Attitudes towards war, killing, and punishment of children among young people in Estonia, Finland, Romania, the Russian Federation, and the USA

Alfred McAlister,1 Patrick Sandström,2 Pekka Puska,3 Anitta Veijo,4 Razvan Chereches,4 & Lenne-Triin Heidmets4



OBJECTIVE: To study the cultural differences in moral disengagement, which lends support to attitudes used to justify violence.
METHODS: We carried out classroom surveys of a total of 3122 students in the USA (Houston, TX, and Washington, DC) and in four European countries — Estonia (Tartu), Finland (Helsinki), Romania (Satu Mare) and the Russian Federation (St Petersburg). Data were also taken from a random sample telephone survey of 341 young adults (aged 18–35 years) in Texas, USA. Ten distinct groups were studied. Seven questions were common to all the surveys, using identical statements about the participants’ agreement with attitudes relating to war, diplomacy, killing, and the punishment of children.
FINDINGS: The US students were more likely than those in Europe to agree with the following statements: ‘‘War is necessary’’ (20% vs 9%), ‘‘A person has the right to kill to defend property’’ (54% vs 17%), and ‘‘Physical punishment is necessary for children’’ (27% vs 10%). Justification of war and killing was less common among females than males in all groups; other differences within the US groups and the European groups were smaller than the differences between the US and European groups.
CONCLUSION: The results confirm the gap between the US and European groups in moral disengagement attitudes and tendencies that could lead to deadly violence.
Keywords Violence; War; Punishment; Attitude/ethnology; Child; Cross-cultural comparison; Europe; Estonia; Finland; Romania; Russian Federation; United States (source: MeSH).

Mots clés Violence; Guerre; Punition; Attitude/éthnologie; Enfant; Comparaison transculturelle; Europe; Estonie; Finlande; Roumanie; Fédération de Russie; Etats-Unis (source: INSERM).

Palabras clave Violencia; Guerra; Castigo; Actitud/etnología; Niño; Comparación transcultural; Europa; Estonia; Finlandia; Rumania; Federación de Rusia; Estados Unidos (fuente: BIREME).




Violence is a major international health problem (1) and public health leaders around the world are eager to learn how deadly conflicts can be prevented (2). To contribute to the search for creative solutions, the Committee on Refugees and Peace of the International Federation of Medical Students’ Associations has organized a programme of research and education to promote and sustain ‘‘cultures of peace’’ (3) among young people in diverse populations. This article summarizes the results of a pilot study in the USA (Texas and Washington, DC) and four European countries — Estonia, Finland, Romania and the Russian Federation.

Aggressive responses to intergroup and international conflicts are partly determined by processes of moral disengagement (4). Through these processes, the perpetration of violence against potential victims is made acceptable by the expression of attitudes that influence personal and collective judgements of choices for resolving conflicts by acts of aggression (5). When moral disengagement occurs, violence is justified by invoking ‘‘rights’’ or ‘‘necessities’’ that provide excuses for the infliction of suffering upon others. Studies of schoolchildren in Italy, for example, have shown that individual differences in the propensity for aggression are related to differences in tendencies towards moral disengagement (6). Moral disengagement also influences group differences in levels of aggression. Studies comparing northern and southern populations in the USA have revealed that public support for lethal aggression (e.g. killing to defend one’s property) was stronger in the southern regions, where homicide rates were higher, than in comparable northern areas (7). Moral disengagement also influences aggression within the family. A survey of several cities in North America and Latin America showed that individuals who endorsed the attitude that ‘‘physical punishment is necessary to raise children properly’’ were more likely to report acts of violence towards their children (8).

High levels of violence in the USA have been attributed to a national ‘‘culture of violence’’, formed by the history of the country and now sustained by the mass media through communication of violent images and ideas (9). Both in their homes and via popular entertainment and journalism, young people in all parts of the USA are exposed to the attitudes of moral disengagement that could lead them to respond aggressively in conflicts between individuals and groups. Global export of violent images through the media are influencing the attitudes of young people around the world. To understand and reduce international violence, we need to study the differences in attitudes that lead to aggression among a wide range of individuals and groups. If significant national variations are revealed by these comparative surveys, the examination and discussion of these findings might help young people to understand and resist the processes of moral disengagement.



Study groups were chosen to reflect diverse populations within selected cities in Europe and the USA, where young people have very different experiences with violence (10). In 1994 the national homicide mortality rate for males aged 14–25 years in the USA was 38.6 per 100 000 population; however, the rate is much higher (70–90 per 100 000) among young African-American males in the cities of Houston, TX, and Washington, DC, where young people were surveyed in the present study. In 1995 the corresponding rate in Estonia was high (39.2 per 100 000) compared with the rates in the other former Soviet Republics on the Baltic (16.9 per 100 000 in Latvia and 11.3 per 100 000 in Lithuania). Homicide rates among males in the 14–25 years age group were much lower in Finland (2.9 per 100 000) and Romania (3.8 per 100 000). The study cities were chosen because of the presence of medical and public health schools and their proximity and feasibility for the research group. The secondary schools that were surveyed represent average socioeconomic levels in the city concerned, with some schools also selected to obtain specific ethnic groups.

Self-administered questionnaires were used to gather data among students aged 13–18 years. Medical students and graduate students of public health distributed and collected the survey and answer sheets in classrooms in each school. Data were collected in accordance with established principles of research on human subjects. The questionnaires were anonymous and the students were free to decline to participate. The response rates (i.e. proportion of students completing the survey) ranged from 0.67 to 0.81. Surveys were conducted in St Petersburg, Russian Federation (n = 243 (101 males, 142 females); response rate = 0.81) and in Helsinki, Finland (n = 322 (104 males, 218 females); the exact response rate was not calculated, but it was estimated to be approximately 100% of the students in attendance). In Estonia, the surveys were conducted among Estonians and ethnic Russians living in Tartu (n = 204 ethnic Estonians (75 males, 129 females); 382 ethnic Russians (144 males, 238 females); response rates not calculated). In Romania, the surveys were conducted among Romanians and ethnic Hungarians living in the border city of Satu Mare (n = 476 ethnic Romanians (218 males and 258 females); and 368 ethnic Hungarians (151 males, 217 females); response rate not calculated). In the USA, students in three major ethnic groups (Anglo-, Hispanic- and African-American) were surveyed in two urban schools in Houston, TX (n = 575 Hispanics (304 males, 271 females); 252 African-Americans (149 males, 103 females); and 115 ‘‘White non-Hispanics’’, termed ‘‘Anglos’’ in this article (54 males, 61 females)). To investigate the attitudes in a US location other than Texas, a survey was conducted among urban students in Washington, DC (n = 61 (ethnic groups not identified) (27 males, 34 females)). The response rates for the US surveys ranged from 0.67 for the schools in Texas to 0.76 in Washington, DC.

Additional data were obtained from young adults aged 18–35 years who participated in a standard telephone survey conducted in Austin and Houston, TX, by the Office of Survey Research, University of Texas. This survey included questions about war and defensive killing, which were identical to those in the students’ questionnaires. The random sample of 1200 participants (response rate = 0.71) included 341 young adults ( n = 211 Anglo (115 males, 96 females); 130 Hispanic (59 males, 71 females); as only 43 young African-Americans were interviewed, this group was not included in the analyses). The surveys in Texas were conducted in the winter of 1997–98, while those in Europe and Washington, DC, were carried out approximately one year later.

The questionnaires asked students to rate their agreement with seven attitude statements related to moral disengagement and the justification of violence. In previous studies, these individual questions have revealed national, group, and gender differences in moral disengagement (8). Three of the statements dealt with international aggression (11) and confidence in the UN and diplomacy: ‘‘War is necessary to settle differences between countries’’; ‘‘We can depend on the UN to settle any conflict we might have with another country’’; and ‘‘We can depend on diplomacy to settle any conflict we might have with another country’’. Previous surveys in the Americas showed that variations in rates of agreement with the necessity of war were associated with national differences in military spending (11). Two of the statements concerned the justification of killing (12, 13): ‘‘A person has the right to kill to defend his or her property’’ and ‘‘A person has the right to kill to defend his or her family.’’ A third question on killing asked if the respondent would approve if ‘‘Someone kills a person who has raped his or her child.’’ In the USA, answers to these questions have been linked to regional differences in homicide rates (7). The last question concerned violence in the home: ‘‘Physical punishment is necessary to raise children properly.’’ Holding this attitude has been tied to reports of parental aggression towards their children (8). Students rated their agreement with the above attitude statements on a five-point scale: ‘‘strongly agree’’, ‘‘somewhat agree’’, ‘‘don’t know or not sure’’, ‘‘somewhat disagree’’, ‘‘strongly disagree’’. Approval of killing a rapist was rated as: ‘‘approve’’, ‘‘don’t approve but understand’’, ‘‘neither approve nor understand’’.

To reduce the likelihood that differences in responses between groups might be due to imperfect translation, the questions and answers were translated from English into the five other languages concerned (Finnish, Estonian, Russian, Romanian, Hungarian) and then back to English independently by two or more students in each country. Some difficulty was encountered with the difference between the meanings of ‘‘strongly agree’’ versus ‘‘somewhat agree’’ and these answer categories were therefore combined for making comparisons between language groups. The results were analysed to compare the responses of students between countries and within each country and c² tests were used to assess the statistical significance of national, ethnic and gender differences. For the reporting of group results, the significance tests were stratified by gender and ethnic or national subgroups, and the data are presented as comparable mean percentages, adjusting for unequal sample sizes within genders and subgroups. Thus, for example, the national proportions we report are equal to the mathematical average of the proportions among males and females in that country. Proportions were also analysed separately for males and females.



The first analyses investigated the participants’ attitudes towards war. Fig. 1, which presents the attitudes to war among males and females in each country, shows that males were consistently more likely than females to agree that war is necessary (20% vs 7%; P < 0.001 in the c2 test and the group mean t test). There was also a large US–European difference in this respect. In analyses that were stratified by gender and national group to control for differences in sample sizes, Eastern European students were significantly less likely than US students to agree with the necessity of war (P < 0.001). The mean proportion was 9% in the US groups vs 20% among the national groups that were studied in Europe. Students in urban Washington, DC, expressed agreement at a rate (23%) that was very close to that in the Texas groups and, despite the small sample size, significantly greater than that in Eastern Europe (P < 0.05). In Texas, the Hispanic students were less likely to endorse war than Anglos and African-Americans combined (15% vs 19–23%; P < 0.05). Among young people in the telephone survey in Texas, 16% of the Hispanics and 27% of the Anglos agreed that war is necessary (P <0.01). There was one significant difference within the European group (P <0.02): endorsement of war was slightly greater among Russian and Finnish students (12%) than among students in the other European countries surveyed (7%).


Attitudes towards the UN were negative in all countries. The highest proportions reporting confidence in the UN were in Finland (39%) and in the USA (34%). Significantly lower rates (22–23%; P < 0.01) were found among ethnic Russians in Estonia and ethnic Hungarians in Romania. Confidence in the UN was lowest (P < 0.05–0.01) among ethnic Estonians (10%), Russians (14%) and Romanians (16%). Confidence in diplomacy was higher than that in the UN among the European groups, with 57% agreeing that it is a dependable way to settle conflicts. There was less confidence in diplomacy in the US study groups, in which only 36% agreed that it was dependable (P <0.001).

Further analyses compared the attitudes towards killing. Fig. 2 shows the proportions of males and females in each country who agreed with the right to kill to defend property. Males were much more likely than females to endorse such killing (44% vs 25%; P <0.001 in the c² test and P <0.005 in the test on group means). The US–European differential was even greater. In comparisons stratified by gender and group, the European students were much less likely than their US counterparts to agree that it is right to kill to defend property. The mean proportion endorsing this right was 17% among the European groups vs 54% among the US groups (P <0.001). Attitudes in Texas were almost identical to those in Washington, DC, and the latter group was significantly different from the European groups combined (P <0.01). A total of 51% of the Hispanics and 42% of the Anglos covered in the telephone survey in Texas agreed with the right to kill (difference not statistically significant). The national groups in Europe only differed in the response to one statement (P <0.001): Romanian students (10%) and ethnic Hungarian students in Romania (15%) were less likely to endorse killing to defend property than the students in the Russian Federation (22%), Finland (23%) and Estonia (25%). In Estonia, ethnic Estonian students were more likely than ethnic Russian students to agree with the right to kill to defend property (29% vs 21%, P <0.05). Among male students in Romania, ethnic Romanians were significantly less likely than ethnic Hungarians to endorse the right to kill to defend property (15% vs 27%, P <0.01).


Agreement with killing in the defence of the family was high in all countries; the highest levels were in the Russian Federation (80%) and the USA (77%), with significantly (P <0.01) lower levels in Estonia (65%) and Finland (62%). In Romania, only 39% agreed with killing in defence of the family (P <0.005 for comparison with other Eastern European groups). Approval of a parent killing the rapist of a child was generally lower than approval of defensive killings (42% in Russian Federation, 35% in the USA (35%), and 35% among ethnic Russians in Estonia). Corresponding rates were significantly lower (P <0.02) in Finland (21%) and among ethnic Estonians in Estonia and ethnic Hungarians in Romania (14% in both groups). Approval of a parent killing the rapist of a child was lowest (P <0.01) among ethnic Romanians (9%).

The question concerning physical punishment of children also revealed a large US–European differential, but less of a difference between genders (Fig. 3). Among students from all five countries, agreement with punishment of children was significantly (P <0.01) higher among males (20%) than females (15%), but this gender difference was not consistent within national or ethnic groups. In analyses stratified by gender and group, European students were less likely than their US counterparts to agree that children need to be punished physically (26% vs 10%; P <0.001). The rate in Washington, DC (21%) was significantly higher than that in the combined European groups (P <0.05). Among the European groups, only 7% of the ethnic Romanians endorsed physical punishment of children. Corresponding rates were slightly, but significantly (P <0.05), higher in the other European countries (10–12%). Among the US students in Texas, significant ethnic group differences were found for attitude towards physical punishment of children (P <0.005 for three-group comparison). The highest agreement was among African-Americans (41%), followed by Hispanics (31%) and Anglos (19%).


Attitudes towards war, killing in defence of property, and the physical punishment of children were correlated at the individual and group level. At the individual level the correlation was 0.24 for war and killing, 0.25 for war and physical punishment of children, and 0.16 for killing and physical punishment of children (P <0.001 for each correlation). At the group level, in which the proportion of agreement in each of the nine groups was treated as a single observation, the correlation between these attitudes was much higher (0.83–0.85).



The students and young adults in the US study groups were much more likely than their counterparts in Europe to approve of attitudes arising from moral disengagement. Variations within the US and European groups were few and generally small, but there were large gender differences across the study groups in terms of attitudes to war and killing. Females were generally less likely to justify fatal violence and the difference, compared with males, was especially large in the European study groups.

Since the present survey was limited to students in selected schools, the results do not provide conclusive evidence for broad national differences in the responses. While the acceptance of fatal aggression may be more common among young adult Anglo-Americans in Texas than in other areas of the USA (7), this could be a reflection of regional norms that are not representative of the nation as a whole. However, the attitudes were strikingly similar among the young people who were surveyed in both Texas and Washington, DC. Even with its small sample size, the group in Washington, DC, was significantly different from the combined European groups in their attitudes towards war and killing. The consistency of results across the five language groups studied shows that the US–European differential in acceptance of violence is probably not due to imperfect translations in the questionnaires. Accurate response rates were not available from all sites and the survey completion rates were low for the US groups. It is possible that differential response rates introduced an error in the study, but this is not a convincing explanation for the national differences. Because more aggressive students are probably less likely to take part in or complete the survey than other students, lower response rates may imply that a higher proportion of less aggressive students took part. However, despite their low response rates, the US groups expressed significantly greater support for violence than the students in Europe.

The use of deadly force in a variety of ways is strongly related to other features in the environment, e.g. availability of handguns, crime rates, and social inequality (7). But even in the economically distressed populations of the Russian Federation and the republics of the former Soviet Union, homicide rates among young people are lower than those among young people in similarly disadvantaged urban populations in the USA. Despite their increasingly similar exposure to violent images through the media, young people in comparable environments in Europe and the USA continue to exhibit very different levels of violence. Our findings may help to explain why youth in different cultures react differently to the same media images. The imitation of aggressive behaviours portrayed through the mass media is highly influenced by perceived peer approval (4). The present survey shows that students in the US study groups were more likely to express approval for killing than students in the European groups. When killing is portrayed through the mass media, the US students can be expected to show a greater tendency towards imitation. Should young people in Europe begin to adopt US attitudes of moral disengagement in approval of killing, the harmful effects of media violence will probably increase sharply.

The present study has shown that student surveys can detect meaningful differences in attitudes towards violence among comparable groups in different countries. The International Federation of Medical Students’ Associations’ Standing Committee on Refugees and Peace is organizing further surveys by international teams of medical and public health students among their peers as well as in universities, secondary schools and the general population. The results of these studies will be used to help young people to understand and resist the processes of moral disengagement. For example, we know that international comparative studies linking the consumption of saturated fats to cardiovascular disease have stimulated social processes leading to widespread dietary reform in Finland and many other countries, and their findings have identified populations where health diets should be maintained (14). By analogy, international comparative studies on the links between moral disengagement attitudes and violence may stimulate and initiate a process of correction. By increasing young people’s awareness of how the processes of moral disengagement influence group and national differences in rates of aggression, we hope to help future generations in many countries to promote cultures of peace and to reform the existing culture of violence.a

Conflicts of interest: none declared.




Attitudes vis-à-vis de la guerre, du droit de tuer et des punitions corporelles infligées aux enfants chez des jeunes d’Estonie, des Etats-Unis d’Amérique, de Fédération de Russie, de Finlande et de Roumanie

OBJECTIF: Etudier les différences culturelles en ce qui concerne le désengagement vis-à-vis des valeurs morales, générateur de points de vue avancés pour justifier la violence.
MÉTHODES: Nous avons réalisé des enquêtes en milieu scolaire sur 3122 jeunes aux Etats-Unis d’Amérique (Houston, TX et Washington, DC) et dans quatre pays d’Europe – Estonie (Tartu), Fédération de Russie (Saint- Pétersbourg), Finlande (Helsinki) et Roumanie (Satu Mare). Des données ont également été recueillies lors d’une enquête téléphonique auprès de 341 jeunes adultes de 18 à 35 ans sélectionnés par tirage au sort au Texas (Etats-Unis d’Amérique). Au total, dix groupes ont été étudiés. Sept questions étaient communes à toutes les enquêtes et demandaient aux participants d’exprimer leur accord ou leur désaccord vis-à-vis des mêmes affirmations concernant la guerre, la diplomatie, le droit de tuer et les punitions infligées aux enfants.
RÉSULTANTS:Les jeunes des Etats-Unis étaient plus souvent d’accord que les jeunes Européens avec les affirmations suivantes : « La guerre est nécessaire » (20% contre 9 %), « On a le droit de tuer pour défendre ses biens » (54% contre 17 %), et « Les punitions corporelles sont nécessaires aux enfants » (27% contre 10 %). Dans tous les groupes, la justification de la guerre et du droit de tuer était moins fréquemment exprimée par les participantes que par les participants ; les autres différences a` l’intérieur des groupes américains et européens étaient plus faibles qu’entre ces groupes.
CONCLUSION: Les résultats confirment l’écart entre les groupes des Etats-Unis et les groupes européens en ce qui concerne les attitudes et tendances de désengagement vis-à-vis des valeurs morales, susceptibles de conduire à une dangereuse violence.


Actitudes frente a la guerra, los asesinatos y el castigo de los niños entre los jóvenes de Estonia, Finlandia, Rumania, la Federación de Rusia y los Estados Unidos

OBJETIVO: Estudiar las diferencias culturales en cuanto a la laxitud moral, que propicia actitudes de justificación de la violencia.
MÉTODOS: Llevamos a cabo encuestas por clases entre un total de 3122 estudiantes en los Estados Unidos (Houston, TX, y Washington, D.C.) y en cuatro países europeos: Estonia (Tartu), Finlandia (Helsinki), Rumania (Satu Mare) y la Federación de Rusia (San Petersburgo). También se obtuvieron datos encuestando telefónicamente mediante muestras aleatorias a 341 jóvenes (edad: 18–35 años) en Texas (EE.UU.). Se estudiaron diez grupos diferenciados. Siete de las preguntas eran comunes a todas las encuestas, empleándose en ellas frases idénticas sobre el acuerdo de los participantes con actitudes relacionadas con la guerra, la diplomacia, los asesinatos y el castigo físico de los niños.
RESULTADOS: Los estudiantes estadounidenses suscribieron con más frecuencia que los europeos las siguientes afirmaciones: « La guerra es necesaria » 20% frente a 9%), « Una persona tiene derecho a matar para defender lo que es suyo » (54% frente a 17%) y « Es necesario castigar físicamente a los niños » (27% frente a 10%). La justificación de la guerra y los asesinatos fue menos común entre las mujeres que entre los hombres en todos los grupos; otras diferencias dentro de los grupos de los Estados Unidos y los grupos de Europa fueron más pequeñas que las diferencias entre los grupos.
CONCLUSIÓN: Los resultados confirman la brecha existente entre los grupos de los Estados Unidos y de Europa en cuanto a las actitudes de laxitud moral que pueden conducir a formas mortíferas de violencia.




1. The world health report 1997 – Conquering suffering, enriching humanity. Geneva, World Health Organization, 1997.        

2. Carnegie Commission on Preventing Deadly Conflict. Final report. New York, Carnegie Corporation, 1998.        

3. A culture of peace and nonviolence within educational institutions. Elements for the launching of an inter-regional project. New York, UNESCO, 1995.        

4. Bandura A. Social foundations of thought and action: a social cognitive theory. Englewood Cliffs, New York, Prentice-Hall, 1986.        

5. Bandura A. Moral disengagement in the perpetration of inhumanities. Personality and Social Psychology Review, 1999, 3: 193–209.        

6. Bandura A et al. Mechanisms of moral disengagement in the exercise of moral agency. Journal of Personality and Social Psychology, 1996, 71: 364–374.        

7. Nisbett RE. Violence and US regional culture. American Psychologist, 1993, 48 (4): 441–449.        

8. Orpinas P. Who is violent? Factors associated with aggressive behaviors in Latin America and Spain. Pan American Journal of Public Health, 1999, 5: 232–244.        

9. Slotkin R. Gunfighter nation. The myth of the frontier in twentieth-century America. New York, Atheneum, 1992.        

10. World health statistics annual. Geneva, World Health Organization, 1996.        

11. McAlister A. International variation in attitudes toward violence. In: Kurtz LR, ed. Encyclopedia of violence, peace and conflict. New York, Academic Press, 1999.        

13. Cohen D, Nisbett RE. Self-protection and the culture of honor: explaining southern violence. Personality and Social Psychology Bulletin, 1994, 20 (5): 551–567.        

14. Puska P et al. The North Karelia project: 20 years’ results and experiences. Helsinki, The National Public Health Institute (KTL), 1995.        



1 Professor, University of Texas (Houston) School of Public Health, Houston, TX, USA.

2 Executive Director, Physicians for Social Responsibility, Finland. Correspondence should be sent to Dr P. Sandström, Savilankatu 2 D, FIN-00250 Helsinki, Finland (email:

3 Professor, National Public Health Institute, Helsinki, Finland.

4 National Officer for Refugees and Peace (NORP), International Federation of Medical Students’ Associations, c/o World Medical Association, Ferney-Voltaire, France.

a More information about this work is available at the following URL:

Ref. No. 99-0125

World Health Organization Genebra - Genebra - Switzerland