No More Learning

31
I. 60 1. 59 1. 60 I. 28
I. 63
I. 58 1. 76 1. 96 2. 08 1. 72
1. 90
2I. 15 0 33. 67 0
23. 70 (57) 3. 79 0? ? (13)4. 25
54. 24 (13) 3. 84 2 1. 65 (12) 3. 44
0 -- 2 ! . 40
72. 30 1. 16
I. 83 2 3. 03
42. 26
33. 20 21. 65 II. 80 I2. 80
62. 45
I2. 80 I 5. 60 3 2. 07 I5. 98 42. 45
0 0
0
0 0
0 0
0 0
1. 00153. 86 8 3. 090 3L13 93. 04 0 -- 123. 93
9 3. 54 (95) 3. 82 (170) 2. 85
POLITICO-ECO~O? v1IC
I D E O L O G Y
A N D
G R O U P
M E M B E R S H I P S
1 8 9
10 3. 02 10 2. 14 23 4. 13
9 5. 02 33 3. 78 17 5. 13 17 4. 11 (191) 4. 88 (288) 3. 69
4. 1158 3. 42 34 4. 73 39 4. 10 (443) 4. 39 (695) 3. 411332 3. 82
27 2. 60 II 2. 04 46 3. 9117
83. 380
00 0 0
0 0 0 0
0 0 0 0
0 0 0 0
74. 86 I3. 300
4. 16
32. 40 22. 700 II. 0053. 34 53. 200 0 74. 64 5 4. 520
13. 40 10 5. 28 2 2. 40 I 2. 80 2 1. 00 2 3. 50 2 3. 90 I 5. 00
0 -- 44. 10 (53) 4. 66 5 5. 64 0 -- 10 5. 44 6 4. 60 (56)? 5. 21
(54) 3. 63 (77) 4. 90 (88) 3. 06 (32) 2. 94 (37) 3. 40
2 3. 70 9 2. 51 2 4. 80 74. 57 0 -- 54. 56
4 5. 15 3 4. 73 (47) 4. 76 II. 20 I1. 20 (24) 4. 92 25. 50 33. 47 (II) 4. 62
593. 83 578 4. 19
"'
,;,
. ,;
'fC:
~
-:. l"'
>- . . . ,
"'
"'- "'-
0
" U,. . :
--
"'
"" c"'"? ? ' -o.
. O" EO
8~
N
. ,"'
2. 77 2. 97
2. 55 2. 76
17 4. 32 13 4. 47 (94) 4. 44 (152) 3. 86
~ for several of the present groups is different, by one or two subjects from the N given for
the same groups in Tables 1? 11 (V), If anN in Tables 12-20 (V) is smaller than theN for the corresponding group in Tables 1-11 (V), it is because one or two subjects who responded to
the PEC scale left blank the whole of page 1 of the questionnaire, In one instance (Extension Class WOmen), two more subjects were available for the analysis of group membership than for
the analysts of scale responses: their questionnaires came in late, aftter the statistics on
scale responses were completed, but still in t. ime to figure in the analysis of group memberships.
of any given political group are not "all alike"; and the fact that an indi- vidual belongs to a particular political group is, in most cases, an insufficient basis for predicting his standing on E.
These intra- and intergroup variations suggest that group membership is not in itself the major determinant of ideology in the individuaL It would appear, rather, that different individuals support a political group for dif- ferent reasons, and that we must understand why an individual chooses to support one group rather than another one. It is incorrect, or at least inade- quate, to say that an individual is prejudiced because he is an Anti-New Deal
(21) 4. 13 (22) 3. 56
(20) 3. 45 (63) 3. 71
(24) (31)
(30) (85)
13 4. 47 6 4. 35 (58) 4. 61
24. 05 I4. 00 (16) 4. II
23. 65 64. 57 (20) 4. 23 (33) 3. 25 43. 05125. II(30)4. 64 (52) 4. 59
~~
(93)
(26) 3. 55
? THE AUTHORITARIAN PERSONALITY
Democrat. He may be an Anti-New Dealer because he was ethnocentric to start with, or, more likely, both the ideology and the group membership must be explained in terms of more basic psychological and social factors. Consideration of these factors may help to explain why some anti-New Deal Democrats are not ethnocentric, and why some New Dealers are.
What of those who profess no preference for any political group (Table rz (V), columns rr-15)? The results for these groups are difficult to inter- pret, but they are suggestive for further inquiry. Four of these groups, the
"undecided," "against all parties," "self-contradictory" and "blank," with means of 3? 9 to 4. 7, are above the over-all mean of 3. 8, while those who consider themselves "nonpartisan" are relatively low on E, with a mean of 3+ Perhaps the most general conclusion to be drawn is that political con- fusion and indifference, as well as opposition to "politics," are associated with greater-than-average ethnocentrism. Since these subjects constitute some ro-r 5 per cent of the present sample, and at least that percentage of the American population, they merit more thorough study.
It was possible indirectly to approach the question of parental influence on subject's ideology by asking for the political party preference of father and of mother. In Table r3(V) the subjects are arranged in groups accord- ing to father's political party preference, and the E mean for each group is shown. 12 The offspring of Republicans are, on the average, slightly less ethnocentric than the offspring of Democrats, the E means being 3? 4I and 3. 68, respectively. Assuming an S. D. of r. 5 for each group, this difference is significant at the 5 per cent level.
These data suggest what everyday observation has often seemed to indi- cate, namely, that people do not necessarily believe what their parents tell them. This hypothesis is neither original nor profound-although we believe that it has profound implications for the understanding of the formation of ideology. It contradicts another commonly held theory, namely that one learns mainly by imitation. The "imitation" theory expects a high correla- tion between parents' ideology and offspring's ideology, on the assumption that one "naturally" (that is, imitatively) takes over parental ideology rela- tively intact. The present data, however, as well as those of many previous studies, e. g. , those discussed by Murphy, Murphy and Newcomb (85), sug- gest that the formation of ideology in the individual is a selective, dynamic process,. in which any ideological pressure from the environment will be accepted or rejected on the basis of the needs and strivings of the individual. Approaching ideological learning in this way, we ask, for example: What kinds of personalities take over intact the views of their parents or other
12 Similar data were obtained in terms of mothers' political preference. These data are not presented here since identical trends were revealed. A theoretical reason for focusing on the father is that politics in the United States seems still to be largely a "paternal" con- cern, just as religion is for the most part the function of the mother in the home.
? TABLE 13 (V)
MEAN A? S OR E SCORESa FOR GROUPS WHOSE FATIIERS HAVE VARIOUS POLITICAL PARTY PREFERENCES
"'
GroUP N
Groups taking forn 78:
u. c. PublicSpeaking Class women 48 u. c. Public Speaking Class Men 20 Extension psychology Class WOmen 17 Professional Women 16
Totals: 101
Groups taking fora 60:
Univ. of Oregon Student Women 18 Univ. of Oregon and Univ. of
California Student WOmen 19 Univ. of Oregon and Univ.
of
California Student Men 21 Totals: 58
Groups taking fora 45:
Maritime School Men 95 Psychiatric Clinic Men :? tl Psychiatric Clinic women 31
Mean N Mean 3. 12 80 3. 53
3. 59 26 3. 45 3. 25 19 3. 64 2. 12 30 2. 56
3. 08 155 3. 34
N Mean N Mean
N 4 2. 73 7 2. 34 140
1. 77 52 3. 36 42 3. ~ 63
2. 88 297
4. 40 47 2. 80 54
3. :? tl 57 3. 37 158
4. 51 178 4. 36 1. 60
3. 60 50 3. 67 1. 59 . . . . .
Totals: 146
9 4. 03 55 4. 11 299 4. 07 1. 63 't":d" l"1 :;. :1 3. 41 5 4. 08 1 2. 30 8 2. 39 18 3. ~ 97 3. 64 754 3. 53 CJl :I: . . . . .
4. 28 86 3. 70 3 4. 03 Over-all totals: 305 3. 68 320
~he following scales were used in the vanous forms:
Form 78: A? S Scale (10 items) Form 60: E Scale (12 items) Form 45: E Scale (10 items)
"'
. . . . \0
Democrat
Republican Socialist O:nnmunist
Other N Mean
None N Mean
Blank N Mean
2. 93
3. 39
24 3. 74 28 3. 19 27 2. 88
2 2. 53 0
3 3. 32 0
3 4
3. 42 1. 38 8 . . -.
3. 24 1. 29
3. 24
3. 19 79 3. 25
2 1. 41 2 2. 16 5
7 2. 55 2 2. 16 12
7 4. 23 32 5. 00 7 1 1. 70 16
2. 93
1. 25
0
4. 39 4. 10 4. 06
42 4. 16 2 4. 70 21 3. 26 2. 70 23 3. 27
"'
Political Preference of Father
. . . . . (""'")'
Over-all 0
I Mean S. D. l"1 (")
0 0 1 2. 30 0
0 z 0
1 5. 90 0
0 0 11. 300 5
3. 32 1. 43
3. 34 1. 48
12. 4. 0 0
2 4. 15 2. 30
0
11. 10 15 1. 30 7 2. 43 30
2. 57 1. 37 3. 18 1. 46
. . . . . 0 l"1 0 t""'
2 2. 50 3
3. 53 71 3. 65 1. 60
l"1
"7
3. 40 1. 36 '""' ('")"
0 3. 18 1. 31 :;. :1
0
~
0
t""' . . . . . .
>
~
'7
CJl
? THE AUTHORITARIAN PERSONALITY
authorities and under what psychological conditions do we find various forms of change or rebellion?
Questions such as those above were raised by the tendency for Republican fathers (presumably more ethnocentric) to have less ethnocentric offspring than did the Democratic fathers. This suggested the hypothesis that "dis- agreement with father" is related to anti-ethnocentrism, regardless of father's political views (see also Levinson and Sanford (71); Murphy, Murphy and Newcomb (85)). The hypothesis was tested by comparing subject's and father's political preference. The results are presented in Table 14(V). The group of subjects whose political preference was the same as their fathers', regardless of party, had a much higher mean (4. 05) than the group of sub- jects who differed from their fathers (mean of 3. 04). The difference is ex-
tremely significant; indeed, it is almost identical with the difference between groups based on liberal vs. conservative party preference (Table 12(V)). In other words, a person's standing on E can be predicted as closely on the basis of his agreement or disagreement with his father's political party prefer- ence (without knowing subject's or father's politics) as it can on the basis of the subject's actual party preference. 13 This is important indirect evidence in favor of hypotheses raised previously (Chapters III and IV), namely, that ethnocentrists tend to be submissive to ingroup authority, anti-ethnocentrists to be critical or rebellious, and that the family is the first and prototypic ingroup. The individual's relation to parental authority, particularly his dis-
position to be submissive or critically independent, appears to be a basic personality trend which partially determines his political party preference and his ideology about group relations.
Data on the average degree of ethnocentrism in various organizations taking Form 40 are presented in Table 15(V). The low-scoring groups, with E means of 1. 20 to 2. 41, are the Labor School men and women, the League of Women Voters, and the Unitarian Church. It is interesting that the working-class Labor School members have an E mean which is con- siderably higher than that for the middle-class members (2-4 to 1. 2), and slightly higher than that for the (middle-class) League of Women Voters and the Unitarian Church. Apparently the middle-class leftists have identified not only with the working class (in their political ideology) but also with
subordinate groups generally. There is, however, some likelihood that working-class individuals may support left-wing political groups without a full acceptance of the underlying social ideology; that is to say, they may
1s The difference would probably have been even greater had we known the fathers' pany-faction preference rather than the simple pany preference. Thus, all New Deal Democratic subjects who gave father's preference as "Democrat" were grouped under "same preference as father. " It is likely, however, that in many of these cases the father was an anti-New Deal Democrat and that thus a real difference between father and son-one which we should expect to accompany lower E scores-was concealed. This inadequacy in measurement makes the obtained differences all the more impressive.
? Totals:
3. 17 11 3. 26
4. 01 158 3. 18 I. 31
4. 52 178 4. 36 1. 60
Groups taking Form 45:
Marit! me School Men
Psychiatric Clinic Men
24 4. 08 27 4. 29 53 4. 62
158 4. 28
12 3. 15 5. 00 18 2. 80 I. 70 8 4. 86 4. 18
50 3. 43 4. 17
14 4. 49 20 6 2. 90
3. 86 7 13 4. 01 16
29 3. 96 32
13 3. 66 10 22 5. 41 14 3. 09 16 9 3. 82 3 4 5. 20 9
57 4. 38 52 113 3. 99 114
4. 51 13 3. 46
3. 39
4.