Measuring Social Distance

Emory S. Bogardus
University of Southern California

This STUDY is supplementary to one already reported upon under the caption "Social Distance and its Origins."[1] Social distance, it may be repeated, refers to the degrees and grades of understanding and. feeling that persons experience regarding each other. It explains the nature of a great deal of their interaction. It charts the character of social relations.

The measurement of social distances is to be viewed simply as a means for securing adequate interpretations of the varying degrees and grades of understanding and feeling that exist in social situations. The. measurement exercise and its results indicate the main points for intensive inquiry into human experiences.

In the experiment described here, a modification of a plan first suggested by Dr. Robert E. Park was used. One hundred and ten persons claiming racial descent as indicated in Table I, took part. These individuals were all mature persons of experience, being of two groups, either young business men, or public school teachers.

Table 1: Racial Descent
English 60   Chinese 3
Germans 31   Swedish 3
Irish 28   Norwegian 3
French 24   Japanese 2
Scotch 22   Jew-German 2
Dutch 15   Spanish 2
Scotch-Irish 17   Jew-Russian 1
Welch 6   Filipino 1
Canadian 4   Pole 1
      Indian 1


Document I is a marked copy of the form that was used. The first sentence was read slowly to the members of each group and its chief points carefully explained. Each per-son was then asked to proceed with the marking, putting a cross under each one of the seven groupings to which he would admit Armenians, for example, and so on down the list of races without stopping.

One of the first questions to be raised is: In how many groupings in our country may the members of any race, (as a class, be admitted, as judged by the ratings of the 110 judges using the arithmetic mean) ? By referring to Document I it will be seen that the Armenians would be admitted by the specific person who made it out, to only one group, namely, the visitors' group, while the English would be admitted in five groups. In the first case the index to the social contact range is 1.00; and in the second instance, 5.00. The social contacts open to the English immigrant are five times as various as those open to the Greek. The Greeks, it may be noted, would be admitted to no groups within the United States, and thus the social contact range (S. C. R.) index in their case would be .00.

Reference to the second column of Table III shows that according to the 110 raters the social contact range varies from 1.18 for the Turks to 4.60 for the English, while the social contact range accorded the Italians is 2.26, which is intermediate between the extremes. A significant correlation is at once obvious between racial membership of the raters and the extent of social contact range is to be noted in Table III.[2] Where the racial membership is low and the range high, as in the case of the Canadians, the relationship of the Canadians to the English and other "high" races among the raters is the chief explanation. Sometimes, as indicated by subsequent interviews with the



Social Distance

According to my first feeling reactions I would willingly admit members of each race (as a class, and not the best I have known, nor the worst member) to on or mor of the classifications under which I have placed a cross (x).

  7 654321
  To close kinship by marriage To my club as personal chums To my street as neighbors To
employment in my occupation in my country
Citizenship in my country As visitors only to my country Would exclude from my country
Armenians           x  
Bulgarians           x  
Canadians x x x x x    
Chinese           x  
Czecho-Slovaks           x  
Danes       x x    
Dutch       x x    
English x x x x x    
French x x x x x    
French Canadians x x x x x    
Finns       x x    
Germans x x x x x    
Greeks             x
Hindus             x
Hungarians           x  
Indians (Amer,)       x x    
Irish x x x x x    
Italians       x x    
Japanese             x
Jew-Germans           x  
Jew-Russians           x  
Koreans             x
Mexicans           x  
Mulattos             x
Negroes             x
Norweigians x x x x x    
Portuguese           x  
Filipinos           x  
Poles           x  
Roumanians x x x x x    
Russians           x  
Serbo-Croatians           x  
Scotch x x x x x    
Scotch-Irish x x x x x    
Spanish           x  
Syrians             x
Swedish       x x    
Turks             x
Welsh       x x    

1. Your fatherís races : English

2. Your motherís races: Scotch-Irish

( 302) raters, a fellow-feeling was aroused primarily by a racial group name, such as French-Canadian.

The S. C. R. index does not indicate merit or traits of the respective races, but rather something of the extent of the social contacts open to each race. The smaller the range of contacts accorded a race, the less, presumably, the opportunities for accommodation and assimilation. The social contact range all indicates something regarding the racial attitudes of the raters. In this way the attitudes of different groups of raters can be compared. For example, the 110 raters participating in this experiment may be divided into two groups, one of the business men and the other of public school teachers. Table II gives a comparison of the S. C. R. indexes accorded certain selected races.

Table II Comparison of Social Contact Range Indexes Accorded Different Races by Business Men and by Public School Teachers (Samples)
Races   By
Business Men
  By Public
School Teachers
Armenians   1.26   1.97
Chinese   0.8   1.61
English   4.33   4.71
Germans   3.03   3.07
Hindus   0.56   1.58
Japanese   0.63   1.71
Negro   0.83   1.58
Turk   0.6   1.51

(303) is excluded from all other group contacts. In other words, the order of the seven-fold classification from left to right seems to constitute (further experimentation is needed) a gradation in social contact distance. If this be true, then an index may be worked out. The "best" index would be the lowest or .00, that is, no groups removed from the individual rater, and the "poorest" would be the highest, 6.00, or six groups removed.

Column III in Table III gives the arithmetic mean of the ratings regarding; social contact distance of the 11.0 persons participating. The Canadians are given the closest "contact" possibilities with a social contact distance (S. C. D.) index of .30, while the Turks are put the farthest away with a S. C. D. index of 4.80. In other words, when we compare the S. C. D. index with the S. C. R. index a high correlation is evident, that is, the best, or highest S. C. R. indexes parallel with minor variations the best, or lowest, S. C. D. indexes.[3] A. long "contact range" is paralleled by a short "contact distance," and vice versa. That is to say, the Canadian immigrant is doubly fortunate, and the Turkish immigrant is doubly unfortunate, for the Canadian immigrant is not only admitted to a large range of group contacts, but he is admitted to the most intimate groups; the Turk, on the other hand, is admitted only to a small range of contacts and these are of the most remote and least intimate types.

Table V Highest and Lowest S.C.D. Indexes (Samples)
Lowest       Highest    
Canadians   0.30   Turk   4.80
English   0.40   Mulatto   4.44
Scotch   0.69   Serbo-Croatians   4.43
Scotch-Irish   0.75   Japanese     4.30
Irish   0.93    


Again, if the 110 raters be divided into two groups of business men and of teachers, the former are found to put nearly all the races a greater distance away than the latter do.

Table VI Comparison of Social Contact Distance Indexes Accorded Different Races by Business Men and by Teachers (Samples)
Rates By Business Men By Teachers
Armenians 4.16 3.27
Chinese 4.83 4.01
English 0.63 0.14
Germans 1.76 1.53
Hindus 5.40 2.97
Japanese 5.16 4.02
Negro 4.63 3.86
Turk 5.23 4.57

Reference to Table III will show as in connection with the S. C. R. indexes that the S. C. D. indexes parallel the racial descent numbers of the 110 raters. The "English descent" raters would not only admit English immigrants to the largest range of contacts but also to contacts of closest intimacy. Serbo-Croatians, for example, are not represented among the raters at all; they are given few contact opportunities and these of remote types. The chief significance of the S. C. D. indexes is that they raise questions regarding the nature of the personal experiences of each of the 110 raters whereby they admit some races to close proximity and put others "far away."

Since the S. C. R. and S. C. D. indexes are largely quantitative, the question arises whether or not a qualitative index may be derived. By assigning arbitrary values to the worth of the social contacts in each of the seven classifications (Document I), namely, a value of 7 for grouping 1; 6, for grouping 2 ; and so on, it is possible to work out what

( 307) may he called a social contact quality (S. C. Q.) index. By referring to Document I it will be seen that the S. C. Q. index for both the Armenians and Bulgarians is 2, but for the English it is 25, the sum of the "quality" units in each of the first five groups. The S. C. Q._ indexes, therefore, might run from 25 (for any race might be put in each. of the first five groupings) down to 1 (the contact quality value given to grouping 7).

Table VII Highest and Lowest Social Quality Indexes (Samples)
Highest     Lowest  
Canadians 22.51   Turks 2.91
Scotch-Irish 23.05   Hindus 3.08
English 22.35   Mulattoes 3.62
Scotch 20.91   Koreans 3.54
Irish 19.38   Negroes 3.84

The arithmetic mean of the S. C. Q. indexes of the 110 raters for each of the races is given in the fourth column of Table III. The five highest. and the five lowest S. C. Q. indexes are given in Table VII, with the Canadians at the top, and the Turks at the bottom, as in the case of S. C. R. indexes. The S. C. Q. indexes take into consideration more factors than either of the other indexes ; they may be considered as representing a summary of the factors included in both the other indexes. As in the case of the other indexes, they call for personal experience data from each of the raters as a means of interpretation. They also require extensive experimentation, especially in connection with the assignment of the values assigned to each of the seven groupings.

Chart I is an attempt to visualize the S. C. R. indexes, the S. C. D. indexes, and the S. C. Q. indexes, in one pic-

( 308) -ture. The range of contacts is indicated by the length of the black lines; the social contact distance by the dotted lines; and the social contact quality by the figures to the right.

 insert Chart I


  1.  Journal of Applied Sociology, January-February, 1925, pp. 215-225.
  2. Because of lack of space, the actual correlation figures are omitted.
  3. Again, for lack of space reasons, correlation figures are not given.

Valid HTML 4.01 Strict Valid CSS2