Back to Journal Cover Page

CURRENT RESEARCH IN SOCIAL PSYCHOLOGY

Volume 7, Number 2
Submitted: October 3, 2001
Resubmitted: October 17, 2001
Accepted: October 31, 2001
Publication date: November 2, 2001

I KNOW WHAT YOU DID LAST MILLENNIUM: ETHNIC STEREOTYPE AND ATTITUDE CHANGE AFTER REMINDING PEOPLE OF HISTORICAL EVENTS

Vasile Cernat
Babes-Bolyai University, Romania

ABSTRACT

This study investigated whether reminding people of centuries-old historical events modifies their ethnic stereotypes and attitudes. Ninety-two Romanian participants completed Romanian history tests describing four types of historical events: (a) Hungarians cooperating with Romanians to overcome common enemies, (b) Romanians being military or territorially aggressed by Hungarians, (c) Romanians defeating non-Hungarian aggressors, (d) Romanians being military or territorially aggressed by non-Hungarian enemies, and a geography test unrelated to Romania or Hungary (control group). After the manipulation phase, participants responded to questions about their affective state, self-esteem, and their attitude and stereotype dimensions of Romanians and Hungarians. Statistical analysis revealed important gender x test conditions interactions and that the experimental manipulation had a significant effect on the ethnic stereotypes of participants, but not on their ethnic attitudes.

[11]
---------------
[12]

INTRODUCTION

Although social psychologists unanimously agree that people’s perception and evaluation of others depend heavily on their pre-existing social knowledge base (Macrae and Bodenhausen 2000, 2001), they have paid little attention to the role of historical knowledge and beliefs in the process of intergroup perception and evaluation. However, what people know or think they know about historical events, personalities, or symbols has a significant impact on theirs thoughts, feelings, and behaviors (Rudmin 1997), or, more specifically, on their ethnic stereotypes and attitudes (Mungiu-Pippidi 1999).

History and Ethnic Conflicts: Romanians and Hungarians
The first ethnic conflict in post-communist Europe involved the Romanian and Hungarian communities of Târgu-Mures, a town situated in the Transylvanian region of Romania. A content analysis of the two local Romanian and Hungarian language newspapers published in that period revealed a high frequency of articles on national history. As a common feature, both newspapers focused on the historical events, personalities, or symbols of the in-group providing a biased presentation of the historical information. However, in the Romanian newspaper the most frequent historical topics involved the territorial integrity and the Hungarian oppressions in the past (Bodó, Cosmeanu, Mátéffy, and Marginean 1995).

The importance of historical beliefs in this ethnic conflict was not fortuitous. While most people all around the world see Transylvania as synonymous with the legend of Dracula, Romanians and Hungarians have other chronically accessible concepts associated with this word. Throughout history, this territory has been disputed between Romanians and Hungarians, who embrace conflicting theories of its history. Although the content of ethnic stereotypes depend on realistic differences between groups on economic, political, cultural, or relational aspects (Peabody 1985, Pope and Linssen 1999), the manner Romanians and Hungarians perceive each other also seem to depend on their beliefs about the history of Transylvania. Many Romanians believe that Hungarians conquered this land from their ancestors and perceive them as aggressors who oppressed Romanians for centuries, whereas many Hungarians believe that Romanians came in Transylvania as shepherds and found here a civilization built by Hungarians, viewing them as uncivilized (see Mungiu-Pippidi 1999, for a detailed account).

Problem and Goal of the Present Study
Many of the historical episodes evoked by the Romanian and Hungarian language newspapers took place hundreds of years ago. Because these newspapers provided a biased view of the historical events, they were often accused that their conduct had an instigating effect. Nevertheless, a question seems justified: is it possible to alter people’s ethnic stereotypes and attitudes by simply reminding them of national historical events that happened hundreds of years ago?

[12]
---------------
[13]

Because stereotypes and attitudes are commonly perceived as strongly associated to the current relations between groups, it seems unlikely that intergroup perception and evaluation could be altered by reminding people of events that happened centuries ago. After all, intergroup perception and evaluation change with time. For example, a recent replication of Princenton trilogy revealed that almost all stereotypes of 10 ethnic and national groups changed in content over a period of 60 years (Madon, Guyll, Aboufadel, Montiel, Smith, Palumbo, and Jussim, in press). This result would be improbable if people would construe their ethnic and national stereotypes based on what they believe that an ethnic or national group did many years ago. However such findings do not imply that stereotypes or attitudes are not associated with historical beliefs, or that they could not be affected by historical beliefs. Moreover, many researches have revealed that even subtle stimuli can significantly alter people’s perception and evaluation of the social world (see Bargh 1994; Macrae and Bodenhausen 2001) or people’s behaviors (Steele and Aronson 1995; Shih, Pittinsky, and Ambady 1999; Stone, Lynch, Sjomeling, and Darley 1999). For example, subjects exposed to pictures of admired Black and disliked White Americans expressed less automatic preference for White over African Americans compared with subjects exposed to pictures of admired White and disliked Black individuals or to non-racial exemplars (Dasgupta and Greenwald, in press). Similarly, subjects that received a "miss-addressed" e-mail were friendlier in their replies when the sender had the same name as a well-known positive exemplar than the sender had the same name as a well-known negative exemplar (Castelli, Zogmaister and Arcuri 2001). Consequently, although it seems unlikely that reminding people of events that happened hundreds of years ago would alter their intergroup perception and evaluation, the results of such contemporary researches on the "sensitive" side of social information processing suggest the opposite.

The present study investigated if short-term changes of ethnic stereotypes and attitudes occur after reminding people of national historical events, some of them similar to the type of events evoked by the Romanian newspaper during the ethnic conflict, others providing counter-stereotypical information (by showing Hungarians cooperating with Romanians), or attenuating information (by reminding participants of more negative historical events related to non-Hungarian groups).

[13]
---------------
[14]

METHOD

Participants and Design
Ninety-two Romanian students at Electromures Highscool of Târgu-Mures voluntarily participated in this experiment. The data from four subjects were discarded because they failed to complete all measures or clearly misunderstood the instructions; the final sample consisted of 88 subjects (59 males and 29 females). The study was of a between-subjects design, with the type of exposure information used as the independent variable, and the subsequent scores on Affectometer 2, Rosenberg’s Self-Esteem, as well as on attitude and stereotype dimensions as the dependent variables. Positive and negative attitudes were evaluated as well as sociability, efficiency, aggression, and extremism stereotype dimensions. Sociability and efficiency were selected because they seem to be the most important dimensions of intergroup perception (Pope and Linssen 1999), whereas aggressivity and extremism were selected because they are particularly relevant for the manner Romanians perceive and evaluate the Hungarians (Mungiu-Pippidi 1999, Stoltz 1999). Affective state and self-esteem were also measured, numerous researches and theories emphasizing their important mediational role in intergroup perception and evaluation (Tajfel 1981; Tajfel and Turner 1986; Greenwald, Banaji, Rudman, Farnham, Nosek, and Mellott, in press; Bodenhausen, Mussweiler, Gabriel, and Moreno, in press).

Materials
History and geography tests, each of them containing five items selected from secondary school textbooks, represented the exposure material. All items of the history tests, except two, described events that happened more than 60 years ago (M = 450 years). The items consisted of a sentence describing an historical event, followed by four letter choices, representing possible years of the events, historical figures involved etc., and participants had to chose the option representing the correct answer. Stereotypes dimensions were assessed using the percentage of shared attribute method, and positive and negative attitudes using 7-point scales that ranged from1-to a small extent to7- to a large extent (see the Appendix). Affectometer 2 (Robinson, Shaver, and Wrightsman 1991, pp. 98-100) and Rosenberg’s Self-Esteem scales (Robinson, Shaver, and Wrightsman 1991, pp. 127-131) have also been used with 7-point scales.

[14]
---------------
[15]

Procedure
Participants were randomly assigned to one of the five conditions: a) positive Hungarian (Romanians and Hungarians fighting together against a common enemy); b) positive non-Hungarian (military successes of Romanians over Turks and Germans); c) negative Hungarian (Hungarian military aggressions against Romanians, losses of territory to Hungarians etc.); d) negative non-Hungarian (Turk and Russian military aggressions against Romanians, losses of territory to these groups etc.); e) geography (control group). Participants were assured they will not receive any grades for their test performance and that their answers will remain anonymous. After participants completed the "knowledge tests" they had to respond to Affectometer 2 and Rosenberg’s Self-Esteem scales and to complete the items that assessed their stereotypes of Romanians and Hungarians and their attitudes toward these two ethnic groups. Finally, participants provided information about their age and gender.

RESULTS

Gender by Test Conditions Analysis of Variance
Although no previous research on how Romanians perceive or evaluate Hungarians revealed important gender differences (Mungiu-Pippidi 1999, Stoltz 1999), the results showed that gender significantly moderated the test conditions effects on some dependent variables. Consequently, the measured variables were submitted to a 2 (gender) by 5 (test conditions) multivariate General Linear Model analysis of variance.

The test conditions had important main effects on two stereotype dimensions: Romanian’s sociability (F(4,86) = 2.46, p = .052) and Hungarian’s efficiency (F(4,86) = 2.25, p = .071). Tukey’s Honestly Significant Differences (HSD) revealed that in the negative Hungarian condition participants perceived Hungarians as more efficient and Romanians as more sociable than in the positive Hungarian condition (p = .040 and .005, respectively). The mean scores are shown in Table 1.

[15]
---------------
[16]

Table 1. Participants mean scores and standard deviations for the measured variables by test conditions.

[16]
---------------
[17]

The gender of participants (Table 2) had significant main effects on three dependent variables. Female participants perceived Hungarians as more sociable than male subjects (F(1,86) = 5.11, p = . 027), reported that more Hungarians are characterized by efficiency than male participants (F(1,86) = 5.07, p = .027), and expressed less negative evaluations of Hungarians than males (F(1,86) = 10.96, p = .001). That is, male participants reported more negative stereotypes and attitudes toward Hungarians than female participants.

Table 2. Male and female participants mean scores by test conditions.

[17]
---------------
[18]

The analysis revealed a significant gender by test conditions interaction effect on Affectometer 2 scores (F(4,86) = 2.54, p = .046). Female participants Affectometer 2 scores were affected only by the presence or the absence of Hungarians, whereas male participants Affectometer 2 scores were affected both by Hungarian’s presence and by the nature of the historical events (Figure 1).

Figure 1. Male and female participants scores for Affectometer 2

The analysis also revealed a weaker (F(4,86) = 2.27, p = .069) but quite similar gender by test conditions interaction effect on self-esteem (Figure 2).

Figure 2. Male and female participants scores for Rosenberg Self Esteem

[18]
---------------
[19]

Gender and test conditions also interacted significantly on Hungarian’s extremism (F(4,86) = 2.84, p = .030). Male participants rated Hungarians as more extremist in the negative Hungarian condition than in the positive Hungarian condition, whereas female participants did the opposite. Furthermore, male participants perceived Hungarians as more extremist in the positive non-Hungarian condition than in the negative non-Hungarian condition, but female participants did the reverse (Figure 3).

Figure 3. Male and female participants scores for Hungarian’s extremism

[19]
---------------
[20]

When affective state was used as a covariate, the analysis revealed significant effects on Romanian’s positive evaluation (F(1,86) = 9.56, p = .003), negative evaluation (F(1,86) = 7.86, p = .006), aggressivity (F(1,86) = 4.61, p = .035), efficiency (F(1,86) = 4.55, p = .036), and sociability (F(1,86) = 9.97, p = .002). If self-esteem was entered as a covariate, the analysis found significant effects on Romanian’s positive evaluation (F(1,86) = 4.52, p = .037) and sociability (F(1,86) = 4.15, p = .045). With no exception, greater affective state or self-esteem were associated with a more positive perception or evaluation of the Romanians.

ANOVA of Male Participants' Results
The test conditions did not have a significant effect on male participants' attitudes. However, their stereotype dimensions were significantly altered: Hungarian’s efficiency (F(4, 57) = 2.67, p = .042), extremism (F(4, 57) = 2.60, p = .046), aggressivity (F(4, 57) = 2.50, p = .054), and Romanian’s sociability (F(4, 57) = 3.79, p = .008), and efficiency (F(4, 57) = 3.36, p = .016). Tukey’s HSD was used to determine where actual significant differences lied when the overall comparison was significant. The following significant differences emerged (Table 3):

1. Male participants perceived Hungarians as more efficient in the negative Hungarian condition (M = 237.03) than in the positive Hungarian condition (M = 137.00);

2. Male participants perceived Hungarians as more aggressive in the negative Hungarian condition (M = 256.21) than in the positive non-Hungarian condition (M = 165.50);

3. Male participants perceived Hungarians as more extremist in the negative Hungarian condition (M = 197.64) than in the negative non-Hungarian condition (M = 117.58);

4. Male participants perceived Romanians as more efficient in the positive non-Hungarian condition (M = 264.93) than in the positive Hungarian condition (M = 184.91;

5. Male participants perceived Romanians as more sociable in the negative Hungarian condition (M = 377.43) than in the positive Hungarian condition (M = 263.09).

[20]
---------------
[21]

Table 3. Multiple comparisons of means of overall significantly altered dimensions (Tukey’s HSD).

Although the overall comparison of means for the affective state of participants was not significant, Tukey’s HSD revealed that the difference between positive non-Hungarian (M = 211.46) and positive Hungarian (M = 184.73) conditions was important (p = .067).

ANOVA of Female Participants Results
Test conditions had almost no effect on female’s attitudes and stereotypes of Romanians and Hungarians, the only measure significantly affected being Hungarian’s extremism (F(4, 24) = 3.42, p = .020). Female participants rated Hungarians as significantly more extremist in the positive Hungarian condition (M = 180.00) and control condition (M = 186.13) than in the positive non-Hungarian condition (M = 82.21).

[21]
---------------
[22]

In-Group Favoritism
Measures of in-group favoritism were computed by subtracting the scores of the out-group from the scores of the in-group. Gender was responsible for one significant difference in this regard: male’s negative evaluation differentiation (M = 11.44) was greater compared with female’s negative evaluation differentiation (M = 4.99), F(1,86) = 6.04, p = .016. Test conditions affected significantly only male participants in-group favoritism on efficiency (F(4,57) = 2.84, p = .033). They seemed to favor the in-group the most in the positive non-Hungarian condition, as post-hoc tests revealed (all Tukey’s HSD had p < .08, except the comparison with the positive Hungarian condition).

In-group favoritism depended more on the affective state of the participants. When affective state was entered as a covariate the analysis revealed significant effects on in-group favoritism on positive evaluation (F(1,86) = 4.18, p = .044), negative evaluation (F(1,86) = 5.290, p = .024) and sociability (F(1,86) = 10.11, p = .002). Self-esteem had significant effects on in-group favoritism on negative evaluation (F(1,86) = 4.40, p = .039) and aggressivity (F(1,86) = 4.57, p = .036).

DISCUSSION

The present study explored the effects of exposure to information about centuries-old national historical events on ethnic stereotypes and attitudes. The results failed to reveal any significant impact of test conditions on the ethnic attitudes expressed by the participants, but showed significant effects of test conditions on their ethnic stereotypes. More specifically, participants rated Romanians as more sociable and Hungarians as more efficient when reminded of Hungarian aggressions in the past than when reminded of Hungarians cooperating with Romanians. Furthermore, test conditions significantly affected how female participants stereotyped Hungarians on extremism, but they significantly altered how male participants stereotyped Hungarians on efficiency, aggressivity, extremism, and Romanians on efficiency and sociability.

One reason results did not reveal a significant impact of the test conditions on participant’s ethnic attitudes could be that recent events, with a potentially more powerful affective impact, were not included in test conditions. The history tests described events that happened centuries ago (M = 450 years) and none of them mentioned Hungarian or non-Hungarian atrocities against Romanians. That the reminded information was "cold" is also revealed by the fact that test conditions did not have significant main effects on the reported affective state of participants. However, it may also be speculated that modifications of ethnic attitudes occurred but have not been detected. A limit of the present study is that it used only explicit measures, which are unable to detect changes at the implicit level of attitudes.

[22]
---------------
[23]

Male participants expressed more negative evaluations toward Hungarians than female participants. They also reported more negative stereotypes of Hungarians than female participants, a result that is consistent with the finding that high-prejudiced persons hold more powerful stereotypes compared with low-prejudiced persons (Lepore and Brown 1997). That female participants expressed significantly less negative attitudes toward Hungarians than male participants could explain why their stereotypes were less affected compared with male participants stereotypes, because category related stimuli affect to a greater degree the social information processing of high-prejudiced persons than the social information processing of low-prejudiced persons (Wittenbrink, Judd, and Park 1997). However, additional factors may be involved. The significant interaction of gender with test conditions on affective state and self-esteem, which showed that females were less affected by the nature (positive or negative) of historical events than males, suggest that the reminded episodes (most of them about military actions) might had distinct meanings for male and female participants.

In the positive Hungarian condition, participants perceived Romanians as significantly less sociable and Hungarians as significantly less efficient than in the negative Hungarian condition. Pope and Linssen (1999) have shown that higher economic status and size of the out-group is associated with decreased in-group favoritism on the status related dimension of competence (efficiency) and with increased in-group favoritism on the alternative dimension of morality (sociability). In the negative Hungarian condition, participants were reminded of historical events describing Hungarian military and territorially aggressions. However, these events also reminded participants of the Hungarian dominance of the in-group in the past. The finding that they perceived the out-group as more efficient, while rating the in-group as more sociable than in the positive Hungarian condition suggests that participants were more sensitive to the information about the status of the out-group than to the nature (aggressive or cooperative) of its behavior. Although the present study failed to reveal significant effects of the test conditions on in-group favoritism, these results agree with the findings of Pope and Linssen (1999). Nevertheless, why participants judged Hungarians as less efficient and Romanians as less sociable in the positive Hungarian condition, which made salient the behavior of the out-group rather than its status? One plausible answer may be that the information provided to participants in this condition (Hungarians cooperating with Romanians) contradicted their negative stereotypes and attitudes toward Hungarians and constrained negatively how they stereotyped the in-group on sociability, whereas out-group’s efficiency was altered as an alternative dimension.

That the information about the status of the out-groups was most important is also revealed by the finding that male subjects judged Romanians as most efficient when reminded of military victories of Romanians against non-Hungarian aggressors, that is, against powerful groups (Turks and Germans). However, the type of behavior performed by the out-group altered male participants stereotypes. Specifically, Hungarians were more negatively stereotyped on aggression and extremism in the negative Hungarian condition than in the conditions unrelated to Hungarians. That is, when reminded of Hungarian aggressions against their national group they perceived them as more aggressive and extremist.

[23]
---------------
[24]

Although test conditions had no significant effects on in-group favoritism, participants with high affective state favored to a greater degree the in-group than participants with low affective state, a result that is consistent with the general findings on the effects of happy and depressive moods on social information processing strategies (see Bodenhausen, Mussweiler, Gabriel, and Moreno, in press). The same pattern was found for participants with high and low self-esteem, except that in this case the differences were less significant. Moreover, results shown that the affective state and self-esteem of participants altered significantly how they rated and evaluated the in-group on several dimensions, but not how they perceived or evaluated the out-group. Interestingly, test conditions altered significantly in-group favoritism only when it also affected almost significantly the affective state of participants. Specifically, male participants favored more the in-group on efficiency in the positive non-Hungarian condition, that is, when they also expressed the greatest affective state.

In conclusion, the results of this study imply that reminding people of events that happened hundreds of years ago may alter their ethnic stereotypes, especially if they are highly prejudiced against the target group. The modifications of ethnic stereotypes seem to depend on the information about the status and the nature of interaction with the out-group. However, it is possible that such centuries-old events are not "hot" enough to alter people’s attitudes and in-group favoritism.

REFERENCES

Bargh, J. A. (1994). "The four horsemen of automaticity: awareness, intention, efficiency, and control in social cognition." In R. S. Wyer Jr, T. K. Srull (Eds.), Handbook of Social Cognition. Vol. 1: Basic Processes, pp. 1-40. Hillsdale, NJ: Erlbaum. 2nd ed.

Bodenhausen, G. V., T. Mussweiler, S. Gabriel and K. N. Moreno (in press). "Affective Influences on Stereotyping and Intergroup Relations." In Forgas, J. P. (Ed.) Handbook of Affect and Social Cognition.

Bodó, Z., M. Cosmeanu, C. Mátéffy and P. Marginean (1995). “Alter/Ego tîrgumuresean,” Altera, 1:77-95.

Castelli, L., C. Zogmaister and L. Arcuri (2001). "Exemplar Activation and Interpersonal Behavior", Current Research in Social Psychology, 6: 33-44, http://www.uiowa.edu/~grpproc.

Dasgupta, N. and A. G. Greenwald (in press). "On the malleability of automatic attitudes: Combating automatic prejudice with images of admired and disliked individuals," Journal of Personality and Social Psychology.

[24]
---------------
[25]

Rudmin, F. W. (1997). "Questions of U.S. hostility towards Canada: A cognitive history of blind-eye perception," Undercurrent, Issue #5.

Greenwald, A. G., M. R. Banaji, A. L. Rudman, S. D. Farnham, B. Nosek and D. S. Mellott (in press) "A Unified Theory of Implicit Attitudes, Stereotypes, Self-Concept, and Self-Esteem," Psychological Review.

Lepore, L. and R. Brown (1997). "Category and stereotype activation: Is prejudice inevitable?" Journal of Personality and Social Psychology, 72:275–287.

Macrae, C. N. and G. V. Bodenhausen (2001), "Social Cognition: Categorical Person Perception," British Journal of Psychology, 92:239-255.

Macrae, C. N. and G. V. Bodenhausen (2000), "Social Cognition: Thinking Categorically about Others," Annual Review of Psychology, 51:93-120.

Madon, S., M. Guyll, K. Aboufadel, E. Montiel, A. Smith, P. Palumbo and J. Jussim (in press). "Ethnic and national stereotypes: The Princeton trilogy revisited and revised". Personality and Social Psychology Bulletin.

Mungiu-Pippidi, A. (1999). Transilvania subiectiva. Bucuresti: Humanitas. An English translation is available at: http://www.osi.hu/ipf/publications/AlinaPP-nation.pdf .

Peabody, D. (1985). National characteristics . New York: Cambridge University Press.

Pope, E and H. Linssen (1999). "In-group favouritism and the reflection of realistic dimensions of difference between national states in Central and Eastern European nationality stereotypes", British Journal of Social Psychology, 38:85-102.

Robinson J. P., P. R. Shaver and L. S. Wrightsman (1991). Measures of personality and social psychological attitudes. San Diego: Academic Press.

[25]
---------------
[26]

Shih, M., T. Pittinsky and N. Ambady (1999). "Stereotype susceptibility: Identity salience and shifts in quantitative performance," Psychological Sciences, 10:80-83.

Steele, C. M. and J. Aronson (1995). "Stereotype threat and the intellectual test performance of African-Americans," Journal of Personality and Social Psychology, 69:797-811.

Stoltz, G. (1999). "Etnotipuri în viziune relationala romani-alti etnici," Revista Româna de Sociologie, 10:299-312.

Stone, J., C. I. Lynch, M. Sjomeling and J. M. Darley (1999). "Stereotype Threat Effects on Black and White Athletic Performance," Journal of Personality and Social Psychology, 77:1213-1277.

Tajfel, H. (1981). Human groups and social categories. Cambridge: Cambridge University Press.

Tajfel, H. and J. C. Turner (1986). "The social identity theory of intergroup conflict". In S. Worchel and W. G. Austen (Eds), Psychology of intergroup relations, pp. 7-24. Chicago: Nelson-Hall.

Wittenbrink, B., C. M. Judd and B. Park (1997). "Evidence for racial prejudice at the implicit level and its relationship with questionnaire measures," Journal of Personality and Social Psychology, 71:262–274.

[26]
---------------
[27]

APPENDIX

Table 4. The pattern matrix of a principal component factor analysis with direct oblimin rotation of stereotype items used in the experiment.

[27]
---------------
[28]

Table 5. The pattern matrix of a principal component factor analysis with direct oblimin rotation of attitude items used in the experiment.

[28]
---------------
[29]

Table 6. Alpha coefficients of stereotype and attitude scales

AUTHOR BIOGRAPHY

Vasile Cernat is a Ph.D. student in social psychology at "Babes-Bolyai" University, Cluj-Napoca, Romania. He also works at "Gheorghe Sincai" Institute for Social and Human Research of the Romanian Academy and at Mediapolis. E-mail: vcernat@fx.ro.

[29]
---------------
[30]