University of Maryland

A very substantial literature has developed in recent years on the nature of prejudicial attitudes in the United States. This interest, particularly in the question of race prejudice, is understandable in view of race relations as a continued source of major social problems in the United States.

Despite the long-standing problems which cultural minorities have faced in many European countries, comparable work on the nature of group prejudice has not developed in Europe. In an era in which immigration restrictions seem to be loosening considerably and economic pressures to increase the flow of cheap labor into many Northern European countries are growing, the public policy implications of societal attitudes toward minority populations would appear to be increasingly relevant. The Scandinavian nations have tended to view racial tensions in the United States, South Africa, and more recently Great Britain in a critical and yet removed manner.¹ Yet little is known about how Scandinavian citizens themselves feel about 'outgroups' and how they might be expected to respond to social situations involving non-Nordic minorities.

1. Methods of Measuring Prejudicial Attitudes

In recent years a number of studies of prejudicial attitudes,² particularly attitudes of whites toward blacks, have been undertaken, most of them based on the development of appropriate scales for assessing the attitudes of one group toward another. The various problems of such scales, however, make it difficult to use them with any degree of confidence for measuring attitudes among groups in either Europe or the United States. First of all, such attitudinal measures must be kept contemporary to be useful, and many of the best known devices are outdated. Secondly, there is little evidence to support claims of scale validity for many of the

most commonly used racial attitude scales. Since the validity of these measures is questionable when they are used hi an American cultural setting, it would seem to be an even more critical problem when such instruments are used in a non setting. A third problem in using most of these standard attitude scales is that among certain groups, particularly university students, there is considerable social reinforcement for being 'tolerant' toward minority groups, especially blacks.³ Since the purpose of many of these measures is often readily apparent, the difficulty in collecting accurate data is substantial when prejudice is something to be hidden from one's peers.

One of the most frequently cited recent studies of prejudicial attitudes, carried out by Rokeach, Smith and Evans,⁴ suggests that belief congruence rather than race prejudice is largely responsible for the behavior of majority groups toward cultural and racial minorities. Earlier studies with American college students by Sedlacek and Brooks, however, suggest that this is not the case⁵ and that the purpose of Rokeach's instrument is too obvious to most respondents, many of whom, therefore, psychologically withdraw from his questionnaire and ignore the racial variable. As a result of this withdrawal, beliefs are measured out of context and the results display a clear lack of attention to race.

To meet the set of problems enumerated above, Sedlacek and Brooks have designed an approach to the study of prejudicial attitudes which should reduce or eliminate these methodological problems. Their studies among American university students to date indicate that this new approach can be viewed as a substantial improvement in measuring prejudicial attitudes accurately.⁶

2. The Situational Attitude Scale

The Situational Attitude Scale (SAS) was developed to measure the degree of prejudice which one group holds for another. Initially, the SAS was used to study the attitudes of whites toward blacks in the United States. To provide a prejudicial context and make withdrawal from the instrument difficult, ten personal and social situations, with some relevance to a racial response, were created (see Table I).

These situations represent instances where one group's actual prejudicial attitudes toward another group might be relevant to the first group's attitude about the particular situation. For each situation, 10 bipolar semantic differential scales were created,⁷ making a total of 100 items in the SAS (see Table II for items). For use with American subjects, two forms of the SAS were developed. Each contained the same situations, bipolar scales, and instructions except that the word 'black' was inserted into each situation in Form B.⁸ The positive pole for each item was varied randomly from right to left to avoid response set.

For use with Danish university students, Forms A and B were translated into
Danish hi as nearly a verbatim manner as possible.9 In addition, a Form C was

developed to measure prejudicial attitudes toward Mediterranean foreign workers (sydlandske fremmedarbejdere). The general opinion of a group of Danish students and teachers consulted on this question was that the sydlandske fremmedarbejdere represent the most direct 'outgroup' threat to Danes at the present time.¹⁰ Although the number of blacks (negre) remains very small in Denmark, the number of Mediterranean foreign workers has increased rapidly in recent years. Their presence in Denmark has been established and may well increase substantially in the next decade if the need for cheap, unskilled labor there continues to grow.

3. The Experimental Group

The three forms of the SAS were administered to 306 students at Copenhagen and Aarhus Universities during regularly scheduled class meetings in nine different classes. The questionnaires were completed anonymously. A total of 32 questionnaires were not included in the analysis because these respondents either left more than 10 items blank or indicated their unwillingness to participate in the experiment. The median scale value (scale 0 to 4; 2 = median) was assigned to any missing item responses, provided there were 10 or less blank items on a questionnaire. The final usable number was 274: 94 Form A's, 90 Form B's, and 90 Form C's.

The SAS was administered by advanced political science students, who were trained in the use of the SAS and had participated in preparing it for use in Denmark ¹¹ The questionnaires were randomly distributed in each class with each participant having an approximately equal chance of receiving any of the three forms. The participants were not aware that there were different forms of the questionnaire. If participants had questions, they were handled on an individual basis so as not to disturb other participants or accidentally expose the fact that different forms were being used. Administration of the instrument required 20-30 minutes.

The participants - all university students - completing the three forms were very similar. There were 223 male and 51 female participants. They were drawn from a number of different institutes in both universities, including architecture, biology, education, English, history, law, mathematics, philosophy, political science, and psychology. Students from both introductory and advanced classes were included and represented a diverse cross-section of Danish university students.¹²

An analysis of variance with form (A, B, or C) and school (Aarhus or Copenhagen) as main effects was conducted. The results indicated that 46 items were statistically significant (.05 level) for form while seven items out of 100 were significant for school and nine items for form by school. Since, according to Sakoda, Cohen and Beall,¹³ nine items out of 100 would be significant by chance, we can conclude that responses varied depending on whether Form A, B, or C was used, but that there were no differences between Copenhagen and Aarhus students or between combinations of form and school.¹⁴

4. Study Results

Table II shows the results when F-values are computed for each combination of forms used in this experiment. First results for Forms A and B are compared. This allows us to measure the degree of difference resulting from the insertion of negre in the social situations used in the SAS. When this is done, 32 of the 100 items show significant differences between the forms.¹⁵ Since we could expect only nine tests in a hundred to be significant by chance, these results suggest that the data for Form B deserve further analysis.¹⁶ When the post hoc results for Forms A and C are compared to measure the effects of sydlandske fremmedarbejdere, 26 of the items show significant differences between the forms.

In order to measure any congruence between the attitudes displayed toward the two outgroups (B and C), F-values were also computed for Forms B and C. While the SAS methodology provides us with a reference or norm group for each experimental group (the respondents to Form A), comparisons between Forms B and C may provide us with further evidence on the types of social situations which elicit similar response patterns from participants and the types of situations which result in basically different responses. When Forms B and C are compared here, 15 situations elicit significantly different responses. Ten of these 15 differences are found to occur in situations I (family moves next door) and V (friend becomes engaged).

There is no empirical evidence produced by the SAS as to the positivity or negativity of either item pole. If goodness or badness is ascribed to either pole on the basis of the social desirability of the dimension involved, however, the significant mean differences within each situation are consistent. For example, if we examine the ten items used in situation 111 (man selling magazines), we find that the insertion of either negre or sydlandske fremmedarbejdere into the situation leads the respondents to feel more positive toward the situation than is true if no group identification is provided (Form A). Form B means are consistently more positive (pro-black) for situations I, 111, IV, V, VI, VII, VIII, and IX. These results are in sharp contrast to the findings for American students on Form B, where results were overwhelmingly anti-black for eight of the 10 situations.¹⁷

What the Danish students reveal here is a substantial and consistent pro-black orientation in their attitude patterns. Like their American counterparts, they approve of blacks in situations involving minimal social contact, such as selling magazines or serving as policemen. But unlike American students they also indicate a substantial bias in favor of blacks for the situations which represent the most intimate types of individual social contact. Danish students would rather have their friend become engaged to a neger; they would rather have a neger join their social group; they would prefer having a neger move in next door to them. They are also less frightened by the loitering men when they are blacks, less disturbed by the youngster who steals when he is black, and more positively disposed toward the campus demonstration when blacks are the demonstrators. Only the situations involving rape and a person standing on the bus find no significant bias in favor of blacks over plain everyday Danes.

When the comparisons between Forms A and C are examined, a substantially different attitude pattern emerges toward the Mediterranean foreign worker. Danish students are positively disposed to Mediterranean foreign workers selling magazines or serving as policemen. They also exhibit less fear of the idea of foreign workers loitering on the corner than they do when just any five young men are loitering there, and they adopt a more passive stance toward the child of a foreign worker who steals in a dimestore. When it comes to those social situations which exhibit a substantial degree of intimacy, however, negative attitudes comparable to those among American students toward blacks surface quite clearly. When their friend becomes engaged to a sydlandsk fremmedarbejder, they are significantly more aggressive, more sad, more angered, and more disgusted by the situation. The evidence on the situations involving the new neighbor (situation I) and the person joining their social group (situation VII) is less dramatic than with the engagement situation, but the results again lean significantly toward the negative pole when the Mediterranean worker is included in the situation. When comparisons between Forms B and C are made, the more negative attitudes toward the Mediterranean foreign worker compared with the blacks are even more evident. The results for situation IX also show a negative attitude toward the Mediterranean student demonstration. While a black student demonstration is judged to be more serious than a regular demonstration with students, indicating

more trust in the blacks and a feeling of being less disturbed by them, the Mediterranean
students' demonstration would be judged comparatively both wrong and
unjustified.

5. Discussion and Interpretation

At a time when serious questions are being raised about the nature of foreign immigration to Denmark, the preliminary results of this study deserve further examination. The Danes have generally been considered a very tolerant and accepting people. Their abhorrence of racial prejudice in the United States, South Africa, and Britain is understandable and admirable. This study shows that Danish students have developed a highly sympathetic and idealized image of black people. But one finds only a handful of blacks living hi Denmark at present, and the few blacks who do live in Denmark are almost never employed in low status jobs. Therefore, such an attitude set must be attributed, at least in part, to the widespread attention which prejudice and discrimination toward blacks in other countries has received in Denmark.18

The influx of Mediterranean peoples into Denmark is a fairly recent phenomenon. Danish immigration policy allows such immigration only to fill jobs which Danes themselves do not desire. This supply of Turks, Yugoslavs, Italians, and Greeks provides Denmark with a significant portion of its unskilled labor in its lowstatus occupations today. Membership in the Common Market and general economic expansion are likely to increase pressures to bring even more foreign laborers into Denmark. Yet these very preliminary results from what may well be one of Denmark's most 'progressive' social groups - its students¹⁹ - suggest that the roots for substantial group conflict are definitely present. Danish society remains one of the most homogeneous among the more developed nations. But the Danes have had their problems with a German minority along their southern border and chose, after World War 11, not to enlarge these problems by raising territorial claims to large portions of Schleswig arid Holstein.²⁰

While preliminary results from a pilot study of 274 Danish university students cannot be used as a basis for policy decisions, the results here suggest that policy concerning the sydlandske fremmedarbejdere could prove crucial to Danish society in the decades to come. It cannot be assumed that mere goodwill and a lack of serious tension at present will prevent difficulties in the future. The study of other cultures suggests that mere social contact will not guarantee either peace or understanding among hostile groups once an outgroup is established within a society ²¹ Such hostility may be in the process of developing in Denmark at present. Even when outgroups enter a society in an equal status capacity, the avoidance of prejudiced attitudes and hostile group attitudes is difficult. In Denmark, the fremmedarbejdere almost always enter the country in an inferior status capacity.

In an important summary of the research on the effects of contact between

ethnic and racial groups, Amir²² concludes that unless contact takes place under favorable conditions it is likely to result in increased intergroup tension and prejudice. Unfavorable conditions include a relatively higher status for one group in the society.

6. Concluding Remarks

This study has definite and obvious limitations. It has been restricted to a limited number of university students. It has attempted to adapt a methodology, originally developed for measuring American race prejudice, to an instrument for measuring a broader range of prejudicial attitudes than that measured in the United States. As a result, some of the situations in the SAS may be less relevant to Denmark, and some of the semantic differential pairs have proved awkward in their translation into Danish.

Despite these limitations, the implications of the results reported here bear further examination, and additional research on these questions would seem justified. Denmark is a consensus-oriented society with a social and political system which minimizes overt conflict among the ruling elites and within the society as a whole. If substantial Mediterranean immigration to Denmark is to continue, programs for strengthening the 'Danish-ness' of immigrants may well be in order. The further education of Danes concerning the norms and cultures of the immigrant groups would also seem to be worth consideration.²³ Planning for such programs now before overt hostilities among Danes and outgroups become serious could prevent substantial minority conflict in Denmark in the decades ahead.

---

NOTES

1. An indication of the Danish attitude toward the American racial problem can be seen in the results of a United States Information Agency study of 1957. In this poll 82 percent of the Danish respondents indicated they thought the treatment of Negroes was 'very bad'. Results reported in Hazel Erskine, 'The Polls: World Opinion of U.S. Racial Problems', Public Opinion Quarterly 32 (1968), p. 302.

2. See for example E. Q. Campbell, 'Some Social Psychological Correlations of Direction of Attitude Change', Social Forces 36 (1958), pp. 335-340; J. M. Fendrich, 'A Study of the Association Among Attitudes, Commitment and Overt Behavior in Different Experimental Situations', Social Forces 45 (1967), pp. 347-355; and H. C. Triandis, 'Exploratory Factor Analysis of the Behavioral Component of Social Attitudes', Journal of Abnormal and Social Psychology 68 (1964), pp. 420-436.

3. See for instance H. Schuman and J. Harding, 'Sympathetic Identification with the Underdog', Public Opinion Quarterly 27 (1963), pp. 230-241; H. Sigall and R. Page, Two Looks at Stereotypes', American Psychological Association Proceedings 5 (1970), pp. 355-356; and W. E. Sedlacek and G. C. Brooks, Jr., 'Social Acceptability in the Measurement of Racial Attitudes', Psychological Reports 29 (1971), pp. 17-18.

4. M. Rokeach, P. Smith and R. Evans, 'Two Kinds of Prejudice or One?' in M. Rokeach, The Open and Closed Mind, New York: Basic Books, 1960, pp. 132-168.

5. See W. E. Sedlacek and G. C. Brooks, Jr., 'Measuring Racial Attitudes in a Situational Context', Psychological Reports 27 (1970), pp. 971-980, and Sedlacek and Brooks, 'The Measurement of Attitudes of \/hites Toward Blacks with Certain Beliefs', Cultural Study Center Research Report No. 7-70, College Park: University of Maryland Press, 1970.

---

5. See W. E. Sedlacek and G. C. Brooks, Jr., 'Measuring Racial Attitudes in a Situational Context', Psychological Reports 27 (1970), pp. 971-980, and Sedlacek and Brooks, 'The Measurement of Attitudes of \/hites Toward Blacks with Certain Beliefs', Cultural Study Center Research Report No. 7-70, College Park: University of Maryland Press, 1970.

6. For reports on the application of their approach to American student groups, see the Sedlacek and Brooks studies cited above.

7. C. E. Osgood, G. Suci and P. A. Tannenbaum, The Measurement of Meaning, Urbana, 111.: University of Illinois Press, 1957.

8. Study by Brooks and Sedlacek shows that the specific prejudiced referent used has no effect in determining responses. See Brooks and Sedlacek, 'Choice of Racial Referent as a Variable in Racial Attitude Measurement', Cultural Study Center Research Report No. 5-71, College Park: University of Maryland, 1971.

9. The authors wish to thank Georg Ginsberg, Tyge Jantzen, and Svend-Aage Mortensen for their assistance in preparing the Danish instruments and in collecting and coding the data. The project could not have been accomplished without their interest and hard work. Copies of the Danish instruments are available from the authors on request.

10. The sydlandske fremmedarbejdere hypothesis represents the collective thoughts of an empirical methods proseminar i:aught at the University of Copenhagen during the spring of 1971 by the senior author while he was a Fulbright-Hays lecturer there.

11. In Aarhus the questionnaires wore distributed by teachers at the Institute of Political Science under the direction of P, Nannestad Olsen. Administrators at Aarhus were fully aware of the purposes of the questionnaires and followed the same instructions followed in Copenhagen.

12. By seeking a heterogeneous group of this type, we were attempting to replicate the Maryland population originally discussed by Sedlacek and Brooks, op.cit., 1970. Students in their first two years at the university are overrepresented in the present study, and students from economics and law constitute a total of 68 percent of all the participants in the study. Previous work by Sedlacek and Brooks suggests that this; is not likely to be of any consequence, and other analyses of the present data indicate that there are no significant differences among the respondents according to their field or length of study.

13. J. M. Sakoda, G. H. Cohen and G. Beall, 'Tests of Significance for a Series of Statistical Tests', Psychological Bulletin 51 (1954), pp. 172-175.

14. The items which were statistically significant between the Copenhagen and Aarhus participants were as follows: 11, 12, 1;>, 14, 18, 25, and 95. Five of the seven items are clustered on situation II where Aarhus respondents were significantly more 'disgusted by', repulsed by', 'saddened by', 'hostile toward', etc. the rape situation than were Copenhagen students. Complete analyses of variance results are available from the authors on request.

15. In an attempt to identify the source of the differences between Forms A, B, and C, a post hoc comparison, using a method developed by Sheffé, was undertaken. The Sheffé method tells us the number of item differences when comparing Forms A versus B, B versus C, and A versus C. Shef Eé's test is conservative in that it requires relatively large differences between item means for significance. Because of the conservative nature of the test, a .10 level of significance is used in all post hoc comparisons as recommended by Sheffé. See H. T. Sheffé, The Analysis of Variance, New York: Wiley, 1959.

16. By comparison, the results for the original University of Maryland experimental group showed 55 items out of 100 to be significant, using the less conservative two-tailed t test at the .05 level.

17. See Sedlacek and Brooks, op.cit., 1970. The two situations where the American students were not anti-black were situations 111 and VI.

18. Actually, a less kind interpretation of the Form B results could also be suggested here. Danish students quite clearly define negre as an 'outgroup'. If they did not, the number of statistically significant responses to Form B would be much smaller. It is quite possible that a type of 'reverse prejudice' or racial paternalism is being revealed here. If this is the case, the introduction of a larger black population in Denmark could possibly lead Denmark down a path similar to the one being followed in Britain and in certain areas of the northern United States where overt racial prejudices among the white population seem to be growing by alarming proportions. Since the situation remains hypothetical, it is very difficult to predict whether Denmark represents a potential Utopia for blacks or simply another potential pit of white racism.

---

19. Several studies of American prejudice suggest that in the American culture, at least, prejudice is least evident among the highly educated, among unmarried people, and among those who are 'social participators'. On all these criteria, students would be potentially among the least prejudiced groups in a society. See, for example, Robin M. Williams, Jr., Strangers Next Door, Englewood Cliffs, N.J.: Prentice-Hall, 1964, pp. 274-276.

20. A brief discussion of the German minority problem can be found in Kenneth E. Miller, Government and Politics in Denmark, Boston: Houghton Mifflin, 1968, pp. 29—34 and 48-50. Miller also maintains that strengthening the 'Danish-ness' of the German border territory has been a policy of the Agrarian Liberals, the Conservatives, and, later, the Independents in the post-World War II era.

21. See for example Williams, op.cit., pp. 296-298.

22. Yehuda Amir, 'Contact Hypothesis in Ethnic Relations', Psychological Bulletin 71 (1969), pp. 319-342.

23. There is overwhelming evidence to suggest that education and acculturation programs lose much of then- potential effect once overt group differences develop. See, for example, the work of Muzafer Sherif, The Psychology of Social Norms, New York: Harper and Brothers, 1936. A recent five-nation study of interpersonal conflict, which includes Denmark and Sweden, concludes that there is far more than just goodwill involved in resolving conflict. The study finds that limitations in man's ability to process information often prevents him from solving problems in a rational manner. See Berndt Brehmer et aL, 'A Cross-National Comparison of Cognitive Conflict', Journal of Cross-Cultural Psychology 1 (1970), pp. 5-20.