UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Cultural variation in unrealistic optimism Heine, Steven J. 1993

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1993_fall_heine_steven.pdf [ 2.94MB ]
JSON: 831-1.0086393.json
JSON-LD: 831-1.0086393-ld.json
RDF/XML (Pretty): 831-1.0086393-rdf.xml
RDF/JSON: 831-1.0086393-rdf.json
Turtle: 831-1.0086393-turtle.txt
N-Triples: 831-1.0086393-rdf-ntriples.txt
Original Record: 831-1.0086393-source.json
Full Text

Full Text

CULTURAL VARIATION IN UNREALISTIC OPTIMISMbySTEVEN J. HEINEB.A., The University of Alberta, 1989A THESIS SUBMITTED IN PARTIAL FULFILLMENTOF THE REQUIREMENTS FOR THE DEGREE OFMASTERS OF ARTSinTHE FACULTY OF GRADUATE STUDIES(DEPARTMENT OF PSYCHOLOGY)We accept this thesis as conformingto the required standardsTHE UNIVERSITY OF BRITISH COLUMBIAAugust 1993© Steven J. Heine, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.Department of 25-■.f 1A-11‘.al The University of British ColumbiaVancouver, CanadaDate^k■-.1r Cn DE-6 (2/88)IIAbstractThe recently proposed notion that self-enhancing biases are significantlyimplicated in mental health is being challenged by cross-cultural research whichsuggests that such biases may be limited to cultures which foster anindependent construal of self. We examined whether individuals from a culturecharacteristic of an interdependent construal of self (Japanese) would show lessunrealistic optimism about potential, future life events than individuals from aculture characteristic of an independent construal of self (Canadian). Canadianrespondents were indeed significantly more unrealistically optimistic than theJapanese, although the Japanese did demonstrate an optimism bias in somelocalized cases. Canadians made more unrealistically optimistic judgments forparticularly threatening events, whereas this was not so for the Japanese. Theweaker unrealistic optimism bias of the Japanese was associated with lower self-esteem, lower dispositional optimism, a more external locus of control, and lessof a tendency to imagine stereotypical people associated with the events. Theresults provide further evidence that self-enhancing tendencies are morecommon for cultures characteristic of an independent construal of self.TABLE OF CONTENTSAbstract^List of TablesAcknowledgements  ^ viIntroduction ^ 1Unrealistic Optimism^ 3The Japanese Case 5Independence versus Interdependence^ 5Cognitive and Motivational Correlates of the Optimism Bias^8Social Comparison^ 9Psychological Control 12Availability of Stereotypes^ 14Self-Esteem and Positive Construals of Self^ 16Self-Enhancement, Psychological Threat, and the Japanese^ 18The Present Study^ 20Method^ 21Respondents^ 21Materials 22Within-Groups Measure^ 23Between-Groups Measure 25Translation of Materials^ 26Results^ 26Comparability of the Samples^ 26Optimism Bias for Individual Events 26Optimism Bias for Item Aggregates^ 28Cultural Differences in the Optimism Bias per Event^29ivCultural Differences in the Optimism Bias for Aggregated Items^30Control, Availability of Stereotypes, and Self-Stereotypes^ 31Correlations between Desirability/Severity and Optimism Bias,Stereotypes, and Self-Stereotypes^ 33Cultural Differences on the Remaining Scales 35Discussion^ 35Unrealistic Optimism^ 35Cultural Differences in Unrealistic Optimism^ 37Control, Stereotypes, and Self-Stereotypes 38Cultural Differences in the Optimism Bias for Positivevs. Negative Events^ 40Unrealistic Optimism as a Form of Coping with Threat^ 41Self-Enhancement and ConstrueIs of Self^ 43Unrealistic Optimism and Positive ConstrueIs of the Self^44General Conclusions and Future Directions^ 47References^ 51List of Tables1. T-Tests for Individual Future Life Events2. T-Tests for Aggregated Events3. Anova and Means per Future Life Event4. Anova for Aggregated Future Life Events and Means5. Anova for Self/Other Estimates6. Anova and Means for Aggregated Control, Stereotype,and Self-Stereotype Items7^Anova for Average Adjusted Correlations BetweenDesirability/Severity, Stereotypes, Self-Stereotypes,Control, and the Optimism Bias8. Anova and Means for the Remaining Scales9. Correlations of Remaining Scales with the Optimism BiasviAcknowledgmentsI thank the members of my committee, Jennifer Campbell and Jim Russell,for their time and thoughtful advice. I would especially like to thank my advisor,Darrin Lehman, for his continual support in guiding me through this and otherrelated projects.Many friends also provided invaluable support along the way. Thanks toLarry Axelrod for his advice and for introducing me to SPSS:X, Chris Davis washis help with statistical matters, Adam Dipaula, liana Katz, and AndrewStarzomski for their helpful advice and support throughout the project, and SaraDettman and Rebecca Tara for their assistance with countless tasks. I alsowould like to thank the translators who worked on this project, Akio Tanaka andMariko Yamada, for their diligent contributions. Lastly I would like to thankNariko for her insights and unwavering support.For decades, classical psychological thought had maintained that arequirement of mental health was that one's perceptions have a firm footing inreality. Accurate perceptions, for example, of one's social environment andabilities, were viewed as critical for one to function in an effective and healthymanner (Jahoda, 1958; Jourard & Landsman, 1980). Recently, however, anumber of theorists have challenged this assumption, arguing that the healthymind is characterized by misperceptions that depart considerably from reality(see Greenwald, 1980; Taylor & Brown, 1988, for reviews). Much contemporarywork on the self has focused upon people's tendencies to distort theirperceptions of the world in a self-enhancing manner. Accuracy, althoughnecessary to a certain extent, is often compromised in favor of flatteringinformation in a typical self-evaluation. For example, people tend to remembertheir past performance as better than it actually was (Crary, 1966), they judgepositive personality attributes to be more appropriate in describing themselvesthan in describing others (Alicke, 1985), and they tend to take credit for success,yet attribute failure to the situation (see, e.g., Greenberg, Pyszczynski, &Solomon, 1982).The prevalence of these illusions, and their tenacity, across a broadspectrum of psychological processes has lead some theorists to argue for theinherent adaptive value of constructing a world view that departs from reality(Myers & Ridl, 1979; Taylor & Brown, 1988). Furthermore, there iscorresponding evidence that links the absence of these self-serving biases withlower self-esteem and mild depression (e.g., Alloy & Ahrens, 1987; Lewinsohn,Mischel, Chaplin, & Barton, 1980). Taylor and Brown (1988) suggest that "Itappears to be not the well-adjusted individual but the individual who experiencessubjective distress who is more likely to process self-relevant information in arelatively unbiased and balanced fashion" (p. 196). Self-enhancing biases thusCultural Variation in Unrealistic Optimism2appear to be necessary luggage for the trek to mental health. These "positiveillusions," as Taylor and Brown coined them, have been found to be associatedwith happiness and contentment (Freedman, 1978), resulting in enhancedabilities of caring about the self and others (Isen, 1984), and effectiveperformance in situations where perseverance is critical (Greenwald, 1980;Taylor & Brown, 1988).Cross-cultural research, however, has raised questions about theuniversality of these biases (see, for example, Markus & Kitayama, 1991a).Research has shown that with certain cultures, some effects attributed to self-enhancing tendencies are significantly lower, if not absent or even reversed(e.g., Kashima & Triandis, 1986; Takata, 1992). The lack of cultural universalityof self-enhancing biases challenges the claim that these illusions are the sinequa non to mental health. An investigation of this cultural variation by comparingthe general mental health of different cultures would, of course, be doomed to beawash in ethnocentric currents. Any measure of mental health for a givenculture cannot be assumed to have the same value or meaning for a differentculture, thus rendering any cultural ranking on this dimension biased andmisleading. A more reasonable approach would be to suggest that the benefitsof maintaining positive illusions presuppose certain cognitive or motivationaltendencies which might be specific to certain cultures. Cross-cultural researchsuggests that various cognitive, emotional, and motivational processes areculturally variant (see Markus & Kitayama, 1991a, for a review), and to theextent that these processes support and sustain self-enhancing biases, thecultural variance of these biases is understandable. Cross-cultural analyseshave thus proven useful in identifying the cultural boundaries of certain self-enhancing biases and in aiding our understanding of the cognitive andmotivational processes that constitute them.Cultural Variation in Unrealistic Optimism3Unrealistic OptimismOne self-serving bias that has proven to be consistently robust in studieswith North American respondents relates to optimism. In general, it appears thatpeople are future-oriented (Gonzales & Zimbardo, 1985), and believe that thepresent is better than the past and the future promises to be even better(Brickman, Coates, & Janoff-Bulman, 1978; for a review see Taylor & Brown,1988). Optimistic expectancies are therefore characteristic of the typical NorthAmerican's outlook.The term "unrealistic optimism," however, has been applied specifically tothe phenomenon represented by people's beliefs that they are more likely toexperience positive events, and less likely to experience negative events, thansimilar others (Weinstein, 1980). It is difficult to ascertain at the individual levelwhether or not a particular person is demonstrating an unrealistically optimisticoutlook on his or her future, as it is indeed possible that this person has a morepromising future than average. However, at a group level, if the discrepanciesbetween self- and other-estimates remain, then it must be the case that peopleare systematically predicting a rosier future for themselves than for others.Possible reasons that may justify a uniquely optimistic future for a givenindividual become untenable at a group level and thus the optimism is labeled"unrealistic" (Weinstein, 1980).Evidence for the unrealistic optimism bias is impressive; it can be foundacross a wide variety of events, and is consistent across age and socioeconomicclasses (Weinstein, 1987). For example, when college students wereadministered an insurance company longevity questionnaire which predictedtheir lifespan from actuarial data, they tended to estimate that they would live tenyears longer than the actuarial prediction (Snyder, 1978). People also tend tobelieve that they are more likely than others to enjoy their job and are less likelyCultural Variation in Unrealistic Optimism4to be fired (Weinstein, 1980). Slovic, Fischhoff, and Lichtenstein (1978)reported that 75 to 90% of drivers felt that they were better than average drivers,suggesting that they believed they were less vulnerable to accidents than others(see also Svenson, 1981).This optimism bias does not seem to be limited to any specific domain ofevents. Given a chance to compare themselves to an average other, people willusually conclude that their futures are relatively "better." However, in general,the more serious a negative event is perceived to be, the larger is thecorresponding optimism bias (Kirscht, Haefner, Kegeles, & Rosenstock, 1966;Taylor et al., 1992). Hence, although unrealistic optimism may be characteristicof any type of event, at least in the case of negative events its magnitudeappears to be dependent on its subjective threat (see also Kunda, 1987). Thissuggests that the foundation of the optimism bias is not simply cognitive, but ismotivational as well.Unrealistic optimism, similar to other self-enhancing biases, has beenargued to be characteristic of the mentally healthy person's thought processes(Taylor & Brown, 1988). Mildly depressed people, or people low in self-esteem,maintain more realistic views of their futures (Alloy & Ahrens, 1987). Biases inone's outlook toward future events thus seem to be associated with thefunctioning of the healthy mind, at least for North Americans. Whetherunrealistic optimism is similar to other self-serving illusions, in that it is subject tocultural determinism, remains an open, empirical question.Although hundreds of articles have been written on the topic of optimism,I have been unable to find any such articles that have explored the extent towhich unrealistic optimism, or even optimism in general, is universal or culturallyspecific. The central argument of this Master's thesis is that the optimism bias isculturally variant.Cultural Variation in Unrealistic Optimism5The Japanese CaseThe literature reveals that, by far, the culture that is most represented incross-cultural comparisons with North Americans is the Japanese. Thefascination, and perhaps envy, of the world at their economic success, and theirstrikingly visible collective nature have probably contributed somewhat to thecurrent interest in comparisons with the Japanese. In addition, Japan occupiesa relatively distinct position among modern cultures of the world (Hofstede,1980). The Japanese people will often be the first to say that their culture isdifferent from any other in the world. Besides unique historical, geographical,and linguistic factors that support these claims of uniqueness (Reischauer,1988), there is empirical, psychological evidence to corroborate this as well. In1980, Geert Hofstede published a multi-cultural study of values. Drawing from asample of workers in forty countries from IBM offices, he extracted fourunderlying value dimensions that reliably discriminated between cultures,namely: Power Distance, Uncertainty Avoidance, Masculinity, and Individualism.The Japanese are on the opposite side from Canada and the U.S. on all four ofthese dimensions, and they occupy the most extreme position in the world ontwo of them (Uncertainty Avoidance and Masculinity). Hence, they are in aparticularly appropriate position to identify cross-cultural differences with NorthAmericans.Independence versus InterdependenceMarkus and Kitayama (1991a) have provided a compelling model tointerpret much of the cross-cultural research that has been conducted thus far.They argue that the various cultures of the world differentially emphasize twotasks relevant to everyday life: independence, i.e., tasks related to agency andautonomy and interdependence, i.e., tasks related to communion and affiliation(Kitayama, 1993). People of cultures in which the former process is primary areCultural Variation in Unrealistic Optimism6said to have an independent construal of self, whereas those who live in cultureswhere the latter process dominates are said to have an interdependent construalof self.Markus and Kitayama define the independent construal of self as beingcharacterized by a bounded and autonomous sense of self that is relativelydistinct from others and the environment. Those with an independent construalof self strive to assert their individuality and uniqueness, and stress theirseparateness from the social world. This construal of self, therefore, placesdemands on the individual to be self-sufficient and in control of the world withwhich it interacts. The social world and surrounding environs provide a stage forthe self to perform on, and subsequently from which to be evaluated (Cousins,1990). For the independent construal of self, others with whom one interactsserve as benchmarks for comparison, in an attempt to ascertain one's relativeworth as an individual (Festinger, 1954). This view is best exemplified by NorthAmerican and Western European cultures.In contrast, the interdependent construal of self is characterized by anemphasis on the interrelatedness of the individual to others and to theenvironment. The self is not considered to be separate and autonomous, and itis only through the contextual fabric of one's social relationships, roles, andduties that the self gains meaning. The interdependent self does not remain aconsistent and inviolate entity that is divorced from context: it must take on theform required by the social situation with which it interacts. As Hamaguchi(1985) put it, "(Selfness) is not a constant like the ego but denotes a fluidconcept which changes through time and situations according to interpersonalrelationships" (p. 302). The social environment does not act so much as a stagefor evaluation, but instead is better construed as sustaining and forming the self.Cultural Variation in Unrealistic Optimism7This construal of self is most representative of Asian cultures, as well as manyAfrican, Latin-American, and Southern European cultures.Markus and Kitayama (1991a) argue that since the self is central to manypsychological phenomena, any phenomenon that implicates the self will beshaped accordingly by that culture's dominant construal of self. Hence, culturescharacteristic of the independent construal of self will show evidence ofmotivations, cognitions, and emotions that affirm the independence andautonomy of the self. Psychological processes within cultures representative ofthe interdependent construal of self, on the other hand, will affirm theinterrelatedness and belongingness of the self.The Japanese culture is particularly appropriate to the definition of theinterdependent culture. Many anthropological studies (see, e.g., Benedict,1946; Reischauer, 1988) stress that one of the most characteristic features ofthe Japanese is the great deal of importance they ascribe to their in-groups.Nakane's (1970) classic sociological analysis similarly traces the Japanesepsychological centre of gravity to the ie, the household, or in-group. As well,Hofstede's (1980) analysis demonstrated that the Japanese were on thecollectivist side of his individualistic-collectivist dimension.Numerous studies have demonstrated that the Japanese do responddifferently from North Americans on a wide variety of psychological processes(e.g., Barnlund & Yoshioka, 1990; Cousins, 1989; Hamilton, Blumenfeld, Akoh, &Miura, 1990; Yamagishi, 1988). Most relevant to the present research arestudies showing that a number of self-enhancing biases that are well-establishedwith North American respondents do not hold for the Japanese (e.g., Kashima &Triandis, 1986; Markus & Kitayama, 1991b; Yamauchi, 1990). This apparentabsence of self-enhancing biases on the part of the Japanese is suggestive ofdifferent motivational tendencies of the interdependent construal of self.Cultural Variation in Unrealistic Optimism8The notion that unrealistic optimism is primarily a product of"independent" cultures would gain plausibility if Canadians demonstrated asignificantly stronger optimism bias than the Japanese. In contrast, similarpatterns of responses between the two cultures might suggest that unrealisticoptimism is based on cognitive or motivational constructs that are consistentbetween the two different construals of self, and is in some way different fromother self-enhancing biases studied thus far.Any difference in unrealistic optimism between cultures would, of course,not exist in a vacuum. It is not enough to demonstrate cultural variation in theoptimism bias without simultaneously exploring the cultural variation of relatedconstructs. This exploration of the contextual net sustaining the optimism biasoffers possible explanations for why these differences exist. A review of thecognitive and motivational correlates of the optimism bias follows. In an attemptto highlight areas that are likely to constitute the basis of any cultural differencesin the optimism bias, the review examines these associated constructs in thecontext of cross-cultural comparisons between North Americans and Japanese.Cognitive and Motivational Correlates of the Optimism BiasWeinstein (1980) was the first to systematically study the underlyingcognitive and motivational factors that could account for the optimism bias over alarge subset of future life events. Since that paper, Weinstein and numerousother researchers (e.g., Perloff & Fetzer, 1986; Zakay, 1984) have furtherexplored these correlates of unrealistic optimism. As the present study isapparently the first to examine the cultural generality of unrealistic optimism,predictions were based on indirect evidence suggestive of cultural differences.Hence, this thesis examined two sets of literature: 1) The cognitive andmotivational constructs that have been shown to be associated with the optimismbias, and 2) the cross-cultural literature that has examined these sameCultural Variation in Unrealistic Optimism9constructs within the Japanese people. The constructs to be discussed aresocial comparison, psychological control, availability of stereotypes, and positiveself-construals. As well, self-enhancement as a means of coping with stress isalso discussed in a cross-cultural context.Social Comparison For four decades a fundamental tenet of social psychologyhas been that because many psychological characteristics cannot be easilyevaluated against an objective standard, people instead evaluate themselves bymaking comparisons with similar others in their social environment (Festinger,1954). People often make downward comparisons, that is, they comparethemselves to relatively disadvantaged others, as this has self-enhancingbenefits (Taylor, Wood, & Lichtman, 1983; Wills, 1981). Wills argues that as thelevel of psychological threat increases, people are more likely to makedownward, self-enhancing comparisons, rather than self-evaluative comparisonswith equal or better off others. This is reflected in the findings of Kirscht et al.(1966), who found that the discrepancy between self- versus other-estimates forvulnerability to diseases increased with the perceived severity of the disease.This suggests that motivational factors are involved with downward comparisons,such that they provide the means to cope with feelings of inadequacy andvulnerability.Perloff and Fetzer (1986) argue that people exhibit an unrealisticallyoptimistic view of their own vulnerability to victimization because they typicallyevoke downward comparison strategies. In the prototypic optimism study,respondents are requested to estimate their perceived chances of experiencingfuture life events relative to the average comparable other, usually the averagesame-sex student at their university. Perloff and Fetzer found that whenrespondents compared themselves to this vague concept of an average other,.they usually brought to mind someone relatively disadvantaged.Cultural Variation in Unrealistic Optimism10This is consistent with other findings that North American respondentstend, in general, to consider themselves to be above average on a wide varietyof abilities (Myers & Ridl, 1979). For instance, 25% of college studentsestimated that they were in the top 1°/0 of the population regarding the ability ofgetting along well with others (Myers, 1987). Perloff (1987) suggested that"insofar as people tend to view themselves as better than average and as moreintelligent than their average peer, the average person may be seen as someonewho is almost by definition, less advantaged, less intelligent, and generallyworse off than oneself" (pp. 222-223). The standard unrealistic optimismparadigm is framed in such a way that comparing themselves with a vague"average other" will apparently lead most "better-than-average" North Americansto engage in downward comparisons.In sharp contrast, the cross-cultural literature has shown that thetendency to see oneself as better than average does not hold for cultures typicalof the interdependent construal of self. For example, Markus and Kitayama(1991b) found that when given a task of estimating one's position relative to thepopulation on a variety of abilities, Japanese students did not view themselvesas better than others. In fact, the Japanese estimates were generally consistentwith what would be expected of realistic estimates that did not involve self-enhancing evaluations. On average, the Japanese consider themselves to beabout average. Perhaps, when comparing themselves to an average student,the Japanese are not as likely as North Americans to employ downwardcomparison strategies. In general, others are seen as existing at an equal,rather than a disadvantaged, level. Markus and Kitayama argue that this isbecause the motivation of the interdependent self is not to stick out from thecrowd, even in a positive way. The motive to fit in, and to belong, is relativelyCultural Variation in Unrealistic Optimism11stronger than it is in the West, so downward comparison strategies might notyield the same psychological satisfaction as has been argued for Westerners.Takata (1992) has provided preliminary evidence to support the claim thatJapanese employ different social comparison strategies than those used byNorth Americans. Schwartz and Smith (1976) showed that American subjectsreadily believed feedback indicating that their performance was superior to thatof another, while they were more reluctant to accept feedback that their relativeperformance was inferior. Using methods identical to those used by Schwartzand Smith, Takata found that Japanese subjects responded in a manneropposite to Americans. That is, Japanese subjects were reluctant to believe thatthey outperformed someone else, but they were more easily convinced abouttheir relative inferior performance. This tendency to trust failure informationmore than success information remained even when the Japanese were told thatthey were competing against a computer program, rather than against a fellowstudent. Takata argues that this self-deprecative tendency, much like self-enhancing biases for Westerners, is a self-esteem maintenance mechanism forthe Japanese. The Japanese are motivated to perceive themselves "not as a'figure,' but as a 'groundw(Takata, 1992. p.5), because it is more favorable to"harmonize" with other people than to be an outstanding individual. Self-enhancing comparisons therefore might be limited to self-esteem maintenancemotivations only for cultures typical of an independent construal of self.The importance of downward comparison to unrealistic optimism, asargued by Perloff and her colleagues, suggests that the relative absence ofdownward comparison strategies in the Japanese, as demonstrated by Takata,will be reflected in lower unrealistic optimism in the Japanese.Cultural Variation in Unrealistic Optimism12Psychological Control A second factor that has been argued to be of importancein explaining unrealistic optimism is that of psychological control. If an event isperceived to be controllable, then a person can bring to mind the series of stepsnecessary to increase the likelihood of a desirable outcome (Weinstein, 1980).When an event is perceived to be under a person's control, and is thuspotentially attainable, one exerts efforts at achieving his or her goal. If the goalis viewed as beyond one's control, the person may give up and turn away(Scheier & Carver, 1987). In this regard the level of perceived control affectsone's expectancies of the future.Perloff (1987) argues that people have an illusion of control whereby theyexaggerate their ability to avoid negative outcomes. Insofar as this tendencymakes the future appear to be under one's control, one should be able tomaintain an optimistic point of view. Similarly, the tendency for nonvictims toperceive themselves as being uniquely invulnerable to threatening events mayreflect a need for personal control (Perloff, 1983). In support of this, Langer(1975) has demonstrated that people typically have an "illusion" of control overchance events. She argues that individuals are motivated to avoid the negativeconsequences that are associated with a perceived loss of control. This illusionof control might result in people overestimating their likelihood of achievingpositive events, and underestimating their likelihood of experiencing negativeevents (Perloff, 1983). Self-serving biases that lead people to believe that theyare more intelligent (Wylie, 1979), and in general "better" than the averageperson (Myers & Ridl, 1979), may be reflected in similar beliefs that one is inmore control of his or her life than the average person.Evidence for the role of control is found consistently in unrealisticoptimism studies (Weinstein, 1982; Zakay, 1984). Weinstein (1980), forexample, found that perceived controllability correlated significantly with theCultural Variation in Unrealistic Optimism13degree of unrealistic optimism that a person maintained for avoiding negativeevents. The contributing role that control plays in unrealistic optimism wasfurther demonstrated in another study by Weinstein (1984). He attempted todelineate the risk factors that people take into account when they are makinglikelihood judgments. He found that respondents were unbiased when theyconsidered hereditary or environmental risk factors. These factors, Weinsteinargued, operate outside people's perceived control. However, whenrespondents considered their actions and psychological risk factors, there was asignificant correlation with their optimism bias. Both actions and psychologicalfactors are associated with perceived controllability, so people appear to believethat if they can control factors that affect their future, they have a lesserlikelihood of experiencing negative events than do others. This suggests thatpeople succumb to the optimism bias, in part, because they tend to note theircontrollable, and henceforth risk-decreasing factors, but they tend to downplayany uncontrollable hereditary or environmental risk-increasing factors that mightbe involved.Weisz, Rothbaum, and Blackburn (1984) argue that Japanese societypromotes a different sense of psychological control than what we areaccustomed to in the West. They argue that in Japan the primary emphasis ison adapting oneself to the demands of an impassive environment. Kojima(1984) suggested that "the Japanese do not think of themselves as exertingcontrol over an environment that is utterly divorced from the self... Rather, theyattempt to regulate the relationship between the self and a complex, oftenconflicting set of environmental demands" (p. 973). Weisz et al. called this focuson controlling one's responses to the environment "secondary control." Incontrast, they argue that the main emphasis in North American culture is onCultural Variation in Unrealistic Optimism14attempting to change the environment to suit one's internal demands. Thissense of internal agency was labeled "primary control."The primary-secondary control distinction is similar to Rotter's distinctionbetween internal and external locus of control. Both primary control and aninternal locus of control imply that an individual can exert control on events inthe environment. Secondary control is similar to an external locus of control inthat it is believed that the outcomes of events is dependent upon forces in theenvironment. Although Weisz et al. only provided sociological andanthropological evidence to support their claims of cultural differences in primaryand secondary control, there is empirical evidence of cultural differencesregarding internal and external loci of control which lend support to theirarguments. The Japanese have been shown to demonstrate a significantlystronger external locus of control, and a significantly weaker internal locus ofcontrol, than Americans (Bond & Tornatzky, 1973; Mahler, 1974). As perceivedcontrol has been shown to be correlated with unrealistic optimism, it is expectedthat the more externally oriented Japanese will show a lower optimism bias thanNorth Americans.Availability of Stereotypes The variable correlated most strongly with unrealisticoptimism in Weinstein's first (1980) study was the availability of a stereotype forthe event. For many events, particularly negative ones, people may have aconception of the kind of person to whom the event is likely to happen.Weinstein argues that stereotypes associated with negative events have an ego-defensive function, as people would rarely consider themselves to berepresentative of the type of person that is vulnerable to misfortune.The correlation between unrealistic optimism and the availability ofstereotypes also suggests the presence of a cognitive basis for unrealisticoptimism. Weinstein (1980) suggests that the "representativeness" heuristicCultural Variation in Unrealistic Optimism15(Kahneman & Tversky, 1972) may account for the relation between optimismbiases and stereotype salience. This heuristic signifies the process of judgingwhether an individual fits into a particular category or not by judging the extent towhich the individual possesses the salient characteristics associated with thestereotype, while ignoring information regarding the base rates of the category.For example, people may have an image of the stereotypical person likely tohave an early heart attack. This image might be of someone who is grosslyoverweight and overly stressed at work. Indeed, these are risk factors for anearly heart attack, however, early heart attacks can, and often do, happen topeople who do not fit this stereotype at all. Most importantly, people fail torealize that relatively few victims of negative events actually fit a givenstereotype, and thus erroneously conclude that they themselves are relativelyinvulnerable to the event because they do not view the stereotype as descriptiveof themselves (Weinstein, 1980).Westerners have a richly defined conception of themselves, so it isdifficult to impose a stereotype on oneself. A relative lack of information aboutothers, on the other hand, makes the stereotypes appear more appropriate incategorizing these others (see, e.g., Nisbett, Krantz, Jepson, & Kunda, 1983,Study 2; Quattrone & Jones, 1980). Thus, North Americans seem relativelyimmune to self-categorization of a negative stereotype, rendering them moresusceptible to unrealistic optimism.The role of stereotypes for the Japanese is arguably less important intheir consideration of others. For example, Kitayama, Markus, Tummala,Kurokawa, and Kato (1990) found that, compared to Americans, Hindurespondents demonstrated a greater awareness of others relative to themselves.This suggests that the cognitive function of stereotypes for Hindu Indians mightbe moderated. Insofar as this finding is typical of people of Eastern cultures inCultural Variation in Unrealistic Optimism16general, it implies that people with an interdependent construal of self have arelatively impoverished conception of themselves, and the application of astereotype to themselves may consequently appear more fitting than it would forthose of an independent self-construal. In contrast, their relatively more detailedconception of others, resulting from their greater attendance to information fromthe social environment, would suggest that stereotypes are not as applicable toothers. If this is the case, then negative stereotypes may appear more self-defining, and less other-defining, for the Japanese than for Canadians. Theperception of invulnerability that one maintains by avoiding stereotypeclassification should thus be moderated for the Japanese rendering them lesssusceptible to unrealistic optimism. Since Weinstein (1980) found a significantcorrelation between stereotype salience and optimism bias only for negativeevents, the hypothesized tendency for the Japanese to see stereotypes as beingmore self-defining, and less other-defining, than North Americans, should beassociated with a particularly lower optimism bias in the case of negative events.Therefore we anticipate that the difference in unrealistic optimism betweenCanadians and Japanese will be more pronounced for negative events than forpositive events.Self-Esteem and Positive Construals of the Self Alloy and Ahrens (1987)provided evidence that depressives and people with lower self-esteem were lessunrealistically optimistic than were normal individuals. This is consistent with thearguments of Taylor and Brown (1988) who suggest that the use of self-enhancing biases is characteristic of the mentally healthy person. It appearsthat positive construals of self, i.e., high self-esteem and low depression, areassociated with the creation of self-enhancing biases that place one in afavorable position relative to others.Cultural Variation in Unrealistic Optimism17Positive construals of the self, however, seem to be culturally variant.Kitayama and Markus (1992) argue that because people in Western cultures areencouraged to identify their own unique features that are desirable, self-evaluative schemata become particularly sensitive to positive information. Thispositivity bias is highly instrumental in discovering and expressing positiveattributes of the self that are necessary in attaining the cultural imperative ofindependence. In contrast, they argue, people in interdependent cultures areconditioned to become particularly sensitive to information about their owndeficiencies, or potential mistakes, so that these "problems" can be correctedand they can maintain and deepen their relationships with others. The role ofthe interdependent individual requires that they adapt and change themselves inorder to foster the harmony of their in-group. People in interdependent culturesthus learn to be particularly aware of negative self-information.Kitayama and his colleagues have provided some empirical support forthese views. A recent study explored the structural complexity for situations inwhich self-esteem was enhanced or reduced for both Americans and Japanese(Kitayama, Markus, Takagi, Sugiman, & Matsumoto, 1992). They found thatwhile Americans have a more differentiated schema for accepting positiveinformation, Japanese had greater structural complexity for negative situationsthat reduced self-esteem (see also Takata, 1992). In line with this, in an earlierstudy we found that the Japanese scored significantly lower on the RosenbergSelf-Esteem measure than did Canadians (Heine, Lehman, Okugawa, &Campbell, 1992).Taylor and Brown (1988) argue that self-serving biases are generallyassociated with positive construals of the self. Since there are reasons tobelieve that the Japanese have a less "positive" construal of self, it makes senseto assume that they should employ fewer self-serving biases. We expect thatCultural Variation in Unrealistic Optimism18the Japanese in this study will again have lower self-esteem scores thanCanadians, and that this will be associated with a lower unrealistic optimismbias.Self-Enhancement, Psychological Threat, and the JapaneseOne of the apparent purposes which self-enhancing biases serve forNorth Americans is aiding in coping with stress (Taylor et al., 1992). In the faceof threatening events, self-enhancing evaluations that place the self in a morefavorable position than that ascribed by an impassive reality can relieve the selfof some of the stress associated with these events. Often, this means makingdownward comparisons with others. Moreover, the more severe the threat, thegreater the need may be for these "positive illusions" to counter the stress of thethreats (Taylor et al., 1992; Wills, 1981).The psychological benefits of self-enhancement, however, might belimited to those with an independent construal of self. The motivations for theindependent self are to maintain the autonomy of the sacrosanct self, to confirmto the individual that they are a worthy person. If this sense of self-sufficiency isthreatened, or worse, if the very existence of the individual is not secure, thenself-enhancing biases might help the individual restore his or her sense ofautonomy and security. Therefore, in the case of the present investigation, weanticipate that Canadians will show more unrealistic optimism for events whichthey perceive as particularly threatening.However, self-enhancing biases might not provide the same palliativereassurances for the interdependent construal of self. Given that the individualin an interdependent society is but a fraction, and does not become whole untilthey have fit in and occupied their proper place within a social unit (Lebra,1976), individuals ought not be motivated to separate themselves from theirsecure position in the group, regardless of the positive reasons for theirCultural Variation in Unrealistic Optimism19separation. Separation from the group might actually imply alienation from theself. Kitayama et al. (1991) found that, while for Americans feelings of pride andsense of achievement were positively correlated with their sense of well-being,for Japanese these were actually negatively correlated with their sense of well-being. Their sense of acceptance from others, on the other hand, was what wasmost highly correlated with their feelings of well-being. Self-enhancement, i.e.,distinguishing oneself as better than others, might similarly be in opposition tothe well-being of the Japanese. Hence, we would not expect self-enhancingbiases to be as common for Japanese as they are for Canadians, nor would theynecessarily be as effective for coping with stress.In support of this, a number of studies have shown that certain self-enhancing biases that are robust in North America exist in an attenuated formwithin cultures typical of an interdependent construal of self. The falseuniqueness bias (Markus & Kitayama, 1991b), the tendency to internalizesuccess and externalize failure, (Chandler, Shama, Wolf, & Planchard, 1981;Kashima & Triandis, 1986; Yamauchi, 1990), and the tendency to have moreconfidence in information that is favorable to the self, than that which isunfavorable (Takata, 1992), have all been shown to either be absent, orreversed for Japanese samples. This evidence suggests that self-enhancingtendencies may be essentially absent in the Japanese motivational repertoire.Whether this tendency is specific to the above biases, or is indicative of ageneral trend, remains an open question until other self-enhancing biases areexamined with Japanese samples.Self-enhancement, then, apparently does not bring the samepsychological satisfaction to the Japanese as it does to North Americans. Self-enhancing evaluations may only serve to isolate the Japanese individual fromhis or her interdependent network. In the face of threat this isolation couldCultural Variation in Unrealistic Optimism20hardly be seen as a coping mechanism. On the contrary, if anything, threatought to motivate the interdependent Japanese to affirm their belongingness tothe group. Rather than helping the Japanese individual to cope with threat, self-enhancement might even exacerbate the negative consequences of thethreatening event. Hence, in contrast to what has been argued for Canadians,we do not expect the Japanese to self-enhance more in the case of particularlythreatening events.The Present StudyThe present investigation examines whether unrealistic optimism exists ata comparable level between Canadian and Japanese respondents. In addition,constructs that have been shown to be related to unrealistic optimism for NorthAmericans were examined with the Japanese as well. Specifically, theconstructs of locus of control, the availability of stereotypes, self-esteem, anddefensive self-enhancement were considered in relation to unrealistic optimism.Unrealistic optimism was assessed in two ways. The most commonparadigm used for unrealistic optimism studies thus far has been a within-groupscomparison, in which respondents make an estimate of the relative likelihoodthat they will experience a future life event in comparison to a similar other (seeWeinstein, 1980, 1982, 1984, 1987). The present study employed thismethodology. In addition, we employed a between-groups paradigm similar tothe method used by Kirscht et al. (1966). Respondents were divided into twogroups. One group made absolute likelihood estimates about their chances ofexperiencing a future life event and the second group made estimates for thechances of a similar other. The average estimates were then compared betweenthe two groups with significant differences representing unrealistically optimisticor pessimistic tendencies. The between-groups methodology provides a"hidden" measure of self-enhancement, because respondents are unaware thatCultural Variation in Unrealistic Optimism21discrepancies between self- and other-estimates are being assessed acrossexperimental groups. The results of these two methodologies were compared.Based on the preceding arguments about differences between thenatures of the independent and interdependent construals of self, we anticipatedthat: 1) Japanese would show less of an optimism bias overall than Canadians;2) the lower level of optimism bias for the Japanese would be particularlypronounced for negative events; 3) perceived threat would be less stronglylinked to unrealistic optimism for Japanese than for Canadians; and 4) therelative lack of optimism bias for the Japanese would be associated with: a) aless internal, and a more external, locus of control, b) fewer availablestereotypes corresponding with each life event, but a greater tendency to seeoneself as fitting a stereotype, c) lower self-esteem, and d) lower dispositionaloptimism.MethodRespondentsA total of 510 respondents participated in the study. They came from fourdifferent sources: 1) a class of introductory psychology students from NagasakiUniversity, a public university in southern Japan (n=112; these students receivedcourse credit for participating in the study); 2) a class of introductory researchmethods students from Ritsumeikan University, a private university in Kyoto, inwestern Japan (n = 84; the study was administered at class time, and wasincluded as part of the lecture); 3) a class of introductory social psychologystudents from the University of British Columbial (n=174; the study wasadministered at class time, and the objectives and results were discussed lateras part of the course material); and 4) students enrolled in the University of1The data were collected in the second class period of the year, so the students had not yetstudied anything relating to self-enhancement, unrealistic optimism, or the like.Cultural Variation in Unrealistic Optimism22British Columbia introductory psychology courses that were contacted throughthe subject pool (n=140; these students received course credit for participatingin the study).As the primary aim of this study was to compare the self-enhancementtendencies of people of Eastern and Western cultures, the cultures of thesamples were polarized. That is, the Canadian sample was separated by ethnicbackground, such that respondents who had the most exposure to Westernculture could be contrasted with a homogenous Japanese sample. To obtainmembership in the Westernized Canadian sample, respondents had to meeteach of the following criteria: 1) The respondent had to be born in eitherCanada or the United States; 2) both of the respondent's parents had to be bornin Canada, the United States, or in a European country; 3) the respondent hadto declare his or her ethnic descent to be that of a European culture; and 4) therespondent had to be between the ages of 18 and 25. A total of 90 respondentssatisfied all of these criteria, and formed the "Canadians of European descent"sample, or "Canadians," for short. Respondents in the Japanese sample wereall between the ages of 18 and 25, and apart from a few students born in otherEast Asian countries, the rest of the sample was Japanese-born. A total of 196students are included in this sample.The Canadian sample consisted of 62.2% females (n=56) compared to66.3% (n=130) for the Japanese sample. As the present study was concernedprimarily with examining cultural differences, and not gender differences, thefemale and male respondents were pooled into one composite sample.MaterialsAll respondents completed a questionnaire packet that consisted of thefollowing measures, some of which are exploratory and are not included in thisMaster's thesis:Cultural Variation in Unrealistic Optimism231) Demographic information (6 items)2) Rosenberg's (1965) Self-Esteem measure (10 items)3) Campbell, Trapnell, Katz, and Lavallee's (1992) Self-Concept Confusionmeasure (12 items)4) Cheek's (1982) Aspects of Identity scale (20 items)5) Levenson's (1972) Locus of Control measure (24 items)6) A measure of modesty calculated by the respondent's total self-rating of adiverse number of ambiguous attributes (8 items)7) Scheier and Carver's (1985) Life Orientation Test (a measure ofdispositional optimism) (12 items)Future Life Events Scale 8) Controllability of future life events (15 items).9) Availability of stereotypes for future life events (15 items).10) Applicability of self-stereotypes for future life events (15 items).11) Ranking of the desirability (5 items) and the severity (10 items) of thepositive, and negative, future life events respectively.12) -Unrealistic optimism events (15 of them) which were measured in thefolVwing manner:Within-Groups Measure A design identical to Weinstein (1982) was used.Respondents were asked "Compared to other UBC/Ritsumeikan/Nagasakistudents--same sex as you--what do you think are the chances that the followingevents will happen to you?" Beneath the description of each event a seven-point rating scale with the following choices appeared: Much below average,below average, slightly below average, average for otherUBC/Ritsumeikan/Nagasaki students of your sex, slightly above average, aboveaverage, and much above average. For purposes of analyses these sevenresponses were assumed to form an equal interval scale and were assigned theCultural Variation in Unrealistic Optimism24values -3 (much below average) through +3 (much above average). Anoptimism or pessimism bias was noted whenever the estimates for a particularevent deviated significantly from zero. A significant negative value implies thatan optimism bias is operating, and a significant positive value implies thepresence of a pessimism bias.The list of future events is a subset of the events that Weinstein has usedin his various studies, plus a few additions that were of interest for the presentstudy.Positive events 1) Enjoy your chosen career.2) Live past 80 years old.3) Own your own home.4) Leave your company for a better job offer.5) Starting salary greater than $30 000/ 2 500 000 yen.Negative events 6) Have a drinking problem.7) Attempt suicide.8) Contract skin cancer.9) Get divorced a few years after marriage.10) Have a nervous breakdown.11) Get AIDS.12) Drop out of university before graduating.13) Do something that will make your family ashamed of you..14) Have a heart attack before the age of 50.15)^Become senile with old age.Items 4 and 13 were added because we expected them to be viewed differentlyacross the two cultures: Item 4, "Leave your company for a better job offer," wasCultural Variation in Unrealistic Optimism25expected to be a more positive event for Canadians, given the strong loyaltyelement in Japanese work organizations. Item 13, "Do something that will makeyour family ashamed of you," was expected to be considered more threateningto the Japanese, due to the important role of family in the Japanese self-concept. Item 11, "Getting AIDS," was seen as a more current item than the item"Getting venereal disease," as originally used by Weinstein. All of the eventsadopted from Weinstein's studies have produced significant unrealistic optimismbiases in past research.Between-Groups Measure  To measure the between-groups factor, two differentversions of the questionnaire were used. In the first version, beneath eachfuture life event that the respondent made a relative likelihood estimate for, theywere also asked to estimate the absolute percentage chance that this eventwould occur to them. In the second version of the questionnaire, beneath eachfuture life event the respondent was asked to estimate the absolute percentagechance that this event would occur to the average same-sex student from theiruniversity. An optimism or pessimism bias was calculated for the between-groups factor whenever the self-estimates were significantly different from thecorresponding other-estimates.The same methods as Weinstein (1980) were employed to assess controland the availability of stereotypes. Respondents were asked on a scale from 1(very controllable) to 5 (not at all controllable) how controllable they felt eachevent was, and they were asked on a scale from 1 (very clear image) to 3 (noimage at all) the extent to which they could imagine a typical person likely toexperience each event. In addition, the applicability of a self-stereotype wasmeasured by asking respondents on a scale from 1 (not at all like me) to 5 (justlike me) the extent to which they felt that they themselves were the type ofperson likely to experience the event. As well, the desirability and severity (i.e.,Cultural Variation in Unrealistic Optimism26threat) of the events were measured by asking respondents to rank the fivepositive events in order of their perceived desirability, and to rank the tennegative events in order of their perceived severity.Translation of MaterialsQuestionnaires were produced both in English and Japanese, andrespondents completed them in their native language. The original Englishversion was translated into Japanese, and then back-translated into English by asecond translator. Any discrepancies between the two English versions werenoted, and three translators worked together on the discrepancies until aconsensus was reached regarding their equivalency.ResultsComparability of the samplesA one-way Manova was conducted on all the major scales included in thisstudy to determine whether the two Japanese samples (i.e., those fromRitsumeikan and Nagasaki Universities) could be pooled into a single sample.The results of this analysis (F = 1.36, p> .10) indicate that the two Japanesesamples did not differ statistically, and thus could be combined.A t-test analysis revealed that the average age of the Japanese samplewas significantly younger than that of the Canadians' ( Japanese M = 19.43,Canadian M = 20.97, 4284) = 4.74, p < .001). A correlational analysis, however,indicates that there were no significant relations between age and total optimismbias for either the Japanese or Canadian samples, rs-0.09, and 0.13,respectively, p> .10 for both), and thus the age difference should not confounda comparison of the optimism bias between cultures.Optimism bias for individual eventsFor the within-group analyses, respondents indicated whether they feltthat their likelihood of experiencing the individual future life events was greaterCultural Variation in Unrealistic Optimism27than, less than, or about the same as that of their peers. An event that has anaverage value that is significantly less than zero demonstrates an optimism bias:that is, respondents feel that they are less likely than their peers to experiencethe negative event (or more likely than their peers in the case of positiveevents). If the event's average value is significantly greater than zero, itdemonstrates a pessimism bias, i.e., respondents feel that they are more likelythan their peers to experience the negative event (or less likely to experiencethe positive event).With respect to the between-groups design, an optimism bias isdemonstrated when respondents receiving the self-estimate version estimatethat their likelihood of experiencing a negative event is significantly lower (orhigher in the case of positive events) than the estimates of the respondentsreceiving the other-estimate version (i.e., estimates of the percentage chancethat the given event will happen to the average same-sex student from therespondent's university).The t-tests representing the significance of an optimism bias for each ofthe 15 events in both the within- and between-groups designs are presented forboth cultural groups in Table 1. Caution should be used when comparing themagnitude of the t-values between cultures, as the Japanese sample is overtwice as large, and hence more powerful, than the Canadian sample.• ***** ••••••••••••••••••••••••• •••••••••Insert Table 1 about here• • ***** • ***** • ***** •• •• • ***** •• ***** ••••For the Canadians, 14 of the 15 items in the within-groups designrevealed a significant optimism bias, thereby replicating Weinstein's earlierstudies with Americans. The one item which did not demonstrate a significantoptimism bias was "Develop skin cancer." In the case of the JapaneseCultural Variation in Unrealistic Optimism28respondents, 12 of the 15 items showed a significant optimism bias.Interestingly, though, 2 of the 15 items, "Live past 80 years old," and "Yourstarting salary will be greater than $30 000/ 2 500 000 yen," showed asignificant pessimism bias.The Canadian results from the between-groups design showed asignificant optimism bias in 9 of the 15 items. Two of the items, "Develop skincancer" and "Become senile with old age," showed a pessimistic tendency butthese did not reach significance. For the Japanese sample, only 3 of the 15between-group items exhibited a significant optimism bias: "Enjoy your career,""Own your own home," and "Have a nervous breakdown." Five of the 15 itemswere answered in the pessimistic direction, with one of these, "Live past 80years old," resulting in a significant pessimism bias.Optimism Bias for Item AggregatesWe next aggregated the items by their valence (5 positive, and 10negative items), and conducted t-tests. Again caution should be taken incomparing the magnitude of the t-values because of the considerably greaterpower of the Japanese group.••••• •• ***** •••• ***** •••••••••••••••••••Insert Table 2 about here••••••••••••••••••••••••••••••••••••••••For the aggregated totals, in the within-group analyses, the Canadiansshowed a strong optimism bias for each of the positive, negative, and total itemaggregates (see Table 2). In the between-group analyses, the Canadiansdemonstrated a similarly strong optimism bias across all aggregates. Canadiansthus showed a consistent and robust optimism bias for all the types of future lifeitems in this study.Cultural Variation in Unrealistic Optimism29The Japanese did not demonstrate such a consistent optimism biasacross item types. They demonstrated a strong optimism bias for the negativewithin-group aggregate, and correspondingly this resulted in a strong optimismbias for the total within-group aggregate as well. However, they showed virtuallyno bias whatsoever in the positive within-group aggregate. In contrast, in thebetween-group analyses they showed an optimism bias for the positiveaggregate, and a non-significant pessimistic tendency for the negativeaggregate, which resulted in a non-significant optimism bias for the total itemaggregate. Even though the Japanese exhibited a significant optimism bias forthe positive aggregate in the between-groups condition, only 2 of the 5 itemsshowed a significant optimism bias, and 1 item showed a significant pessimismbias. The significant effect for this aggregate is due to the unusually largeoptimism bias demonstrated for the item "You will enjoy your chosen career."Thus, the Japanese only appear to be unrealistically optimistic for negativewithin-group items, and, to a questionable extent, positive between-group items.Cultural Differences in the Optimism Bias per EventThe following analyses reveal differences between Canadians andJapanese regarding the magnitude of the optimism bias for each of the events inthe study. A summary of the means for all the events is presented in Table 3.••••••••••••••••••••••••• ***** ••••••••••Insert Table 3 about here•• ••• •••••••••• ***** •• ***** •••••••••••• •For the within-groups design, Canadians were significantly moreoptimistic than the Japanese on 12 of the 15 items. On only 1 of the 15 items,"You will develop skin cancer," were the Japanese respondents more optimisticthan the Canadians, but this did not approach significance.Cultural Variation in Unrealistic Optimism30For the between-groups design, Canadians were significantly moreoptimistic than the Japanese on 5 of the 15 items. On only 1 of the 15 items,"You will become senile with old age," was the Japanese mean more optimisticthan the Canadian mean, but this did not approach significance.Cultural Differences in the Optimism Bias for Aggregated ItemsTable 4 presents the aggregates of the items on the basis of valence, andtheir means are compared across the two cultures. The values in the withincolumns represent the average optimism bias per event. Negative numbersrepresent optimistic responses. The values in the between columns representthe average difference between self and other estimates. Again, negativenumbers represent optimistic responses.• • •••••• • ••••• **** * •••••••••••••••• •••••Insert Table 4 about here•• • • • • *********** • • • • • • • • • • • • • • • • • • • • • • •Canadians showed a significantly greater tendency to exhibit an optimismbias compared to the Japanese in all aggregates in both the within- andbetween-groups designs. Canadians were consistently more unrealisticallyoptimistic than Japanese in all conditions.To better understand between-groups results, the respondents' actualpercentage estimates were compared for the aggregated items (see Table 5).•• ********************** • • ••• • ••••• *****Insert Table 5 about hereCanadians believe that positive future events are significantly more likely tohappen to themselves than do Japanese. Canadians also believe that positivefuture events are significantly more likely to happen to others than do Japanese.As well there is a significant interaction which shows that Canadians' estimatesCultural Variation in Unrealistic Optimism31for others are more similar to the same estimates of the Japanese than are theirself-estimates. Hence, the significant difference in the optimism bias betweencultures for the positive events in the between-groups analysis is mostly theresult of Canadians expecting positive events to happen to themselves morethan the Japanese did.With respect to negative events, a significant interaction emerged. TheJapanese believe that negative events are marginally more likely to happen tothemselves than do Canadians, whereas they believe that negative events aresignificantly less likely to happen to others than do Canadians. Thus, comparedto the Japanese, there is a marginal tendency for the Canadian self-estimates tobe more favorable to themselves, whereas their estimates for others aresignificantly less favorable.Control, Availability of Stereotypes, and Self-stereotypesRespondents indicated on a 5-point scale the extent to which they felt thateach event was under their control. Corroborating the theoretical suggestion ofWeisz et al. (1984), Canadians viewed 14 of the 15 events to be significantlymore under their control than did the Japanese. The Japanese reportedsignificantly more control for 1 of the 15 events: "You will become senile with oldage." The aggregated items show that Canadians demonstrated significantlymore control than the Japanese across all aggregates (see Table 6).• ••••• ••• ******** •••••••••••••••••••••••Insert Table 6 about here• • ***** • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • •Replicating Weinstein (1980), control was significantly correlated with unrealisticoptimism for negative events for both Japanese and Canadians (within-respondent rs = -0.35, and -0.30, respectively, p < .01 for each), although to alesser degree than he had found (r = -0.67).Cultural Variation in Unrealistic Optimism32The availability of stereotypes was measured by asking respondents toindicate on a 3-point scale the extent to which they could imagine a typicalperson likely to experience the event. Canadians were significantly more able toimagine the type of person likely to experience the events for 13 of the 15events. Consequently, Canadians showed a significantly greater tendency toimagine stereotypical people across all aggregates (see Table 6). In contrast tothe highly significant correlation between stereotype availability and optimismbias for negative events (r = -0.76) found by Weinstein (1980), the within-respondent correlations did not reach significance for either of the culturalgroups in this study (is = -0.11, and -0.04 for Japanese and Canadiansrespectively, p> .10 for each).2Self-stereotypes were measured by asking respondents to indicate on a5-point scale the extent to which they felt that they were the type of person likelyto experience a particular event. Canadians made significantly more self-stereotypes than the Japanese for each of the 5 positive events. The Japanesemade significantly more self-stereotypes than Canadians for 7 out of the 10negative events. Table 6 shows that Canadians demonstrated significantly moreself-stereotypes for positive events, while Japanese demonstrated significantlymore self-stereotypes for negative events. The within-respondent correlationbetween self-stereotypes and optimism bias was significant for both culturalgroups, and for both positive and negative events (rs = -0.66 and -0.62 forJapanese and Canadians respectively for positive items, and is = 0.68 and 0.75for negative items, p < .01 for each).2These figures were calculated by averaging the correlations obtained for each respondent.Weinstein's correlations were calculated by correlating the mean values for bias, control, andstereotype salience for each item. In the present study, there were only 10 negative events, asopposed to 24 events in Weinstein's study, so the method used by him would not be reliable forthis study. The correlations obtained using his method for the Canadians are -0.50, and 0.07 forcontrol and stereotype salience respectively).Cultural Variation in Unrealistic Optimism33Correlations between Desirability/Severity and Unrealistic Optimism,Stereotypes, Self-Stereotypes, and ControlA within-respondent's correlational design was used to correlate therankings of desirability and severity for the positive and negative items,respectively with unrealistic optimism, stereotype availability, tendencies to self-stereotype, and perceived controllability. The average within-respondents'correlations for each culture were compared.••••••••••••••••••••••••••••••••••••••••Insert Table 7 about here• • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • • •• •Both Japanese and Canadians show a significant within-respondentcorrelation between the ranking of desirability and an optimism bias (z = -0.53and -0.45 respectively, p < .01 for each; see Table 7). Thus, the more desirablethe event, the more likely respondents show an optimism bias. Thesecorrelations were not significantly different. However, given that there wasconsiderable consensus on the desirability rankings between respondents, andthat these correlations are based on only 5 items, it is possible that thecorrelations merely reflect unrealistically optimistic tendencies associated with.particular items, and thus one must be careful in drawing conclusions.The Japanese show no correlation (z = -.01, p >.10) between the rankingof severity and an optimism bias, while the Canadian correlation (z = -0.18, p <.01) is significant; the more severe the event is perceived by Canadians, themore likely they are to respond in an unrealistically optimistic manner. Thedifference between the correlations of the two cultures is also significant (F(1,269) = 8.22, p < .01), indicating that the magnitude of Canadians' optimism biasis tied to the perceived severity (or threat) of the event, while this is not true forthe Japanese.Cultural Variation in Unrealistic Optimism34In the case of stereotype availability, both Japanese and Canadiansdemonstrated a significant within-respondent correlation between desirabilityand the salience of a stereotype ( z = -.30 and -.40 respectively, p < .01 forboth). These correlations are not significantly different. The Japanese show asmall yet significant correlation between severity and stereotype availability (z =-.05, p < .05), while the Canadian correlation is also significant (z = -.22, p <.01). The Canadian correlation is significantly larger than the Japanese (F(1,260) = 10.99, p < .01), indicating that the more severe an event is perceived tobe by Canadians the more likely they are to report that they possess a clearimage of the type of person likely to experience it, whereas this tendency is lesspronounced for the Japanese.With respect to self-stereotypes, both Japanese and Canadiansdemonstrate a significant within-respondent correlation between desirability andthe tendency to see oneself as likely to experience the event (z = -.47 and -.39respectively, p < .01 for both). These correlations are not significantly different.The Japanese showed no correlation between perceived severity and self-stereotyping (z = .03, p> .10), while the Canadian correlation is significant (z =.23, p < .01). The difference between these correlations is also significant (F(1,272) = 10.58, p.< .01). Hence, the more severe an event is perceived to be byCanadians the less likely they are to report that they are the type of person likelyto experience it, whereas this tendency is absent for the Japanese.Lastly, with regards to the correlations with control, both Japanese andCanadians show a significant correlation between desirability and the perceivedcontrollability of the event (both Japanese and Canadian zs = -0.32, p < .01 forboth). The Japanese showed a small yet significant correlation betweenperceived severity and controllability (z = -.08, p < .01), while the Canadiancorrelation is also significant (z = -.33, p < .001). The Canadian correlation isCultural Variation in Unrealistic Optimism35significantly larger than the Japanese correlation (F(1, 274) = 24.68, p < .001)indicating that both cultural groups, but especially the Canadians, will state thatthey have more control for particularly threatening events.Cultural Differences on the Remaining ScalesRespondents also completed the Scheier and Carver (1985) LifeOrientation Test, Rosenberg's (1965) Self-Esteem scale, Campbell et al.'s(1992) Self-Concept Confusion scale, and Levenson's (1972) IPC control scale.••••••••••••••••••••••••••••••••••••••••Insert Table 8 about here•••• •• • ***** ••• •• • •• • • ***** • ••••••••••••Canadians scored significantly higher on the dispositional optimism, self-esteem,and internal locus of control scales than did the Japanese. The Japanesescored significantly higher on the self-concept confusion, external and luckcontrol scales than did Canadians. These differences were robust for each ofthe scales (see Table 8). Table 9 shows the correlations of these scales witheach other and with the total within-groups optimism bias.••••••••••••••••••••••••••••••••••••••••Insert Table 9 about here••••••••••••••••••••••••••• •••••••••••••DiscussionUnrealistic OptimismThe results of this study support the hypothesis that unrealistic optimismoccurs to a greater extent with Canadian students than it does with Japanese.However, Japanese respondents did show a significant optimism bias inlocalized areas. As the results for the within-group analyses of the individualitems in Table 1 indicate, the unrealistic optimism effect documented byCultural Variation in Unrealistic Optimism36Weinstein in his series of studies with American respondents (1980, 1982, 1984,1987) was replicated with Canadian respondents. One single item, "Sometimein the future you will develop skin cancer" failed to replicate with the Canadiansample. This is perhaps a reflection of the growing concern of Canadians, asfrequently reported in the media, that everyone is potentially at risk for thisdisease. The fact that this item had the highest mean for Canadians for thestatement "I am the type of person likely to experience this event," furthersupports this reasoning. Overall, then, the within-group results show aconsistent and reliable unrealistic optimism bias for Canadians.The between-groups analyses similarly confirm a strong optimism bias forthe Canadians. The lower power of this test makes it difficult to assess whetherthe optimism bias is different for Canadians in between-group analyses than it isfor the standard within-group analyses, yet as the aggregated item analyses inTable 2 show, it too is a highly reliable effect. The unrealistic optimism bias,then, does not require direct comparisons of self versus other to elicit an effect.Canadian respondents do not simply operate with an unrealistically optimisticsocial comparison heuristic that dictates that one's future is relatively better thanthat of a given comparison other. Table 5 shows that their absolute likelihoodestimates are similarly formatted to fit an unrealistically optimistic template.Thus, Canadians seem to view the world as a place where good things are likelyto happen to them, and bad things will most likely not, while a similar optimism isnot as strong for the fate of their peers.The demonstration of an unrealistic optimism bias for the Japanese wasnot as unambiguous. On the one hand, in the within-groups paradigm, 12 of the15 items showed a significant optimism bias for the Japanese. However, anexamination of the aggregated items reveals that the Japanese only showed anoptimism bias in the case of negative events. They demonstrated virtually noCultural Variation in Unrealistic Optimism37bias for positive events. Conversely, in the between-groups comparison thereverse pattern emerged. Only 3 of the 15 items showed a significant optimismbias, and the aggregated analyses reveal that in the between-groups paradigm itis the positive items, not the negative items, that demonstrate a significantoptimism bias3. Moreover, even the optimism bias for the positive aggregate islargely the result of one extreme item, "Enjoy your career," and is thus possiblyunreliable. The overall similarities between the self- and other-estimates for theJapanese show that they are not viewing the world in a way that makes themconsistently appear better than average. This indirect comparison reveals thatthe Japanese think of themselves, for the most part, to be about average.In contrast to Markus and Kitayama's (1991b) study of false uniqueness,and Takata's (1992) study of success orientation, then, it must be noted that theJapanese did demonstrate some self-enhancing tendencies in that theyexhibited significant optimism biases in certain conditions. Nevertheless, theinconsistent pattern of the Japanese optimism ratings suggests that theirestimates were not indiscriminately influenced by a reflexive self-enhancingtendency, as those of the Canadians appear to be, but rather, that in specificsituations, unrealistically optimistic judgments are made. These situations arenot ones in which the events are particularly threatening (see Table 7), nor arethey consistently positive or consistently negative events. And they are notconsistent within relative or absolute likelihood estimates either. The erraticpattern of the Japanese optimism. bias suggests that although they can act in aself-enhancing manner, they do so only in certain situations, the parameters ofwhich remain unclear.3The Canadians demonstrated a similar pattern to this reversal: namely, in the within-groupsdesign they show more bias for negative than for positive events, but for the between-groupsdesign they show more bias for the positive events. This is apparently because the base ratesfor the negative events are lower such that the absolute differences between self- and other-estimates for the negative events are smaller.Cultural Variation in Unrealistic Optimism38Cultural Differences in Unrealistic OptimismThe main hypothesis of this study, that the Japanese would show lessunrealistic optimism than Canadians, was strongly supported. In 12 out of the15 within-group comparisons the Canadians exhibited significantly more of anoptimism bias. In no case did the Japanese demonstrate a significantly strongerbias. There was a highly significant difference for the aggregated items, both forpositively and negatively valenced items. With respect to the between-groupcomparisons, while only 5 of the 15 items showed a significantly greateroptimism bias for Canadians, in 9 out of the remaining 10 comparisons, themean of the Canadian optimism bias was nonsignificantly higher than that of theJapanese. In sum, whether the items were positive or negative, and whether thedesign was within- or between-groups, Canadians were significantly moreunrealistically optimistic than Japanese.Control, Stereotypes, and Self-StereotypesWe found that the lower optimism bias in the Japanese sample wasindeed associated with a lower internal locus of control. The Japanesedemonstrated a significantly lower degree of internal locus of control, andlikewise, a higher degree of external locus of control, both with respect topowerful others and to luck. An internal locus of control was significantlycorrelated with the total optimism bias for both cultural groups (rs = -.26 and -.34respectively, p < .001 for each), however, neither of the two external locus ofcontrol measures demonstrated a significant correlation with the total optimismbias for either Japanese or Canadians. As well, in 14 out of the 15 optimismitems, the Canadians indicated that they felt that the events were significantlymore under their control than did the Japanese, and this measure of control wasalso significantly correlated with unrealistic optimism.Cultural Variation in Unrealistic Optimism39We argued that the Japanese should be less able to imaginestereotypical people associated with future life events because the other-oriented nature of the interdependent construal of self means that they shouldpossess a more refined conception of others, and thus not be as likely to imposestereotypes over this detailed image (Nisbett et al. 1983, Study 2; Quattrone &Jones, 1980). Conversely, we argued that since the attention of the Japaneseshould be more focused toward others, they should have a relativelyimpoverished self-conception, thus making the employment of a self-stereotypemore likely for them, compared to Canadians. The hypothesis that Canadianswould be more likely to imagine stereotypical people associated with particularevents was strongly supported by the data. Canadians were significantly morelikely than the Japanese to hold stereotypes in 13 out of the 15 events. Thisresulted in a highly significant difference in stereotype salience compared withthe Japanese, across both positive and negative aggregates.Interestingly, then, whether an event is positive or negative the Japanesereport that they are less able to imagine a typical person likely to experience it.This is consistent with the notion that the Japanese have a surplus ofinformation about others compared to Canadians, and that this reduces theirtendency to employ stereotypes. The Japanese apparently are relativelyresistant to the imposition of simplifying stereotypes over the elaborate imagesthat they possess of others. This is in line with Markus and Kitayama's (1991a)portrayal of the interdependent self which suggests that the Japanese do indeedfocus more attention toward others. It is also in line with the findings ofKitayama et al. (1990) who found that Hindu Indians appear to have a greatercognitive awareness of others, compared to Americans.Regarding self-stereotypes, the results were not as straightforward. TheJapanese had significantly higher self-concept confusion scores, indicating aCultural Variation in Unrealistic Optimism40less clear and consistent self-conception. Although, compared to Canadians,the Japanese were much more likely to endorse self-stereotypes for the negativeitems, they were much less likely to endorse self-stereotypes for the positiveitems. This measure of self-stereotypes thus seems confounded with thevalence of the items, and might more appropriately be seen as measuring self-enhancing tendencies. To agree with the statement that you are the type ofperson who is likely to experience a positive event, and to disagree that you arethe type of person likely to experience a negative event, is in itself a self-enhancing way of thinking. The strong correlations between self-stereotypesand the optimism bias are evidence of this relation. Given the argument that theJapanese should be less likely to self-enhance than Canadians, the results arehighly consistent. It does not appear possible to measure the tendency to holdself-stereotypes without introducing the confound of self-enhancement in themanner in which the question was framed in the present study.These results indicate that for positive events, Canadians are better ableto imagine a stereotypical person likely to experience that event, and they alsofeel that they are a person similar to that stereotype. Conversely, for negativeevents, although Japanese are less able to imagine a stereotypical person likelyto experience the event, they still are more likely to state that they are similar tothat stereotypical person.Cultural Differences in the Optimism Bias for Positive vs. Negative EventsIn Weinstein's studies, both control and stereotype salience were highlycorrelated with the tendency to be unrealistically optimistic for negative futurelife events. Because the Japanese were predicted to have a moderated senseof internal control, and of stereotype availability, it was expected that they wouldbe particularly less optimistic than Canadians with regards to negative events.In fact, the comparison of the magnitude of the difference in optimism biasCultural Variation in Unrealistic Optimism41between positive and negative items is nót straightforward. In the within-groupsanalysis, the Japanese demonstrated no optimism bias for positive events,whereas they showed a substantial optimism bias for negative events, thuscontradicting the above prediction. In the between-groups analysis, however,the Japanese demonstrated a significant, albeit unreliable, between-groupsoptimism bias for positive events, but no optimism bias for negative events. Thispattern contrasts with the Canadians' highly significant optimism bias in allareas, thus lending some support to the above prediction. There does notappear to be a consistent pattern, however, of the optimism bias differencebetween positive and negative events.One potential explanation for why the optimism bias between the twocultures was not consistently larger for the negative events as predicted involvesthe obtained correlations between optimism bias and control, and betweenoptimism bias and stereotype salience, for negative events. Because weexpected the Japanese would have lower control and stereotype salience scoresthan Canadians, we anticipated that their optimism bias would be particularlylower than Canadians in the case of negative events, given the substantialrelation that Weinstein (1980) had obtained for these events. In the presentstudy there was no relation between stereotype availability and unrealisticoptimism for negative events for either cultural group, and the relations betweencontrol and unrealistic optimism for negative events were much smaller thanthose found by Weinstein. It is thus not surprising that the significant differencesbetween Canadians and Japanese on control and stereotype salience were notassociated with greater differences between the cultures on negative events ascompared to positive events.Unrealistic Optimism as a Form of Coping with ThreatCultural Variation in Unrealistic Optimism42This study confirms the prevalence of an unrealistically optimistic way ofthinking, especially for North Americans, and to a limited extent for Japanese aswell. Taylor and Brown (1988) argue that the optimism bias is adaptive, and thatit aids the ability to cope effectively with stress (see also Taylor, 1989).Unrealistic optimism has been linked to effective coping, in part, because thebias is correlated with the degree of threat of the event (see Kirscht et al., 1966).Taylor et al. (1992) state that "while illusions of invulnerability may be generallyadaptive and protect people from the minor negative experiences of daily life,illusions may become especially important and exaggerated in people facingsevere threats as a method of dealing with the threat" (pp. 469- 470). Theoptimism bias has thus been argued to be a defense mechanism -- being able toimagine that one's future is better than the average other means that one will notbe struck by the same calamities as the average other. In the present study, asignificant correlation was obtained for Canadians between the ranked severityof a negative event and its corresponding optimism bias. This providesevidence that the optimism bias is a means of coping with threat. In the case ofthe Japanese, however, the correlation was virtually nil. Regardless of theperceived severity of the event, the Japanese were equally likely to show, or notshow, an optimism bias. In hindsight, it would have been preferable to measureperceived severity on an absolute, as opposed to a ranking, scale, so that thecomparisons between cultures could have been based on the same standard.Nevertheless, the Japanese did not show any correlation between rankedseverity and bias, suggesting that the optimism bias is not activated by threat forthem. It appears that unrealistic optimism does not serve as a defensemechanism for the Japanese.The correlations between severity and stereotype availability, self-stereotyping, and control provide additional support to this possibility. AsCultural Variation in Unrealistic Optimism43perceived threat increased, Canadians were more likely than the Japanese toimagine stereotypical people associated with future negative life events. Threatis apparently a sufficient motivator for Canadians to conjure up images ofvulnerable others -- others who are distinctly different from themselves, as theirtendency to avoid self-stereotyping for negative events indicates. This tendencywas significantly less pronounced for the Japanese, suggesting that theJapanese are not as motivated to seek vulnerable others in the face of threat.Similarly, when perceived threat increased Canadians were less likely toreport that they felt that they were the type of person likely to experience futurenegative life events. This tendency was also absent for the Japanese. Last, asperceived threat increased, Canadians were more likely than the Japanese tostate that the event was under their control. Canadians are thus more motivatedthan the Japanese to believe that they have control over threatening events.Perhaps by possessing "illusions of control" over threatening events one is ableto dispel the anxiety that one is potentially vulnerable to those events. Given themore external locus of control of the Japanese, it follows that they are not aslikely as the Canadians to report that these threatening events are under theircontrol.Hence, the above correlations indicate that the Japanese do not respondto threat in the same manner as Canadians. They do not tend to view thesethreatening events as things that are highly controllable and only happen tostereotypical victims. Apparently, in contrast to Canadians, threat does notinduce the Japanese to engage in a defensive self-protective way of thinking.Self-Enhancement and Construals of SelfSelf-enhancing comparisons are the norm for individuals with anindependent construal of self, especially when their self-esteem is threatened(Wills, 1987). They provide the self-flattering information necessary to bolster aCultural Variation in Unrealistic Optimism44sagging ego. However, for those with an interdependent construal of self, thethreats to self-esteem are likely to be of a different nature. Threats to the selfmay often suggest that the individual is different in ways that are preventing himor her from intertwining with the social fabric of the group. Comparisons thatexacerbate this difference could hardly be of much help in restoring a sense ofbelonging. The interdependent self is expected instead to seek confirminginstances of his or her belongingness with the group. Thus information thatsuggests that one is average, as opposed to "better than average," is likely to domore to support self-esteem.The results of the present study are in support of this argument.Unrealistic optimism did not increase in the face of threat for the Japanese, as itdid for the Canadians. It appears to be unrelated to self-protection for theJapanese. Perhaps these types of threat do not jeopardize the integrity of theinterdependent self in the same manner that they do for the independent self. Athreat to the individual amounts to a threat to the self, for people in independentcultures. It might be the case for Japanese, that a threat to the individual, doesnot threaten the interdependent network sustaining the Japanese self, and thusself-protective measures are not invoked. We might find that the Japaneseengage in self-protective measures instead when their sense of belongingnessto the group is threatened. Further research that explores the kinds of eventsthat are most threatening to the interdependent self, and their reactions to thatthreat, is necessary to resolve this issue.Unrealistic Optimism and Positive Construals of the SelfPast research has demonstrated that positive construals of the self andself-esteem are associated with both unrealistic and dispositional optimism(Alloy & Ahrens, 1987; Scheier & Carver, 1985). Similarly, in the present studysignificant correlations were obtained between self-esteem and both unrealisticCultural Variation in Unrealistic Optimism45optimism and dispositional optimism for both cultures (Japanese is = 0.53, and0.40, Canadian is = 0.77, and 0.46, for correlations between self-esteem and theLife Orientation Test, and between self-esteem and the total within-groupsoptimism bias, respectively). These relations are reflected in the significantlyhigher self-esteem, dispositional optimism, and unrealistic optimism scores ofthe Canadians compared to the Japanese The Japanese do appear to have aless positive construal of their self than that of the Canadians. The Japaneseare apparently more likely to admit to negative information about themselves.Self-esteem, or more specifically, what the Rosenberg measure assumesreflects self-esteem, appears to flourish in a Canadian cultural environment,whereas its growth is relatively hampered in Japanese culture. A positiveconstruel of self, then, seems representative of cultures characteristic ofindependent selves.In addition, Canadians felt that each of the positive future life events weremore likely to happen to themselves than did the Japanese. The aggregate totalof the positive events shows that the Canadians also felt that these positiveevents were more likely to happen to others than did the Japanese. Thedifferent base rates of the occurrences of these events in the two culturesrequires us to interpret these differences with caution, but base rates alonecannot account for the resultant distribution. For example, the Japanese havethe longest life expectancy in the world, and on average outlive Canadians bymore than 2 years (United Nations, 1991), yet for the item "You will live past 80years old," the Canadians estimated a greater likelihood of this event happeningboth for themselves and for others. Hence, apparently the drama of subjectiveCanadian life is, to a greater extent than the Japanese, a series of positiveevents for themselves, and even for others.Cultural Variation in Unrealistic Optimism46These results provide further empirical support that the interdependentconstrual of self is not as likely to pursue positive information, or shun negativeinformation, as is the independent construal of self. Kitayama (1993) arguesthat the experience of good feelings, i.e. the kind associated with optimism, isakin to identifying positive features of the self, and is thus highly self-affirmingfor those with independent selves. Western culture encourages people to seekpositive aspects of their identity, to establish their worth as individuals, andthereby accomplish the cultural task of independence. In contrast, the demandsfor the interdependent construal of self are to be as harmonious a member of thein-group as possible, and thus people are encouraged to become particularlysensitive to information that suggests that they as individuals are interfering withthe integrity of the group. Negative information about the self is highlyinstrumental in allowing the interdependent individual to correct his or herdeficiencies, thereby deepening relations with others, and achieve the culturalimperative of interdependence (Kitayama & Markus, 1992).In one sense, the results of this study provide further support to Taylorand Brown's (1988) contention that positive illusions are associated withpsychological well-being. The Japanese demonstrated less unrealistic optimismas well as a less positive construal of self. Insofar as a positive construal of selfis associated with "well-being ," this may be evidence for the relation betweenillusions and well-being at the cultural, as opposed to the individual, level.However, it is extremely difficult to accept the notion that Western culturehas a monopoly on psychological well-being. A more reasonable conclusion isthat the measures of "well-being" used in this and other studies are based onWestern conceptions of mental health and do not generalize well to Easterncultures. The cultural differences in self-enhancing tendencies and positiveconstruals of self that were found in this study suggest a different way toCultural Variation in Unrealistic Optimism47interpret Taylor and Brown's argument. Perhaps the components of "well-being"(e.g., happiness, positive affect), discussed by Taylor and Brown, and others,are achieved when the individual satisfies the cultural standards of selfhood .For people with an independent construal of self, realization of the cultural idealrequires that one believes that he or she is competent as an individual. Withoutany objective standards of competence, social comparison theory suggests thatwe determine our worth by sizing ourselves up to others (Festinger, 1954).Hence, believing that one is better than the average other (or in the case of thisstudy, believing that one's future is better than the average other), is tantamountto believing that one has self-worth in an independent culture. Self-enhancingbiases might thus be seen as the necessary tools to construct the sense of selfvalued by Western culture, thereby achieving well-being.However, for the interdependent self, well-being is not based on feelingsof individual competence, but on feelings of belongingness (Kitayama et al.,1991). Since well-being is not as tied to individual competence, it is also lessbound to self-enhancing assessments of one's competence. Therefore,although the pattern of results in this study shows self-enhancing tendenciesand measures of well-being going hand in hand for both cultural groups, thismay be due to the fact that our definition of "well-being" was derived in aWestern context. If we used measures of "well-being" based on feelings ofbelongingness we might be able to show that it is self-effacement, and not self-enhancement, that is critical to "mental health." Further research on the relationbetween well-being and positive illusions, employing different cultural definitionsof "well-being," will prove fruitful.General Conclusions and Future DirectionsThe hypotheses of the present study, as formulated by contrasting theunrealistic optimism literature with the cross-cultural literature, were, for the mostCultural Variation in Unrealistic Optimism48part, supported by the results. The Japanese were less unrealistically optimisticthan Canadians, and this was associated with lower self-esteem, lowerdispositional optimism, a more external locus of control, and less of an ability toassociate stereotypical people with optimism events. Unrealistic optimism hasthus been shown to be, similar to other self-enhancing biases researched thusfar, influenced by the cultural environment of the respondent. Cultures thatfoster an interdependent construal of self (e.g., Japan) greatly curtail themanifestation of an optimism bias. In addition, while Canadians demonstrated asignificant tendency to be more unrealistically optimistic in the face ofparticularly threatening events, the Japanese were unaffected. Thus, whileunrealistic optimism appears to be a defense mechanism for the Canadians, itapparently does not serve this purpose for the Japanese.Given our questionnaire format, a possible confound between culture andresponse set exists. It is possible that a modesty tendency, or a tendency toendorse responses towards the centre of a scale on the part of the Japanesecould be responsible for the significant differences in unrealistic optimism.However, there are a number of indicators that cast doubt on this interpretation.First is the absence of any reliable correlations for the Japanese samplebetween threat and optimism bias, stereotypes, self-stereotypes, and control. Ifthe obtained differences in unrealistic optimism were solely the result ofresponse sets, then we would expect that their optimism judgments wouldparallel those of the Canadians, but at a lower level. However, the Canadiansexhibited an increase in unrealistic optimism, availability of stereotypes, andperceived control, along with a decrease in self-stereotypes, as perceived threatincreased, but this was not the case for the Japanese. In addition, theconstructs associated with unrealistic optimism, i.e., internal locus of control,availability of stereotypes, and positive construals of self, were all lessCultural Variation in Unrealistic Optimism49pronounced for the Japanese. This pattern would also not be expected ifresonse sets were the only factors behind the obtained differences in unrealisticoptimism scores. Last, behavioral experiments, such as Takata's (1992) work onsocial comparison, show that the Japanese reluctance to self-enhance is notlimited to questionnaire data, but extends to the behavioral realm as well.Further research that demonstrates behavioral differences between cultures isnecessary to dispel arguments that differences in cross-cultural questionnairestudies are merely the result of cultural differences in response sets.On the basis of the present findings, future research on the presence ofunrealistically optimistic tendencies in the Japanese might be improved bydirecting attention to five specific issues: First, a more extensive list of positiveand negative items should be employed to spotlight any types of events thatmight consistently show a particularly pronounced bias. This list should begenerated by Japanese respondents to ensure that the events are all ofconsiderable importance to them. Second, an absolute measure of theperceived desirability and severity (or threat) of the events should be employedto avoid correlations based on a forced distribution of rankings. This, inconjunction with an expanded list of events, would result in more reliablemeasures of the relation between threat and optimism bias. Third, it isnecessary to explore the types of events that are perceived to be threatening bythose with interdependent construals of self. Once this is established, the self-protective measures employed by the interdependent self could be examinedand compared to the self-protective measures (e.g., self-enhancement), used byNorth Americans. Fourth, the presence or absence of other self-enhancingbiases not yet explored with the Japanese (e.g., false-consensus effect,hindsight bias) should be established, to determine the circumstances underwhich the Japanese do or do not self-enhance. And fifth, it is crucial to exploreCultural Variation in Unrealistic Optimism50other cultural notions of well-being, and their relation to self-enhancement, orself-effacement.Cultural Variation in Unrealistic Optimism51ReferencesAlicke, M. D. (1985). Global self-evaluation as determined by the desirability andcontrollability of trait adjectives. Journal of Personality and SocialPsychology, 49, 1621-1630.Alloy, L. B., & Ahrens, A. H. (1987). Depression and pessimism for the future:Biased use of statistically relevant information in predictions for selfversus others. Journal of Personality and Social Psychology, 52, 366-378.Barnlund, D. C., & Yoshioka, M. (1990). Apologies: Japanese and Americanstyles. International Journal of Intercultural Relations, 14, 193-206.Benedict, R. (1946). The Chrysanthemum and the Sword. Boston: HoughtonMiller.Bond, M. H., & Tornatzky, L. G. (1973). Locus of control in students from Japanand the United States: Dimensions and levels of response. Psychologia,16, 209-213.Brickman, P., Coates, D., & Janoff-Bulman, R. (1978). Lottery winners andaccident victims: Is happiness relative? Journal of Personality and SocialPsychology, 35, 917-927.Campbell, J. D., Trapnell, P. D., Katz, I., & Lavallee, L. (1992). Personality andself-knowledge: Development of the self-concept confusion scale andexamination of its personality correlates. Unpublished manuscript,University of British Columbia.Chandler, T. A., Shama, D. D., Wolf, F. M., & Planchard, S. K. (1981).Multiattributional causality: A five cross-national samples study. Journal ofCross-Cultural Psychology, 12, 207-221.Cultural Variation in Unrealistic Optimism52Cheek, J. M. (1982). The Aspects of Identity Questionnaire: Revised scalesassessing personal and social identity. Unpublished manuscript,Wellesley College.Cousins, S. D. (1989). Culture and selfhood in Japan and the U.S. Journal ofPersonality and Social Psychology, 56, 124 -131.Cousins, S. D. (1990). Culture and social phobia in Japan and the United States.Unpublished doctoral dissertation, University of Michigan.Crary, W. G. (1966). Reactions to incongruent self-experiences. Journal ofConsulting Psychology, 30, 246 -252.Festinger, L. (1954). A theory of social comparison processes. Human Relations,7, 117-140.Freedman, J. (1978). Happy people: What happiness is, who has it, and why.New York: Harcourt Brace Jovanovich.Gonzales, A., & Zimbardo, P. G. (1985, March). Time in perspective. PsychologyToday, pp. 21-26.Greenberg, J., Pyszczynski, T., & Solomon, S. (1982). The self-servingattributional bias: Beyond self-presentation. Journal of ExperimentalSocial Psychology, 18, 56 -67.Greenwald, A. G. (1980). The totalitarian ego: Fabrication and revision ofpersonal history. American Psychologist,35, 603-618.Hamaguchi, E. (1985). A contextual model of the Japanese: Toward amethodological innovation in Japan studies. Journal of JapaneseStudies, 11, 289 -321.Hamilton, V. L., Blumenfeld, P C Akoh, H. & Miura, K. (1990). Credit andblame among American and Japanese children: Normative, cultural, andindividual differences. Journal of Personality and Social Psychology, 59,442-451.Cultural Variation in Unrealistic Optimism53Heine, S. J., Lehman, D. R., Okugawa, 0., & Campbell, J. D. (1992).The effectsof culture on self-implicated processes: A comparison of Canadians andJapanese. Ritsumeikan Social Sciences Review, 28, 3, 29 -38.Hofstede, G. (1980). Culture's consequences: International differences in work-related values. Beverly Hills, CA: Sage.Isen, A. M.(1984). Toward understanding the role of affect in cognition. In R.Wyer & T. Srull (Eds.), Handbook of social cognition (pp. 174-236).Hillsdale, NJ: Erlbaum.Jahoda, M. (1958). Current concepts of positive mental health. New York: BasicBooks.Jourard, S. M., & Landsman, T. (1980). Healthy personality: An approach fromthe viewpoint of humanistic psychology (4th ed.). New York: Macmillan.Kahneman, D., & Tversky, A. (1972). Subjective probability: A judgement ofrepresentativeness. Cognitive Psychology, 3, 430-454.Kashima, Y., & Triandis, H. C. (1986). The self-serving bias in attributions as acoping strategy: A cross-cultural study. Journal of Cross -CulturalPsychology, 17, 83-97.Kirscht, J. F., Haefner, D. P., Kegeles, S. S., & Rosenstock, I. M. (1966). Anational study of health beliefs. Journal of Health and Human Behavior, 7,248-254.Kitayama, S. (1993). Sociocultural inflluences in basic psychological processes.NIMH Behavioral Science Task Force, Socio-Cultural InfluencesSubcommittee, Position Paper.Kitayama, S., & Markus, H. R. (1992). Construal of the Self as Cultural Frame:Implications for Internationalizing Psychology. Paper presented at theSymposium for Internationalization and Higher Education.Kitayama, S., Markus, H.R., & Kurokawa, M. (1991). Unpublished manuscript.Cultural Variation in Unrealistic Optimism54Kitayama, S., Markus, Takagi, Sugiman, & Matsumoto. (1992). Unpublisheddata.Kitayama, S., Markus, H., Tummala, P., Kurokawa, M., & Kato, K. (1990). Cultureand self-cognition. Unpublished manuscript.Kojima, H. (1984). A significant stride toward the comparative study of control.American Psychologist, 39, 972-973.Kunda, Z. (1987). Motivated inference: Self-serving generation and evaluation ofcausal theories. Journal of Personality and Social Psychology, 53, 636-647.Langer, E. (1975). The illusion of control. Journal of Personality and SocialPsychology, 32, 311-328.Lebra, T. S. (1976). Japanese patterns of behavior. Honolulu: University ofHawaii Press.Levenson, H. (1972). Distinctions within the concept of internal-external control:Development of a new scale. Proceedings of the 80th Annual A.P.A.Convention, 7, 259-260. (summary)Lewinsohn, P. M., Mischel, W., Chaplin, W., & Barton, R. (1980). Socialcompetence and depression. The role of illusory self-perceptions. Journalof Abnormal Psychology, 89, 203-212.Mahler, I. (1974). A comparative study of locus of control. Psychologia, 17, 135-139.Markus, H., & Kitayama, S. (1991a). Culture and the self: Implications forcognition, emotion, and motivation. Psychological Review, 98, 224-253.Markus, H., & Kitayama, S. (1991b). Cultural variation in the self-concept. InG.R. Goethals & J. Strauss (Eds.), Multidisciplinary perspectives on theself. New York: Springer-Verlag.Myers, D. (1987). Social psychology (2nd ed.). New York: McGraw-Hill.Cultural Variation in Unrealistic Optimism55Myers, ID. G., & Ridl, J. (1979). Can we all be better than average? PsychologyToday,13, 89-98.Nakane, C. (1970). Japanese society. Los Angeles: University of CaliforniaPress.Nisbett, R. E., Krantz, D. H., Jepson, C., & Kunda, Z. (1983). Study 2: The use ofstatistical heuristics in everyday inductive reasoning. PsychologicalReview, 90, 339-363.Perloff, L. S. (1983). Perceptions of vulnerability to victimization. Journal ofSocial Issues, 39, 41-61.Perloff, L. S. (1987). Social comparison and illusions of invulnerability tonegative life events. In C. R. Snyder & C. E. Ford (Eds.), Coping withnegative life events: Clinical and social psychological perspectives. NewYork: Plenum.Perloff, L. S., & Fetzer, B. K. (1986). Self-other judgements and perceivedvulnerability to victimization. Journal of Personality and SocialPsychology,50, 502-510.Quattrone, G. A., Jones, E. E. (1980). The perception of variability within in-groups and out-groups: Implications for the law of small numbers. Journalof Personality and Social Psychology, 38, 141-152.Reischauer, E. 0. (1988). The Japanese Today: Change and Continuity.Harvard: Belknap.Rosenberg, M. (1965) Society and the adolescent self-image. Princeton, NJ:Princeton University Press.Scheier, M. F., & Carver, C. S. (1985). Optimism, coping, and health:Assessment and implications of generalized outcome expectancies.Health Psychology, 4, 219-247.Cultural Variation in Unrealistic Optimism56Scheier, M. F., & Carver, C. S. (1987). Dispositional optimism and physical well-being: The influence of generalized outcome expectancies on health.Journal of Personality, 55, 169-209.Schwartz, J. M., & Smith, W. P. (1976). Social comparison and the inference ofability difference. Journal of Personality and Social Psychology, 34, 1268-1275.Slavic, P., Fischhoff, B., & Lichtenstein, D. (1978). Accident probabilities andseat belt usage: A psychological perspective. Accident Analysis andPrevention, 10, 281-285.Snyder, C. R. (1978). The "illusion" of uniqueness. Journal of HumanisticPsychology, 18, 33-41.Svenson, 0. (1981). Are we all less risky and more skillful than our fellowdrivers? Acta Psychologica, 47, 143-148.Takata, T. (1992, June). Self-Deprecative Social Comparison in Japan. Paperpresented at the Conference on Emotion and Culture, Eugene, OR.Taylor, S. E. (1989). Positive illusions: Creative self-deception and the healthymind. New York: Basic Books.Taylor, S. E., & Brown, J. D. (1988) Illusion and well-being: A socialpsychological perspective on mental health. Psychological Bulletin, 103,193-210.Taylor, S. E., Kemeny, M. E., Aspinwall, L. G., Schneider, S. G., Rodriguez, R.,& Herbert, M. (1992). Optimism, coping, psychological distress, and high-risk sexual behavior among men at risk for AIDS. Journal of Personalityand Social Psychology ,63, 460-473Taylor, S. E., Wood, J. V., & Lichtman, R. R. (1983). It could be worse: Selectiveevaluation as a response to victimization. Journal of Social Issues, 39, 19-40.Cultural Variation in Unrealistic Optimism57United Nations (1991). Life expectancy at birth. In United Nations DemographicYearbook. New York: United Nations.Weinstein, N. D. (1980). Unrealistic optimism about future life events. Journalof Personality and Social Psychology, 39, 806-820.Weinstein, N. D. (1982). Unrealistic optimism about susceptibility to healthproblems. Journal of Behavioral Medicine, 5, 441-460.Weinstein, N. D. (1984). Why it won't happen to me: Perceptions of risk factorsand susceptibility. Health Psychology, 3, 431-457.Weinstein, N. D. (1987). Unrealistic optimism about susceptibility to healthproblems: Conclusions from a community-wide sample. Journal ofBehavioral Medicine, 10, 481-500.Weisz, J. R., Rothbaum, F. M., & Blackburn, T. C. (1984). Standing out andstanding in: The psychology of control in America and Japan. AmericanPsychologist, 39, 955-969.Wills, T. A. (1981). Downward comparison principles in social psychology.Psychological Bulletin, 90, 245-271.Wills, T. A. (1987). Downward comparison as a coping mechanism. In C. R.Snyder & C. E. Ford (Eds.), Coping with negative life events: Clinical andsocial psychological perspectives. New York: Plenum.Wylie, R. C. (1979). The self-concept: Theory and research on selected topics.(Vol. 2). Lincoln, NB: University of Nebraska Press.Yamagishi, T. (1988). Exit from the group as an individualistic solution to the freerider problem in the United States and Japan. Journal of ExperimentalSocial Psychology, 24, 530-542.Yamauchi, H. (1990). Actor and observer attributions by Japanese subjects forsuccess and failure in non-competitive situations. Psychologia, 33, 212-219.Cultural Variation in Unrealistic Optimism58Zakay, D. (1984). The influence of perceived event's controllability on itssubjective occurrence probability. The Psychological Record, 34, 233-240.Cultural Variation in Unrealistic Optimism59Table 1T-Tests for Individual Optimism EventsItem JapaneseWithin^BetweenCanadiansWithin^Between1^Become an alcoholic 14.6** 0.92 9.56** 2.93**2p Enjoy Career 8.56** 3.87** 12.19** 5.75**3^Attempt suicide 13.11** 0.29 15.68** 1.524^Develop skin cancer 2.20* -1.86(P) -0.40(P) -0.47(P)5p Live past 80 -6.6** P -2.51*(P) 3.09** 0.906^Get divorced 6.39** -0.27(P) 11.12** 5.60**7p Own your own home 3.62** 2.20* 8.59** 5.88**8^Have nervous breakdown 5.48** 2.15* 6.06** 0.549^Get AIDS 11.2** 0.89 13.85** 3.12**10 Drop out of university 11.9** 0.04 14.02** 3.57**11p Leave company for better job -1.28(P) 1.63 6.33** 3.08**12p Starting salary > $30 000 -5.23**(P) 0.91 2.23* 2.74**13^Make family ashamed 11.27** -0.52(P) 8.88** 3.83**14^Have early heart attack 4.07** -0.24(P) 9.17** 0.3215^Become senile 2.33* 1.05 5.72** -0.58(P)Note: (P) - indicates that the item was responded to in a pessimistic manner* - p < .05** - p < .01Cultural Variation in Unrealistic Optimism60Table 2T-Tests for Aggregated EventsItem^Japanese^ CanadiansWithin^Between^Within^BetweenPositive (5)^0.223 2.11* 10.63**^5.57**Negative (10) 15.53**^-0.46(P)^18.28**^3.26**All (15)^13.94**^0.63 19.56**^5.89**Note: (P) - indicates that the items were responded to in a pessimistic manner.** - p < .01* - p < .05Cultural Variation in Unrealistic Optimism61Table 3Anova and Means per Future Life EventItem Japanese^CanadiansWithin^Between^Within Between1^Become an alcoholic -1.55 1.96 -1.56 9.522p Enjoy Career -0.82 -10.47 -1.27** -19.123^Attempt suicide -1.49 0.7 -2.07** 4.444^Develop skin cancer -0.16 -5.33(P) 0.06(P) -1.88(P)5p Live past 80 0.51(P1 6.78(P) -0.45** -4.44*6^Get divorced -0.66 -0.72(P) -1.52** 21.50**7p Own your own home -0.29 -6.50 -1.12** -21.69**8^Have nervous breakdown -0.59 -5.52(P) -0.99* 2.349^Get AIDS -0.92 2.37 -1.81** 7.6010 Drop out of university -1.36 0.10 -1.98** 13.42**11p Leave company for better job 0.13(P) -5.14 -0.75** -14.0112p Starting salary > $30 000 0.42(P) -2.96 -0.32** -13.6713^Make family ashamed -1.18 - 1.10(P) -1.29 14.86**14^Have early heart attack -0.34 -0.60(P) -1.29** 1.0515^Become senile -0.16 2.97 -0.76** -2.47(P)Note: The numbers in the within columns represent the average relativelikelihood estimates for each respondent on a (-3 to +3) 7-point scale. Negativenumbers represent an optimistic tendency. The numbers in the betweencolumns represent the mean percentage estimates of the average same-sexstudent from the respondents' university less the mean percentage estimates forself. Positive values represent optimistic tendencies for the negative events,and pessimistic tendencies for the positive events. Stars in any columnrepresent a significantly greater optimistic tendency than the other cultural groupfor that particular event.(P) - indicates that the items were responded to in a pessimistic manner.** - Anova between cultures is significant at p < .01* - Anova between cultures is Significant at p < .05Cultural Variation in Unrealistic Optimism62Table 4Anova for Amireqated Future Life Events and MeansItem^Japanese^CanadiansWithin^Between Within^BetweenPositive (5)^-0.01^-3.64^-0.77^-14.79Negative (10) -0.84^0.30(P)^-1.32^-7.23All (15)^-0.56^-0.65^-1.15^-9.43Note: (P) - indicates that the items were responded to in a pessimistic manner.Positive Within: F(1,279) = 93.588 , p < 00 1•^,^•^•Between: F(1,272) = 14.056, p < .001Negative Within: F(1,290) = 26.610, p < .001Between: F (1^7.728,^p < .01,275)=All Within: F(1,279) = 65.391,^p < .001Between: F(1,270) = 21.899, p < .001Cultural Variation in Unrealistic Optimism63Table 5Anova for Self/Other EstimatesItem^Japanese^CanadiansSelf^Other^Self^OtherPositive^47.07^43.46* 67.60^52.81**Negative 20.68^20.39^18.06^25.29*** (Within culture analysis) self # other, p < .05**(Within culture analysis) self # other, p < .01Between culture analyses:Positive:^Self:^F(1,136) = 110.538, p < .001Other:^F(1,133) = 17.214, p < .001Interaction: F(1,276) = 11.743, p < .001Negative:^Self:^F(1,136) = 1.8,^p >.10Other:^F(1,136) = 6.852,^p < .01Interaction: F(1,278) = 7.760,^p < .01Cultural Variation in Unrealistic Optimism64Table 6Anova and Means for Aqprepated Control, Stereotype, and Self-StereotypeItems Japan CanadaPositive Control 15.36 18.89 F(1,277) = 93.004, p < .001Negative Control 31.85 37.05 F(1,277) = 52.04,^p < .001Total Control 47.23 56.04 F(1,277) = 82.852, p < .001Positive Stereotype 9.19 11.74 F(1,275) = 91.182, p < .001Negative Stereotype 16.91 20.54 F(1,273) = 55.35,^p < .001All Stereotypes 26.08 32.30 F(1,272) = 84.218, p < .001Positive Self-Stereotype 14.76 19.42 F(1,276) = 267.32, p < .001Negative Self-Stereotype 22.65 19.87 F(1,277) = 20.935 , p < 001Note: The positive and negative self-stereotypes are not combined because theirvalences are in the opposite direction, and would thus confound anycomparisons.Cultural Variation in Unrealistic Optimism65Table 7Anova for Average Adiusted+ Correlations Between Desirability/Severity, Stereotypes,Self-Stereotypes, Control, and the Optimism Bias Positive EventsDesirability Stereotypes Self-Stereo. Control Opt. BiasDesirability^1.00 -0.30** -0.47** -0.32** -0.53**Stereotypes^-0.40** 1.00 0.51** 0.59** -0.34*Self-Stereo.^-0.39** 0.20 1.00 0.59** -0.62**Control^-0.32**Opt. Bias^-p. 45**0.68**-0.63*0.68**-0.62**1.00-0.43**-0.40**1.00Negative EventsSeverity Stereotypes Self-Stereo. Control Opt. BiasSeverity 1.00 -0.05* 0.03 -0.08** -0.01Stereotypes -0. 22 ** 1.00 -0.12** 0.40** -0.11*Self-Stereo. 0.23** -0.07 1.00 -0.39** 0.68**Control -0.33** 0.39** -0.40** 1.00 -0.35**Opt. Bias -0.18** -0.04 0.75** -0.30** 1.00Note: Correlations in the upper right triangle are for the Japanese sample, andcorrelations in the lower left triangle are for the Canadian sample.Underlined correlations are significantly different between cultures (p <.01).- Correlations were first converted to Fisher Z-scores to approximate anormal distribution.* - z#0,p<.05z#0,p<.01Cultural Variation in Unrealistic Optimism66Table 8Anova and Means for the Remaining ScalesScale Japan CanadaLife Orientation Test 24.64 28.00 F(1,281) = 34.9,^p < .001Rosenberg Self-Esteem 32.26 39.37 F(j,258) = 86.14,^p < .001Self-Concept Confusion 36.99 32.60 F(1,258) = 25.66,^p < .001IPC^- Internal Control 27.18 29.29 F(1,258) = 15.25,^p < .001IPC^- External Control 21.72 18.80 F(1,258) = 37.90, p < .001IPC^- Luck Control 23.95 19.36 F(1,258) = 84.77, p < .001Cultural Variation in Unrealistic Optimism67Table 9Correlations of Remaining Scales with the Optimism BiasOpt. Bias Self-Esteem SCC LOT Int. Control Ext. ControlOpt. Bias 1.00 -0.40** 0.38** -0.24** -0.26** 0.05Self-Esteem -0.46** 1.00 -0.46** 0.53** 0.29** -0.29**scc 0.52** -0.63** 1.00 -0.24** -0.23** 0.32**LOT -0.32** 0.77** -0.50** 1.00 0.20** -0.29**Int. Control -0.41** 0.40** -0.40** 0.36** 1.00 -0.12Ext. Control 0.41** -0.46** 0.47** -0.36** -0.23* 1.00Luck Control 0.34** -0.40** 0.44** -0.37** -0.44** 0.56**Note: Correlations in the upper right triangle are for the Japanese sample, andcorrelations in the lower left triangle are for the Canadian sample.Underlined correlations are significantly different between cultures (p <.06).r 0, p < .05r 0, p < .01


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items