk condition made lower estimates (M = 42.78, SD = 29.60) than did partic ipants in the common risk condition (M = 56.91, SD = 32.44). That is, participants judge d that they were at greater risk of receiving a bill when the billing error wa s common than when the billing error was relatively rare. For the purpos e of illustrating the findings of my second hypothesis, I separated participants into high and low n eed groups using a median split of their 20 30 40 50 60 70 Low NeedHigh Need Financial NeedRisk Estimates Rare Common Figure 1. Risk estimates as a f unction of risk condition and need.

PAGE 27

19 responses to my inventory of need (median = 5.4). Consistent with Hypothesis 2, high need participants made higher estimates (M = 55.63, SD = 29.03) than did low need participants (M = 43.52, SD = 33.36). That is, participants who were financially needy judged they were more likely to receive a bill than did participants who were not financially needy. Contrary to Hypothesis 3, high need participants made higher estimates than did low need participants regardless of whether the objective likelihood of receiving a bill was high or low. That is, participants who were financially needy judged that they were at greater risk of receiving a bill both when the risk was low and when it was high. Although this finding is consistent with my prediction for the rare risk condition, in the common risk condition I expected participants to make similar predictions regardless of need. We examined the first three hypotheses statistically using simultaneous multiple regression procedures in which Need (after centering), Risk (rare or common), and the Need by Risk interaction were entered as predictors. Analysis of the risk estimates revealed the predicted main effects of Risk, F(1, 176) = 12.12, p < .001, eta-squared = .06, and Need, F(1, 176) = 8.38, p < .01, eta-squared = .05, but did not reveal the predicted need by risk interaction, F(1, 176) = .03, p = .85, eta-squared = .0002. We tested two additional hypotheses: 4) high need participants would be pessimistic in the rare risk condition but would be realistic in the common risk condition, and 5) low need participants would be realistic in both the common risk and low risk conditions. To test these hypotheses, I compared participants estimates to the base rates I provided (either 20% or 80%). Figure 2 presents the results of these analyses. In all

PAGE 28

20 conditions of need and risk, participants made estimates that differed significantly from the base rate. Specifically, both high need, t(50) = 7.18, p < .0001, and low need, t(40) = 3.25, p < .01, participants in the rare event condition estimated that their likelihood of receiving a bill was greater than the 20% base line they received (M = 49.6, SD = 29.4 and M = 34.2, SD = 27.9, respectively). Likewise, both high need, t(39) = -3.90, p = < .001, and low need, t(47) = -5.51, p < .0001, participants in the common event condition estimated that their likelihood of receiving a bill was less than 80% (M = 63.4, SD = 27.0 and M = 51.5, SD = 35.8, respectively). In short, all participants in the rare event condition were significantly pessimistic and all participants in the common event condition were significantly optimistic. High Need Low Need 80% Risk Estimate 63.4 49.6 51.5 34.2 20% 80% 20% Baseline Figure 2. Mean risk estimates as a function of risk condition and need. In sum, participants who were told that their risk of receiving a bill was 80% made higher personal risk estimates than did participants who were told that their risk was

PAGE 29

21 20%. Furthermore, financially needy students in both risk conditions estimated their risk to be higher than did non-needy participants. Finally, participants in the rare risk condition were pessimistic and those in the common risk condition were optimistic, in both cases regardless of need. However, needy students were more pessimistic in the rare risk condition and less optimistic in the common risk condition than the non-needy students. Are the Results Due to Outliers? Two possible explanations for the present findings deserve investigation. A first possible explanation for my findings is that they resulted from the responses of a few outlying participants. Because participants could respond using a scale ranging from 0 to 100, there is considerable opportunity for variability in responses and thus a possibility that the findings were the result of outliers. To test this possibility, I conducted a Chi-square analysis of the number of participants who were optimistic, pessimistic, and realistic in each risk condition. Table 1 presents the frequencies of low need and high need participants who were optimistic, realistic, and pessimistic, relative to the baseline of 20% or 80%. I categorized estimates below the baseline (i.e., below 20% for rare event and below 80% for common events) as optimistic, estimates equivalent to the baseline (20% in the rare event condition, 80% in the common event condition) as realistic, and estimates above the baseline as pessimistic. I compared the proportion of low need and high need participants who were optimistic, realistic, and pessimistic separately for the rare and common event conditions using chi-square analyses. The estimate frequencies for needy and non-needy participants differed significantly for rare events, 2 (2, N = 92) = 16.78, p < .001, but did not differ significantly for common events, 2 (2, N = 88) = 1.78, p > .20. I

PAGE 30

22 next compared the proportion of participants in the common and rare event conditions who were optimistic, realistic, and pessimistic separately among needy and non-needy participants. The estimate frequencies for rare and common events differed significantly among both needy participants, 2 (2, N = 91) = 62.37, p < .0001, and non-needy participants, 2 (1, N = 89) = 49.83, p < .0001. In sum, although the pattern of findings did not differ significantly in one case, these findings suggest that the results of need and risk were not due to a few outlying responses. Second, I found that people in the rare event condition were generally pessimistic, estimating a risk greater than 20%, and that participants in the common event condition were generally optimistic, estimating a risk less than 80%. It is possible that the high level of pessimism for rare events and the high level of optimism for common events is simply the result of participants in the rare condition having greater room to be pessimistic than optimistic, and the participants in the common condition having greater room to be optimistic than pessimistic. For example, in the rare event condition, participants could be pessimistic by choosing any risk level greater than 20% -a choice of 80 possible values. However, they could be optimistic by choosing any risk level less than 20% -a choice of only 20 possible values. In short, the difference in the extent to which participants in the rare and common events conditions displayed optimism and pessimism may be an artifact of how much opportunity they had to display optimism and pessimism. To examine this possibility, I conducted Chi-square analysis on the number of participants in the common event condition that provided responses above and below 20%, and the number of participants in the rare event condition that provided responses

PAGE 31

23 above and below 80%. The Chi-square analyses examined whether 80% of estimates in each condition fell into the eighty-percentage-point space and 20% in to the twenty-percentage point space. In other words, I compared the expected frequencies (80% or 20% of estimates falling above or below the baseline) with the actual estimate frequencies. Estimates that equaled the base line were excluded from these analyses because they provided no information about systematic errors in responses. If participants were merely responding randomly or supplying responses where they had the greatest room to respond, then in the rare risk condition, 80% of responses should exceed 20% and 20% of responses should fall below 20%. Conversely, in the common risk condition 80% of responses should fall below 80% and 20% of responses should fall below 20%. Table 2 displays the number of estimates above and below 20% and 80% for each condition. The distribution of responses was consistent with the random responding explanation in all four cases. For common events, neither the estimates of participants high in need, 2 (1, N = 32) = 2.53, p = .11, nor the estimates of participants low in need, 2 (1, N = 39) = .78, p > .20, differed from the expected pattern. Likewise, for rare events, neither the estimates of participants high in need, 2 (1, N = 43) = .98, p > .20, nor the estimates of participants low in need, 2 (1, N = 29) = 2.21, p = .14, differed significantly from the expected pattern. In sum, the distribution of responses was consistent with random responding in all four cases. This distribution of responses suggests that random responding could be responsible for participants risk estimates. Mediation Analyses We included several measures of participants reactions (i.e., anxiety, disappointment, and future thinking) to the possibility of receiving a financial blow with an eye toward investigating possible mediators of the effect of need on personal risk

PAGE 32

24 estimates. Using procedures recommended by Baron and Kenny (1986), I first examined whether the three potential mediators predic ted participants predictions. Correlation analyses revealed significant relationships between participants risk estimates and anxiety, r (180) = .32, p < .0001, disappointment, r (179) = .20, p < .01, and future thinking, r (180) = .25, p < .001, indicating that particip ants who felt more anxious, expected to experience more disappointment and were thinking more about the future made greater risk estimates. Furthermore, financial need correlated significantly with anxiety, r (180) = .40, p < .0001, disappointment, r (179) = .68, p < .0001, and future thinking, r (180) = .79, p < .0001, indicating that needy st udents, compared to non-needy students were feeling more anxious, pred ict that they would experience greater disappointment if they recei ved a bill, and were thinking more about the future consequences of receiving a bill. We then conducted three separated regr ession analyses, one each for the three possible mediators. In each case, the mediat or was entered into the model first, followed by need, risk condition and the need by risk co ndition interaction term. In all three cases, when the mediator was added, need no longer predicted participants risk estimates, all F s(1, 175) < 1, all p s > .35, but the mediator did, all F s(1, 175) > 8.13, p < .01. Of note, the mediators in no way mediated the effect of risk condition on participants estimates. Risk condition remained unchanged as a sign ificant predictor of participants risk judgments in all three cases, all F s(1, 175) > 11.16, p < .01. The fact that all three variables completed mediated th e effect of need on risk judgments suggests that the three mediators are all tapping a common underlying process.

PAGE 33

25 Table 1. Frequency analyses of optimistic, realistic, and pessimistic risk estimates ________________________________________________________________________ Optimistic Realistic Pessimistic ______________________________________________________ Condition Frequency % Frequency % Frequency % ________________________________________________________________________ Common Event High Need 22 55% 8 20% 10 25% Low Need 29 60% 8 17% 11 23% ________________________________________________________________________ Rare Event High Need 6 12% 8 16% 37 73% Low Need 9 22% 12 29% 20 49% ________________________________________________________________________

PAGE 34

26 Table 2. Frequency analyses of risk estimates ________________________________________________________________________ Condition Above Baseline Below Baseline Frequency % Frequency % ________________________________________________________________________ Common Event High Need Observed 10.0 31% 22.0 69% Expected 6.4 20% 25.6 80% Low Need Observed 10.0 26% 29.0 74% Expected 7.8 20% 31.2 80% ________________________________________________________________________ Rare Event High Need Observed 37.0 86% 6.0 14% Expected 34.4 80% 8.6 20% Low Need Observed 20.0 69% 9.0 31% Expected 23.2 80% 5.8 20% ________________________________________________________________________

PAGE 35

CHAPTER 4 DISCUSSION The goal of this study was to examine whether people brace differently when an event is common vs. rare. I predicted that participants would only brace for rare events, displaying pessimism about the likelihood of receiving a bill, and that only participants for whom a bill was particularly consequential would brace. The data generally supported the predictions, although some findings were unexpected. Looking first at the effect of risk level (Hypothesis 1), participants who believed that 20% of students would receive a bill made lower personal predictions than did participants who believed that 80% of students would be affected. Furthermore, in line with Hypothesis 2 and previous findings that people are pessimistic about outcomes that are important to them, participants high in financial need made higher risk estimates than did low need participants. Comparing risk estimates to the base line of 20% or 80% yielded an unexpected pattern of results. In the rare risk condition, I predicted that only high need participants would be pessimistic. However, the results show participants were pessimistic, regardless of need, about the likelihood of a rare event. Importantly, high need participants were more pessimistic than were low need participants. Nevertheless, low need participants were also bracing in their estimates. In the common risk condition, I predicted that all participants would be realistic about their risk of receiving a bill. However, the results show that participants were actually optimistic, regardless of need, 27

PAGE 36

28 about the likelihood of a common event. Further analyses confirmed that these patterns were not a result of outliers. Possible Explanations How do I explain this pattern of optimism and pessimism? One possible explanation for the findings is that participants who were low in financial need were simply less engaged in the procedures because the consequences were unimportant to them, resulting in muted effects of risk level on their estimates. Although presumably low need participants saw the event as less important than did high need participants, the results do not point to this difference as the sole player in my findings. Only in the rare risk condition could I describe low need participants estimates as a muted version of the effect found with high need participants. In the common risk condition, participants low in need actually deviated more from the base line than did high need participants, ruling out the possibility of a muted effects explanation. A second possible explanation lies in the finding that people consistently overestimate the risk of rare events and underestimate the risk of common events (Fischoff, 1981; Johnson &Tversky, 1983; Lichtenstein, Slovic, Fischoff, Layman, & Combs, 1978; Pulford & Colman, 1996; Slovic, 1987; Weinstein, & Lyon, 1999; Rothman, Klein, & Weinstein, 1996; Brandsttter, Khberger, & Schneider, 2002). This research suggests that when people do not know the base rate for an event, they make predictions based on available knowledge and what seems reasonable. People often have some sense of the true base rate, but this sense is imperfect. As a consequence, they underestimate how rare the rare events are and overestimate how common the common events are. If they knew the true base rates, their estimates would conform more to the base rates. Thus, perhaps students simply overestimated the likelihood of receiving a bill

PAGE 37

29 when the risk was 20% and underestimated when the risk was 80% because they were making their best estimates based on their available knowledge of the event. However, this explanation falls apart in light of the fact that the participants in this study had knowledge of the exact base rate. Moreover, analyses of the distribution of participants who were optimistic, realistic, and pessimistic confirms that participants were not merely overestimating their risk when the event was rare and underestimating their risk when the event was common. Third, perhaps participants were simply using the availability heuristic when making their estimates. With the availability heuristic, judgments of likelihood are based on how easily examples come to mind. Although available information such as past billing problems with the registrars office may account for the differences in estimates between needy and non-needy participants, it does not explain the fact that both needy and non-needy participants overestimated their risk for a rare outcome and underestimated their risk for a common outcome. Alternatively participants may have employed the availability heuristic in conjunction with an anchoring and adjustment process. Specifically, participants may have used the base rates provided to them then adjusted their estimates from the base rates based on their experience. Thus, both needy and non-needy participants may have had some experience with billing problems and so they adjusted their estimates somewhat from the base rate. Prior research, however, has found that needy students experience more billing problems than do non-needy students (Shepperd et al., 2000). The greater experience may have prompted greater upward adjustment from 20% in the rare risk condition and less adjustment from 80% in the common risk condition. Although this explanation seems plausible, it does not explain

PAGE 38

30 why participants, particularly participants high in need, supplied estimates below 80% in the common risk condition. Fourth, perhaps participants had difficulty understanding the probabilities of 20% and 80%. Instead of judging their risk in terms of percentages, perhaps participants were thinking in binary terms, such that they interpreted a 20% risk as I wont receive a bill and an 80% risk as I will receive a bill. Again, this explanation is easily refuted. If, in fact, participants were thinking in binary terms, they would have made estimates of 0% and 100%. Examination of the data reveals that less than 11% of participants made an estimate of 0%, and fewer than 6% of participants made estimates of 100%, hardly evidence for a binary thinking explanation. A final possible explanation for the findings is that the estimates were randomly distributed and that the distribution of estimates reflects nothing more than people supplying estimates where they had the most room to estimate. Thus, participants were pessimistic in the rare event condition because they had more room to be pessimistic than optimistic. Conversely, people were optimistic in the common event condition because they had more room to be optimistic. As I noted earlier, the distribution of responses does not rule out this explanation. However, it is difficult to conceive why financial need would differentially impact random responses. Thus, although chance may play a role in the results, it seems that chance cannot fully account for all of the findings. In addition, the distribution of optimistic, pessimistic, and realistic responses does not support an explanation of the findings as due to outliers. The Role of Disappointment I proposed at the outset of this study that differences in peoples risk estimates for rare and common events would result from differences in their expectations of

PAGE 39

31 disappointment. Previous studies show that people experience disappointment to the extent that their expectations exceed their outcomes (van Dijk & van der Pligt, 1997). People can avoid feeling disappointed in one of two ways: a) improve outcomes, or b) lower expectations. In this study, the outcome, a bill of $178, was out of the participants control. Therefore, keeping their expectations of a positive outcome low was the only option available to mitigate possible disappointment. Thus, perhaps participants were pessimistic about rare events to avoid being caught off guard. In contrast, perhaps participants did not need to be pessimistic about common events, for which their expectations were low from the start. Do my results support this proposal? Admittedly, the findings are mixed in their support for the role of disappointment. On the one hand, participants reported expectations of feeling disappointment did not mediate the effect of risk condition on personal estimates. In fact, expected disappointment seemed to be a facet of financial need rather than a response to the risk manipulation. Thus, concerns with disappointment did not differentiate risk estimates for rare and common events. On the other hand, a close look at the pattern of risk estimates may bring disappointment back into the picture. Ignoring, for a moment, participants self-reported expectations for disappointment, the pattern of results conceptually supports the initial predictions. Concerning rare events, participants were pessimistic across levels of financial need, with high need participants displaying stronger pessimism than low need participants. Although I initially predicted realism from low need participants, their slight pessimism is understandable when one takes into account that the bill was for $178, a considerable sum of money for a college student. In other words, perhaps all

PAGE 40

32 participants wanted to be prepared to some degree when the bill could catch them off guard. Concerning common events, participants were optimistic across levels of financial need, with low need participants displaying stronger optimism than high need participants. Although I initially predicted realism from all participants in the common risk condition, varying degrees of optimism are not inconsistent with an explanation of disappointments role in risk predictions. In this explanation, I propose that people do not want to be surprised by a negative event. When the baseline is high enough, people no longer need to embrace a pessimistic outlook to prepare themselves for disappointment because reality does that for them. In fact, people may even become slightly optimistic, taking advantage of the affective benefits of a positive outlook while still remaining relatively prepared for the worst. For example, participants who estimated 70% in the common risk condition may have concluded that this estimate sufficiently prepared them for the possibility of receiving a bill while offering the subjective comfort that their risk, while high, was still lower than the risk of others. In light of this adjustment to my original model of disappointment and risk, participants estimates in the common risk condition fit quite well. Participants high in need sacrificed only a small degree of preparedness and became only slightly optimistic, while low need participants, for whom the consequences of the bill would be minor, adopted a relatively highly optimistic outlook. In sum, a number of explanations fail to account for the results of this study. First, low need participants did not display a muted version of the high need participants pattern of estimates. Second, participants did not simply overestimate low probabilities

PAGE 41

33 and underestimate high probabilities. Third, participants were not responding to the salience of the bill. Fourth, participants did not interpret their risk of receiving a bill in binary terms. Fifth, the pattern of results was not due the presence of outliers. Sixth, although initial analyses suggests that the distribution of estimates might simply arise from participants were responding where there was the most room to respond, this explanation does not explain why participants high in need rated their risk as greater than did participants low in need. Finally, I propose an explanation based on the differing role of disappointment in predictions for rare and common events. Although this explanation is tentative and not without inconsistencies (i.e., the self-reported expectations of disappointment), it does capture the complete picture of my findings in a way that other explanations fail to do. Limitations and Implications The present study addressed a previously unexplored question about bracing for bad news. Although the results begin to paint a picture of the role of commonality and importance in peoples perceptions of risk, I acknowledge several limitations of this study. As just discussed, the role of disappointment requires clarification in order to form conclusions about why people brace differently for common and rare events. I used a single item measure of expected disappointment, and this single item may have been insufficient to capture the complex experience of affective prediction. In fact, disappointment mediated the effect of financial need, suggesting that my measure of disappointment (along with anxiety and thoughts about the future) may have tapped something more related to importance than to an expected affective experience. Examining disappointment more fully would be an important goal for future research in this area.

PAGE 42

34 In this study, financial need served as my measure of the importance of consequences. Although I believe that need was a valid measure of importance, future studies should seek to replicate my findings using an importance manipulation. Furthermore, this study used 20% and 80% as the definitions of rare and common, respectively. Commonality is a relative concept, and one could make the argument that 20% isnt very rare, or that 80% isnt common enough. I chose 20% and 80% for practical reasons, and it is the job of future research to determine the generality of my findings with varying degrees of commonality. Conclusion Do people brace sensibly? At first glance, my findings appear to suggest that people do not. Why would people be pessimistic about events that are unlikely to occur and optimistic about near certainties? Upon further examination, a possible answer emerges: perhaps people merely need to prepare enough for future outcomes. To illustrate this point, I return to the Washington sniper. Although people were at extremely low risk of becoming the next victim, they were prepared for the worst. Given the potential consequences of being unprepared, bracing for the worst served to keep people mindful of their risk. In contrast, consider the life of an Israeli living in Jerusalem. For the Israeli, danger lies in every unattended bookbag and every ride on the bus. In order to maintain sanity and proceed with necessary daily activities, the Israeli may actually embrace optimism in an otherwise paralyzing situation. Were the Washington, DC, residents being realistic? No. Is the optimistic Israeli being realistic? No. However, it seems possible that each form of unreality serves the purpose at hand.

PAGE 43

APPENDIX A QUESTIONNAIRE A By completing this survey, you are giving consent for your answer to be used by researchers at the University of Florida. Your participation is voluntary and no compensation will be provided. You may skip any question you are uncomfortable answering. Your answers will be anonymous. Thank you for your help. 1. To what extent are you on a tight financial budget? 1 2 3 4 5 6 7 8 9 10 11 not on a extremely tight budget tight budget 2. How much difficulty do you have making financial ends meet? 1 2 3 4 5 6 7 8 9 10 11 no extreme difficulty difficulty In processing the summer semester tuition and fees, the Office of the Registrar discovered that 20% of students were underbilled by $178 for a one time Information Technology fee imposed this semester. The fee was not itemized on the bills so it is unlikely you would know if you were underbilled. This billing error did not result from any problems with financial aid. Thus, students receiving financial aid are no more likely than students not on financial aid to be affected by this billing error. The registrar's office intends to bill students affected by this error for the difference. The bills will be sent in 1 to 2 weeks. Because scholarships, grants, and loans have already been disbursed, students affected by this error will be required to pay the Registrars office directly. Students who fail to pay the difference will have their records flagged and they will not be permitted to graduate or register for the next semester. Read each of the following items. Circle the number that best indicates how you feel right now, at this moment Not at SomeModerately Very all what so much so 1. calm 1 2 3 4 2. nervous 1 2 3 4 3. anxious 1 2 3 4 4. relaxed 1 2 3 4 5. worried 1 2 3 4 Please respond to the following questions: 3. Were you aware of this situation? Yes No 4. Using a scale from 0 to 100%, what is the probability that you will receive a $178 bill from the Registrars office? Please give your true gut feeling in providing this estimate. _______ % 5. If you received the bill, how much would this impact your life? 1 2 3 4 5 6 7 8 9 10 11 little impact great impact 6. If you received the bill, how disappointed would you be? 1 2 3 4 5 6 7 8 9 10 11 not disappointed very disappointed 7. If you received the bill, to what extent would this experience affect your budget? 1 2 3 4 5 6 7 8 9 10 11 little effect great effect 8. If you received the bill, what effect would this experience have on your finances? 1 2 3 4 5 6 7 8 9 10 11 little effect great effect 9. To what extent were you thinking ahead about difficulties the bill would present in the immediate future ? 1 2 3 4 5 6 7 8 9 10 11 not at all very much 10. Do you have pre-paid tuition? Yes No 11. Your sex (circle one): Male Female 12. Are you currently a Bright Futures Scholarship recipient or do receive some other form of scholarship that covers your tuition expenses? Yes No 35

PAGE 44

36 13. If you estimated that your probability is less than 20% on item 5, check the reason below that best applies to you. Choose only ONE response. I estimated that my probability of receiving a bill is LESS than 20% because: _____ a. I know pretty well what my tuition and fee charges should be and I would have noticed an error before. _____ b. I feel pretty lucky and believe that this sort of error wont happen to me. _____ c. When these sorts of errors have occurred in the past, it usually happened to other people but not me. _____ d. I think its wise to be optimistic and not to imagine the worst. After all, if you think bad things will happen to you, then they often do happen. _____ e. Other (please specify): _______________________________________________ __________________________________________________________________ __________________________________________________________________ 14. If you estimated that your probability is greater than 20% on item 5, check the reason below that best applies to you. Choose only ONE response. I estimated that my probability of receiving a bill is GREATER than 20% because: _____ a. I know pretty well what my tuition and fee charges should be and I had already suspected or detected the error myself. _____ b. I am bracing for the worst. Bad news feels worse when it is unexpected. Im expecting a bill so Ill be ready for it. _____ c. I always seem to get hit by unexpected expenses or bills. Im sure this is just another instance. _____ d. The university has made mistakes on my bills in the past and they have probably made a mistake in my case again. _____ e. Other (please specify): _______________________________________________ __________________________________________________________________ __________________________________________________________________ 15. If you estimated that your probability is equal to 20% on item 5, then place a check here: ___

PAGE 45

APPENDIX B QUESTIONNAIRE B By completing this survey, you are giving consent for your answer to be used by researchers at the University of Florida. Your participation is voluntary and no compensation will be provided. You may skip any question you are uncomfortable answering. Your answers will be anonymous. Thank you for your help. 1. To what extent are you on a tight financial budget? 1 2 3 4 5 6 7 8 9 10 11 not on a extremely tight budget tight budget 2. How much difficulty do you have making financial ends meet? 1 2 3 4 5 6 7 8 9 10 11 no extreme difficulty difficulty In processing the summer semester tuition and fees, the Office of the Registrar discovered that 80% of students were underbilled by $178 for a one time Information Technology fee imposed this semester. The fee was not itemized on the bills so it is unlikely you would know if you were underbilled. This billing error did not result from any problems with financial aid. Thus, students receiving financial aid are no more likely than students not on financial aid to be affected by this billing error. The registrar's office intends to bill students affected by this error for the difference. The bills will be sent in 1 to 2 weeks. Because scholarships, grants, and loans have already been disbursed, students affected by this error will be required to pay the Registrars office directly. Students who fail to pay the difference will have their records flagged and they will not be permitted to graduate or register for the next semester. Read each of the following items. Circle the number that best indicates how you feel right now, at this moment Not at SomeModerately Very all what so much so 1. calm 1 2 3 4 2. nervous 1 2 3 4 3. anxious 1 2 3 4 4. relaxed 1 2 3 4 5. worried 1 2 3 4 Please respond to the following questions: 3. Were you aware of this situation? Yes No 4. Using a scale from 0 to 100%, what is the probability that you will receive a $178 bill from the Registrars office? Please give your true gut feeling in providing this estimate. _______ % 5. If you received the bill, how much would this impact your life? 1 2 3 4 5 6 7 8 9 10 11 little impact great impact 6. If you received the bill, how disappointed would you be? 1 2 3 4 5 6 7 8 9 10 11 not disappointed very disappointed 7. If you received the bill, to what extent would this experience affect your budget? 1 2 3 4 5 6 7 8 9 10 11 little effect great effect 8. If you received the bill, what effect would this experience have on your finances? 1 2 3 4 5 6 7 8 9 10 11 little effect great effect 9. To what extent were you thinking ahead about difficulties the bill would present in the immediate future ? 1 2 3 4 5 6 7 8 9 10 11 not at all very much 10. Do you have pre-paid tuition? Yes No 11. Your sex (circle one): Male Female 12. Are you currently a Bright Futures Scholarship recipient or do receive some other form of scholarship that covers your tuition expenses? Yes No 37

PAGE 46

38 13. If you estimated that your probability is less than 80% on item 5, check the reason below that best applies to you. Choose only ONE response. I estimated that my probability of receiving a bill is LESS than 80% because: _____ a. I know pretty well what my tuition and fee charges should be and I would have noticed an error before. _____ b. I feel pretty lucky and believe that this sort of error wont happen to me. _____ c. When these sorts of errors have occurred in the past, it usually happened to other people but not me. _____ d. I think its wise to be optimistic and not to imagine the worst. After all, if you think bad things will happen to you, then they often do happen. _____ e. Other (please specify): _______________________________________________ __________________________________________________________________ __________________________________________________________________ 14. If you estimated that your probability is greater than 80% on item 5, check the reason below that best applies to you. Choose only ONE response. I estimated that my probability of receiving a bill is GREATER than 80% because: _____ a. I know pretty well what my tuition and fee charges should be and I had already suspected or detected the error myself. _____ b. I am bracing for the worst. Bad news feels worse when it is unexpected. Im expecting a bill so Ill be ready for it. _____ c. I always seem to get hit by unexpected expenses or bills. Im sure this is just another instance. _____ d. The university has made mistakes on my bills in the past and they have probably made a mistake in my case again. _____ e. Other (please specify): _______________________________________________ __________________________________________________________________ __________________________________________________________________ 15. If you estimated that your probability is equal to 20% on item 5, then place a check here: ___

PAGE 47

LIST OF REFERENCES Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182. Brandsttter, E., Khberger, A., & Schneider, F. (2002). A cognitive-emotional account of the shape of the probability weighting function. Journal of Behavioral Decision Making, 15, 79-100. Bruine de Bruin, W., Fischhoff, B., Millstein, S. G., & Halpern-Flesher, B. L. (2000). Verbal and numerical expressions of probability: "It's a fifty-fifty chance." Organizational Behavior & Human Decision Processes, 81, 115-131. Carroll, P., Dockery, K., & Shepperd, J. A. (2003). Exchanging optimism for pessimism in personal predictions. Unpublished manuscript. Fischoff, B. (1981). Acceptable risk. New York: Cambridge University Press. Gilovich, T., Kerr, M., & Medvec, V. H. (1993). Effect of temporal perspective on subjective confidence. Journal of Personality and Social Psychology, 64, 552-560. Greening, L., & Chandler, C. C. (1997). Why it cant happen to me: The base rate matters, but overestimating skill leads to underestimating risk. Journal of Applied Social Psychology, 27, 760-780. Ho, G. Y., Bierman, R., Beardsley, L., Chang, C. J., & Burk, R. D. (1998). Natural history of cervicovaginal papillomavirus infection in young women. New England Journal of Medicine, 338(7), 423-428. Jemmot, J. B., Ditto, P. H., & Croyle, R. T. (1986). Judging health status: Effects of perceived prevalence and personal relevance. Journal of Personality & Social Psychology, 50, 899-905. Kahneman, D., & Tversky, A. (1972). Subjective probability: A judgment of representativeness. Cognitive Psychology, 3, 430-454. Klein, W. M., & Kunda, Z. (1994). Maintaining self-serving social comparisons: Biased reconstruction of ones past behaviors. Personality and Social Psychology Bulletin, 19, 732-739. 39

PAGE 48

40 Lee, C. (1989). Perceptions of immunity to disease in adult smokers. Journal of Behavioral Medicine, 12, 267-277. Lichtenstein, S., Slovic, P., Fischoff, B., Layman, M., & Combs, B. (1978). Judged frequency of lethal events. Journal of Experimental Psychology: Human Learning and Memory, 4, 551-578. MacLeod, C., & Campbell, L. (1992). Memory accessibility and probability judgments: An experimental evaluation of the availability heuristic. Journal of Personality and Social Psychology, 63, 890-902. McKenna, F. P. (1993). It wont happen to me: Unrealistic optimism or illusion of control. British Journal of Psychology, 84, 39-50. McKenna, F. P., Warburton, D. M., & Winwood, M. (1993). Exploring the limits of optimism: The case of smokers' decision making. British Journal of Psychology, 84, 389-394. Mellers, B. A. (2000). Choice and the relative pleasure of consequences. Psychological Bulletin, 126, 910-924. Mellers, B. A., Schwartz, A., Ho., K., & Ritov, I. (1997). Decision affect theory: Emotional reactions to the outcomes of risky options. Psychological Science, 8, 423-429. Nisan, M. (1972). Dimension of time in relation to choice behavior and achievement Orientation. Journal of Personality and Social Psychology, 21, 175-182. Nisan, M. (1973). Perception of time in lower-class black students. International Journal of Psychology, 8, 109-116. Pulford, B. D., & Colman, A. M. (1996). Overconfidence, base rates and outcome positivity/negativity of predicted events. British Journal of Psychology, 87, 431-445. Rothman, A. J., Klein, W., & Weinstein, N. D. (1996). Absolute and relative biases in estimations of personal risk. Journal of Applied Social Psychology, 26, 1213-1236. Sackett, A. M. (2002). Optimism and accuracy in performance predictions: An experimental test of the self-protection hypothesis. Unpublished Masters Thesis, Yale University, New Haven, CT. Sanna, L. J. (1999). Mental simulations, affect, and subjective confidence: Timing is everything. Psychological Science, 10, 339-345. Shepperd, J. A., Findley-Klein, C., Kwavnick, K. D., Walker, D., & Perez, S. (2000). Bracing for loss. Journal of Personality and Social Psychology, 78, 620-634.

PAGE 49

41 Shepperd, J. A., & McNulty, J. K. (2002). The affective consequences of expected and unexpected outcomes. Psychological Science, 13, 85-88. Shepperd, J. A., Ouellette, J. A., & Fernandez, J. K. (1996). Abandoning unrealistic optimism: Performance estimates and the temporal proximity of self-relevant feedback. Journal of Personality and Social Psychology, 70, 844-855. Slovic, P. (1987). Perception of risk. Science, 236, 280-285 Slovic, P., Fischoff, B., & Lichtenstein, S. (1982). Facts versus fears: Understanding perceived risk. In D. Kahneman, P., Slovic, & A. Tversky (Eds.), Judgment under uncertainty: Heuristics and biases (pp.463-489). New York: Cambridge University Press. Taylor, K. M., & Shepperd, J. A. (1998). Bracing for the worst: Severity, testing and feedback as moderators of the optimistic bias. Personality and Social Psychology Bulletin, 24, 915-926. Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty. Science, 185, 1124-1131. Tversky, A., & Kahneman, D. (1982). Evidential impact of base rates. In D. Kahneman, P. Slovic, & A. Tversky (Eds.), Judgment under uncertainty: Heuristics and biases (pp.153-160). New York: Cambridge University Press. Van Dijk, W. W., & van der Pligt, J. (1997). The impact of probability and magnitude of outcome on disappointment and elation. Organizational Behavior and Human Decision Processes, 69, 277-284. Van Dijk, W. W., Zeelenberg, M., & van der Pligt, J. (1999). Not having what you want versus having what you do not want: The impact of type of negative outcome on the experience of disappointment and related emotions. Cognition and Emotion, 13, 129-148. Weinstein, N. D., & Lyon, J. E. (1999). Mindset, optimistic bias about personal risk and health-protective behavior. British Journal of Health Psychology, 4, 289-300. Zeelenberg, M., van Dijk, W. W., & Manstead, A. S. R. (1998). Reconsidering the relation between regret and responsibility. Organizational Behavior and Human Decision Processes, 74, 254-272. Zeelenberg, M., van Dijk, W. W., Manstead, A. S. R., & van der Pligt, J. (2000). On bad decisions and disconfirmed expectancies: The psychology of regret and disappointment. Cognition and Emotion, 14, 521-541.

PAGE 50

BIOGRAPHICAL SKETCH Katharine Dockery, born Katharine Sweeny, was born in 1980 in Burlington, Vermont. In 2002 she received a Bachelor of Science degree in psychology at Furman University, graduating summa cum laude. She then began her graduate education in social psychology at the University of Florida, where she will receive her Master of Science degree in December of 2003 and intends to receive her Doctorate of Philosophy in 2006. 42


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20110113_AAAAAJ INGEST_TIME 2011-01-13T07:04:15Z PACKAGE UFE0002622_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 3099 DFID F20110113_AAAHUN ORIGIN DEPOSITOR PATH dockery_k_Page_02thm.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
563a8ac94b44f97ed4beb2804d019859
SHA-1
867a2142f33d5e27d5d1965931b7b11cd77a667f
5670 F20110113_AAAHUO dockery_k_Page_02.QC.jpg
9cc9054d92769a9ba2ac3571001d0628
43dfb96733e1461420e1ae1ab00989e281bf76e7
51731 F20110113_AAAHPQ dockery_k_Page_33.jpg
39996c75bcc453c3a52c037b5a0aeac8
fc78a23cd6c0acdda2333fea256f63b4e18b6676
20998 F20110113_AAAHUP dockery_k_Page_03.QC.jpg
a07e4262dcc9d38463d2683e3a76fce0
6caa369e59bc7979cd371b56d9c801d64fd86906
62825 F20110113_AAAHPR dockery_k_Page_34.jpg
5f355b7197e9f5a490753f27a4270f56
6fffcedbe398ddcf331844624d94f04043d6bbda
80882 F20110113_AAAHUQ dockery_k_Page_04.QC.jpg
3f931aa73e129123f7bc8eda74c20d20
d80840b126d537ab70a0bafce07c5446175304c3
173814 F20110113_AAAHPS dockery_k_Page_35.jpg
a124cebf421a8755178f7277efdba4c6
26ac683f782aa96ca7630ed05092c7762f300d95
30976 F20110113_AAAHUR dockery_k_Page_05thm.jpg
d0a48851c6731604d0885c86ecbcbebc
d9f14e32d11bf8ebdd12b6a200ce1da340ea942f
211150 F20110113_AAAHPT dockery_k_Page_36.jpg
f72192feab15e874444ca68ac6953852
a49b98749c63cd3cf3f41f0a869a2d3d278e5c91
40813 F20110113_AAAHUS dockery_k_Page_05.QC.jpg
8073a1c395f2adad78592b9aa456d10d
bd38f3dbf58b14fd511b6aff2af7fc15f200e1a2
212580 F20110113_AAAHPU dockery_k_Page_37.jpg
5d896ecdef363f35e19672e2de188645
7efc0210a2fcd8be226b4a031860f8a3392fc7e8
30746 F20110113_AAAHUT dockery_k_Page_06thm.jpg
748094517976c4d247defb2f1d234459
0e12c68c1119653fb71307670c7c82aadc7c9ced
210936 F20110113_AAAHPV dockery_k_Page_38.jpg
bd6340654c7f66f63c2d63803ad4baf6
100ace8a52f97c9c773ce768505e7353f5d6a120
30720 F20110113_AAAHUU dockery_k_Page_07thm.jpg
bd83492093c062d514e7c496f686badf
b27d6127132acfacc4ac3cc8def206ed184b43dd
210069 F20110113_AAAHPW dockery_k_Page_39.jpg
625a93814a21d303f05ecf10b32feae0
d3e1bbc05ae658f1a9cda50cdd6fbd866b52150e
205222 F20110113_AAAHPX dockery_k_Page_41.jpg
e2340972e97e1431feacb4c194d54a7e
3c5ac4ab6337aa742e8bb5cf15f58e9bdec7d271
39005 F20110113_AAAHUV dockery_k_Page_07.QC.jpg
867a3eab81ad2fea086c12023039959e
1d012efb3f43b4cdab1335bbe1df05f7ed1e9cca
24139 F20110113_AAAHNA dockery_k_Page_24thm.jpg
25cb24a5eca9e844d30e54c3bf160f14
a502d451a2fe90741cd98a0ac4fd9c47f33b0b85
190456 F20110113_AAAHPY dockery_k_Page_42.jpg
dede1ff5b16e2137c92292973c842ba6
b821f6cf8f7fe90d5aa63646e0d46559a58dd446
18386 F20110113_AAAHUW dockery_k_Page_08thm.jpg
b40f1af0c1ed4e24494751de40fa8a85
cd88d702699252a77dbc768ab0504e70e40118b4
24848 F20110113_AAAHNB dockery_k_Page_14thm.jpg
b8b64fc08482377caab68b041b9db02f
27fe983c7f0800ce8fc30c4c8a79932891f69c21
196320 F20110113_AAAHPZ dockery_k_Page_43.jpg
19671e5f3574fabc1b9ab724a58060dc
b8576d047f064f1b5ad568a659e77289667299b7
57169 F20110113_AAAHUX dockery_k_Page_08.QC.jpg
cbb7846f1ac2be045ba10363721c83f3
308bc2762d32367798d494392aca684c64cdcdf3
50742 F20110113_AAAHNC dockery_k_Page_29.pro
de8627e84d09c810ada8f3efa21ef4b8
6d5a1f5ee4e4452d0dc3ae6a02502dd043ad7bae
22411 F20110113_AAAHUY dockery_k_Page_09thm.jpg
b1ca9bfe483eee093ea19afaad7cd2eb
5cddc4951aec83f91ae25fa6ca7dfd022a1aefb3
75459 F20110113_AAAHND dockery_k_Page_27.QC.jpg
db4b42220c06ceb0141d70aea5f77417
ed84956ddbc8079c783a55dbf9e7abfa8aedf147
52471 F20110113_AAAHSA dockery_k_Page_17.pro
8d22f68b5bcb6b3cf77200cd08fb8b7b
8cb72b2ad471e53f8fe7eda525072c2360ba348c
81200 F20110113_AAAHUZ dockery_k_Page_10.QC.jpg
9106314e76b7aefefce58ae900231f4a
b8d6af2d0cf9c2c527c13039f727a3e4469bceb1
21162 F20110113_AAAHNE dockery_k_Page_25thm.jpg
7d0de93282d67ce4e72c9d48f5f22cb6
2d7a7ee615e69de6d6df9577e903c3ccda4b8a81
51180 F20110113_AAAHSB dockery_k_Page_18.pro
43adf13895632e06233b169bc5c935c5
e4eba5a341d74d88eee852e8c3a07450341f6c73
41203 F20110113_AAAHNF dockery_k_Page_04thm.jpg
573fd20da366376829957a54c607d203
79e06f35e38b8c63e717d7b2db8426905e377969
45656 F20110113_AAAHSC dockery_k_Page_19.pro
97f4a3e63038986531fc5ea3642bca4d
d54dd5ec546bc879a5631dc18ae132d69c604ad1
71457 F20110113_AAAHNG dockery_k_Page_09.QC.jpg
e77bf614b652ff94422d87d746c401a7
cdfe40bf48e5cd00cb342f41c5c744cb6bb5f93a
59942 F20110113_AAAHXA dockery_k_Page_46.QC.jpg
9e2795aed8cc936939ab122fdefb74d3
f8693573e5c8c9ee6e824996767408b1180baa3c
46018 F20110113_AAAHSD dockery_k_Page_20.pro
07a96ff6071bc8c378ddac0667897098
1bb9975a818cfdd173d58189e280d4db7b27df7f
79049 F20110113_AAAHNH dockery_k_Page_36.QC.jpg
394f1a66587ea557272a0682559e2801
1e08ad3589a10eb90f98f9fb5f1ed2f48262cb67
21635 F20110113_AAAHXB dockery_k_Page_47thm.jpg
08693be582be4f54186ac3b9553307fd
3d9ca29b4032429a69e6079a5e2b115dd2d8868d
23240 F20110113_AAAHSE dockery_k_Page_21.pro
4d15a432ca57d302826b3ea364f29550
f31bee2182bea09dcb8e8121e50a0844144c433c
7754 F20110113_AAAHNI dockery_k_Page_03thm.jpg
ada41eee019d95223a733b1c746069a0
2335e750bb445f0b90ff1ebe1ee09469dd038d65
25490 F20110113_AAAHXC dockery_k_Page_48thm.jpg
06cbf84767981e7136055603a28a2c67
91b3dfba11aebc587cd159d73c2e177093fb6653
43281 F20110113_AAAHSF dockery_k_Page_25.pro
0c2f815873f4ecdbe0801be9e49edae2
dee323dc258b674a60ee5f450b4e967aa4454806
199861 F20110113_AAAHNJ dockery_k_Page_32.jpg
b05b50737e60e50100ee91dd10f48cbe
03dece17dd8f941df6a90d3c36d1d181528285bc
83063 F20110113_AAAHXD dockery_k_Page_48.QC.jpg
ee89696e5fc0fb8763813c125d317a3c
f91f7aa9c8e9d036d4097691d8f3f4f066f770ea
25602 F20110113_AAAHSG dockery_k_Page_26.pro
cecdb606ba4057971321901cada34bf1
917a8acf129c716597021297dab51bcd52160073
1917 F20110113_AAAHNK dockery_k_Page_27.txt
d88e11bacb32c73026accc3d86aec8c2
f6d320053864b1f0150e72ecf8fa0c8e6c75015b
23674 F20110113_AAAHXE dockery_k_Page_49thm.jpg
33f06346eeb057db50d8d9c3b687a45e
8b0526a3440ffbe19c1bea1033807d8ff744f248
48494 F20110113_AAAHSH dockery_k_Page_27.pro
08bd3fa978631239d195a8902f2a48aa
c908af1b02533c22c9e667728b2634a2aff97f9d
78265 F20110113_AAAHNL dockery_k_Page_39.QC.jpg
94ea636964d9be9e5321546ee067e11e
f0e11d0c3078b359844f15a75b6b74d7e94f2c72
81407 F20110113_AAAHXF dockery_k_Page_49.QC.jpg
89280f704541bba70b2872fcc913ad93
21de8f0ad4331109924889249b8c55f79a66e140
32182 F20110113_AAAHSI dockery_k_Page_28.pro
580c04b965dee075cb5339b54d80d582
78e97269c86d9a404c8dc4193a746360ab991c3f
2036 F20110113_AAAHNM dockery_k_Page_11.txt
15ba8153fd7d4052c6384c9bc0f3196d
bc670e7b61b9f9c27323f71db4283522af809840
21831 F20110113_AAAHXG dockery_k_Page_50.QC.jpg
7a20864cc7e9f0e072bd835fdfcc7862
722ca3524116dd347630f3aaefbeccd8ae79c950
49657 F20110113_AAAHSJ dockery_k_Page_30.pro
d918f22afbf9db57362f4f3cd141c6b2
aa52dd7560d68264654315e2c4ac1b02da2d649a
23289 F20110113_AAAHNN dockery_k_Page_42thm.jpg
70ae0313ee6040a8793f16d2bd5dc224
27edb2c5f74e08893531a716fab161d5c89502c8
50356 F20110113_AAAHXH UFE0002622_00001.mets FULL
e266a26ec4f9a4c758772d89073725c8
fd64793ba1217cf0d7ae94806d1a3395f60e8d32
52218 F20110113_AAAHSK dockery_k_Page_31.pro
135cdc71c7b7947c5bdb0785778caafa
3a6b730977e0935de8e539ba6eedd7274eec694f
47965 F20110113_AAAHSL dockery_k_Page_32.pro
ff43060245470bb92caa4a29e1242082
30bc5c621d207fa395dbe88162281c957581d908
196794 F20110113_AAAHNO dockery_k_Page_45.jpg
521766f2b4bc2fa94edf6ce4aae325d3
7c53e2e919eb69a32968feecca293f40f8254a9d
11195 F20110113_AAAHSM dockery_k_Page_33.pro
761c3ead6809f4cc760093e4150e1568
14ddd58f9c429db00e85687bee079fb8a7e68d37
19518 F20110113_AAAHNP dockery_k_Page_33.QC.jpg
62f3a33e9a3eded3c933dea868c87cf1
3eb8ce7a494eb551d160efe9a94fd0cc73a75c69
14859 F20110113_AAAHSN dockery_k_Page_34.pro
2d30a560f7d115409c6fd9ef53b6026d
102cc40eede29fd2623aad0fccefe656e7ed53e7
76669 F20110113_AAAHNQ dockery_k_Page_40.QC.jpg
ac9b3adb359052ce14c58d3c9fe90d6b
80de48f48634a1a0b56322ac60aad21fc69ab22f
42008 F20110113_AAAHSO dockery_k_Page_35.pro
da7c8b4217a9727cd79e2af52147bf55
29a5bb00d36eb65d5a3e13c6c365f4cec1fc9116
1053954 F20110113_AAAHNR dockery_k_Page_31.tif
ae805156dbf8f228a7e3d32cb8e91381
ea70691acbe984a701c5724a1b11c72f10fe0964
50647 F20110113_AAAHSP dockery_k_Page_37.pro
2e53e4eb4a7f02392b77278c30776c8c
0b10adfe29e917aa780cbbc43fe45eaf8627f4c1
2021 F20110113_AAAHNS dockery_k_Page_38.txt
d9dd50da0142b17316a8180e6e32a033
a6e1347dbbb9c6897ae60277bb5b061fa00b9c1c
50360 F20110113_AAAHSQ dockery_k_Page_38.pro
72f3f4db60725ef285c1bfd6cd2fbfba
a926b575cb23366be1d1a7832b57d5947b0d9cae
F20110113_AAAHNT dockery_k_Page_01.tif
119aa6fa8ab4be290440ab63ee04578b
175391901e70cb3c2f0087fe8e98d2c304bcc003
F20110113_AAAHSR dockery_k_Page_39.pro
02c143faa1b07271c457cb08ef10e4e5
e73439e4f9ff1f437b84d45a7bf73acacf66420f
373 F20110113_AAAHNU dockery_k_Page_07.txt
1ecdffa3c1d9a63e32a4b10677ffb17e
fd8a10e3b1f84ba280800c544342bf6bfe91e8de
49000 F20110113_AAAHSS dockery_k_Page_40.pro
f6699f9fdb43a4eb35ba4f2607f5c0a4
5d548a3a919bd81d0400e49d2e43024452dcc3a8
F20110113_AAAHNV dockery_k_Page_37.tif
559b35824b9f012f335c5f623fa7bb60
7b1959c35866e0d8ec4179a17dfac8222102d6b1
179936 F20110113_AAAHNW dockery_k_Page_25.jpg
3d88a95428bff85e124dce0e0489b22e
d34a0548508cfa988f358582182a8d0d6b542082
49984 F20110113_AAAHST dockery_k_Page_41.pro
c3c35e8a6c66f82cd5520ed8158990a9
caab9a784702512ab71824a597d150a6cebd9545
90411 F20110113_AAAHSU dockery_k_Page_43.pro
461ea0c6fb4db734021d3e83521242c6
89cad7c98df29bfbef4d415232f197373e42d02e
8423998 F20110113_AAAHNX dockery_k_Page_28.tif
248c8b28839db6ee5564b892499d0ef4
aff6561975ee532d1fdaf58e6d28d0150e7196e5
43398 F20110113_AAAHSV dockery_k_Page_44.pro
3b06035934075521d845b0a0377a4f7f
f35af942089c45c72ee5745d60d87cbf52e2dc20
38621 F20110113_AAAHNY dockery_k_Page_06.QC.jpg
312f415cac12ab79bdbd7aec419d23b8
3b79da9bcd63d5722cc986ce83659c46f259ae1d
90371 F20110113_AAAHSW dockery_k_Page_45.pro
e87d0cee3eff68359c06544e0eba5c5d
f72c327b26425e9a5eca9ae6300013eb0fd7a9b7
F20110113_AAAHNZ dockery_k_Page_38.tif
94a7fed23428242078d27b30d64b6f90
fc3576461ff213fa77200959e8eab0313b32d8ed
51782 F20110113_AAAHSX dockery_k_Page_47.pro
e938d7725bd2d603e06b0968ac9dd368
8c5b286ff3acc58789664bade0948342d5d48960
162323 F20110113_AAAHQA dockery_k_Page_44.jpg
39adcf2eefc5f52b51c2a5248681f944
53c78380bfde6fb909cd113f4a28cf41b3ca9fb4
58161 F20110113_AAAHSY dockery_k_Page_48.pro
000c88a506ab6edbf693543e3a239524
f7d88e5857fb5e8dafd51e673928ffd6fb2295f0
162241 F20110113_AAAHQB dockery_k_Page_46.jpg
8a0a184b6dde82d508cdfd94c1edcddb
262cc7ae1efc50c216be85cf665cce9caf0a8448
11802 F20110113_AAAHSZ dockery_k_Page_50.pro
dfd9ca96cad40bc4083975f1f380b5c9
3a5ad1fa19271f3c4cc825800328d58c4d45b446
206629 F20110113_AAAHQC dockery_k_Page_47.jpg
1924d7e5e8b291b5b9ca23cf1cc7f886
82221a8ae1da2db8b7d9447bf7bf81d39ff8b481
264046 F20110113_AAAHQD dockery_k_Page_48.jpg
a03f4510244848e099956284460af613
26ea3df3ce4efc1b322cebcba5887268e59e160e
24705 F20110113_AAAHVA dockery_k_Page_11thm.jpg
7265e1bb5f71181173ccadbf1f9e8d71
98ebea9914e54ab37583bd262ddfd166ab7f3e70
58633 F20110113_AAAHQE dockery_k_Page_50.jpg
7c3432ab3dc39ce8a3f504c0060f6351
e1ba2ce168d1f83e0a54aa417387d0f806ebccfd
77548 F20110113_AAAHVB dockery_k_Page_11.QC.jpg
b0a6dfed1f5f6371cdbd35b9c774ef4e
b4c12a56fe457657e9e03cb62b95f884179ffb0e
F20110113_AAAHQF dockery_k_Page_02.tif
c3bbfe7849a2f46a5235f226c11f212d
ee9387f914b056dde9a8d9e9a8d49c1252ee26cf
80206 F20110113_AAAHVC dockery_k_Page_12.QC.jpg
d0011a68103543c1681da4cc31dcd189
17cf4906b3b638f86edd3233d01aef09e0da6c22
F20110113_AAAHQG dockery_k_Page_03.tif
bc4de5bb399d8594ebe90f2c492d8f51
6f424256852a80019a42cfcba10aad1adab86c66
25384 F20110113_AAAHVD dockery_k_Page_13thm.jpg
7fdf1c231b2d9373b5156d3bdef76059
8a9e49158945cf0e0e7746b3650e11ec559b51fd
25271604 F20110113_AAAHQH dockery_k_Page_06.tif
e4849605cb4dca177bd16bdb79191623
fa1bdc4450b3eadfe4832034af756c45643e9175
80143 F20110113_AAAHVE dockery_k_Page_13.QC.jpg
756f6c25d3e132707ebf345cf933062d
4cc2450625758a086498e87cf83317d53426d11e
F20110113_AAAHQI dockery_k_Page_07.tif
ad5f11ad2761cd635e5d8fc7932e1c13
5732c90f9dc559485b2200002647231fb38c66c4
81248 F20110113_AAAHVF dockery_k_Page_14.QC.jpg
c57d9cc12d1220c44854854790204580
8e10e8111fac61bffc4be746170b1b6e4e669963
F20110113_AAAHQJ dockery_k_Page_08.tif
c207728f50cbb72eccf2a6e687d01fa2
3226c2b74659ebd25513c3892019d1539aee042b
23611 F20110113_AAAHVG dockery_k_Page_15thm.jpg
3359f18c1087393ea956f832a71b0d2c
cc903d045d5f3fd356b135fdc4dd397227cf08da
F20110113_AAAHQK dockery_k_Page_09.tif
fdad0f25d83acff8b153f8f7dcf05092
aaccda17a44427c80a011875d4004813a8ad67b5
74061 F20110113_AAAHVH dockery_k_Page_15.QC.jpg
ae9d7a7db75e7a2e6375a19634664efe
1a64aa48623abc3b32f313fe31e0610f07b2430a
F20110113_AAAHQL dockery_k_Page_10.tif
9c7094a2776f8ee412cb87530739fbf9
fec748276f40e2e48a10ebb33b0e8bd02cb84497
24369 F20110113_AAAHVI dockery_k_Page_16thm.jpg
ba59bbad8785d53153e1f725d438caf1
70eeb4154bc4b23fed567c2b2905d95d049f9a2a
F20110113_AAAHQM dockery_k_Page_11.tif
1c1922f39a713fb78681ba2465914854
dbd7c28e32b18b942b3277c984796f5444d874e4
77138 F20110113_AAAHVJ dockery_k_Page_16.QC.jpg
654cdf1821763840ab4cb1049ea5db4d
ff709e59661793ddf64c753b44b8826222707116
F20110113_AAAHQN dockery_k_Page_13.tif
258810bacf232ef89387c912223cae5f
ddcc4f7db87eaebb651e5efe0757cc8385dc17c2
F20110113_AAAHLP dockery_k_Page_39.tif
f051b28c313658e6404ce55e4006d7b5
8d25f8cdeef42097319d386bc5072f22017d6487
83044 F20110113_AAAHVK dockery_k_Page_17.QC.jpg
c58c6723965a8590d64cbc7e86998bed
6e0d563988f7a02a222e5e56f527beec19eec378
F20110113_AAAHQO dockery_k_Page_14.tif
7a8bbc9bf8f12025be1ea2218dffd170
96ecbec83a4ff661044ca2c9ddcd6a27385a6153
25958 F20110113_AAAHLQ dockery_k_Page_10thm.jpg
eec41b0609a2d058b869e2f2e1f40897
a42488e52b4899ce55b2f8bbea890b096d1cc983
69422 F20110113_AAAHVL dockery_k_Page_19.QC.jpg
12610c2b6d26bd32a090b9bb15883165
b225df8fb483485ed200bf9d7269c99909406f89
F20110113_AAAHQP dockery_k_Page_15.tif
bc15b4029c8cee514afe22f64bd7bbdf
b54a9cc329c0683e8e8d083ae6dc23a38d629fc5
25058 F20110113_AAAHLR dockery_k_Page_17thm.jpg
b1d38c88e78fabe17524232440caf00d
5c9035afcc535cc555e1d8f38f29d8255c4f89d4
23292 F20110113_AAAHVM dockery_k_Page_20thm.jpg
9e7c37e822819824ab0506cc5a9c5f11
d6fa5e25204b7cb9a35eb5a4fc1573e43a740a40
F20110113_AAAHQQ dockery_k_Page_16.tif
d27ca7f90b420c8889d5a24b5a4571a4
da846b534e750628c106a3b6c024e38d7bd9fd27
70290 F20110113_AAAHLS dockery_k_Page_05.jpg
ebaba22dd72bb1836519696253b99f56
af58ffac2285e4e7d97e267fccedfa758771170e
12713 F20110113_AAAHVN dockery_k_Page_21thm.jpg
b3f9ccadde6fc747a607c11b71ea9657
4722fda098210c97a4a64a0f2cb9e696aa41ee16
1863 F20110113_AAAHLT dockery_k_Page_20.txt
8c21721aee1f3d8632eef102421428cc
f49b841ff8dc21772a5db40c7daa95f6e288cd44
40016 F20110113_AAAHVO dockery_k_Page_21.QC.jpg
a28d762b16fef928347d6e5ce2726faf
3b086626865f53dd7d106c20917346e33b7f799c
F20110113_AAAHQR dockery_k_Page_17.tif
2340909a8e67a97ffba799ec6d92d34b
089d7705216f580da7c435c59deeccae69810ba9
67442 F20110113_AAAHLU dockery_k_Page_45.QC.jpg
fadd00fa8ca2ff28c2008eab8fe015a1
51769ca9b01c53343b86603f8189c422fc6a1d99
17575 F20110113_AAAHVP dockery_k_Page_22thm.jpg
cbfb8bfd0c0de37f899d825a05805dca
016ba6cde6813a5b076889d08755e9c5d86ff764
F20110113_AAAHQS dockery_k_Page_19.tif
2c655ab9eb07f3bfc048f433116d89d3
518f2531ba0120356c159033696743debb6797b0
1981 F20110113_AAAHLV dockery_k_Page_39.txt
0fa010b445b5fc94f28522cf38eb8bdd
b6b21c8adbfbfd39ba8f50e584d60e6fb53235c5
55168 F20110113_AAAHVQ dockery_k_Page_22.QC.jpg
1849670d2c2dd31b4ecf34239c8d215d
93932b6679e0b70c464461f9bd5d78edaf814be7
69090 F20110113_AAAHLW dockery_k_Page_47.QC.jpg
21845a616d11a5036ee41b2e9b6d9e84
a77e8730228137052c81ef4fd5d16be93a9efe57
25370 F20110113_AAAHVR dockery_k_Page_23thm.jpg
80b4da5ec1378af9e1de1eac1835d2f3
c82e0039832c2b7f72547b6f61165034db9d8f54
F20110113_AAAHQT dockery_k_Page_20.tif
ad8504326c8360d686726941aef1b0b6
6ca2a4bdbf76771aedac3a1ad8febae9460d7455
22024 F20110113_AAAHLX dockery_k_Page_19thm.jpg
102dd106324db1b8cea6f764ad4823f7
a29510c8ce3d7b00a664f56230cc802d59a28e61
79330 F20110113_AAAHVS dockery_k_Page_23.QC.jpg
0ab2230447a916697ecec9175aa5ec9e
6cee81f2962f0e7d42e953f6fa6e8510fd2b24f9
F20110113_AAAHQU dockery_k_Page_22.tif
449d4e60ac780aa26a53d57a5836f29c
1eaf70fc09007e85694ca919d934ac85b5543be5
2393 F20110113_AAAHLY dockery_k_Page_48.txt
c4085f80eec87c8fea75f045f342beda
184138925c09be6a064904b306a60cb6cc810ac1
77432 F20110113_AAAHVT dockery_k_Page_24.QC.jpg
7e20b571b2cc9e25df7fbdd752febbe5
7ec958aefdd61629559efb1444fbd2d8b872f701
F20110113_AAAHQV dockery_k_Page_24.tif
51e957e4eeeb565431f0f53d6b1879de
c06f4fa00d3353498632d87e5fddb5fc7ee43c04
2085 F20110113_AAAHLZ dockery_k_Page_17.txt
ee8bfeef6c26801ec7cfd033d15b6077
2b262bebf82992a5461b1468555097ccc7aa7e81
62962 F20110113_AAAHVU dockery_k_Page_25.QC.jpg
4e4321008414ecbbfeb0490b23c79d4c
d1867ec79074dbfce409c941a2736e0c0b9e1d3a
F20110113_AAAHQW dockery_k_Page_25.tif
1ad1dda68a433cda7239689dd469ba0c
244c7f5c1b2d66816c3d50530bbf19f27d5e92e1
26418 F20110113_AAAHVV dockery_k_Page_26thm.jpg
2e015a226281d7a713f0ce8a6e987bd0
975cd4a1d8660ffa0f36b59b17183e48a95634f6
F20110113_AAAHQX dockery_k_Page_26.tif
ed5f66c5c9769f342c2dc6756edb6ca0
e418ac28d784a1fe6070fc9bc296a36f2c7659be
80458 F20110113_AAAHOA dockery_k_Page_18.QC.jpg
f1c438fe02b8280697e01c1dc4665a4e
85d661d2a70d91533aefd04834c7561a2af40b03
F20110113_AAAHQY dockery_k_Page_27.tif
abde32b29ae417d03f60f654e73b9e76
83c145b2a1cb1720d3b370dc3593c129a1d5b0f8
57350 F20110113_AAAHVW dockery_k_Page_26.QC.jpg
f14ff9c1217a965cabd748b84f6a8bef
60fa696e42ecaa97966b0e14b7999ae8e22b4961
69569 F20110113_AAAHOB dockery_k_Page_20.QC.jpg
087058da77ef82e8aca71233bfb9e803
3ec103104dd3de2176e1832e49c20b7278ca81e4
F20110113_AAAHQZ dockery_k_Page_29.tif
dfe3254772f13e223d99becaf6b61a80
e579a269d562b3dc22be07a1dbb11a644935eaba
23977 F20110113_AAAHVX dockery_k_Page_27thm.jpg
520dab020d412da7d1c23729070d1eb8
02c5b58f62187091f49501751f538b29b55fafe2
51785 F20110113_AAAHOC dockery_k_Page_23.pro
541c608524e7cda6f65cd40a6dcc4b0a
db23f920ca96444a6ae55c3671e7cb9e6dc17391
457 F20110113_AAAHTA dockery_k_Page_01.txt
a778dfcbb2c1c162f245778bf0c0b9fa
ef1021e5c9ee4219e6b29038c2b404df8442f8f1
25737 F20110113_AAAHVY dockery_k_Page_28thm.jpg
5eed67a2eb578fe3e7723fc86e317fad
a74bf3640f037a99f0e1d7d15efcd625f1bbb725
50666 F20110113_AAAHOD dockery_k_Page_36.pro
805ddfdcc637c1194d818718dde80c6c
9e8be377b7738984e500510f21387cce715e4046
107 F20110113_AAAHTB dockery_k_Page_02.txt
1dc6caaeef4314b039c8f12e44af9494
c816db575e1d9e43d06f98565afbb6c709b36fc6
23425 F20110113_AAAHVZ dockery_k_Page_29thm.jpg
931d74951c3ff56a9c0afbb84fa642de
5582ef12e40e79825e165000cc306c6d553849d5
52924 F20110113_AAAHOE dockery_k_Page_14.pro
f2e6ca532076f6d3e5686fe09dfe39b7
35472b256ec49e32055faeb41404598eb15e320d
2640 F20110113_AAAHTC dockery_k_Page_04.txt
f74da0efff047743ffb88585ac183c38
7ff968383edd91f1125551330a3bc2ecf28e95d0
630 F20110113_AAAHOF dockery_k_Page_33.txt
f0eba0a3d3a4fb3ecb07fe19e9ef2825
579a75d41af26b19b4c5216c7c95bef97e30f3f1
680 F20110113_AAAHTD dockery_k_Page_05.txt
dc7951d4297887dc3bf355613af29313
97771483266ae17c635e2d8cd77a58a9349bb037
25016 F20110113_AAAHOG dockery_k_Page_12thm.jpg
d7d06f8088b946890b3c64a1f74002db
4eb4a5f4223f4417f94589f7c710e3eba150f257
375 F20110113_AAAHTE dockery_k_Page_06.txt
fc5cf1e91a331984fe4b53db7060c8ad
09753659a66467517775ed068a497e98f69cd291
7677 F20110113_AAAHOH dockery_k_Page_50thm.jpg
0512a11a07c43f5ef965d6d237073b14
3dc09a102e34b0501e143cbad3e01b306e32ba8f
1659 F20110113_AAAHTF dockery_k_Page_08.txt
07518457d481c750afab29475f9a503d
df56b5c00ddb2944273d6f34e5fe45cd34b98513
131077 F20110113_AAAHOI dockery_k_Page_26.jpg
f5fc3505164d319b51567d3987e94c22
e9272f35884511b5a94444c0f76e8c46dafffee1
2047 F20110113_AAAHTG dockery_k_Page_10.txt
f7de12b83a46c75d19ab125de9b11dfa
70cc4f63715ab25f3085790162260dcdcfc0995e
6957 F20110113_AAAHOJ dockery_k_Page_01thm.jpg
df8698b5a5663cf485a1496f1fd9c294
19079f2bb0094da705edd7604b9838adbb80330b
2015 F20110113_AAAHTH dockery_k_Page_12.txt
ea05323dacb63f0b318b559d518a0dbf
dd34995b7b9c9c9f2d42ab4d855456696dbc3a07
248378 F20110113_AAAHOK dockery_k_Page_49.jpg
18900954eb7fd55dfb90af79f1a5d976
a7838d4367208767dfdfc361277349ea48df5412
2003 F20110113_AAAHTI dockery_k_Page_13.txt
5aa703a140aca1131d80722e00979c54
81b635d87f93e1721d996e5bfa3ff4fb749fec02
50842 F20110113_AAAHOL dockery_k_Page_13.pro
fe0a5c65fd11a4037a44c80853baa810
152bcaac49767f43cfb284986c2375c4aa220345
2092 F20110113_AAAHTJ dockery_k_Page_14.txt
5316a3b3523ffeee9fb874f93e1d5c8d
3e74b2c8c988a51911cb5c7e2b76662dc5762dbb
67942 F20110113_AAAHOM UFE0002622_00001.xml
9f55bc9884cb710ad4ff49cba26e7539
415189b54d76002ee2e25e1aae45cb927c89fbe3
1895 F20110113_AAAHTK dockery_k_Page_15.txt
77e58f3d5dc4f4e622123093695b1ce6
606fc1b4fbd69e15a9d4207b045bd3c9c4428412
1926 F20110113_AAAHTL dockery_k_Page_16.txt
c436d395f4b525e78f9a936dac9cd517
aea4dc80f670fe70aeb1e0ca5768d275eff42eb7
2014 F20110113_AAAHTM dockery_k_Page_18.txt
a04eebd4e5b5bd4239513655823d21ee
016c3fc56f46468e5d1ef6d41903bdbab38e875b
1812 F20110113_AAAHTN dockery_k_Page_19.txt
ac8a989c7d383896b59d23421d06caca
db7350b909ea5f98ecc8e93087af87f5be1dc819
14331 F20110113_AAAHOP dockery_k_Page_02.jpg
76f6bd56cb03f13b8f477ed4f0e34cd9
79ea1b26ba51efe3af861aca5ece47c43a4ac1d5
936 F20110113_AAAHTO dockery_k_Page_21.txt
aab3af721b7e69cb83a14be5e4c2ce13
7380b0fd81f1477e9cf6a39697da6e68fe3c1e60
52010 F20110113_AAAHOQ dockery_k_Page_03.jpg
bb72d791de2d7c7410e094af81fafec2
c3fb90ff52882ea317de4f3de3b14c708e53a990
1588 F20110113_AAAHTP dockery_k_Page_22.txt
5a6463df7417058287a289de60e4bf01
d2c7c85ab3fe51fd6a9affe46619703faa238cae
204818 F20110113_AAAHOR dockery_k_Page_04.jpg
713733bb14051e9cb2140cf7ca6f5851
3833b504224eb8af41c6b86fd95b59040e25693d
58259 F20110113_AAAHOS dockery_k_Page_06.jpg
4237c7927c5369feb2dbba67635e434b
766e7587ef0e18bf196e1b2a946e7741d09912ba
2044 F20110113_AAAHTQ dockery_k_Page_23.txt
7edac31db87c17483ebd761412524c2a
76101ab1ab524eb6be77140e37afaeb3d248003f
58801 F20110113_AAAHOT dockery_k_Page_07.jpg
1eec1568cb13e937c6877eea03ce1288
0f06df56faefdbf2c68e59afb99de2c3c96d32e1
1988 F20110113_AAAHTR dockery_k_Page_24.txt
524740c85ef55004d4d89dc37c4a04f4
050fe25bf929d514a62974fdcb44865d7141e6e8
163853 F20110113_AAAHOU dockery_k_Page_08.jpg
49bb01e7cd3f11c9122599b3ce7f624f
582b4f0d366f5517d5f19aba1268cb4d5ed5f074
1827 F20110113_AAAHTS dockery_k_Page_25.txt
cd33bf9bc156b3127c18cf04c8b4c08b
50ef6c7c31fe9e025d18e418c7a5c0cc44104353
190203 F20110113_AAAHOV dockery_k_Page_09.jpg
c35da23f5a40ec9d3e6957b0652acdce
f90eff51b02e8a586c79362013e7e39cf8d0be48
1126 F20110113_AAAHTT dockery_k_Page_26.txt
61f6a47f5f3271bc2fe277abfa76d415
318c7696bb563bcbfb6dd840fd21cd7698591fc8
218862 F20110113_AAAHOW dockery_k_Page_10.jpg
c47e98846286a9adc33898e6df5af06b
1b8af982e070feaab664bb68517a497b2738323a
210359 F20110113_AAAHOX dockery_k_Page_11.jpg
541a8c0cdf48cbc1c8747e3d28864d50
e996168a0a81ea0cbc52cfa2bcf1bbadeeaf19bb
1961 F20110113_AAAHTU dockery_k_Page_30.txt
1182a74ebd6a095fa3eef75eab59ae78
5d35da06fa11aed295d21bfe91089e77baf9ca59
F20110113_AAAHMA dockery_k_Page_45.tif
9299a2e13242873b629cb1bf7f05992b
3c9455bb61bbd212572c1b67ae9de27dc1ae81b7
211866 F20110113_AAAHOY dockery_k_Page_12.jpg
4de4388ed8aba5bc11ef6dab0dc628ca
49f815666876be7d082ae7b621bee6458804ef9c
2090 F20110113_AAAHTV dockery_k_Page_31.txt
dfc238c799f16009ee6787000469b228
7312669d7546afcaab872f6b9fbac0397fadc488
F20110113_AAAHMB dockery_k_Page_12.tif
2916084e399972230788b77284bf0862
83f515e06755203e46de07611a59c45170baf916
212674 F20110113_AAAHOZ dockery_k_Page_13.jpg
b2430b75d2061d9e6ff7f7994cea9477
469aa018ba55a6b1a6bc5ec35732f2bdfb73adbf
1891 F20110113_AAAHTW dockery_k_Page_32.txt
d9925fc23a7b903c808e028b49368c42
f7918d76fcc0337291eed98d31b1672e60b2ce05
50217 F20110113_AAAHMC dockery_k_Page_24.pro
14883d0010c03d12ef65e1d14ae9ff46
acd496edc6829d46ee06741a17d4fcedabd179ae
914 F20110113_AAAHTX dockery_k_Page_34.txt
e4476e6ec6d2eef8d1546101314307ae
6c211c7ca6b363f9a1da105ddb59c20749d8d85f
F20110113_AAAHRA dockery_k_Page_30.tif
1cb14380bb8bbf24d1f2e140144b8927
21d452af1c1ada14eb8cc18c8639ec93dc7568b2
1744 F20110113_AAAHTY dockery_k_Page_35.txt
15973ebfffa9363ca1f413d7144e35bc
80704d6686b35fa633fcffaa6a407774cb6a911d
45323 F20110113_AAAHMD dockery_k_Page_09.pro
3c8d67188ff2d2b6950b082dc9288381
7f4788d9328e1ec6c527e959b2b8fe265b4bdf7c
F20110113_AAAHRB dockery_k_Page_33.tif
49cc7f283905817b710d54123b630042
a5b43d3caf6f98abebc29ac741d1aad736c868f8
2018 F20110113_AAAHTZ dockery_k_Page_36.txt
98a83d5f71de88480d0be49e11d47fa3
6600056042e744436cc0a75380a8106c5ca4a16b
F20110113_AAAHME dockery_k_Page_32.tif
75f26fdaf838edd6a1ebf05e07bcb095
290527475caadcca57139338452664b1c3471f50
F20110113_AAAHRC dockery_k_Page_34.tif
bbd9ebf76dc4df97342d512ad1a31417
705a93eb6ed88c7977114d6c5f8b9bf95ce37449
F20110113_AAAHMF dockery_k_Page_18.tif
929c5480e3742ea162d2cd06b08a91c1
4b58dc43cfd501bd9cc720912a5cce71302a5761
76979 F20110113_AAAHWA dockery_k_Page_29.QC.jpg
2ecfad394e1eabd229d5739021186aeb
a7f5e4c81554d4e9b3e149dbcae9b14146bc7909
F20110113_AAAHRD dockery_k_Page_35.tif
948690f2317fdcaa45a618890de79874
de4f5cfba7bb782c036c8e410902084703a6d52f
1859 F20110113_AAAHMG dockery_k_Page_09.txt
73161ead25fe915ebcb43adc72130511
5b18b186ad62840076c880ac9a4fdd40a8fa4992
24330 F20110113_AAAHWB dockery_k_Page_30thm.jpg
afc3e87c50d0a9637827ddc142116647
40a001ea00b4f97cc1a1fd7a21d1813b01969da2
F20110113_AAAHRE dockery_k_Page_36.tif
dae42266628961b117a58b9b8e73bcbe
4bfbcf8ee49fcc301ae20a7950f61073277e52c2
1631 F20110113_AAAHMH dockery_k_Page_28.txt
263537eb2bd06a6484ba1f9615a92d8a
1cbaa763b97d92f76a6ec28c0032776a89a3ac6b
78583 F20110113_AAAHWC dockery_k_Page_30.QC.jpg
af64dab6dd7607bf85fcd74427d57810
f3db8cf47013681329e716e845c1eba93b9cd84c
F20110113_AAAHRF dockery_k_Page_41.tif
49e9124a20161af7283236b2ccabcb82
a30c7ef73e35d3452520495d346bbd6ee1c254dc
34819 F20110113_AAAHMI dockery_k_Page_22.pro
49b396e0268b23be4398def8c9dc3051
ce0594a830f154f6a1cbeebc54cbddc2475c25cc
25414 F20110113_AAAHWD dockery_k_Page_31thm.jpg
2942f314f15b242322b9803fdb84fffa
1f618f71cdb93c805cc5c33852c44bd71d08059d
F20110113_AAAHRG dockery_k_Page_42.tif
11372d8cb5d5130d1740fc9132f7b144
60e3c6ebd33a17e26e3a2aad946ecffda261bea3
F20110113_AAAHMJ dockery_k_Page_23.tif
edafc520f80e599df7c991a007201364
6aeaddf8bde162b19edf4841b5b65ac45795dd7f
84555 F20110113_AAAHWE dockery_k_Page_31.QC.jpg
5cb9d0c55876ffddf2487e79101d4b5c
f833c4d321908f0c0e8eef695b90f6acdc7b7b43
F20110113_AAAHRH dockery_k_Page_43.tif
194994909d481a15347ae98144112a5a
2e23eefa9a309d4dbd6eb7b8ccd22a1de34070b5
58403 F20110113_AAAHMK dockery_k_Page_49.pro
57ca7c2f0586cb6b5d3bdc83ef817721
c0c92e798b0e9517d4367e5050a3950cd089019e
23403 F20110113_AAAHWF dockery_k_Page_32thm.jpg
465439ed547629b75d984d8e1b53becf
c4ce1dc0e265da03548f53164e574bef96d3fe30
F20110113_AAAHRI dockery_k_Page_44.tif
681ebb7a52b3f05a3ed0a815be1f4327
f33cb3fed03637fd6b3d70bbe76bf866dbaacfaa
46020 F20110113_AAAHML dockery_k_Page_42.pro
9cc0ab9e10eafb0ab8b8845ebcf7d388
14731d96e8bdb04624c185d9307bff1841334d09
74374 F20110113_AAAHWG dockery_k_Page_32.QC.jpg
5f7e7c445601e6ced9ac645e4973b526
f34279047b771f20b4a3f536e16cc86be60cdb76
F20110113_AAAHRJ dockery_k_Page_47.tif
b7760a6a9eb034118e4fa030a533cdbb
12b1c9a2a18e8451b401bc23c5a301161790e8e3
54607 F20110113_AAAHMM dockery_k_Page_01.jpg
34d61fe8901ca38ac86acd03da6df2f5
320807f5a3e8fe1cb7d4b27cd429105ac5b66634
8371 F20110113_AAAHWH dockery_k_Page_33thm.jpg
122b7caf5f043dd8e0e36e3a3d6d20e4
ddf84e5c9837d0771383d816160abe5e6dc5a038
F20110113_AAAHRK dockery_k_Page_48.tif
f34642385057209788820e2e282aca70
d7d76ae07a860363e3ca6b6698d072b0872a1474
10293 F20110113_AAAHWI dockery_k_Page_34thm.jpg
27c36cff09731e42f34ec4a7dc060f85
5225d107901f837d6491df19a8ae481cbe1dcc8d
F20110113_AAAHRL dockery_k_Page_49.tif
1431adffe2a755070b0ed236545d5407
7fea4f80521da70247cf17f627bf8e9fbfb6d542
F20110113_AAAHMN dockery_k_Page_05.tif
9bb27d749d6ad1283fe5ad8e2351c25e
b1e651fc78c137b5f907022e79893a45534975bb
24825 F20110113_AAAHWJ dockery_k_Page_34.QC.jpg
c749da49c9b5375cdc709c76e348d406
b8de7d1033475e1215b67f8b0c2c3a956451b69c
F20110113_AAAHRM dockery_k_Page_50.tif
da574b228d879ac0b0faa4d06c210432
c58aeb7f0aec002892b68b19c35c799eb48a440b
2019 F20110113_AAAHMO dockery_k_Page_29.txt
c98f16ef877b43cd07b7c4b9f55536e3
d37a80ee9bccc3fff74c9c58f72c3a0a394a709d
20947 F20110113_AAAHWK dockery_k_Page_35thm.jpg
4c622a88d27ab8a47130deb7061a3984
aad70a956373f4dcf7c81515a297bcf5882f4c49
8160 F20110113_AAAHRN dockery_k_Page_01.pro
fb0272525adf847ae2d215fb2d535262
7f1ea57083db0bec13593be72bf0b78e9d709d75
25783 F20110113_AAAHMP dockery_k_Page_18thm.jpg
1583438eaaf4a29fa16d46be520246ba
22225cf2c2d7ca023b9bc1046068080dd9f23f09
66892 F20110113_AAAHWL dockery_k_Page_35.QC.jpg
1eea52a6d6c98acde4ed538c082e6a51
4418fb0d8b3162186979c87faf5039f0a17413cf
1097 F20110113_AAAHRO dockery_k_Page_02.pro
0079adc7d1a48715cd073e3e5e2fe48d
20bc38abb9296ba486c51c565a9826f0a79591f0
18423 F20110113_AAAHMQ dockery_k_Page_44thm.jpg
481c7690100f26a3cb1f10edb3378790
48f14b638e0a25a3a41a8493a5756fde67fc4556
10112 F20110113_AAAHRP dockery_k_Page_03.pro
68a03bbc99408aaf96dfb5551afb934c
83baeb3bb2dcf1c8bf850f5fe52b414ebbdeda10
F20110113_AAAHMR dockery_k_Page_46.tif
9345de02c8989d1df5c746ed25e9b992
9ee107e7972230c2f525377304629eeeb7a003d2
24677 F20110113_AAAHWM dockery_k_Page_36thm.jpg
39b989a7c7866eca9940d9bcc505d2dc
92a49c1d348b2aa0714a4747caa340e72ea75adc
62928 F20110113_AAAHRQ dockery_k_Page_04.pro
9a8771ba5ed3192a929f25fa90393e70
e69215a69e3a0aed0327d82011e5380f286f413d
451 F20110113_AAAHMS dockery_k_Page_03.txt
41465e7ec71dabccd078e73f7ed10629
e1f136fe46fae8d345b26167baf561085fca7162
24813 F20110113_AAAHWN dockery_k_Page_37thm.jpg
9944143bdeddaa96f0bb784f371a7b7e
56562a90f12dbf1007e045034b4996c63da0a027
13665 F20110113_AAAHRR dockery_k_Page_05.pro
701f8f108f34e86f367350ef54a8408e
266b5c37cfe2ed79f2be721c26b9051dcfed5963
F20110113_AAAHMT dockery_k_Page_21.tif
33b5fbffba7c58b0f9afed9dfb882667
b73f27b50233ad007ca40ee398999613a6d7ad53
79603 F20110113_AAAHWO dockery_k_Page_37.QC.jpg
3dc6c0c6d4e4ccd2b1feae328f74ee9f
07e8dfadd9efb345fe54ad1c811bddac100960f9
F20110113_AAAHMU dockery_k_Page_40.tif
090d7ab5f75fb2f08aa1b6cae90e5f90
a36a77118a18f68b52900cdfddfdb1d10207d49b
24629 F20110113_AAAHWP dockery_k_Page_38thm.jpg
e820b5737aaf8c4a6828f865474d0cc9
3f491c8c8c83a5b23f38aed7123f76f76ff96bf8
7200 F20110113_AAAHRS dockery_k_Page_06.pro
6fe136a9e5c71b00c68a89b7c4c02bdf
13b3f02f810711edb2e50d1c9b322a63d5bb738b
F20110113_AAAHMV dockery_k_Page_46.pro
184a766f9e5bffdf0a1d4a07ea801ad6
c22afbda711ffdeeb187ea2e7696cadb39b24390
77127 F20110113_AAAHWQ dockery_k_Page_38.QC.jpg
dba1c4a494674221b1c816e5cddae635
42713f6b157ff40826c69792d3d645016cc3834c
7426 F20110113_AAAHRT dockery_k_Page_07.pro
c6f229d34125f8f787f6ef909c379beb
364252e26e0dd02622ade404e1c0727161a0c7ce
24241 F20110113_AAAHMW dockery_k_Page_41thm.jpg
7b13624d2b57f3b1e52e1a2aa5ab6ce3
9fa23a87bb5aca4915f24e0bf87460d8455d4adb
24547 F20110113_AAAHWR dockery_k_Page_39thm.jpg
f23733e47263e7fdd048cdba7ae3eeb1
b0c099ab5d6078ca08e842663f7f938f8ca951f2
37169 F20110113_AAAHRU dockery_k_Page_08.pro
0597a8de4530141cc49648c2ec037bdd
51748e74faacffcf5b7d410e5d7e2a490f54c95e
F20110113_AAAHMX dockery_k_Page_04.tif
d0524c8d46adade78996a3cc40e95510
9b07aadab50f5a8a39c0a8ea78d466ada7c946f9
24157 F20110113_AAAHWS dockery_k_Page_40thm.jpg
547d30ddee79ace360ac09e19a470642
ee7ca8e97173ebe1ec90e572874df9577b2358aa
52198 F20110113_AAAHRV dockery_k_Page_10.pro
c82fb758f3f83eba8c22a0e019395000
5bdfca83ea01726f2a32b28c4766999db7832a42
59904 F20110113_AAAHMY dockery_k_Page_28.QC.jpg
8f9a7c86edc684f6e160d354cd4cba05
57849acb33de36e44866f28921f345c70261285d
76666 F20110113_AAAHWT dockery_k_Page_41.QC.jpg
6a928260ee44ced60e7e8b0176da1b57
1776c6ea88e835e1e3bed7e62fa0488aae22f32d
50996 F20110113_AAAHRW dockery_k_Page_11.pro
4155bf7db96bf4f03870463d06cac84c
b069cb4ffb892443fdb64a3fce6f691d4db86a16
203007 F20110113_AAAHMZ dockery_k_Page_40.jpg
7a41ad6c3d2aba54d71db0640ef2604a
d7117595ee27cd427285f3ddc0799fbe859a93f0
71376 F20110113_AAAHWU dockery_k_Page_42.QC.jpg
c5a90ca9591c42c6791c2b5b0919dabf
04caab4bb3715841f163e15fa5030877aa84b0a7
51076 F20110113_AAAHRX dockery_k_Page_12.pro
a86b78c47fa690d7125ba075e6858ebb
ef1ef47c681ace41cea463fcfd3b3e5c7104b0f2
20977 F20110113_AAAHWV dockery_k_Page_43thm.jpg
6be2c733e61e53f349237120924bf25e
124a2f7bf6a6b7b1da58ff02d081853c1cb1e9e4
48011 F20110113_AAAHRY dockery_k_Page_15.pro
810d3335fcc5ce091246b1913fcb888e
355871e74c7246219e527c937b1df2dab2584c11
68618 F20110113_AAAHWW dockery_k_Page_43.QC.jpg
2e35881045b570f575aaa84dcb28a1e8
65868057fed2dcb4e0d319939f2c1fb5a9ef8b97
217931 F20110113_AAAHPA dockery_k_Page_14.jpg
7d3905bdcde484d507b8be2dca7a3ff5
4cb87dc692c8f7880769c0dbfbe7cb2f83a4339f
48735 F20110113_AAAHRZ dockery_k_Page_16.pro
4923685a7d206cf2eb00e2fdb27ae2a4
ca1a55ea86a12ffb5ed9b489cb9b3b32886fdefe
201015 F20110113_AAAHPB dockery_k_Page_15.jpg
9f4e6d83891435a7a796ea1e19f71344
5b13ba379f1fb6e189d9346b3a7044f346edd013
59908 F20110113_AAAHWX dockery_k_Page_44.QC.jpg
dfc0fe56327be4becf8836b7d0fc32c5
047afc76446e0d13c06cbe55984e16d6310240d1
205461 F20110113_AAAHPC dockery_k_Page_16.jpg
d6762c28517cf4263b61a31fbc416c4d
7dd5416f5dcfe3a45687870c7f51e1dd81c8af9d
F20110113_AAAHUA dockery_k_Page_37.txt
47400e26e380c834d149f7e671e1df70
0543dc9c83a20f7a1edd0710c5c5a7aad24e8bef
21192 F20110113_AAAHWY dockery_k_Page_45thm.jpg
a4f34f1852c3349ad898b8378aa12ecb
7fb1272936b42dac5a38393d14696b6d9bed7584
220352 F20110113_AAAHPD dockery_k_Page_17.jpg
0ab0a02df4f78b318fb761def499cfab
9288cd53d8718bea51334abf874d17f8f772ed82
1934 F20110113_AAAHUB dockery_k_Page_40.txt
9107dec498f606fe9fd70df98ef307ff
47c71897c64ac0b008c09f1558c6cf56f6ce3567
18409 F20110113_AAAHWZ dockery_k_Page_46thm.jpg
004e89626f8873cc95748972eafcaa5f
67fffc1758c32ecdd1c08191ce6c39373b150ed5
211883 F20110113_AAAHPE dockery_k_Page_18.jpg
cc502f82c88bd4dfa48d02ec55932035
01cc33685599fb375b2954252eb62e5d56f38a9e
1993 F20110113_AAAHUC dockery_k_Page_41.txt
823e9b93cbcaee2e5a1fe291825723cb
b40fe0c1a1bca1d1cb5ea7e219a8d510ed783837
190206 F20110113_AAAHPF dockery_k_Page_19.jpg
df39b87ab35d8d93b9dbe129a99f28fb
45d5932d04292d12cd673bbd8c79fdf8ca02c3b6
1856 F20110113_AAAHUD dockery_k_Page_42.txt
ae8a99c99d84f6c23c1a6a0eb072f244
30ff6d015fc1f16efce2d19c6c638a56c4806086
191996 F20110113_AAAHPG dockery_k_Page_20.jpg
caeaacdc937faeb71ce92420913520d7
4dae162617c5774e763188fd902111f6af717527
4175 F20110113_AAAHUE dockery_k_Page_43.txt
9c7773faee4c7d975eea11e64f14fb0c
7de772f5d7ba9fdd7e03abd86af1bc03abeeb078
103787 F20110113_AAAHPH dockery_k_Page_21.jpg
57709b11f2ae4b5875e9e37d7dc0503c
c5cb4f71a2763bd1d30c0301ac965de7f0851f46
1781 F20110113_AAAHUF dockery_k_Page_44.txt
bf00e61a2020d6e1257fdfb32d8198ef
dcf5c168f3801d321052479886381c008e5bae09
151936 F20110113_AAAHPI dockery_k_Page_22.jpg
54e9492b74bdb95c6e834c49f2aa14d4
bee0941362e8241329ce50022422f4ffeed5b195
4222 F20110113_AAAHUG dockery_k_Page_45.txt
c453abd4cd6619a78e9a2efe22a50478
ee0423bf05b27272bfadc6ca0abab2db8ac79546
210202 F20110113_AAAHPJ dockery_k_Page_23.jpg
28cbbedc93209946e049a6aa3ead9f97
9aacaea9dca51988e6109262993b528ac1e4ebf8
F20110113_AAAHUH dockery_k_Page_46.txt
1d226cbe3db8c9cbce2627ac4c14a1e8
4cf1e92f75fe0c91e446ff7e6da4046fe7bc7512
206720 F20110113_AAAHPK dockery_k_Page_24.jpg
385fb68cea18ed4de8a8fb357f396f59
efa45cebe3a9f42cded93e4c53f381f09ebbbea6
2109 F20110113_AAAHUI dockery_k_Page_47.txt
4ff759b8a72a48be1602f62c6fae0aa1
55bd3343442b62b9c3782382e7ec1f803f49e689
204938 F20110113_AAAHPL dockery_k_Page_27.jpg
0af6d632e381fd75ebb112080f1a71ce
ef78529f29e7ecc51d79537dc109a98591409957
2378 F20110113_AAAHUJ dockery_k_Page_49.txt
ca2c81a1378cee5a4b784322f76428ff
b44c868e0f59a83592a9781e32515a210dd84343
145925 F20110113_AAAHPM dockery_k_Page_28.jpg
fd05c61973ea1f2a859373424396c752
85f7b5b4f2cadd5c7d232625db4bf22f6d3a22c0
491 F20110113_AAAHUK dockery_k_Page_50.txt
fe78075ac5fbe1743e8d05ee0aaba453
6ad10aa6ec5e539e66d6b11bcea20c71131e1cc0
205990 F20110113_AAAHPN dockery_k_Page_29.jpg
f05e50dd931ac462545fc5b00ae3fe82
d0d0424001e0d61ab6c3f9234fe558eefdee50d9
771573 F20110113_AAAHUL dockery_k.pdf
c2b8c5dddc8c5f5bef4fcc30f3976e10
9c2bc4b46a56141872aaa4591158ea51997a5d04
206690 F20110113_AAAHPO dockery_k_Page_30.jpg
01ae5b6810418d6c6ae444fb23927299
c1001453129d6322cf20b4aec18568d3782507a7
19308 F20110113_AAAHUM dockery_k_Page_01.QC.jpg
69c04cb309cf89e4e09846992afa8592
2d28fda03b364e0072629790b4514640ddbd3787
219956 F20110113_AAAHPP dockery_k_Page_31.jpg
ce9d95a4b19669350d177527586bf3d8
2953a03665094c7b66388f61292566394fbf9c8c


Permanent Link: http://ufdc.ufl.edu/UFE0002622/00001

Material Information

Title: Do People Brace Sensibly? Risk Judgments, Outcome Importance, and Risk Prevalence
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0002622:00001

Permanent Link: http://ufdc.ufl.edu/UFE0002622/00001

Material Information

Title: Do People Brace Sensibly? Risk Judgments, Outcome Importance, and Risk Prevalence
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0002622:00001


This item has the following downloads:


Full Text












DO PEOPLE BRACE SENSIBLY?
RISK JUDGMENTS, OUTCOME IMPORTANCE, AND RISK PREVALENCE
















By

KATHARINE DOCKERY


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2003

































Copyright 2003

by

Kate Dockery















ACKNOWLEDGMENTS

I thank Dr. James Shepperd for his constant guidance and encouragement and for

never allowing me to do less than my best work. I thank my husband and my parents for

their unconditional love and support, without which this process would have been

difficult and painful. Finally, I thank my God, "who is able to do immeasurably more

than all we ask or imagine."
















TABLE OF CONTENTS
page

A C K N O W L E D G M E N T S ......... .................................................................................... iii

LIST OF TABLES ........... ................... .............. ....... ....... vi

L IST O F F IG U R E S .... ...... ................................................ .. .. ..... .............. vii

ABSTRACT .............. .......................................... viii

CHAPTER

1 IN TR OD U CTION ............................................... .. ......................... ..

Judging Likelihood .................. .......... .......... .................. ....... .. ...... .. 3
Avoiding Disappointment as a Source of Judgment Errors............... ...................6
Outcom e Likelihood and Bracing........................................... .......................... 9
Overview and H ypotheses .......................................................... ............... 12

2 M ETHOD S ..................................... ................................. ........... 14

O v e rv iew .........................................................................................14
P a rtic ip a n ts ........................................................................................................... 1 4
P ro c e d u re ....................................................................................................... 1 4

3 R E S U L T S ........................................................................................................1 7

Risk Estimates ........................................... .......... 17
Are the Results Due to Outliers? ..............................................21
M edition A naly ses ..............................................................23

4 D ISC U S SIO N ............................................................................... 27

Possible Explanations ................................. ........................... .... ...... 28
The R ole of D isappointm ent .......................................................................... ........ .... 30
Limitations and Implications ................................. ................................. 33
C conclusion ...................................................................................................... ....... 34







iv










APPENDIX

A Q U E ST IO N N A IR E A ........................................................................ ...................35

B Q U E ST IO N N A IR E B ........................................................................ ...................37

L IST O F R E FE R E N C E S ............................................................................. .............. 39

B IO G R A PH IC A L SK E TCH ...................................................................... ..................42

















































v
















LIST OF TABLES

Table p

1 Frequency analyses of optimistic, realistic, and pessimistic risk estimates .............25

2 Frequency analyses of risk estim ates ............................................ ............... 26
















LIST OF FIGURES

Figure page

1 Risk estimates as a function of risk condition and need. ........................................ 18

2 Mean risk estimates as a function of risk condition and need...............................20















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

DO PEOPLE BRACE SENSIBLY?
RISK JUDGMENTS, OUTCOME IMPORTANCE, AND RISK PREVALENCE

By

Katharine Dockery

December 2003

Chair: James A. Shepperd, PhD
Major Department: Psychology

Previous research shows that people become pessimistic about impending bad news

to "brace for the worst." The current study examined whether the commonality and

importance of an event moderates bracing. Students learned about a billing error that

would result in an unexpected bill for either 20% (rare event) or 80% (common event) of

the students at their university. Students in the common event condition made higher

personal risk estimates than did students in the rare event condition. Financially needy

students also made higher risk estimates than did non-needy students. Comparing risk

estimates to the base rates provided to participants revealed that students in the rare event

condition were pessimistic about their risk of receiving a bill, with the financially needy

students making the most pessimistic estimates. In contrast, students in the common

event condition were optimistic about their risk, with non-needy students making the

most optimistic estimates. The discussion explores several possible explanations for

these findings.














CHAPTER 1
INTRODUCTION

During the fall of 2002, a series of sniper attacks terrorized life in the greater

Washington, DC, region. People living in the area felt unable to leave their houses or

send their children to school for fear that they might be the next victims. By the time the

assailants were apprehended on October 25, 2002, they had left in their wake ten fatalities

and three injuries-out of approximately 4.4 million people living in the counties where

the attacks occurred. The chances of any given person being attacked by the snipers were

less than 0.00003%, yet activity in Virginia and Maryland was paralyzed until the snipers

were caught. Although the objective likelihood that any given person would fall prey to

the snipers' bullet was quite low, the behavior of people living in the area (e.g.,

withholding their children from school, postponing trips to gas stations) suggests that

people felt otherwise, perceiving their risk as quite high. What made people so

drastically overestimate their risk?

Numerous studies find that people do not fully account for base rate information

(Tversky & Kahneman, 1982), give undue weight to salient examples (MacLeod &

Campbell, 1992; Slovic, Fischoff, & Lichtenstein, 1982), have a poor understanding of

small numbers (Tversky & Kahneman, 1974), and rely on stereotypes (Kahneman &

Tversky, 1972) when estimating their risk. Undoubtedly, these errors in thinking

contributed to an overestimation of personal risk of being attacked. In this thesis I

propose an additional reason people might overestimate their risk. Specifically, I propose

that in some instances people may overestimate their risk to avoid negative feelings such









as disappointment associated with having bad events come as a surprise. That is, people

may overestimate their risk by making unduly pessimistic predictions in an attempt to

"brace for the worst" (Carroll, Dockery, & Shepperd, 2003). The pessimism of residents

of the Washington DC area about the risk of a sniper attack may have been motivated in

part by an attempt to brace for the worst possible scenario. In fact, people who are

bracing themselves may consistently overestimate the risk of rare events in their effort to

prepare themselves for the blow of a negative outcome.

This research examines whether the commonality of an event moderates bracing. I

specifically address whether people brace more for rare events or for common events.

On the one hand, people may brace more in anticipation of negative events that are

common because of the greater objective likelihood of experiencing common event,

which prompts more worry and fear of disappointment. For example, during an epidemic

when large numbers of people are testing positive for a disease, people objectively face a

greater risk of catching the disease and thus should be more preoccupied with the disease.

In contrast, people should be less concerned about a rare disease for which their objective

risk is lower. Thus, one might expect that people are more pessimistic, relative to the

base rate, about common negative events. On the other hand, it is also possible that

people brace more in anticipation of negative events that are rare. Research shows that

the intensity of disappointment depends on the unexpectedness of a negative event. The

more unexpected the event, the greater the disappointment people experience (van Dijk,

Zeelenberg, & van der Pligt, 1999). Because bracing serves to reduce or avoid negative

feelings such as disappointment, people may be more inclined to brace for rare events,

which have greater capacity for producing disappointment, than for common events.









Indeed, people may not only brace more for rare events, they may also prepare more in

other ways, a finding that would have important implications regarding risk-taking

behaviors. For example, travelers who mistakenly believe that they are likely to crash in

a plane may take to the far more dangerous highways. The current study tests the

competing possibilities regarding the influence of commonality on bracing.

Judging Likelihood

Although people often attempt to predict future outcomes (e.g., the likelihood that

rain will spoil a picnic, that stock prices will rise, that traffic will be lighter on one route

or another), try as they might they often predict inaccurately (Tversky & Kahneman,

1974). In fact, people often fail to adequately consider information about the actual

frequency of an event, or its base rate. The base rate for an event, or more accurately,

people's perception of the base rate, can influence estimates of an event's likelihood.

However, people are often unaware of the base rate for an event, and even available base

rate information does not guarantee accurate estimates. People often neglect such

information, attending instead to less informative details of a situation (Tversky &

Kahneman, 1982).

This phenomenon is related to the process of anchoring and adjustment, another

example of base rate use in probability judgments. When people employ the anchoring

and adjustment heuristic, they make estimations by accounting for present information

and then adjusting for differentiating characteristics (Tversky & Kahneman, 1974).

Although anchoring and adjustment can lead to better predictions when people

objectively consider relevant information, people often do a poor job of using such

information or rely on biased perceptions of their capacity to avoid negative outcomes.

For instance, participants in one study read scenarios in which the base rates for several









events (car accident, burglary, pet illness, parachute failure, and raffle win) were either

high or low and the target was the participant or another person. For every scenario,

participants estimated a greater likelihood when the base rate was high than when the

base rate was low. However, the estimations for the negative events were consistently

lower when the participant was the target of the estimate than when another person was

the target (Greening & Chandler, 1997). On the whole, people inadequately consider

base rate information when estimating personal likelihood.

Researchers have identified several cognitive biases that can prompt errors in

judgments about base rates. First, people sometimes misjudge probability by overusing

the availability heuristic, a shortcut in decision making in which people base judgments

of likelihood on how easily they can generate examples of the event. For instance,

people often overestimate the likelihood of a fatal plane crash or homicide because they

can easily bring to mind poignant instances of these events from the media (Slovic,

Fischoff, & Lichtenstein, 1982). The availability heuristic results in overestimation of

probabilities for very salient events (often due to recency, personal experience, or media

coverage) and underestimation for less salient events, regardless of the actual rate of

occurrence. One study demonstrated use of the availability heuristic by objectively

measuring "availability" and relating it to perceived probability. The results showed that

quicker speed of retrieval of an occurrence of the event corresponded to higher

probability judgments for the event (MacLeod & Campbell, 1992).

Second, people sometimes misjudge probability because they overuse the

representativeness heuristic, a shortcut in decision making that involves prototype

matching. Specifically, people have prototypes for entities such as events, things, and









people, and base judgments of likelihood that a particular example falls within a category

on how closely the example matches the prototype. In other words, people perceive their

risk for an event as higher when they closely match their stereotypes of someone who

would experience that event. For example, people's judgments of their likelihood of

being mugged are influenced by how similar they think they are to their stereotype or

prototype of the typical mugging victim. The more similar people perceive they are to

the stereotype, the more likely they think they will be mugged.

People also hold stereotypes about the likelihood of certain kinds of events. For

example, if people believe a negative event is serious, they will also perceive it as rare.

The converse is also true if people believe an event is rare, they will perceive it as more

serious (Jemmott, Ditto, & Croyle, 1986). Although these beliefs may often be true (e.g.,

diseases that cause death are usually relatively rare), it is not always the case. For

example, the human papilloma virus (HPV) can have serious consequences including

infertility and cervical cancer, yet over 40% of college-age women are infected with the

virus (Ho, 1998). Clearly, assumptions about the correlation between severity and risk

can be misleading and even dangerous.

Third, people sometimes err in judgments of likelihood based on biased

perceptions of control, perceiving that desirable items are more likely to occur when they

are controllable than when they are uncontrollable. More specifically, people

underestimate their risk for certain events when they believe that they can control the

outcome. In addition, people tend to overestimate their personal control (McKenna,

1993; Klein & Kunda, 1993), leading them to be overly optimistic about their ability to

avoid certain negative events. In a study by Greening and Chandler (1997) described









earlier, people perceived their likelihood of experiencing a variety of negative events as

lower than another target's likelihood. For example, most people believe they are better

than average drivers, and estimate their chances of getting in a car accident as less than

average (McKenna, 1993). However, by definition most people are average drivers, and

such flawed logic is likely a result of an overestimation of personal control. Thus,

probability judgments based on perceptions of personal skill or control are likely to be

underestimations. In addition, people may believe that they can control not only their

actions, but also their outcomes. For example, smokers may believe they are less likely

to get smoking related illnesses than other smokers because, unlike other smokers, they

will not be smokers in the future (McKenna, Warburton, & Winwood, 1993; Lee, 1989).

In summary, a variety of factors can cause people to make errors in estimates of

an event's base rate, or probability. First, people overuse the availability heuristic, over

relying on salient examples to give them clues as to the likelihood of an event. Second,

people overuse the representativeness heuristic, judging likelihood based on how closely

an event or person matches their prototype for the event or person. Third, people

underestimate their risk for negative events and overestimate the likelihood of desirable

events when they overestimate how much control they have over their actions or the

outcomes.

Avoiding Disappointment as a Source of Judgment Errors

The prior sources of error in likelihood judgment represent cognitive errors that

arise from discounting or ignoring information that is hard to comprehend (underusing

the base rate), or from overusing common shortcuts in making judgments (the availability

and representativeness heuristics), or from attempts to calibrate judgments based on

personal information (misperceptions of control). I propose another source of error that









is more motivational in flavor arising from a desire to reduce or avoid negative emotions

such as disappointment.

Consistent with previous researchers, I view disappointment as the experience of

outcomes falling short of expectations ( Van Dijk, Zeelenberg, & van der Pligt, 1999; van

Dijk & van der Pligt, 1997; Zeelenberg et al., 2000). People feel disappointed about

shattered hopes or expectations, not simply about negative outcomes. In this way,

disappointment can be distinguished from the similar emotions of sadness, anger,

frustration, and regret (Zeelenberg et al., 2000). Disappointment is specifically

associated with absence of a positive, hoped for outcome, whereas sadness, frustration,

and anger are generally associated with the presence of a negative outcome (van Dijk,

Zeelenberg, & van der Pligt, 1999). Disappointment can be further distinguished from its

close relative, regret, in that people feel regret over actions and disappointment over

outcomes, regardless of the precipitating actions (Zeelenberg, van Dijk, & Manstead,

1998).

The intensity of disappointment depends primarily how people's outcomes

compare with their expectations. The more expectations exceed outcomes, the more

intense the disappointment (van Dijk & van der Pligt, 1997). If people set their sights

very high in anticipation of feedback, they are likely to be disappointed by almost any

outcome. For example, a student who expects a perfect score on an exam will almost

certainly be disappointed. On the other hand, very poor outcomes also make

disappointment likely. A student receiving a failing grade is also likely to experience

disappointment. Decision affect theory (DAT; Mellers et al., 1997) discusses the

emotional consequences of the relationship between expectations and outcomes, noting









that how people feel about an outcome depends on the value of the unobtained outcome

(was it better or worse than the outcome obtained?) and the perceived likelihood of

receiving the outcome. For example, in a study presenting participants with gambling

tasks with varying possible outcomes, subsequent affect depended not only on absolute

loss or gain, but also on the possible alternative outcome and the probability of winning

(Mellers et al., 1997, Study 1).

Given that the experience of disappointment depends on the relationship between

outcomes and expectations, how can people avoid feeling disappointment? Even at the

moment of truth, when all control over outcomes is gone, people can still moderate their

disappointment over a negative outcome by lowering their expectations (Carroll,

Dockery, & Shepperd, 2003; Shepperd, Oullette, & Fernandez, 1996, experiments 2 & 3;

see also Gilovich, Kerr, & Medveck, 1993; Sackett, 2002; Sanna, 1999). This process of

bracing for the worst results in pessimistic likelihood estimates at the point of feedback.

For example, students in one study estimated their scores on an exam at four points in

time: one month before the exam (Time 1), immediately after the exam (Time 2), one

hour before receiving their exam grades (Time 3), and immediately before receiving the

grades (Time 4). Students shifted, first from optimism before the exam to realism

immediately after the exam was completed. The realism persisted at Time 3, when the

students did not expect feedback for another hour. However, when the graded exams

were being distributed, the students became pessimistic in their predictions (Shepperd et

al., 1996).

Furthermore, not all outcomes are of equal importance. For example, a test for

HIV could have far greater consequences than a test for Strep throat. Do people brace









equally for all events, regardless of importance? Research shows that people, in fact,

become more pessimistic about an outcome with severe consequence than they do for one

with only mild consequences (Taylor & Shepperd, 1998). Importance can also be in the

eye of the beholder. What seems life-threatening to a high school student, for instance, is

usually insignificant to a mature adult. If people see an outcome as unimportant or

inconsequential for any reason, they will likely maintain realism or even optimism.

In summary, people experience disappointment when outcomes fall short of

expectations. People can avoid disappointment by either changing their outcomes or

change their expectations. At the moment of truth, however, people can no longer

change the outcome. They can, however, change their expectations, avoiding

disappointment by making a pessimistic prediction. However, people will only embrace

pessimism to avoid disappointment over outcomes they view as important.

Outcome Likelihood and Bracing

Do people brace more for common events or for rare events? On first blush it

might appear that reliance on the availability heuristic would lead people to brace more

for common events. As discussed earlier, judgments of frequency are based on how

easily examples come to mind. The pervasiveness of common events is likely to

facilitate access to examples of common events, prompting greater estimates of frequency

reliance on the availability heuristic leads to concern over salient events. For example,

people who lived in the days of the influenza virus were certainly aware of the high

numbers of fatalities and would likely have rated their chances of contracting the illness

as high. People would have heard stories the influenza victims, accounts of how the virus

devastated families, and what could happen to them if they did not take the appropriate

precautions. However, although this argument is intuitively appealing, in reality the









commonality and availability of an event are distinct. As stated earlier, people frequently

underestimate the likelihood of very common events. Thus, although some common

events are also highly available in memory and the consequences easily imagined,

equally often availability and commonality are unrelated.

Availability aside, people may brace more for common events because they have

difficulty understanding probabilities. Instead of thinking of likelihood in terms of

percentages or probabilities, people may think about likelihood in binary terms -- an

event will either happen or it won't. Accordingly, for low probability events, people may

conclude that because the event is rare, it will not happen. For high probability events,

however, people may conclude that the event is inevitable even though there is some

chance that it will not happen. Binary thinking would thus lead people to see common

events as inevitable and rare events as entirely avoidable.

While some evidence suggests that people will brace more for common events

than for rare events, other evidence suggests that people will brace more for rare events

than for common events. First, a great deal of research shows that people underestimate

the likelihood of common events and overestimate the likelihood of rare events (Fischoff,

1981; Johnson & Tversky, 1983; Lichtenstein, Slovic, Fischoff, Layman, & Combs,

1978; Pulford & Colman, 1996; Slovic, 1987; Weinstein, & Lyon, 1999; Rothman, Klein,

& Weinstein, 1996; Brandstatter, Ktuhberger, & Schneider, 2002). For example, people

underestimated their risk for common diseases, such as the human papilloma virus and

chlamydia, and overestimated their risk for rare diseases such as chronic liver disease and

cirrhosis (Rothman et al., 1996, see also Lichtenstein et al., 1978). The availability

heuristic may account in part for these errors in estimation in that media attention to rare









diseases may prompt people to overestimate the base rate and, as a consequence,

overestimate their risk (Slovic et al., 1982). Recent events make this point clear.

Following the events of September 11, 2001, people became afraid to travel by air.

Although, the risk of dying in a plane crash is extremely low, the vivid plane crashes

appeared to have created the perception that travel by plane is highly dangerous and that

fatal plane crashes are widespread.

Second, the disappointment literature reveals that people feel negative outcomes

are particularly aversive when unexpected (Mellers et al., 1997; Shepperd & McNulty,

2002). The intensity of disappointment depends not only on the negativity of the

outcome, but also on the expectations regarding the outcome (van Dijk & van der Pligt,

1997), and people would naturally have positive expectations when a negative event is

unlikely to occur. In other words, if a negative event has a known base rate of 1%,

people will probably be fairly certain that the event will not happen to them. On the other

hand, if the negative outcome were to occur, it would be extremely disappointing because

it is so rare and unexpected.

In sum, the question of how the commonality of an event moderates bracing has

two possible answers. The first, and most intuitive prediction is that people brace more

for common negative events. This possibility is supported by the possibility that people

think in binary terms rather than considering probabilities and statistics. The second,

opposing prediction is less intuitive and proposes that people brace more for rare negative

events, and is supported by research on disappointment and probability judgments of

common and rare events.









Overview and Hypotheses

The present study examines whether people brace more for common vs.

uncommon negative outcomes by exploring responses to news of a possible financial

setback. Specifically, participants believed that either 20% or 80% of students at their

university would be affected by an error in the registrar's office that would result in those

students receiving an unexpected bill.

Hypothesis 1: I hypothesized that participants in the common risk (80%) condition

would estimate their likelihood of receiving a bill as greater than would participants in

the rare risk condition (20%). Consequently, I expected to find a main effect of risk

condition on risk estimates, with the mean of the common risk condition greater than the

mean of the rare risk condition.

Hypothesis 2: In line with the finding that people will be pessimistic about

outcomes that are important to them, I hypothesized that, when averaging across the

common and rare events, participants who were financially needy would estimate their

likelihood of receiving a bill as greater than would participants who were not needy.

Consequently, I expected to find a main effect of need on risk estimates, with the mean of

the high need participants greater than the mean of the low need participants.

Hypothesis 3: Among participants receiving news of a rare billing error, I predicted

that high need participants would be more likely than low need participants to estimate

that they would receive a bill. Among participants receiving news of a common billing

error, I predicted that high need and low need participants would not differ in their risk

judgments. These sets of means would manifest themselves as an interaction of need and

risk level.









Hypothesis 4: Based on the literature on disappointment, I hypothesized that high

need participants would be pessimistic in the rare risk condition but would be realistic in

the common risk condition. Consequently, I predicted that a dependent t-test comparing

the predictions of low need participants would reveal that predictions in the rare risk

condition would differ significantly from the 20% baseline, but that predictions in the

common risk condition would not differ significantly from the 80% baseline.

Hypothesis 5: I hypothesized that low need participants would be realistic in both

the common risk and rare risk conditions. Consequently, I predicted that dependent t-

tests comparing the predictions of low need participants in the rare and common risk

conditions with their respective baselines (either 20% or 80%) would reveal no

differences from the baselines.














CHAPTER 2
METHODS

Overview

The methods were adapted from those used Shepperd et al. (2000). This

experiment examined students' reactions to the prospect of a financial loss after learning

that a high or low percentage of students would be experience the loss. The greater

disappointment experienced when a loss is unexpected led me to predict that people

would be more pessimistic about the prospect of a loss, relative to the baseline, when

they were told that few students would be affected than when they knew most students

would be affected. Moreover, I expected that the effect would be due to greater

pessimism on the part of financially needy students.

Participants

Introductory psychology students (N= 234) participated voluntarily as part of

three class sections and were randomly assigned to either the high likelihood (80%) of

receiving a bill condition or the low likelihood (20%) of receiving a bill condition.

Procedure

Participants in three classes received a description of a recently discovered billing

error that would affect either 20% or 80% of the student body (see appendix A).

Participants in each situation learned that students affected by the error would receive a

$178 bill in three to four weeks and that failure to pay the bill would result in their

records being flagged.









Participants then completed a questionnaire regarding their reactions to the news

of the billing error. First, immediately after learning of the billing error, participants

responded to five adjectives assessing state anxiety (calm, nervous, anxious, relaxed,

worried). Participants responded to each item with how they felt "right now, at this

moment," using a four-step scale (1 = not at all; 4 = very much so). These items were

summed (after reverse coding) and divided by five to produce a measure of anxiety,

range = 1 to 4, M= 2.06, SD = 0.76, Cronbach's a = .85.

Second, participants completed the primary dependent measure, estimating their

likelihood of receiving the bill. This item asked participants to use a 0 to 100% scale to

estimate the probability that they would receive a bill.

Third, previous findings reveal that only financially needy students braced at the

prospect of a financial blow (Shepperd et al., 2000). As a consequence, in the present

study I assessed financial need by having participants complete the same six items

assessing financial need used by Shepperd et al. (2000). Specifically, participants

indicated (a) the extent to which they were on a tight financial budget (1 = not on a tight

budget; 11 = extremely tight budget), (b) how much difficulty they had making ends meet

(1 = extreme difficulty; 11 = no difficulty), (c) how much the bill would impact their lives

(1 = little impact; 11 = great impact), (d) what effect the bill would have on their finances

(1 = little impact; 11 = great impact), (e) how dependent they were on financial aid (1 =

not at all dependent; 11 = very dependent), and (f) the extent a bill would affect their

budget (1 = not at all; 11 = a great deal). Three of these items (a, b and e) assessed the

extent to which participants faced financial challenges and the remaining two items

assessed the financial consequences of receiving the bill. I combined the five items to









form a single index of need because collectively the five items provided a more complete

picture of financial neediness. These five items were summed, after a reverse coding

item b, and divided by five to form a single index with a potential range of 1 to 11, M=

5.63, SD = 2.70, Cronbach's a = .93. Of note, the first two items were assessed prior to

the description of the billing error, and the last three items were assessed after the

description of the error. It is possible that the risk manipulation influenced responses to

the last three financial need items. However, an independent t-test revealed no difference

between participants in the rare and common event conditions on their responses to these

three items, t(178) = 1.43, p > .15. Participants were also asked if they currently were

receiving a Bright Futures Scholarship (unique to public universities in Florida), if their

tuition was pre-paid, and whether they were paying in-state or out-of-state tuition.

Fourth, participants reported the extent to which they were thinking about the

financial difficulties that they would experience in the immediate future as a result of

receiving a bill (1 = not at all; 11 = a great deal) and how disappointed they would be if

they received a bill (1 = not disappointed; 11 = extremely disappointed). These two

items were included to address possible processes mediating any effect of financial need

on personal estimates. Finally, participants indicated if they estimated their personal risk

to be more or less than the 20% or 80% given in the scenario, why they gave that

estimate. They were given several options to choose from, including the option to write a

reason other than the ones provided.

Of note, one version of the questionnaire included several additional items.

However, these items did not yield significant results and are irrelevant to the purposes of

this study.














CHAPTER 3
RESULTS

From the initial pool of 234 participants I omitted from analyses data from 54

participants because they doubted the authenticity of the billing error, a fact that was

evident from their responses to the item, "I estimated that my probability of receiving a

bill was...because...". More specifically, I omitted data from 28 participants because

they indicated that they knew their bill well and would have noticed the error, 5

participants because they indicated that scholarships would cover all expenses, 8 because

they did not believe that a tuition error had occurred, and 13 due to a clerical error that

rendered their responses unusable. Of note, including the responses of these participants

did not change my basic findings.

As noted earlier, data were collected from three classes. Preliminary analyses

including class as a variable in the model revealed no main effects or interactions

involving class. Consequently, I collapsed across class in all subsequent analyses.

Risk Estimates

Did risk judgments vary as a function of need and risk level? Figure 1 displays the

probability estimates as a function of need (high or low) and risk condition (rare or

common). My first three hypotheses were 1) participants in the common risk (80%)

condition would estimate their likelihood of receiving a bill as greater than would

participants in the rare risk condition (20%), 2) participants who were financially needy

would estimate their likelihood of receiving a bill as greater than would participants who

were not needy, across risk conditions, and 3) among participants receiving news of a









rare billing error, high need participants would be more likely than low need participants

to estimate that they would receive a bill, but among participants receiving news of a

common billing error, high need and low need participants would not differ in their risk

judgments.

Consistent with Hypothesis 1, participants in the rare risk condition made lower

estimates (M = 42.78, SD = 29.60) than did participants in the common risk condition (M

= 56.91, SD = 32.44). That is, participants judged that they were at greater risk of

receiving a bill when the billing error was common than when the billing error was

relatively rare. For the purpose of illustrating the findings of my second hypothesis, I

separated participants into high and low need groups using a median split of their


Rare Common

70


60
u,

5 50
E

LU
4 40


30


20


Low Need


High Need


Financial Need


Figure 1. Risk estimates as a function of risk condition and need.









responses to my inventory of need (median = 5.4). Consistent with Hypothesis 2, high

need participants made higher estimates (M = 55.63, SD = 29.03) than did low need

participants (M = 43.52, SD = 33.36). That is, participants who were financially needy

judged they were more likely to receive a bill than did participants who were not

financially needy.

Contrary to Hypothesis 3, high need participants made higher estimates than did

low need participants regardless of whether the objective likelihood of receiving a bill

was high or low. That is, participants who were financially needy judged that they were

at greater risk of receiving a bill both when the risk was low and when it was high.

Although this finding is consistent with my prediction for the rare risk condition, in the

common risk condition I expected participants to make similar predictions regardless of

need.

We examined the first three hypotheses statistically using simultaneous multiple

regression procedures in which Need (after centering), Risk (rare or common), and the

Need by Risk interaction were entered as predictors. Analysis of the risk estimates

revealed the predicted main effects of Risk, F(1, 176) = 12.12, p < .001, eta-squared =

.06, and Need, F(1, 176) = 8.38, p < .01, eta-squared = .05, but did not reveal the

predicted need by risk interaction, F(1, 176) = .03, p = .85, eta-squared = .0002.

We tested two additional hypotheses: 4) high need participants would be

pessimistic in the rare risk condition but would be realistic in the common risk condition,

and 5) low need participants would be realistic in both the common risk and low risk

conditions. To test these hypotheses, I compared participants' estimates to the base rates

I provided (either 20% or 80%). Figure 2 presents the results of these analyses. In all









conditions of need and risk, participants made estimates that differed significantly from

the base rate. Specifically, both high need, t(50) = 7.18, p < .0001, and low need, t(40)=

3.25, p < .01, participants in the rare event condition estimated that their likelihood of

receiving a bill was greater than the 20% base line they received (M= 49.6, SD = 29.4

and M= 34.2, SD = 27.9, respectively). Likewise, both high need, t(39) = -3.90, p = <

.001, and low need, t(47) = -5.51, p < .0001, participants in the common event condition

estimated that their likelihood of receiving a bill was less than 80% (M= 63.4, SD = 27.0

and M= 51.5, SD = 35.8, respectively). In short, all participants in the rare event

condition were significantly pessimistic and all participants in the common event

condition were significantly optimistic.

D High Need

D Low Need




80% I I


-z

,, -i


63.4

49.6
51.5

34.2


Lk A
80% 20%
Baseline


Figure 2. Mean risk estimates as a function of risk condition and need.

In sum, participants who were told that their risk of receiving a bill was 80% made

higher personal risk estimates than did participants who were told that their risk was









20%. Furthermore, financially needy students in both risk conditions estimated their risk

to be higher than did non-needy participants. Finally, participants in the rare risk

condition were pessimistic and those in the common risk condition were optimistic, in

both cases regardless of need. However, needy students were more pessimistic in the rare

risk condition and less optimistic in the common risk condition than the non-needy

students.

Are the Results Due to Outliers?

Two possible explanations for the present findings deserve investigation. A first

possible explanation for my findings is that they resulted from the responses of a few

outlying participants. Because participants could respond using a scale ranging from 0 to

100, there is considerable opportunity for variability in responses and thus a possibility

that the findings were the result of outliers. To test this possibility, I conducted a Chi-

square analysis of the number of participants who were optimistic, pessimistic, and

realistic in each risk condition.

Table 1 presents the frequencies of low need and high need participants who were

optimistic, realistic, and pessimistic, relative to the baseline of 20% or 80%. I categorized

estimates below the baseline (i.e., below 20% for rare event and below 80% for common

events) as optimistic, estimates equivalent to the baseline (20% in the rare event

condition, 80% in the common event condition) as realistic, and estimates above the

baseline as pessimistic. I compared the proportion of low need and high need

participants who were optimistic, realistic, and pessimistic separately for the rare and

common event conditions using chi-square analyses. The estimate frequencies for needy

and non-needy participants differed significantly for rare events, X(2, N= 92) = 16.78, p

< .001, but did not differ significantly for common events, (2, N= 88) = 1.78, p > .20. I









next compared the proportion of participants in the common and rare event conditions

who were optimistic, realistic, and pessimistic separately among needy and non-needy

participants. The estimate frequencies for rare and common events differed significantly

among both needy participants, X2(2, N= 91) = 62.37, p < .0001, and non-needy

participants, X2(1, N = 89) = 49.83, p < .0001. In sum, although the pattern of findings did

not differ significantly in one case, these findings suggest that the results of need and risk

were not due to a few outlying responses.

Second, I found that people in the rare event condition were generally pessimistic,

estimating a risk greater than 20%, and that participants in the common event condition

were generally optimistic, estimating a risk less than 80%. It is possible that the high

level of pessimism for rare events and the high level of optimism for common events is

simply the result of participants in the rare condition having greater room to be

pessimistic than optimistic, and the participants in the common condition having greater

room to be optimistic than pessimistic. For example, in the rare event condition,

participants could be pessimistic by choosing any risk level greater than 20% -- a choice

of 80 possible values. However, they could be optimistic by choosing any risk level less

than 20% -- a choice of only 20 possible values. In short, the difference in the extent to

which participants in the rare and common events conditions displayed optimism and

pessimism may be an artifact of how much opportunity they had to display optimism and

pessimism.

To examine this possibility, I conducted Chi-square analysis on the number of

participants in the common event condition that provided responses above and below

20%, and the number of participants in the rare event condition that provided responses









above and below 80%. The Chi-square analyses examined whether 80% of estimates in

each condition fell into the eighty-percentage-point "space" and 20% in to the twenty-

percentage point "space." In other words, I compared the expected frequencies (80% or

20% of estimates falling above or below the baseline) with the actual estimate

frequencies. Estimates that equaled the base line were excluded from these analyses

because they provided no information about systematic errors in responses. If participants

were merely responding randomly or supplying responses where they had the greatest

room to respond, then in the rare risk condition, 80% of responses should exceed 20%

and 20% of responses should fall below 20%. Conversely, in the common risk condition

80% of responses should fall below 80% and 20% of responses should fall below 20%.

Table 2 displays the number of estimates above and below 20% and 80% for each

condition. The distribution of responses was consistent with the random responding

explanation in all four cases. For common events, neither the estimates of participants

high in need, X2(1, N= 32) = 2.53, p = .11, nor the estimates of participants low in need,

X(1, N= 39) = .78, p > .20, differed from the expected pattern. Likewise, for rare events,

neither the estimates of participants high in need, (1, N = 43) = .98, p > .20, nor the

estimates of participants low in need, (1 1,= 29)= 2.2, p = .14, differed significantly

from the expected pattern. In sum, the distribution of responses was consistent with

random responding in all four cases. This distribution of responses suggests that random

responding could be responsible for participants' risk estimates.

Mediation Analyses

We included several measures of participants' reactions (i.e., anxiety,

disappointment, and future thinking) to the possibility of receiving a financial blow with

an eye toward investigating possible mediators of the effect of need on personal risk









estimates. Using procedures recommended by Baron and Kenny (1986), I first examined

whether the three potential mediators predicted participants' predictions. Correlation

analyses revealed significant relationships between participants' risk estimates and

anxiety, r(180)= .32, p < .0001, disappointment, r(179)= .20, p < .01, and future

thinking, r(180) = .25,p < .001, indicating that participants who felt more anxious,

expected to experience more disappointment, and were thinking more about the future

made greater risk estimates. Furthermore, financial need correlated significantly with

anxiety, r(180)= .40, p < .0001, disappointment, r(179)= .68, p < .0001, and future

thinking, r(180) = .79, p < .0001, indicating that needy students, compared to non-needy

students were feeling more anxious, predict that they would experience greater

disappointment if they received a bill, and were thinking more about the future

consequences of receiving a bill.

We then conducted three separated regression analyses, one each for the three

possible mediators. In each case, the mediator was entered into the model first, followed

by need, risk condition and the need by risk condition interaction term. In all three cases,

when the mediator was added, need no longer predicted participants' risk estimates, all

Fs(1, 175) < 1, all ps > .35, but the mediator did, all Fs(1, 175) > 8.13, p < .01. Of note,

the mediators in no way mediated the effect of risk condition on participants' estimates.

Risk condition remained unchanged as a significant predictor of participants' risk

judgments in all three cases, all Fs(1, 175) > 11.16,p < .01. The fact that all three

variables completed mediated the effect of need on risk judgments suggests that the three

mediators are all tapping a common underlying process.









Table 1. Frequency analyses of optimistic, realistic, and pessimistic risk estimates


Optimistic Realistic Pessimistic



Condition Frequency % Frequency % Frequency %


Common Event

High Need 22 55% 8 20% 10 25%

Low Need 29 60% 8 17% 11 23%


Rare Event

High Need 6 12% 8 16% 37 73%

Low Need 9 22% 12 29% 20 49%









Table 2. Frequency analyses of risk estimates


Condition Above Baseline Below Baseline

Frequency % Frequency %


Common Event

High Need

Observed 10.0 31% 22.0 69%

Expected 6.4 20% 25.6 80%

Low Need

Observed 10.0 26% 29.0 74%

Expected 7.8 20% 31.2 80%





Rare Event

High Need

Observed 37.0 86% 6.0 14%

Expected 34.4 80% 8.6 20%

Low Need

Observed 20.0 69% 9.0 31%


23.2 80%


Expected


5.8 20%














CHAPTER 4
DISCUSSION

The goal of this study was to examine whether people brace differently when an

event is common vs. rare. I predicted that participants would only brace for rare events,

displaying pessimism about the likelihood of receiving a bill, and that only participants

for whom a bill was particularly consequential would brace. The data generally

supported the predictions, although some findings were unexpected.

Looking first at the effect of risk level (Hypothesis 1), participants who believed

that 20% of students would receive a bill made lower personal predictions than did

participants who believed that 80% of students would be affected. Furthermore, in line

with Hypothesis 2 and previous findings that people are pessimistic about outcomes that

are important to them, participants high in financial need made higher risk estimates than

did low need participants.

Comparing risk estimates to the base line of 20% or 80% yielded an unexpected

pattern of results. In the rare risk condition, I predicted that only high need participants

would be pessimistic. However, the results show participants were pessimistic,

regardless of need, about the likelihood of a rare event. Importantly, high need

participants were more pessimistic than were low need participants. Nevertheless, low

need participants were also bracing in their estimates. In the common risk condition, I

predicted that all participants would be realistic about their risk of receiving a bill.

However, the results show that participants were actually optimistic, regardless of need,









about the likelihood of a common event. Further analyses confirmed that these patterns

were not a result of outliers.

Possible Explanations

How do I explain this pattern of optimism and pessimism? One possible

explanation for the findings is that participants who were low in financial need were

simply less engaged in the procedures because the consequences were unimportant to

them, resulting in muted effects of risk level on their estimates. Although presumably

low need participants saw the event as less important than did high need participants, the

results do not point to this difference as the sole player in my findings. Only in the rare

risk condition could I describe low need participants' estimates as a "muted" version of

the effect found with high need participants. In the common risk condition, participants

low in need actually deviated more from the base line than did high need participants,

ruling out the possibility of a "muted effects" explanation.

A second possible explanation lies in the finding that people consistently

overestimate the risk of rare events and underestimate the risk of common events

(Fischoff, 1981; Johnson &Tversky, 1983; Lichtenstein, Slovic, Fischoff, Layman, &

Combs, 1978; Pulford & Colman, 1996; Slovic, 1987; Weinstein, & Lyon, 1999;

Rothman, Klein, & Weinstein, 1996; Brandstatter, Kthberger, & Schneider, 2002). This

research suggests that when people do not know the base rate for an event, they make

predictions based on available knowledge and what seems reasonable. People often have

some sense of the true base rate, but this sense is imperfect. As a consequence, they

underestimate how rare the rare events are and overestimate how common the common

events are. If they knew the true base rates, their estimates would conform more to the

base rates. Thus, perhaps students simply overestimated the likelihood of receiving a bill









when the risk was 20% and underestimated when the risk was 80% because they were

making their best estimates based on their available knowledge of the event. However,

this explanation falls apart in light of the fact that the participants in this study had

knowledge of the exact base rate. Moreover, analyses of the distribution of participants

who were optimistic, realistic, and pessimistic confirms that participants were not merely

overestimating their risk when the event was rare and underestimating their risk when the

event was common.

Third, perhaps participants were simply using the availability heuristic when

making their estimates. With the availability heuristic, judgments of likelihood are based

on how easily examples come to mind. Although available information such as past

billing problems with the registrar's office may account for the differences in estimates

between needy and non-needy participants, it does not explain the fact that both needy

and non-needy participants overestimated their risk for a rare outcome and

underestimated their risk for a common outcome. Alternatively participants may have

employed the availability heuristic in conjunction with an anchoring and adjustment

process. Specifically, participants may have used the base rates provided to them then

adjusted their estimates from the base rates based on their experience. Thus, both needy

and non-needy participants may have had some experience with billing problems and so

they adjusted their estimates somewhat from the base rate. Prior research, however, has

found that needy students experience more billing problems than do non-needy students

(Shepperd et al., 2000). The greater experience may have prompted greater upward

adjustment from 20% in the rare risk condition and less adjustment from 80% in the

common risk condition. Although this explanation seems plausible, it does not explain









why participants, particularly participants high in need, supplied estimates below 80% in

the common risk condition.

Fourth, perhaps participants had difficulty understanding the probabilities of 20%

and 80%. Instead of judging their risk in terms of percentages, perhaps participants were

thinking in binary terms, such that they interpreted a 20% risk as "I won't receive a bill"

and an 80% risk as "I will receive a bill." Again, this explanation is easily refuted. If, in

fact, participants were thinking in binary terms, they would have made estimates of 0%

and 100%. Examination of the data reveals that less than 11% of participants made an

estimate of 0%, and fewer than 6% of participants made estimates of 100%, hardly

evidence for a binary thinking explanation.

A final possible explanation for the findings is that the estimates were randomly

distributed and that the distribution of estimates reflects nothing more than people

supplying estimates where they had the most room to estimate. Thus, participants were

pessimistic in the rare event condition because they had more room to be pessimistic than

optimistic. Conversely, people were optimistic in the common event condition because

they had more room to be optimistic. As I noted earlier, the distribution of responses does

not rule out this explanation. However, it is difficult to conceive why financial need

would differentially impact random responses. Thus, although chance may play a role in

the results, it seems that chance cannot fully account for all of the findings. In addition,

the distribution of optimistic, pessimistic, and realistic responses does not support an

explanation of the findings as due to outliers.

The Role of Disappointment

I proposed at the outset of this study that differences in people's risk estimates for

rare and common events would result from differences in their expectations of









disappointment. Previous studies show that people experience disappointment to the

extent that their expectations exceed their outcomes (van Dijk & van der Pligt, 1997).

People can avoid feeling disappointed in one of two ways: a) improve outcomes, or b)

lower expectations. In this study, the outcome, a bill of $178, was out of the participants'

control. Therefore, keeping their expectations of a positive outcome low was the only

option available to mitigate possible disappointment. Thus, perhaps participants were

pessimistic about rare events to avoid being caught off guard. In contrast, perhaps

participants did not need to be pessimistic about common events, for which their

expectations were low from the start. Do my results support this proposal? Admittedly,

the findings are mixed in their support for the role of disappointment. On the one hand,

participants' reported expectations of feeling disappointment did not mediate the effect of

risk condition on personal estimates. In fact, expected disappointment seemed to be a

facet of financial need rather than a response to the risk manipulation. Thus, concerns

with disappointment did not differentiate risk estimates for rare and common events.

On the other hand, a close look at the pattern of risk estimates may bring

disappointment back into the picture. Ignoring, for a moment, participants' self-reported

expectations for disappointment, the pattern of results conceptually supports the initial

predictions. Concerning rare events, participants were pessimistic across levels of

financial need, with high need participants displaying stronger pessimism than low need

participants. Although I initially predicted realism from low need participants, their

slight pessimism is understandable when one takes into account that the bill was for

$178, a considerable sum of money for a college student. In other words, perhaps all









participants wanted to be prepared to some degree when the bill could catch them off

guard.

Concerning common events, participants were optimistic across levels of financial

need, with low need participants displaying stronger optimism than high need

participants. Although I initially predicted realism from all participants in the common

risk condition, varying degrees of optimism are not inconsistent with an explanation of

disappointment's role in risk predictions. In this explanation, I propose that people do

not want to be surprised by a negative event. When the baseline is high enough, people

no longer need to embrace a pessimistic outlook to prepare themselves for

disappointment because reality does that for them. In fact, people may even become

slightly optimistic, taking advantage of the affective benefits of a positive outlook while

still remaining relatively prepared for the worst. For example, participants who estimated

70% in the common risk condition may have concluded that this estimate sufficiently

prepared them for the possibility of receiving a bill while offering the subjective comfort

that their risk, while high, was still lower than the risk of others. In light of this

adjustment to my original model of disappointment and risk, participants' estimates in the

common risk condition fit quite well. Participants high in need sacrificed only a small

degree of "preparedness" and became only slightly optimistic, while low need

participants, for whom the consequences of the bill would be minor, adopted a relatively

highly optimistic outlook.

In sum, a number of explanations fail to account for the results of this study. First,

low need participants did not display a muted version of the high need participants'

pattern of estimates. Second, participants did not simply overestimate low probabilities









and underestimate high probabilities. Third, participants were not responding to the

salience of the bill. Fourth, participants did not interpret their risk of receiving a bill in

binary terms. Fifth, the pattern of results was not due the presence of outliers. Sixth,

although initial analyses suggests that the distribution of estimates might simply arise

from participants were responding where there was the most room to respond, this

explanation does not explain why participants high in need rated their risk as greater than

did participants low in need. Finally, I propose an explanation based on the differing role

of disappointment in predictions for rare and common events. Although this explanation

is tentative and not without inconsistencies (i.e., the self-reported expectations of

disappointment), it does capture the complete picture of my findings in a way that other

explanations fail to do.

Limitations and Implications

The present study addressed a previously unexplored question about bracing for

bad news. Although the results begin to paint a picture of the role of commonality and

importance in people's perceptions of risk, I acknowledge several limitations of this

study. As just discussed, the role of disappointment requires clarification in order to form

conclusions about why people brace differently for common and rare events. I used a

single item measure of expected disappointment, and this single item may have been

insufficient to capture the complex experience of affective prediction. In fact,

disappointment mediated the effect of financial need, suggesting that my measure of

disappointment (along with anxiety and thoughts about the future) may have tapped

something more related to importance than to an expected affective experience.

Examining disappointment more fully would be an important goal for future research in

this area.









In this study, financial need served as my measure of the importance of

consequences. Although I believe that need was a valid measure of importance, future

studies should seek to replicate my findings using an importance manipulation.

Furthermore, this study used 20% and 80% as the definitions of rare and common,

respectively. Commonality is a relative concept, and one could make the argument that

20% isn't very rare, or that 80% isn't common enough. I chose 20% and 80% for

practical reasons, and it is the job of future research to determine the generality of my

findings with varying degrees of commonality.

Conclusion

Do people brace sensibly? At first glance, my findings appear to suggest that

people do not. Why would people be pessimistic about events that are unlikely to occur

and optimistic about near certainties? Upon further examination, a possible answer

emerges: perhaps people merely need to prepare "enough" for future outcomes. To

illustrate this point, I return to the Washington sniper. Although people were at

extremely low risk of becoming the next victim, they were prepared for the worst. Given

the potential consequences of being unprepared, bracing for the worst served to keep

people mindful of their risk. In contrast, consider the life of an Israeli living in

Jerusalem. For the Israeli, danger lies in every unattended bookbag and every ride on the

bus. In order to maintain sanity and proceed with necessary daily activities, the Israeli

may actually embrace optimism in an otherwise paralyzing situation. Were the

Washington, DC, residents being realistic? No. Is the optimistic Israeli being realistic?

No. However, it seems possible that each form of unreality serves the purpose at hand.



















APPENDIX A
QUESTIONNAIRE A

By completing this survey, you are giving consent for your answer to be used by researchers at the University of Florida.
Your participation is voluntary and no compensation will be provided. You may skip any question you are uncomfortable
answering. Your answers will be anonymous. Thank you for your help.


1. To what extent are you on a tight financial
budget?

1 2 3 4 5 6 7 8 9 10 11
not on a extremely
tight budget tight budget


2. How much difficulty do you have making financial
ends meet?

1 2 3 4 5 6 7 8 9 10 11
no extreme
difficulty difficulty


In processing the summer semester tuition and fees, the Office of the Registrar discovered that 20% of students were
underbilled by $178 for a one time Information Technology fee imposed this semester. The fee was not itemized on the
bills so it is unlikely you would know if you were underbilled. This billing error did not result from any problems with
financial aid. Thus, students receiving financial aid are no more likely than students not on financial aid to be affected by
this billing error. The registrar's office intends to bill students affected by this error for the difference. The bills will be sent
in 1 to 2 weeks. Because scholarships, grants, and loans have already been disbursed, students affected by this error will
be required to pay the Registrar's office directly. Students who fail to pay the difference will have their records flagged
and they will not be permitted to graduate or register for the next semester.


Read each of the following items. Circle the number
that best indicates how you feel right now, at this
moment.

Not at Some- Moderately Very
all what so much so


1. calm


1 2 3 4


2. nervous 1 2 3 4


3. anxious


2 3 4


4. relaxed 1 2 3 4

5. worried 1 2 3 4

Please respond to the following questions:

3. Were you aware of this situation? Yes
No

4. Using a scale from 0 to 100%, what is the
probability that you will receive a $178 bill from the
Registrar's office? Please give your true "gut
feeling" in providing this estimate.
__ %

5. If you received the bill, how much would this
impact your life?

1 2 3 4 5 6 7 8 9 10 11
little great
impact impact


6. If you received the bill, how disappointed would
you be?

1 2 3 4 5 6 7 8 9 10 11
not very
disappointed disappointed

7. If you received the bill, to what extent would this
experience affect your budget?

1 2 3 4 5 6 7 8 9 10 11
little great
effect effect

8. If you received the bill, what effect would this
experience have on your finances?

1 2 3 4 5 6 7 8 9 10 11
little great
effect effect

9. To what extent were you thinking ahead about
difficulties the bill would present in the immediate
future?

1 2 3 4 5 6 7 8 9 10 11
not at very
all much

10. Do you have pre-paid tuition? Yes No


11. Your sex (circle one): Male Female

12. Are you currently a Bright Futures Scholarship
recipient or do receive some other form of
scholarship that covers your tuition expenses?
Yes No


t










13. If you estimated that your probability is less than 20% on item 5, check the reason below that
best applies to you. Choose only ONE response.

I estimated that my probability of receiving a bill is LESS than 20% because:

Sa. I know pretty well what my tuition and fee charges should be and I would have
noticed an error before.

Sb. I feel pretty lucky and believe that this sort of error won't happen to me.

Sc. When these sorts of errors have occurred in the past, it usually happened to other
people but not me.

d. I think it's wise to be optimistic and not to imagine the worst. After all, if you think
bad things will happen to you, then they often do happen.

e. Other (please specify):







14. If you estimated that your probability is greater than 20% on item 5, check the reason below
that best applies to you. Choose only ONE response.

I estimated that my probability of receiving a bill is GREATER than 20% because:

Sa. I know pretty well what my tuition and fee charges should be and I had already
suspected or detected the error myself.

b. I am bracing for the worst. Bad news feels worse when it is unexpected. I'm
expecting a bill so I'll be ready for it.

Sc. I always seem to get hit by unexpected expenses or bills. I'm sure this is just
another instance.

Sd. The university has made mistakes on my bills in the past and they have probably
made a mistake in my case again.

e. Other (please specify):


15. If you estimated that your probability is equal to 20% on item 5, then place a check here:



















APPENDIX B
QUESTIONNAIRE B

By completing this survey, you are giving consent for your answer to be used by researchers at the University of Florida.
Your participation is voluntary and no compensation will be provided. You may skip any question you are uncomfortable
answering. Your answers will be anonymous. Thank you for your help.
2. How much difficulty do you have making financial
1. To what extent are you on a tight financial budget? ends meet?


1 2 3 4 5 6 7 8 9 10 11
not on a extremely
tight budget tight budget


1 2 3 4 5 6 7 8 9 10 11
no extreme
difficulty difficulty


In processing the summer semester tuition and fees, the Office of the Registrar discovered that 80% of students were
underbilled by $178 for a one time Information Technology fee imposed this semester. The fee was not itemized on the
bills so it is unlikely you would know if you were underbilled. This billing error did not result from any problems with
financial aid. Thus, students receiving financial aid are no more likely than students not on financial aid to be affected by
this billing error. The registrar's office intends to bill students affected by this error for the difference. The bills will be sent
in 1 to 2 weeks. Because scholarships, grants, and loans have already been disbursed, students affected by this error will
be required to pay the Registrar's office directly. Students who fail to pay the difference will have their records flagged
and they will not be permitted to graduate or register for the next semester.


Read each of the following items. Circle the number
that best indicates how you feel right now, at this
moment.

Not at Some- Moderately Very
all what so much so


1. calm


1 2 3 4


2. nervous 1 2 3 4


3. anxious


2 3 4


4. relaxed 1 2 3 4

5. worried 1 2 3 4

Please respond to the following questions:

3. Were you aware of this situation? Yes
No

4. Using a scale from 0 to 100%, what is the
probability that you will receive a $178 bill from the
Registrar's office? Please give your true "gut
feeling" in providing this estimate.
__ %

5. If you received the bill, how much would this
impact your life?

1 2 3 4 5 6 7 8 9 10 11
little great
impact impact


6. If you received the bill, how disappointed would
you be?

1 2 3 4 5 6 7 8 9 10 11
not very
disappointed disappointed

7. If you received the bill, to what extent would this
experience affect your budget?

1 2 3 4 5 6 7 8 9 10 11
little great
effect effect

8. If you received the bill, what effect would this
experience have on your finances?

1 2 3 4 5 6 7 8 9 10 11
little great
effect effect

9. To what extent were you thinking ahead about
difficulties the bill would present in the immediate
future?

1 2 3 4 5 6 7 8 9 10 11
not at very
all much

10. Do you have pre-paid tuition? Yes No


11. Your sex (circle one): Male Female

12. Are you currently a Bright Futures Scholarship
recipient or do receive some other form of
scholarship that covers your tuition expenses?
Yes No


t










13. If you estimated that your probability is less than 80% on item 5, check the reason below that
best applies to you. Choose only ONE response.

I estimated that my probability of receiving a bill is LESS than 80% because:

Sa. I know pretty well what my tuition and fee charges should be and I would have
noticed an error before.

Sb. I feel pretty lucky and believe that this sort of error won't happen to me.

Sc. When these sorts of errors have occurred in the past, it usually happened to other
people but not me.

d. I think it's wise to be optimistic and not to imagine the worst. After all, if you think
bad things will happen to you, then they often do happen.

e. Other (please specify):







14. If you estimated that your probability is greater than 80% on item 5, check the reason below
that best applies to you. Choose only ONE response.

I estimated that my probability of receiving a bill is GREATER than 80% because:

Sa. I know pretty well what my tuition and fee charges should be and I had already
suspected or detected the error myself.

b. I am bracing for the worst. Bad news feels worse when it is unexpected. I'm
expecting a bill so I'll be ready for it.

Sc. I always seem to get hit by unexpected expenses or bills. I'm sure this is just
another instance.

Sd. The university has made mistakes on my bills in the past and they have probably
made a mistake in my case again.

e. Other (please specify):


15. If you estimated that your probability is equal to 20% on item 5, then place a check here:















LIST OF REFERENCES


Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in
social psychological research: Conceptual, strategic, and statistical considerations.
Journal ofPersonality and Social Psychology, 51, 1173-1182.

Brandstatter, E., Kuhberger, A., & Schneider, F. (2002). A cognitive-emotional account
of the shape of the probability weighting function. Journal of Behavioral Decision
lMaking. 15, 79-100.

Bruine de Bruin, W., Fischhoff, B., Millstein, S. G., & Halpern-Flesher, B. L. (2000).
Verbal and numerical expressions of probability: "It's a fifty-fifty chance."
Organizational Behavior & Human Decision Processes, 81, 115-131.

Carroll, P., Dockery, K., & Shepperd, J. A. (2003). Exchanging optimism for pessimism
in personal predictions. Unpublished manuscript.

Fischoff, B. (1981). Acceptable risk. New York: Cambridge University Press.

Gilovich, T., Kerr, M., & Medvec, V. H. (1993). Effect of temporal perspective on
subjective confidence. Journal ofPersonality and Social Psychology, 64, 552-560.

Greening, L., & Chandler, C. C. (1997). Why it can't happen to me: The base rate
matters, but overestimating skill leads to underestimating risk. Journal ofApplied
Social Psychology, 27, 760-780.

Ho, G. Y., Bierman, R., Beardsley, L., Chang, C. J., & Burk, R. D. (1998). Natural
history of cervicovaginal papillomavirus infection in young women. New England
Journal of Medicine, 338(7), 423-428.

Jemmot, J. B., Ditto, P. H., & Croyle, R. T. (1986). Judging health status: Effects of
perceived prevalence and personal relevance. Journal ofPersonality & Social
Psychology, 50, 899-905.

Kahneman, D., & Tversky, A. (1972). Subjective probability: A judgment of
representativeness. Cognitive Psychology, 3, 430-454.

Klein, W. M., & Kunda, Z. (1994). Maintaining self-serving social comparisons: Biased
reconstruction of one's past behaviors. Personality and Social Psychology
Bulletin, 19, 732-739.






40


Lee, C. (1989). Perceptions of immunity to disease in adult smokers. Journal of
Behavioral Medicine, 12, 267-277.

Lichtenstein, S., Slovic, P., Fischoff, B., Layman, M., & Combs, B. (1978). Judged
frequency of lethal events. Journal ofExperimental Psychology: Human Learning
andMemory, 4, 551-578.

MacLeod, C., & Campbell, L. (1992). Memory accessibility and probability judgments:
An experimental evaluation of the availability heuristic. Journal ofPersonality
and Social Psychology, 63, 890-902.

McKenna, F. P. (1993). It won't happen to me: Unrealistic optimism or illusion of
control. British Journal ofPsychology, 84, 39-50.

McKenna, F. P., Warburton, D. M., & Winwood, M. (1993). Exploring the limits of
optimism: The case of smokers' decision making. British Journal ofPsychology,
84, 389-394.

Mellers, B. A. (2000). Choice and the relative pleasure of consequences. Psychological
Bulletin, 126, 910-924.

Mellers, B. A., Schwartz, A., Ho., K., & Ritov, I. (1997). Decision affect theory:
Emotional reactions to the outcomes of risky options. Psychological Science, 8,
423-429.

Nisan, M. (1972). Dimension of time in relation to choice behavior and achievement
Orientation. Journal of Personality and Social Psychology, 21, 175-182.

Nisan, M. (1973). Perception of time in lower-class black students. International
Journal of Psychology, 8, 109-116.

Pulford, B. D., & Colman, A. M. (1996). Overconfidence, base rates and outcome
positivity/negativity of predicted events. British Journal of Psychology, 87, 431-
445.

Rothman, A. J., Klein, W., & Weinstein, N. D. (1996). Absolute and relative biases in
estimations of personal risk. Journal ofApplied Social Psychology, 26, 1213-1236.

Sackett, A. M. (2002). Optimism and accuracy in performance predictions: An
experimental test of the self-protection liypoh,,,wi Unpublished Masters Thesis,
Yale University, New Haven, CT.

Sanna, L. J. (1999). Mental simulations, affect, and subjective confidence: Timing is
everything. Psychological Science, 10, 339-345.

Shepperd, J. A., Findley-Klein, C., Kwavnick, K. D., Walker, D., & Perez, S. (2000).
Bracing for loss. Journal of Personality and Social Psychology, 78, 620-634.









Shepperd, J. A., & McNulty, J. K. (2002). The affective consequences of expected and
unexpected outcomes. Psychological Science, 13, 85-88.

Shepperd, J. A., Ouellette, J. A., & Fernandez, J. K. (1996). Abandoning unrealistic
optimism: Performance estimates and the temporal proximity of self-relevant
feedback. Journal ofPersonality and Social Psychology, 70, 844-855.

Slovic, P. (1987). Perception of risk. Science, 236, 280-285

Slovic, P., Fischoff, B., & Lichtenstein, S. (1982). Facts versus fears: Understanding
perceived risk. In D. Kahneman, P., Slovic, & A. Tversky (Eds.), Judgment under
uncertainty: Heuristics and biases (pp.463-489). New York: Cambridge
University Press.

Taylor, K. M., & Shepperd, J. A. (1998). Bracing for the worst: Severity, testing and
feedback as moderators of the optimistic bias. Personality and Social Psychology
Bulletin, 24, 915-926.

Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty. Science, 185, 1124-
1131.

Tversky, A., & Kahneman, D. (1982). Evidential impact of base rates. In D. Kahneman,
P. Slovic, & A. Tversky (Eds.), Judgment under uncertainty: Heuristics and biases
(pp.153-160). New York: Cambridge University Press.

Van Dijk, W. W., & van der Pligt, J. (1997). The impact of probability and magnitude
of outcome on disappointment and elation. Organizational Behavior and Human
Decision Processes, 69, 277-284.

Van Dijk, W. W., Zeelenberg, M., & van der Pligt, J. (1999). Not having what you want
versus having what you do not want: The impact of type of negative outcome on
the experience of disappointment and related emotions. Cognition and Emotion,
13, 129-148.

Weinstein, N. D., & Lyon, J. E. (1999). Mindset, optimistic bias about personal risk and
health-protective behavior. British Journal of Health Psychology, 4, 289-300.

Zeelenberg, M., van Dijk, W. W., & Manstead, A. S. R. (1998). Reconsidering the
relation between regret and responsibility. Organizational Behavior and Human
Decision Processes, 74, 254-272.

Zeelenberg, M., van Dijk, W. W., Manstead, A. S. R., & van der Pligt, J. (2000). On bad
decisions and disconfirmed expectancies: The psychology of regret and
disappointment. Cognition and Emotion, 14, 521-541.















BIOGRAPHICAL SKETCH


Katharine Dockery, born Katharine Sweeny, was born in 1980 in Burlington,

Vermont. In 2002 she received a Bachelor of Science degree in psychology at Furman

University, graduating summa cum laude. She then began her graduate education in

social psychology at the University of Florida, where she will receive her Master of

Science degree in December of 2003 and intends to receive her Doctorate of Philosophy

in 2006.