| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||
RESEARCH ARTICLE |
1 School of Psychology, Georgia Institute of Psychology, Atlanta, Georgia.
2 Center for Family and Demographic Research, Bowling Green State University, Ohio.
3 Department of Psychology, Portland State University, Oregon.
Address correspondence to Fredda Blanchard-Fields, School of Psychology, Georgia Institute of Psychology, Atlanta, GA 30332-0170. E-mail: fb12{at}prism.gatech.edu
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Recent findings have shown that older adults display stronger dispositional biases than do young adults. For example, older adults are more likely to attribute the cause of a negative event outcome to an actor's traits than to influences related to the situational context (Blanchard-Fields, 1994
; Blanchard-Fields & Beatty, 2005). In the present study, we asked how pervasive age differences in social judgment biases are across cultures; that is, we asked whether cross-cultural differences appear when a life-span developmental perspective is taken.
Researchers often measure dispositional biases by using the attitude-attribution paradigm. After reading a short essay on a controversial topic, participants are told that the essay was written either by choice or as an assignment; they are then asked to infer the writer's true attitude about the topic. Even in the assigned condition, American participants exhibit a tendency to infer that the target's beliefs were consistent with those expressed in the essay; this is known as a correspondence bias (Gilbert & Malone, 1995; Jones & Harris, 1967).
Older Americans show the correspondence bias more so than young adults do (Blanchard-Fields & Horhota, 2005; Follett & Hess, 2002). Gilbert, Pelham and Krull (1988) suggest that dispositional attributions represent an early stage in attributional processing that is relatively automatic and effortless. Effortful processing is required to consider contextual information and correct one's initial judgment. Given that older adults may have fewer resources to draw from (Zacks, Hasher, & Li, 2000), it is possible that these declines may force older adults to rely on more easily accessible information, which in Western cultures is typically an initial dispositional response (Chen & Blanchard-Fields, 2000). Alternatively, we have found evidence that age differences can be eliminated when older adults engage an attitude-attribution paradigm in which situational information is made salient by an increase in the plausibility of the behavioral constraint (Blanchard-Fields & Horhota).
Cross-cultural research on the correspondence bias (Choi & Nisbett, 1998; Morris & Peng, 1994) suggests that people in collectivist (e.g., Asian) and individualistic (e.g., American) cultures differ in using situational information to make causal attributions. Asian young adults tend to be less dispositionally biased than their American counterparts because they tend to rely more on contextual factors when making judgments, rather than personal characteristics of the individual, particularly when situational information is salient (Choi & Nisbett; Masuda & Kitayama, 2004; Miyamoto & Kitayama, 2002). Individuals in collectivist cultures (e.g., Chinese) embrace the idea that nothing is isolated and everything is connected (i.e., the principle of relationship or holism). In this case, individuals cannot be responsible for their actions alone (Morris & Peng). Thus, situational or contextual information becomes a more salient causal factor than individualist or personal information. In contrast, people in individualistic cultures (e.g., Americans) more readily hold the individual responsible for his or her own action. Such cultural beliefs and values are typically learned and reinforced through an acculturation process (e.g., through mass media). We suggest that if it is a well socialized phenomenon that Asian cultures are more prone to reflect on such situational information, we would expect that Chinese individuals of all ages would be less susceptible to the correspondence bias than would Americans.
We propose two hypotheses. First, Americans will show a stronger correspondence bias than Chinese will. Second, we expect age-related differences in both samples; however, the American older adults will be more dispositionally biased whereas the Chinese older adults will be more situationally biased than their respective young counterparts.
|
METHODS |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
. Ninety-four young Americans (aged 18–25 years, M = 20.27, SD = 1.22) and 89 older Americans (aged 60–80 years, M = 68.76, SD = 4.45) participated. In that study, we recruited young adults from an undergraduate psychology student pool. Each young adult received one credit toward a course participation requirement. We recruited older adults from a participant pool; they received $15 remuneration. We found typical age differences in vocabulary by using the Advanced Vocabulary Test (Ekstrom, French, Harman, & Dermen, 1976), with older adults outperforming young adults on this task (p <.01; M = 23.45, SD = 6.57 for young adults; M = 16.87, SD = 5.26 for older adults).
Chinese sample
Eighty-nine young adults (aged 17–31 years, M = 21.25, SD = 2.30) and 84 older adults (aged 58–74 years, M = 65.65, SD = 2.85) participated. We recruited young adults from a university and older adults from a Chinese metropolitan area. All participants received ¥10 for participation. We used the Chinese version of the Revised Wechsler Adult Intelligence Scale (Cai, Chen, & Zhou, 1996) for the vocabulary test. Young adults outperformed older adults on this task (p <.01; M = 54.58, SD = 7.37 for young adults; M = 50.54, SD = 5.12 for older adults).
Materials
We had the materials translated into Chinese and backtranslated into English (Brislin, 1970).
Essays and Attitude Questionnaire
Participants received an essay that was either favorable or unfavorable toward capital punishment. Development of the stimuli has been discussed elsewhere (Blanchard-Fields & Horhota, 2005). In addition, it should be noted that capital punishment is a controversial issue in both cultures (Wang & Zhao, 2006). Participants rated the degree to which the writer of the essay was in favor of or opposed to capital punishment (1 = very much opposed, 7 = very much in favor) and were instructed to give a written explanation for their response.
Procedure
We used the same procedure for American and Chinese samples. After participants had completed the demographics form and vocabulary test, the experimenter handed out the essays. This ensured that participants did not look over the instructions and essay in advance. Participants were asked to read the essay and estimate the writer's actual position on the issue. Half of the participants were told that the writer chose the position to advocate in the essay, whereas the remaining participants were told that the writer was assigned a position to advocate. All participants then read the essay and indicated the writer's true attitude about capital punishment and wrote a brief reason for their rating. Participants were debriefed and received remuneration.
|
RESULTS |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
and Follett & Hess, 2002). We rescaled the 7-point attitude measure into a 4-point extremity scale by calculating the absolute difference from the neutral point on the original scale. Therefore, on the rescaled attitude scale, lower numbers reflect a less biased rating (i.e., they are closer to neutral) and higher numbers reflect a correspondent rating in the direction of the essay. (Note that the extremity score does not account for the fact that some individuals may have made responses in the opposite direction, i.e., rating strongly in favor for an unfavorable target. Few individuals gave opposite responses and the pattern of significant results remained the same after we removed data from these individuals.)
We conducted a 2 (age) x 2 (culture) x 2 (choice) analysis of variance on the rescaled target attitude ratings. A three-way Age x Culture x Choice interaction, F(1, 348) = 3.74, p =.05, 
2 =.01 emerged, warranting further decomposition for specific hypotheses testing. Figure 1 depicts the recoded attitude ratings as a function age, culture, and target choice.
|
2 =.10, with American young adults producing more extreme ratings than Chinese young adults. There was also a main effect of choice, F(1, 179) = 13.33, p <.01, 2 =.07, with participants producing more extreme attitude ratings in the choice condition compared with the no-choice condition. There was no interaction between choice and culture, p >.60.
In older adults, there was a main effect of culture, F(1, 169) = 65.89, p <.01, 2 =.10, and a main effect of choice, F(1, 169) = 22.92, p <.01, 2 =.12. We qualified these main effects by a significant two-way interaction between culture and choice, F(1, 169) = 5.47, p <.05, 2 =.03. American older adults did not adjust their attitude ratings in the no-choice compared with the choice condition, t(66.37) = –1.89, p >.05. However, Chinese older adults did adjust their ratings in the no-choice compared with the choice condition, t(75.78) = –4.51, p <.01. These results suggest that Chinese participants of all ages and American young adults showed less correspondence bias than did older adult Americans.
To further examine our second hypothesis, we decomposed the three-way interaction by looking at each culture separately. In both Chinese and American cultures, we found main effects of choice, F(1, 169) = 22.45 and F(1, 179) 13.94, respectively (ps <.01). All individuals reported more dispositional ratings in the choice as compared with the no-choice condition. However, a different pattern of age differences emerged in the two samples. In the American sample, there was a main effect of age, F(1, 179) = 13.86, p <.01, with older adults providing more dispositional ratings than young adults. However, in the Chinese sample, there was no main effect of age, F(1, 169) =.002, p >.96, with both age groups providing equal ratings that were less dispositional than those of the American sample. There were no Age x Choice interactions, although there was a trend in the Chinese sample (p <.11). Although this interaction did not reach significance, Figure 1 shows that the trend was in the hypothesized direction, with Chinese older adults showing slightly more situational responses in the no-choice condition compared with the young adults.
Qualitative Responses
Participants also provided a written explanation of their ratings of target attitude. Two raters for each culture coded the responses into eight categories (see Blanchard-Fields & Horhota, 2005 for the coding scheme). The interrater reliability was 84% for the American sample and 90% for the Chinese sample. Three Chinese participants did not complete the qualitative ratings. Therefore, we examined the information for 167 Chinese participants and all of the American participants. The qualitative data are presented as a percentage of the sample broken down by culture and age.
Both American and Chinese participants most frequently mentioned that their target attitude ratings were influenced by the type of arguments presented in the essay (more than 70% of participants). However, of greater interest was the large discrepancy between the American and Chinese samples in terms of the number of individuals reporting that the situational constraint on the writer impacted their attitude rating. Only 4 Americans (3 young, 1 old) reported considering the situational constraint on the target when rating the target's attitude, whereas 39 Chinese (16 young, 23 old) reported this factor as influential. These reports suggest that Chinese participants placed greater importance on the situational constraint than did the American participants in the sample.
|
DISCUSSION |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
A cognitive aging approach would predict that older adults would display a stronger correspondence bias in the no-choice condition than would young adults, regardless of culture, as a result of their decreased processing resources. This explanation was not supported by our data. Rather, our data suggest that an acculturational explanation is more tenable. Chinese participants, regardless of age, show a preference to examine the situational pressures on the behavior of the target, and both young and older Chinese showed the same pattern of adjustment in their attributional ratings. Furthermore, the qualitative responses provided evidence that Chinese individuals acknowledged the importance of the target's situational constraint far more frequently than did members of our American sample.
Beliefs and values may influence attributional processing in general and cross-cultural age differences more specifically. Along these lines, Funder (1995) argues that it is a "fundamental evaluation mistake" to define an individual's evaluative judgment solely in terms of the cognitive processes by which it is made without considering the empirical evidence the individual has gathered from the social world that demonstrates its validity. Similarly, when age differences are examined in a cross-cultural context, it is important to consider the influence of the content-specific and spontaneous nature of responding. As a result of a lifelong accumulation of cultural experience, older adults may be more likely than young adults to internalize cultural-specific modes of attribution.
In the case of older Americans, their spontaneous mode of responding may be to attribute cause to dispositional characteristics of the target character in a situation, which is due to a lifelong experience of an individualistic orientation. In order to adjust this initial judgment, one must make the contextual information salient to them in a socially meaningful manner. Support for this idea comes from studies showing that when there is a plausible motivation for the target's behavior, then older adults can correct their judgments to be less biased than in a standard attitude-attribution paradigm (Blanchard-Fields & Horhota, 2005). For older Americans to correct their attributions, the constraint must provide a meaningful reason why a person would contradict his or her own beliefs. In the case of Chinese older adults, the meaningful nature of the situation need not be reinforced, given that acknowledging situational influences represents a naturally occurring manner in which to approach any judgment situation as well as a lifelong experience of a collectivist orientation.
Finally, although the non-universality of age differences suggests that universal cognitive decline is not the best explanation for our findings, it should be noted that we did not directly test a cognitive hypothesis. Future research must directly test the two competing hypotheses (cognitive and cultural) in order for researchers to better understand the conditions under which age differences emerge and when they do not.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received for publication April 4, 2007. Accepted for publication July 27, 2007.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
| ||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
