The E means in Table 19(V) do not vary c"onsistently in
relation
to father's income.
Adorno-T-Authoritarian-Personality-Harper-Bros-1950
The large but heterogeneous group-mostly women-with no income is also relatively high on E (4.
10).
Among the Working-Class Men (Form 40), E mean decreases as income increases up to $5,ooo, after which the E mean goes up again.
Among the Middle-Class Men (Form 40) and the University of Oregon and University of California Student Men (Form 6o), on the other hand, the $5,ooo-$w,ooo group is by far the lowest on E.
Thus, there is no simple relation between income and ethnocentrism, and the relation between income and ideology may well be different for the middle class as compared with the working class.
It is of some interest that in the two groups of nonstudent women (Psychiatric Clinic and Middle Class) those with no income were considerably more ethnocentric on the average than those with some income.
The lower E mean in women who work may be due to their economic position; it is more likely, however, that the personality trends which lead to nonethnocentrism tend to produce also the willingness or desire to have gainful work.
It seemed that expected income might yield a better measure of economic aspirations (and perhaps of class identification) than did present income. The mean E scores for groups divided on the basis of expected income (ten years from now) are presented in Table IS (V). The two largest groups are those expecting $5,ooo-$w,ooo and $3,ooo-$3,9oo, in that order. The over-all totals for all samples combined show that E mean gradually increases as expected income increases. However, the data for individual samples reveal a more complex state of affairs. The E mean for the three highest income groupings ($4,ooo and above) is greater than that for the low income groupings mainly because the ethnocentric Maritime School samples (Forms 45 and 40) form the bulk (5o-7o per cent) of these groupings. It will be noted that within each Maritime School sample there is no clear-cut relationship between expected income and E mean. Similarly, there are no consistent trends in the other samples. Thus, for Form 78 the $4,ooo-$w,ooo levels are slightly but not significantly more ethnocentric than the lower levels, but the lowest E mean is for the "above $w,ooo" level. For Form 6o, on the other hand, the variations in E mean are small (2. 97 to 3? 34) and unrelated to income. Dif- ferences among samples are, therefore, much greater than differences among actual or expected income levels. 16 These results have little if any bearing
16 Similar negative results were obtained in preliminary (unpublished) studies of E in relation to desired income.
? 200
THE AUTHORITARIAN PERSONALITY TABLE 18 (V)
MEAN A-S OR E SCORESa FOR GROUPS HAVING VARIOUS LEVELS OF EXPECTED YEARLY INCOME
Group
Groups taking Form 78:
U. C. Public Speaking Class Women U. C. Public Speaking Class Men Extension Psychology Class Women Professional Women
Totals:
Groups taking Form 60:
University of Oregon Student Women
Univ. of Oregon and Univ. of California Student Women Univ. of Oregon and Univ. of California Student Men
Totals:
Groups taking Form 45:
Maritime School Men Psychiatric Clinic Men
Totals
Group taking Form 40:
Maritime School Men OVer-all totals:
Expected Income Below $2,000-
$2,000 2,900 N Mean N Mean
on the theory that economic forces play a basic role in creating a setting for the development of ethnocentrism; but they provide evidence against the hypothesis that economic level and economic motives per se operate as major psychological forces impelling the individual in an ethnocentric or anti- ethnocentric direction.
A further hypothesis to be considered is that prejudice is determined by the economic level of the parents. Stated most simply: "A person growing up in a rich family is more likely to be prejudiced than one growing up in a middle- or low-income family. " In order to make a partial test of this hypoth- esis, a question regarding father's income was included in the questionnaire, and the mean A-S or E score was obtained for groups representing several income levels. The data are presented in Table 19(V). The number of cases
6 3. 08 26 0 5 0 4
14 2. 91 11 20 2. 97 46
2. 99 3. 74 3. 80 2. 65
3. 06
3. 27 3. 30 3. 07
3. 27
4. 16 3. 30
3. 94
5. 22 3. 37
2 2. 99 6 3. 90 2 1. 99
9 9 2
10 3. 34 20
3 5. 17 9 0 3
3 5. 17 12
1 5. 20 16 34 3. 34 94
? POLITICO-ECONOMIC IDEOLOGY AND GROUP MEMBERSHIPS 20I
$3,000- $4,000- 3,900 4,900 N Mean N Mean
21 3. 18 6 4. 32 9 3. 00 10 3. 52 4 2. 10 3. 80
14 2. 84 3 2. 07 48 2. 95 20. 3. 56
6 3. 54 1 2. 66 16 2. 91 4 4. 03 12 2. 75 7 2. 70
34 2. 97 12 3. 14
27 4. 57 14 4. 52 83. 5934. 73
35 4. 34 17 4. 55
Expected Income $5, 000- Above
10,000 $10,000 Blank NMean NMean NMean N
None over-all Mean N Mean S . D .
3. 11 140 3. 32 1. 43 52 3. 34 1. 48 42 3. 40 1. 36 63 2. 57 1. 37
3. 11 297 3. 18 1. 46
28 5. 23 20 5. 05 44 5. 06 10 145 3. 73 69 4. 16 193 4. 09 31
113. 62 22 3. 55 4 2. 35 4 2. 10
41 3. 31
1 2. 82
31. 91 24 3. 12
28 2. 98
70 4. 46 10 3. 45
0 56 1 1. 20 5 0 29 3 3. 07 14
4 2. 60 104
0 19 0 8 4 3. 03 6
4 3. 03 33
12 4. 38 2 1 3. 40 0
3. 45 14 2. 76 0 3. 66 0 2. 04 0
3. 28 14
3. 50 9 3. 71 8 3. 03 0
3. 47 17
3. 95 41 25
3. 95 66
45 3. 33 142
3. 55 2. 95
47 3. 42 1. 38 54 3. 24 1. 29 57 2. 93 1. 25
80 4. 34 13
4. 30 2
5. 60 0 4. 34 139
3. 27158 3. 18 1. 31
3. 98178 4. 36 1. 60 3. 70 50 3. 67 1. 59
3. 87228 4. 20 1. 62
4. 84 164 5. 08 1. 76 4. 03 847 3. 82
aThe following scales were used in the various forms:
Form 78: A-S Scale (12 items)
Form 60: Form 45: Form 40:
E Scale E Scale E Scale
(10 items) (10 items) (5 items)
in the various income levels provides another indication of the largely middle- class character of the total sample. The $5,ooo-$ro,ooo group was the largest, with 205 cases. The $z,ooo-$z,9oo and $3,ooo-$3,900 groups, which during 1944-46 would probably have been the largest in the general population, had 154 and r86 cases, respectively. A disproportionately large number, 55, were in the "$ro,ooo and above" group. The interpretation of these data is com- plicated by the fact that the father's income now may not b~ what it was
? 202
THE AUTHORITARIAN PERSONALITY TABlE 19 (V)
MEAN A-S OR E SCmESa F<Jt GROUPS WHOSE FATHERS HAD VARIOlE Il'? 0MES
Groups taking Form 78:
u. c. Public Speaking Class Women u. c. Public Speaking Class Men Extension Psychology Class Women Professional Women
Totals:
Groups taking Form 60:
University of Oregon Student Women
Univ. of Oregon and Univ. of California Student Women Univ. of Oregon and Univ. of California Student Men
Totals:
Groups taking Form 45:
Maritime School Men Psychiatric Clinic Men Psychiatric Clinic Women
Totals:
Groups taking Form 40:
George Washington University Student Women Maritime School Men
Middle-Class Women
Middle-Class. Men
Working-Class Men Totals:
Over-all totals:
$2,000- 2,900 N Mean
17 3. 35 6 3. 33 2 4. 25 6 2. 30
31 3. 20
7 3. 08
9 3. 17 11 3. 28
27 3. 19
26 4. 09 4 2. 88 4 3. 70
34 3. 90
7 3. 20 34 5. 48 9 3. 29 8 3. 35 4 5. 40
during the subject's childhood. It should also be noted that almost half of the subjects left this question blank; it is not possible to say how much this has influenced the results.
The E means in Table 19(V) do not vary c"onsistently in relation to father's income. They show negligible and unsystematic variations (from 3? 77 to 3. 92) among the various levels below $w,ooo. However, the group whose fathers earned $ro,ooo per year and above is significantly less ethno- centric than the combined lower income levels (means of 3? 35 and 3. 84
Below $2,000 N Mean
8 2. 94 2 5. 45 1 4. 20 2 5. 65
13 3. 84
5 3. 45 0
6 2. 60
11 2. 99
12 3. 88 6 3. 62 5 4. 36
23 3. 92
2 4. 80 6 5. 57 7 3. 46 3 2. 87 5 4. 00
23 4. 16
70 3. 84 154 3. 92
62 4. 62
? 28 2. 99 9 10 3. 18 5 7 3. 17 0 3 2. 40 0
48 3. 02 14
8 3. 17 0 93. 144 8 3. 17 5
25 3. 16 9
31 4. 48 12 6 3. 03 2 5 1. 66 2
42 3. 94 16
13. 4. 11 10 30 5. 01 14 10 3. 30 5
3. 83 45 3. 34 140
64. 63 5 12 4. 00 0
4. 89 164 3. 96 154 4. 00 69 3. 41 59
4. 08 578 3. 84 1332
5. 08 1. 76 3. 64 1. 96 3. 89 2. 08 3. 83 1. 72
4. 19 1. 90 3. 82
71 4. 40 34
POLITICO-ECONOMIC IDEOLOGY AND GROUP MEMBERSHIPS
203
$3,000- $4,000- $5,000- Above Blank OV er-all 3,900 4,900 10,000 $10,000
NMean NMean NMean NMean NMean N Mean
S. D.
1. 43 1. 48 1. 36 1. 37
1. 46
1. 38 1. 29 1. 25
1. 31
1. 60 1. 59 1. 60
1. 63
3. 97 22 3. 42 13 4 10
3. 77 49
10 3. 13 11 2. 56 8
2. 81 29
4. 68 28 3. 25 7 3. 95 6
4. 41 41
3. 64 35 4. 73 29 1. 84 9 4. 56 10
3 3. 96 86 186 3. 77 73 3. 88 205
3. 32 11 3. 39 5 2. 63 1 2. 23 5
3. 06 22
3. 03 1 3. 34 0 2. 66 2
3. 05 3
4. 68 8 2. 97 0 2. 75
4. 11 9
4. 14 7 5. 13 5 2. 64 5 4. 12 4 5. 67 0
4. 37 21 3. 82 55
3. 34 11 4. 70 27 2. 34 37
3. 42 120
5. 31 16 21 2. 49 17
3. 43 54
3. 54 6f 25 1. 60 48
3. 32 134
3. 97 58 4. 60 46 1. 72 109 2. 40 33
35 3. 29 281 3. 35 589
3. 03 52 3. 43 42 2. 59 63
3. 10 297
3. 80 47 3. 28 54 3. 01 57
3. 32 3. 34 3. 40 2. 57
3. 18
3. 42 3. 24 2. 93
3. 18
4. 36 3. 67 3. 65
4. . 07
aThe following scales were used in the varions forms: Form 78: A-S Scale (10 items)
Form 60: Form 45: Form 40:
E Scale (12 items) E Scale (10 items) E Scale ( 5 items)
3. 35
4. 40
4. 18
3. 92 71
4. 19 299
158
178 50
4. 12 132 4. 04 1. 58
? THE AUTHOlUT AlUAN PERSONALITY
respectively). Whether this lower mean holds for all individuals whose fathers are in this income group, or only for those individuals who get into organized groups such as those tested, is not clear. Further study may reveal that the lower E mean characterizes those individuals who were born in wealthier families but who tend-presumably for emotionally important reasons-to gravitate toward middle- or working-class groups, occupations, and ideologies. We are led to suspect, on the basis of results in numerous areas, that upward class mobility and identification with the status quo cor- relate positively with ethnocentrism, and that downward class mobility and
identification go with anti-ethnocentrism.
A final "socioeconomic background" factor studied was father's occupa-
tion. Table 2o(V) gives the mean E score for various groups based on occupation of father. The most common occupations (N = 136-169) were: Labor (skilled and unskilled), white collar, and big business-managerial. Small business and professional groups were next in size (N = 95 and 90), and in order of decreasing size we find farmers, engineers, government offi- cials, and religious (ministers, etc. ). With regard toE mean, there are only three groups which deviate more than o. 3 points from the over-all mean of 3. 86. The offspring of engineers are significantly above average, with a mean of 4. 36. On the other hand, the offspring of fathers with religious or govern- ment occupations are well below average (3. 20 and 3. 25). For all other occu- pations differences are minor and even smaller than the differences from sample to sample for any one occupation. No occupational grouping is con- sistently high or consistently low in every sample. Even in the case of fathers with big business and managerial occupations, the E mean varies considerably; it is sometimes below, sometimes above that for the test group from which it was taken. Thus, we find particularly low E means for this occupational group in the Extension Psychology Class and Professional Women, and a relatively high mean for the George Washington University Women (rela- tive to the other occupational groupings in each sample). These variations suggest, as do the data above, that ethnocentrism in the individual is not significantly correlated with many of the socioeconomic groupings which
are commonly assumed (by many social scientists as well as by laymen) to be direct, immediate determinants of ethnocentrism. It is the meaning of the group to the individual rather than membership per se, that helps us to predict his stand on ethnocentrism and other issues.
On the basis of the group membership data presented in this section (Tables 12 (V)-20(V) ), certain hypotheses can tentatively be drawn. Per- haps the first lesson to be learned concerns the danger of stereotyped think-
ing about groups. No broad grouping in this study showed anything approaching ideological homogeneity. This is not presented as a startling discovery but rather as a sober reminder to those who assume a close relation
? Groups
i""". S . ! l~. o
N Mean N Mean N Mean N Mean N Mean
"'
N Mean
N Mean
N Mean
N Mean
N Mean
N Mean
N Mean S. D. 0 I
Groups td ing Fora 78:
u. c. Public Speaking Class Women u.
It seemed that expected income might yield a better measure of economic aspirations (and perhaps of class identification) than did present income. The mean E scores for groups divided on the basis of expected income (ten years from now) are presented in Table IS (V). The two largest groups are those expecting $5,ooo-$w,ooo and $3,ooo-$3,9oo, in that order. The over-all totals for all samples combined show that E mean gradually increases as expected income increases. However, the data for individual samples reveal a more complex state of affairs. The E mean for the three highest income groupings ($4,ooo and above) is greater than that for the low income groupings mainly because the ethnocentric Maritime School samples (Forms 45 and 40) form the bulk (5o-7o per cent) of these groupings. It will be noted that within each Maritime School sample there is no clear-cut relationship between expected income and E mean. Similarly, there are no consistent trends in the other samples. Thus, for Form 78 the $4,ooo-$w,ooo levels are slightly but not significantly more ethnocentric than the lower levels, but the lowest E mean is for the "above $w,ooo" level. For Form 6o, on the other hand, the variations in E mean are small (2. 97 to 3? 34) and unrelated to income. Dif- ferences among samples are, therefore, much greater than differences among actual or expected income levels. 16 These results have little if any bearing
16 Similar negative results were obtained in preliminary (unpublished) studies of E in relation to desired income.
? 200
THE AUTHORITARIAN PERSONALITY TABLE 18 (V)
MEAN A-S OR E SCORESa FOR GROUPS HAVING VARIOUS LEVELS OF EXPECTED YEARLY INCOME
Group
Groups taking Form 78:
U. C. Public Speaking Class Women U. C. Public Speaking Class Men Extension Psychology Class Women Professional Women
Totals:
Groups taking Form 60:
University of Oregon Student Women
Univ. of Oregon and Univ. of California Student Women Univ. of Oregon and Univ. of California Student Men
Totals:
Groups taking Form 45:
Maritime School Men Psychiatric Clinic Men
Totals
Group taking Form 40:
Maritime School Men OVer-all totals:
Expected Income Below $2,000-
$2,000 2,900 N Mean N Mean
on the theory that economic forces play a basic role in creating a setting for the development of ethnocentrism; but they provide evidence against the hypothesis that economic level and economic motives per se operate as major psychological forces impelling the individual in an ethnocentric or anti- ethnocentric direction.
A further hypothesis to be considered is that prejudice is determined by the economic level of the parents. Stated most simply: "A person growing up in a rich family is more likely to be prejudiced than one growing up in a middle- or low-income family. " In order to make a partial test of this hypoth- esis, a question regarding father's income was included in the questionnaire, and the mean A-S or E score was obtained for groups representing several income levels. The data are presented in Table 19(V). The number of cases
6 3. 08 26 0 5 0 4
14 2. 91 11 20 2. 97 46
2. 99 3. 74 3. 80 2. 65
3. 06
3. 27 3. 30 3. 07
3. 27
4. 16 3. 30
3. 94
5. 22 3. 37
2 2. 99 6 3. 90 2 1. 99
9 9 2
10 3. 34 20
3 5. 17 9 0 3
3 5. 17 12
1 5. 20 16 34 3. 34 94
? POLITICO-ECONOMIC IDEOLOGY AND GROUP MEMBERSHIPS 20I
$3,000- $4,000- 3,900 4,900 N Mean N Mean
21 3. 18 6 4. 32 9 3. 00 10 3. 52 4 2. 10 3. 80
14 2. 84 3 2. 07 48 2. 95 20. 3. 56
6 3. 54 1 2. 66 16 2. 91 4 4. 03 12 2. 75 7 2. 70
34 2. 97 12 3. 14
27 4. 57 14 4. 52 83. 5934. 73
35 4. 34 17 4. 55
Expected Income $5, 000- Above
10,000 $10,000 Blank NMean NMean NMean N
None over-all Mean N Mean S . D .
3. 11 140 3. 32 1. 43 52 3. 34 1. 48 42 3. 40 1. 36 63 2. 57 1. 37
3. 11 297 3. 18 1. 46
28 5. 23 20 5. 05 44 5. 06 10 145 3. 73 69 4. 16 193 4. 09 31
113. 62 22 3. 55 4 2. 35 4 2. 10
41 3. 31
1 2. 82
31. 91 24 3. 12
28 2. 98
70 4. 46 10 3. 45
0 56 1 1. 20 5 0 29 3 3. 07 14
4 2. 60 104
0 19 0 8 4 3. 03 6
4 3. 03 33
12 4. 38 2 1 3. 40 0
3. 45 14 2. 76 0 3. 66 0 2. 04 0
3. 28 14
3. 50 9 3. 71 8 3. 03 0
3. 47 17
3. 95 41 25
3. 95 66
45 3. 33 142
3. 55 2. 95
47 3. 42 1. 38 54 3. 24 1. 29 57 2. 93 1. 25
80 4. 34 13
4. 30 2
5. 60 0 4. 34 139
3. 27158 3. 18 1. 31
3. 98178 4. 36 1. 60 3. 70 50 3. 67 1. 59
3. 87228 4. 20 1. 62
4. 84 164 5. 08 1. 76 4. 03 847 3. 82
aThe following scales were used in the various forms:
Form 78: A-S Scale (12 items)
Form 60: Form 45: Form 40:
E Scale E Scale E Scale
(10 items) (10 items) (5 items)
in the various income levels provides another indication of the largely middle- class character of the total sample. The $5,ooo-$ro,ooo group was the largest, with 205 cases. The $z,ooo-$z,9oo and $3,ooo-$3,900 groups, which during 1944-46 would probably have been the largest in the general population, had 154 and r86 cases, respectively. A disproportionately large number, 55, were in the "$ro,ooo and above" group. The interpretation of these data is com- plicated by the fact that the father's income now may not b~ what it was
? 202
THE AUTHORITARIAN PERSONALITY TABlE 19 (V)
MEAN A-S OR E SCmESa F<Jt GROUPS WHOSE FATHERS HAD VARIOlE Il'? 0MES
Groups taking Form 78:
u. c. Public Speaking Class Women u. c. Public Speaking Class Men Extension Psychology Class Women Professional Women
Totals:
Groups taking Form 60:
University of Oregon Student Women
Univ. of Oregon and Univ. of California Student Women Univ. of Oregon and Univ. of California Student Men
Totals:
Groups taking Form 45:
Maritime School Men Psychiatric Clinic Men Psychiatric Clinic Women
Totals:
Groups taking Form 40:
George Washington University Student Women Maritime School Men
Middle-Class Women
Middle-Class. Men
Working-Class Men Totals:
Over-all totals:
$2,000- 2,900 N Mean
17 3. 35 6 3. 33 2 4. 25 6 2. 30
31 3. 20
7 3. 08
9 3. 17 11 3. 28
27 3. 19
26 4. 09 4 2. 88 4 3. 70
34 3. 90
7 3. 20 34 5. 48 9 3. 29 8 3. 35 4 5. 40
during the subject's childhood. It should also be noted that almost half of the subjects left this question blank; it is not possible to say how much this has influenced the results.
The E means in Table 19(V) do not vary c"onsistently in relation to father's income. They show negligible and unsystematic variations (from 3? 77 to 3. 92) among the various levels below $w,ooo. However, the group whose fathers earned $ro,ooo per year and above is significantly less ethno- centric than the combined lower income levels (means of 3? 35 and 3. 84
Below $2,000 N Mean
8 2. 94 2 5. 45 1 4. 20 2 5. 65
13 3. 84
5 3. 45 0
6 2. 60
11 2. 99
12 3. 88 6 3. 62 5 4. 36
23 3. 92
2 4. 80 6 5. 57 7 3. 46 3 2. 87 5 4. 00
23 4. 16
70 3. 84 154 3. 92
62 4. 62
? 28 2. 99 9 10 3. 18 5 7 3. 17 0 3 2. 40 0
48 3. 02 14
8 3. 17 0 93. 144 8 3. 17 5
25 3. 16 9
31 4. 48 12 6 3. 03 2 5 1. 66 2
42 3. 94 16
13. 4. 11 10 30 5. 01 14 10 3. 30 5
3. 83 45 3. 34 140
64. 63 5 12 4. 00 0
4. 89 164 3. 96 154 4. 00 69 3. 41 59
4. 08 578 3. 84 1332
5. 08 1. 76 3. 64 1. 96 3. 89 2. 08 3. 83 1. 72
4. 19 1. 90 3. 82
71 4. 40 34
POLITICO-ECONOMIC IDEOLOGY AND GROUP MEMBERSHIPS
203
$3,000- $4,000- $5,000- Above Blank OV er-all 3,900 4,900 10,000 $10,000
NMean NMean NMean NMean NMean N Mean
S. D.
1. 43 1. 48 1. 36 1. 37
1. 46
1. 38 1. 29 1. 25
1. 31
1. 60 1. 59 1. 60
1. 63
3. 97 22 3. 42 13 4 10
3. 77 49
10 3. 13 11 2. 56 8
2. 81 29
4. 68 28 3. 25 7 3. 95 6
4. 41 41
3. 64 35 4. 73 29 1. 84 9 4. 56 10
3 3. 96 86 186 3. 77 73 3. 88 205
3. 32 11 3. 39 5 2. 63 1 2. 23 5
3. 06 22
3. 03 1 3. 34 0 2. 66 2
3. 05 3
4. 68 8 2. 97 0 2. 75
4. 11 9
4. 14 7 5. 13 5 2. 64 5 4. 12 4 5. 67 0
4. 37 21 3. 82 55
3. 34 11 4. 70 27 2. 34 37
3. 42 120
5. 31 16 21 2. 49 17
3. 43 54
3. 54 6f 25 1. 60 48
3. 32 134
3. 97 58 4. 60 46 1. 72 109 2. 40 33
35 3. 29 281 3. 35 589
3. 03 52 3. 43 42 2. 59 63
3. 10 297
3. 80 47 3. 28 54 3. 01 57
3. 32 3. 34 3. 40 2. 57
3. 18
3. 42 3. 24 2. 93
3. 18
4. 36 3. 67 3. 65
4. . 07
aThe following scales were used in the varions forms: Form 78: A-S Scale (10 items)
Form 60: Form 45: Form 40:
E Scale (12 items) E Scale (10 items) E Scale ( 5 items)
3. 35
4. 40
4. 18
3. 92 71
4. 19 299
158
178 50
4. 12 132 4. 04 1. 58
? THE AUTHOlUT AlUAN PERSONALITY
respectively). Whether this lower mean holds for all individuals whose fathers are in this income group, or only for those individuals who get into organized groups such as those tested, is not clear. Further study may reveal that the lower E mean characterizes those individuals who were born in wealthier families but who tend-presumably for emotionally important reasons-to gravitate toward middle- or working-class groups, occupations, and ideologies. We are led to suspect, on the basis of results in numerous areas, that upward class mobility and identification with the status quo cor- relate positively with ethnocentrism, and that downward class mobility and
identification go with anti-ethnocentrism.
A final "socioeconomic background" factor studied was father's occupa-
tion. Table 2o(V) gives the mean E score for various groups based on occupation of father. The most common occupations (N = 136-169) were: Labor (skilled and unskilled), white collar, and big business-managerial. Small business and professional groups were next in size (N = 95 and 90), and in order of decreasing size we find farmers, engineers, government offi- cials, and religious (ministers, etc. ). With regard toE mean, there are only three groups which deviate more than o. 3 points from the over-all mean of 3. 86. The offspring of engineers are significantly above average, with a mean of 4. 36. On the other hand, the offspring of fathers with religious or govern- ment occupations are well below average (3. 20 and 3. 25). For all other occu- pations differences are minor and even smaller than the differences from sample to sample for any one occupation. No occupational grouping is con- sistently high or consistently low in every sample. Even in the case of fathers with big business and managerial occupations, the E mean varies considerably; it is sometimes below, sometimes above that for the test group from which it was taken. Thus, we find particularly low E means for this occupational group in the Extension Psychology Class and Professional Women, and a relatively high mean for the George Washington University Women (rela- tive to the other occupational groupings in each sample). These variations suggest, as do the data above, that ethnocentrism in the individual is not significantly correlated with many of the socioeconomic groupings which
are commonly assumed (by many social scientists as well as by laymen) to be direct, immediate determinants of ethnocentrism. It is the meaning of the group to the individual rather than membership per se, that helps us to predict his stand on ethnocentrism and other issues.
On the basis of the group membership data presented in this section (Tables 12 (V)-20(V) ), certain hypotheses can tentatively be drawn. Per- haps the first lesson to be learned concerns the danger of stereotyped think-
ing about groups. No broad grouping in this study showed anything approaching ideological homogeneity. This is not presented as a startling discovery but rather as a sober reminder to those who assume a close relation
? Groups
i""". S . ! l~. o
N Mean N Mean N Mean N Mean N Mean
"'
N Mean
N Mean
N Mean
N Mean
N Mean
N Mean
N Mean S. D. 0 I
Groups td ing Fora 78:
u. c. Public Speaking Class Women u.
