E. Differences in Mean F-Scale Score Among Various Groups

We may turn now to a consideration of the mean F-scale scores of different groups. These means have been set forth in Table 12 (VII). It is well to recall here what was stated at the beginning (Chapter I, C),^[16] that since no steps were taken to insure that each group studied was actually representative of a larger section of the population, we are in no position to generalize from the present results on mean scores, however suggestive they might be. (A large-scale community study would be necessary in order to produce a sound estimate of the relative amounts of fascist potential in different sections of the general population. The F scale, we believe, is worthy to be used in such a study, though it would have to be modified somewhat in order to be suitable for groups with little education.) It seems well to recall, too, that the group with which a subject filled out the questionnaire does not necessarily represent a group membership that is significant for the present study. The differences with which we are here concerned are not very large, while the variability within each group is marked. Only rarely is the difference between two groups greater than one S.D. In our view, we should find large group differences in mean F score only when membership in a group has some psychological significance, and this does not seem to be true of most of the present groups. (A study of the F-scale score in relation to group membership factors such as those covered by page 1 of the questionnaire (income, religion, etc.) would probably be rewarding. In view of the high correlation between F and E we should expect results generally similar to those found in the case of the latter scale, but discrepancies would be particularly interesting.) Nevertheless, some important sociological and psychological differences among the present groups are known to exist – indeed some of these groups have been described as »key« groups – and, if the F scale is valid, we should expect differences in mean score that are intelligible in the light of our general theory.

Table 12 (VII)

Mean F-Scale Scores of Groups Taking the Several Forms of the Questionnaire

Group N Mean S.D.

Form 78:

Public Speaking Class Women 140 3.94 .71

Public Speaking Class Men 52 3.72 .57

Extension Class Women 40 3.75 .70

Professional Women 63 3.43 .86

Over-all mean, Form 78 295 3.71 .71

Form 60:

Univ. of Oregon Student Women 47 3.32 .86

Univ. of Oregon and Univ. of

California Student Women 54 3.39 .96

Univ. of Oregon and Univ. of

California Student Men 57 3.82 .93

Oregon Service Club Men 68 3.74 .81

Oregon Service Club Men (A Form only) 60 3.25 .71

Over-all mean, Form 60 286 3.50 .85

Form 45:

Testing Class Women 59 3.62 .99

San Quentin Men Prisoners 110 4.73 .86

Psychiatric Clinic Women 71 3.69 1.30

Psychiatric Clinic Men 50 3.82 1.01

Over-all mean, Form 45 290 3.96 1.04

Form 40:

George Washington Univ. Women 132 3.51 .90

California Service Club Men 63 4.08 1.03

Middle-Class Women 154 3.62 1.26

Middle-Class Men 69 3.69 1.22

Working-Class Women 53 3.86 1.67

Working-Class Men 61 4.19 1.18

Los Angeles Women 130 3.49 1.13

Los Angeles Men 117 3.68 1.17

Over-all mean, Form 40 779 3.76 1.20

Forms 40 and 45:

Employment Service 106 3.74 1.04

Men Veterans

Maritime School Men 343 4.06 .77

Over-all mean, Forms 40 and 45 449 3.90 .90

Over-all mean, Four Forms (78, 60, 45, 40): 2099 3.78

Of all the fourteen groups taking Form 40–45, the San Quentin Inmates obtained the highest mean score, 4.73. This mean is significantly different (C.R. = 3.2) from that of the next highest scoring group, the Working-Class Men, whose mean is 4.19. Between the San Quentin group and the lowes4t scoring group of men (Los Angeles Men, M = 3.68) the difference is very marked (C.R. = 7.8). In view of all that has been written concerning the close affinity of criminality and fascism, these results should not be surprising. Since the findings on the »key« San Quentin group are analyzed in detail in Chapter XXI,^[17] further discussion here is unnecessary.

Service Club Men and Working-Class Men do not differ significantly in mean F score. This will come as a surprise only to those who have become accustomed to explaining all important differences in social attitudes on the basis of socioeconomic group membership, and who look to the working man as the main carrier of liberal ideas. It is true, of course, as a matter of economic and social fact, that the crucial role in the struggle against increasing concentration of economic power will have to be played by the working people, acting in accordance with their self-interest, but it is foolhardy to underestimate the susceptibility to fascist propaganda within these masses themselves. For our part, we see no reason to suppose that the authoritarian structures with which we are concerned would be any less well developed in the working class than in other segments of the population. If it be argued that our sample of working-class men might be an unusually reactionary one, the answer is that approximately half of this sample come either from the militantly »liberal« United Electrical Workers Union (C.I.O.) or from classes at the California Labor School, and that there is no reason to suppose that men from the United Seaman's Service or new members of the I.L.W.U. – who constitute the remainder of the sample – are more conservative than working men generally. For that matter, the extremely high scoring San Quentin Inmates come in very large part from the working class, and there is good reason to suppose that their general outlook depends upon their background as well as upon the circumstance of their being in prison.

It appears that differences among the present groups of men depend more upon the factor of contact with liberal organizations and liberal thought than upon socioeconomic group membership. This is the basis on which we would explain the relatively low means of the Middle-Class Men (3.69) and the Los Angeles Men (3.68), both of which are significantly different (beyond the 5 per cent level) from that of the Service Club Men (4.08). The Middle-Class Men and the Service Club Men are quite similar with respect to economic and occupational status; the difference between them that is reflected in their F-scale mean lies, most probably, in whatever it is that disposes the former to appear at a meeting of the P.T.A. or the layman's league of a Presbyterian Church or at evening classes at the California Labor School, and the latter at a Service Club luncheon. This, in our opinion, is primarily a psychological matter; the difference lies in the degree of something which may be labelled, for the moment, a disposition toward liberalism or progressivism or humanitarianism. The Los Angeles Men, it will be recalled, were recruited primarily from the University and the movie communities. Thus, though their socioeconomic status was certainly no lower than that of the Service Club Men in the San Francisco area, the setting in which they were found was one of greater liberalism. The Maritime School Group, made up predominantly of men with working-class and lower middle-class antecedents who are out to raise their status, belongs on the basis of its mean (4.06) with the Service Club Men and the Working-Class Men, while the Psychiatric Clinic Men (M = 3.82) and the Employment Service Veterans (M = 3.74), who probably are more heterogeneous with respect to either class status or liberal affiliations, have intermediate positions in the rank order of means.

It has been pointed out that the fact of the men in our total sample having a higher mean than the women is due primarily to the presence in the male sample of the outstandingly high scoring groups that have just been considered. The present data show that where social group membership is constant, the means for men are not significantly different from those of women. Thus, in the case of the Working-Class Women and the Working-Class Men, the C.R. is only 1.22, while the differences between men and women in the Psychiatric Clinic, the Los Angeles and the Middle-Class groups are practically negligible. It is to be noted, however, that in each case the men are slightly higher, and that in a larger sample the difference might become significant.

Among the women's groups, the only difference that approaches significance is that existing between the Working-Class Women (M = 3.86), on the one hand, and the George Washington University Women (M = 3.51) and Los Angeles Women (M = 3.49) on the other. If a true difference exists, the explanation would seem to be the same as that advanced in the case of some of the men's groups: that the latter groups of women have been in closer touch with liberal trends.

It is of some interest to consider group differences in mean F score in relation to the mean E score of these same groups. In general, groups that score highest on F tend to score highest on E also. The most notable discrepancies occur in the cases of the George Washington Women, who are relatively much higher on E (M = 4.04) than on F (M = 3.51), and the Working-Class Men, who are slightly higher on F (M = 4.19) than on E (M = 3.92). It seems probable that in the case of this group of women, we have to deal with a regional difference: many observers have noted that there is more prejudice in the East than in the West. It may be, therefore, that although these College women were relatively liberal as a group, they were led by the prevailing climate of opinion to go fairly high on E. This is in keeping with the fact that the correlation between F and E in this group was one of the lowest obtained.

The group of Working-Class Men is the only one in which the mean E score is lower than the mean F score. This is probably attributable to the success of indoctrination in antidiscrimination which occurs in the »liberal« unions to which a majority of these subjects belong. Apparently, however, this indoctrination did not go so far as to modify those attitudes centering around authoritarianism, which are more pronounced in this group than in most others. One might say that if this indoctrination were dispensed with, or if propaganda having an opposite direction were substituted for it, then the results from this group would fall into line with all the others.

It has often been suggested that working-class people are relatively uninhibited in expressing the prejudice that they have and that this does not go very deep, while middle-class people are more restrained in giving vent to their – often deeper – prejudice. That nothing to support this formulation is to be found in the present data may be due most largely to the fact that our ethnocentric statements were for the most part fairly restrained, i.e., formulated in such a way that a pseudodemocratic person could agree with them and still maintain the illusion that he was not prejudiced.