The Measurement and Motivation of Atypical Opinion In A Certain Group

Floyd Henry Allport and D.A. Hartman
Syracuse University


Terms denoting political attitudes, such as `conservative,' `liberal,' `radical,' and `reactionary,' have long been familiar both in popular usage and in the language of political science. Though sufficiently understood for ordinary discourse, their use is likely to lead to a confusion between a political opinion and the type of person who holds the opinion. There is considerable agreement as to what is meant by a radical view; but is there such a thing as a radical type of personality? If there is, we need a method for the measurement and identification of such individuals. If not (and psychologists are becoming increasingly suspicious of type classifications), we must ask what psychological characteristics in individuals are the most likely to produce a radical trend in political and social convictions. The same observation holds for the other familiar attitudes upon public questions.

A logical procedure would seem to be, first, to measure the distribution of public opinion in a representative sample, and secondly, to select from the various regions of this distribution (conservative, radical, and the like) a sufficient number of individuals for detailed study of the motives and traits of their personalities which give rise to the opinions they hold. This paper will be devoted to a preliminary report on the results of such a study.

The first step, namely, the measurement of the opinion of the

(736) group requires a special technique. It will be clear that the conventional method of ascertaining opinion, the arbitrary vote for or against a proposal, is adapted for practical rather than scientific purposes. Carefully graded and standardized scales are needed for any measurement whether physical or psychological. In order to construct such a set of scales the following method was adopted. Seven concrete issues of current interest were chosen, dealing, respectively, with the League of Nations, the qualifications of President Coolidge, the distribution of wealth, the legislative control of the Supreme Court, prohibition, the Ku Klux Klan, and graft in politics. Sixty students, upperclassmen, were asked to write their personal views on the various phases of these questions. The resulting opinions on each issue were then carefully sifted and the distinct and relevant views were assembled. Keeping the issues separate, these views were printed on slips of paper and arranged independently by six judges, teachers of political science and psychologists, in order of their logical position in a scale ranging from one extreme on the issue in question to the opposite extreme. The average rank assigned to each statement was taken as its final rank in the completed scale.[2]

The scales, so constructed, were given out in quiz-section groups to the freshman class (College of Liberal Arts, Syracuse University) whose members were required to take the introductory course in "Responsible Citizenship." Instructions were given to each student to read the scale carefully, and check the one statement in each of the seven issues which most nearly coincided with his or her own view. With each issue there was provided a place for the student to check the certainty of his opinion on a graphic scale of five steps ranging from `extremely uncertain' to `extremely certain.' A similar scale was provided for checking the degree of intensity of interest or feeling upon the question concerned. The number of opinions obtained in this manner was, as an average for the various issues, 367.

The graphic representation of the results for the prohibition

(737) question (figure 4) will illustrate both the scale and its use as a means of measuring the distribution of opinion. The scale, the steps of which are represented along the base line, begins with the statement that "the present prohibition amendment and interpretative statute are satisfactory, and enforcement should be made more severe." This view is represented in column i, at the left; and the number of subjects accepting it, as shown proportionally by the height of the column, is 56, comprising a percentage of the entire group, of 15.5. At the opposite end of the scale, column XIII, we have the view that "the open saloon should be universally permitted." It has only two adherents, comprising 0.6 of one per cent of the group. The second statement, column ii, is the same as the first, except that a more uniform enforcement is called for. Fourteen and seven-tenths per cent hold this view. The next statement, column iii, calls for upholding the prohibition laws, since, though not wholly successful now, they will be successful in the next generation. This is the modal view of the class, having 28.9 per cent of adherents. The fourth step, suggesting that minor changes will be needed from time to time, was unpopular (1.1 per cent). Steps five and six uphold the principle of prohibition, though admitting that enforcement is difficult or impossible. With step seven we pass over from prohibition to the side of license. The dotted line indicates the point at which the shift in meaning occurs. Steps seven to thirteen represent, successively-state option (vii), home brew of wines and beers (viii), government stores of wines and beers (X), local option (X), beers and wines in cafés (XI), government stores for all liquors (XII), and the open saloon (XIII).

The graph as a whole pictures the overwhelming sentiment for prohibition within this group. It shows also the distribution upon all shades of opinion from one extreme to the other, the small pro-license minority arranging themselves in a little distribution curve of their own, which, except for column x, is practically normal.

The League of Nations issue (figure 1) splits the class into two modes, possibly upon party lines. The scale ranges from

(738) complete adherence to a policy of isolation (column I), on the left, to not only joining the League but helping to abolish nationalism, and form a world state (column XII). The first mode (column ii), comprising 39.7 per cent of the class, votes for Monroe Doctrine, friendly coöperation with other nations, and staying out of the League. Intermediate steps range through referendum votes, triple alliances, international conferences, World Court, entrance to League without responsibility under Article X, to the dotted line which brings us to entrance with full responsibility but upon authority of a referendum (VIII). The second mode (column IX), comprising 24 per cent, stands for unqualified entrance with responsibility, adjusting defects in the League after we enter.

Opinions on the qualifications of Mr. Coolidge for the Presidency (figure 2) are distinctly favorable. The scale ranges from "perfectly fitted for the office," on the left to "sure to bring with him a corrupt government" on the right. None held the latter view. The eulogy becomes milder up to column v, which includes those who think he had not yet had a chance to show his ability; the break to the atypical side coming at column vi which represents him as "a little too conservative." The opinions on this question were obtained before the presidential election.

The proposal of empowering Congress to overrule decisions of the Supreme Court on nonconstitutionality of enactments produces two distinct modes (figure 3). The scale ranges from "the proposal is a menace" (column i) to "Congress should have power to over-rule any decision of the Supreme Court" (column XI). The major mode centers in the view (column iii), that "even though the Supreme Court has made mistakes, none of its power should be given over to Congress." The steps which group themselves about the minor mode represent, successively, a two-thirds decision of the Court on nonconstitutionality (VIII), modification of the Supreme Court (IX), and full adoption of the La Follette plan (x).

The scale on the question of distribution of wealth (figure 5) ranges in five steps from praise of an extreme individualistic and capitalistic scheme (column i) to the utmost socialism and

(739) paternalism (column v). The overwhelming mode (65 per cent) takes the view of recognizing a problem in the present status, but opposing government ownership, and favoring equal distribution of opportunity rather than of wealth. We shall return later to a closer analysis of opinion upon this issue.

In figure 6, relating to the seriousness of the corruption in government disclosed by the recent oil scandals, the curve approaches more nearly than any other to the normal type of distribution. The mode (iii), comprising 47 per cent of the subjects, represents the opinion that there has been a limited amount of corruption, bringing out weak points in our system of government, but that neither the government as a whole, nor the major political parties, are thereby discredited.

The attitudes in regard to the Ku Klux Klan (figure 7) are widely distributed. A distinct mode, at the extreme left, (i) believes the Klan a menace, and needful of suppression. Steps four to seven, inclusive, show a large number, more tolerantly disposed, who believe the Klan unAmerican (IV), unAmerican, but not as bad as painted (V), unwise in its methods (VI), or in a class with other fraternal orders needing legal regulation (vii). A small minority believe it useful (viii), necessary (ix), or absolutely right (x).

With these pictures of the `lay' of opinion before us we may now select whatever portions are of interest for intensive study. Our present treatment deals with the extremes of the scale, sometimes spoken of as reactionary and radical, respectively. These groups are also those whose opinions are in the minority.[3] For this purpose we shall classify them both under the more general term of atypical opinion. Some of the more general symptoms of 'atypicality' at the extremes will first be shown. We shall then proceed to a closer study of the personnel which is atypical upon a single issue, namely, the distribution of wealth. This question was selected both for its traditional significance in radicalism and because it gives us a body of sharply distin-

(740) -guished opinion which is both atypical in frequency, and extreme in its logical content.

We wish to emphasize from the start the limits of conclusions to be drawn from the present data. A college group, of course, is not a representative sample of a typical community. The Syracuse group is also probably too homogeneous, and lacking in the clear-cut radical element to be found in other communities. Our purpose is merely to show the possibilities of the method and outline a few factors which we might expect, upon further investigation, to find operative in larger and more representative groups.


One of the most interesting facts of motivation within our group is one which applies equally to the extremists at both ends of the scale. Let us begin with an a priori interpretation and test it later by the facts. First, it is a safe assumption that much, if not most, popular thinking is based upon emotional urges. Secondly, in any issue upon which the mass of people divide into opposing camps, there is probably some truth upon each side, otherwise so many `reasonable people' would not be converted to one side or the other. The full truth, therefore, probably lies somewhere between the two extremes. Those who approach the very ends of the scale tend to single out one fact and exaggerate it at the expense of a more comprehensive survey of all relevant facts. The hypothesis that these persons are either very dull or ignorant can not be sustained. Another explanation must be found. It is suggested that the atypical extremists are actuated in their thinking by partially repressed emotional drives, and that they develop a method for concealing from themselves and others the fact that their opinion is determined rather by wishes than by the process of reason. Dogmatic certainty and moral conviction are the means adopted to offset, of course unconsciously, the challenge that so extreme a view as theirs should be carefully analysed. To put it another way, if one wishes to hold a certain belief which happens to be of an

(741) extreme sort, one must have a strong conviction. Otherwise, one can not feel justified in holding the view in the face of the great majority who think differently. On a priori grounds we should expect, therefore, that those who stand at the atypical extremes would express the greatest degree of certainty in their opinions.

Now let us look at the facts. In the flat-shaped graph below each curve of distribution, we have plotted, beneath each step of the scale, the average certainty which was felt by the persons who chose the view represented by the step in question. The possible range of certainty scores is from 1 to 5; and the units of vertical distance express the position of the average point of certainty upon this scale.

Conforming to our expectation, we find in the question of the League of Nations (figure 1) that the two extremes, steps I and XII, show the highest certainty of opinion, the averages being 3.1 and 2.9 respectively. At the reactionary end we also find three successive certainty steps upward in direction opposite to the decreasing extremeness of views I, II, and III.

In the graph for the qualifications of Mr. Coolidge (figure 2), there is a slight but steady rise in certainty as we go toward the favorable extreme. The other extremity does not show the tendency, but it may be significant that the average level of certainty for the minority against Coolidge is higher than that of the favoring majority.

In the question of legislative control of the Supreme Court (figure 3), while the greater mode has the higher level of conviction, still our law of certainty toward extremes receives very clear support. Those who consider the proposal a positive menace, the reactionaries, are again of the highest certainty in their opinion. The curve drops for the modal column and rises toward the right extreme of the first mode. Toward the radical end, certainty again rises in four steps paralleling the increase in extremeness of the last five steps of the scale.

In regard to prohibition (figure 4), if we except column x, which is based on only one case, the strength of conviction is clearly lower in the mid-region of the scale and higher toward the extremes. Three well-marked certainty increases accompany the

(742) trend toward absolute prohibition, and two run parallel with the tendency toward the open saloon.

In the distribution of wealth (figure 5); the reactionary wing express the highest confidence in their. opinion, the score being 3.2. The radical extreme does not follow suit; but the great modal position yields less certainty than the two more radical steps just to the right.

Opinion on graft in politics (figure 6) carries at its extremes decidedly the strongest conviction. The step-wise increase is also suggested in columns III, II, and I.

In the curve for the Klan issue (figure 7) we find on the left extreme that three out of four places are high in certainty (though not regularly so); the middle is again low; and the four last steps, which favor the Klan and 100 per cent Americanism, show the usual regular increase of certainty.

We thus find evidence in support of our theory of the relation between atypical opinion and strength of conviction. Reactionary and radical, strong `pro' and rabid `anti' are alike in the fact that they are more certain in their opinions than those who lie at a mid-region of the scale. Some may accept our facts but not our interpretation, and say that the right opinion may really lie at the extreme, and therefore strength of conviction is justified. The man feels strongly because he knows he is right. To this we may answer simply that both extremes of the scale can not be right at the same time. Certainty of the kind here exhibited is not an index of objective truth, but an accompaniment of increasing distortion of truth through narrowed emphasis upon one phase. The man thinks he is right because he feels strongly.

An additional fact of some interest remains to be mentioned. The ratings given for intensity of feeling upon the various questions were found to correlate so closely with the ratings for certainty (coefficient above 0.90), that there was little to be gained by plotting separate distributions. Intensity and certainty go together. Intensity of feeling is thus shown to be correlated with increasing extremeness of view at both ends of the scale.

The political implications of these results, if verified by a

(743) more extensive investigation, are not difficult to see. From the ranks of the atypical, because of their certainty and intensity of conviction, we may expect that there will appear, not only the most aggressive interest in the, ballot, but also powerful leaders borne in upon some wave of extremist agitation. The well-known rule of minorities is thus elucidated by experimental findings.

After certainty of conviction, perhaps the most interesting condition of atypicality disclosed by our results is sex. Since a brief statement must suffice, it may be said that the women of the group avoided the extreme positions on the scale, and formed a higher distribution than the men at the mode or modes of the curve. The men predominated, in proportion to their total number, at both extremes. The only exception to this rule occurred in the prohibition question, where a distinctly greater percentage of women than men chose the end positions in defense of prohibition. The conservative tendency of the women was shown most strongly in the questions of legislative control of the Supreme Court and attitude on the Ku Klux Klan.[4]


In order to come nearer to an understanding of individuals representing different regions of the opinion scale, several group-test and rating forms were given to the freshman class at convocation hours. These included (1) a self-rating study of personality, covering, through the method of checking described degrees of the traits, such fields as home environment, economic status, religion, political opinions of parents, emotionality, ideals, vocational interests, self-reliance, leadership, attitudes toward the sexual relation, insight into self, and social and moral adjustments; (2) a study of personal attitudes (moralistic, conservative, optimistic, cynical, scientific, etc.) conducted through the checking of agreements or disagreements with statements expressing these attitudes in different fields of social relations; and (3) a modified form of Professor Woodworth's

(744) psychoneurotic inventory. Finally, individual interviews were conducted with three small samples of subjects representing respectively the reactionary, conservative, and radical points of view.

We shall discuss briefly certain results of the personality-rating study, referring to a few related findings from the other group-test material. The incidence of the various traits has been studied with reference to their frequencies, respectively, in the left extreme, the mode, and the right portions of the opinion curve for the distribution of wealth (figure 5). (a) The atypicals at the left end of the scale represent the opinion that the present form of wealth distribution is fair and wise. The poor are necessarily poor because of inferior ability or laziness. Ability should be rewarded. Government ownership, or legislative control of wealth will bring chaos. For convenience, rather than with a sense of exactness, we may call this the reactionary position. (b) The great typical group of column a believe that the present system, though not wholly satisfactory, is the best so far devised. Government ownership would discourage initiative and, foster corruption. Opportunity, but not wealth, should be equally distributed. This we shall call the conservative view. (c) The last three steps (iii, iv, v) owing to the small number of cases at the extreme, have been combined into one group. These steps represent increasingly the policy of heavier taxation upon wealth up to the point (column v), of abolishing large fortunes, the increasing governmental control of utilities, and the equalization of reward for services. We may call this combination of steps the radical standpoint.

Unavoidable circumstances limited the number of cases to the following: reactionary group, 21; conservative group 125; and radical group, 35. Though this number is not large enough to be finally conclusive, it may be remarked that the distributions secured from the reactionary group, having only 21, appear to be as regular as those of the conservative group which contains 125 persons. A few of the more suggestive results appear in figures 8 to 16. Upon the base line are shown, by thé small letters, the various degrees of the traits indicated, the maximum

(745) indicated by a plus sign, the minimum by a minus sign. The three groups of individuals are represented as follows: the reactionary with a line composed of long and short dashes; the conservative with a solid line; and the radical with a short-dashed line. The vertical distances indicate the per cent of the group rating themselves as possessing the degree of the trait shown on the base line. The well-known and characteristic errors of self-rating must be borne in mind in the interpretation of these curves.

Figure 9 shows the religious tendencies of the three groups. The ordinate at a corresponds to the rating that `religion plays a vital part in one's life;' b signifies only an `occasional thought' about religion; while c is the position of those who take religion `merely as a matter of form.' It will be seen, both by the large per cent of the radical group (short-dashed line) who rated themselves at a and the small per cent of the radical group rating at c, that for this group religion is a more vital thing than for either of the others. The conservatives stand next; while the reactionary group is lowest in its interest in religion. It should be emphasized that in the distribution of wealth curve the percentage of women who stand at the radical extreme is only slightly less than that of the men; whereas at the reactionary end the men are in marked predominance. This fact may partly account for the relatively low rating of the reactionaries in religious interest.

Figure 10 shows that both reactionary and radical groups rate themselves as distinctly less rapid in talking and walking than do the conservatives. They show, moreover, a similarity in figure 8, `tendency toward emotionality,' where they both rate themselves as less emotional than the conservatives rate themselves. It is impossible, of course, to tell whether these ratings represent facts or psychological 'over-corrections.' Radical and reactionary are also on the same side, rather than opposite sides, of the conservative in reliance upon their own opinion. Figure 11 shows a distinctly higher percentage of both these groups who form their opinions without reference to others (a), and a lower percentage who rely slightly (b), or considerably

(746) (c) upon others for their thinking. The reactionaries are particularly self-reliant.

Figure 12 shows, however, a difference in the self-rating of the atypical groups on expansiveness (that is, the tendency to air one's opinions). The reactionary group follows the conservative distribution, while the radical element, strangely enough, considers itself as decidedly less expansive.

In regard for the approval of others, the reactionary portion again shows its more independent quality. In figure 13 this group is lower than the others in percentage who admit a continual (a), or a moderate (b), catering to the good will of others, and is higher in the percentage who do not care what others think about them (d). Whether genuine or compensatory in origin, this self-rating shows a tendency toward a personal ideal of social conduct which is highly suggestive.

Figure 14 pictures an attempt to measure degree of insight, or appreciation, of one's own level of intelligence, a trait which has acquired some significance in personality study. A rough comparison was made between the individual's self-rating (relative to the whole class) in general intelligence and his actual relative standing in score in the Freshman intelligence test. The ordinate at a shows the percentage of the group overestimating their intelligence, at b those whose estimation was about correct, and at c those who under-estimated their intelligence. It is clear that the reactionary group have the poorest insight, the strongest predilection for over-estimating their ability, and the least likelihood of doing themselves an injustice. The radical group is somewhat better balanced, but still higher than the conservative, and therefore resembling the reactionary in over-self-evaluation. We may perhaps suppose that the very modest conservative element has full insight, with perhaps an exaggerated sense of inferiority for recognized defects; while the atypical element, particularly the reactionaries, repress their feelings of inferiority and assume a compensatory and overcorrectional attitude of self-assurance. Further data are, however, required to substantiate this hypothesis.

The reactionary, as shown in figure 15, has the least objection

(747) to being tested by psychological methods. The radical group is much more sensitive on this score, and shows the greatest reluctance; while the conservative group is intermediate.

Adjustment to the conventional moral code, plotted in figure 16, helps to confirm our suspicion that, in spite of certain differences, there is a similar foundation beneath the motives of reactionary and radical. It will be seen that both lie on the same side of the conservative in a smaller percentage who accept without difficulty the prevailing moral standards (b), and likewise in a greater percentage who consider themselves not bound by this code, but substitute their own notions of right and wrong (d).

There remain a few facts from the results of the group testing which are not shown in the charts. Owing to the tentative nature of the measurements of personality ands interpretation available these findings are not presented in quantitative form, but are intended merely as suggestions. We may first mention those which differentiate the two atypical groups from each other. Failures to check estimates of certain traits may be regarded as due either to resistance against self-analysis or to lack of insight, that is to say, lack of the necessary knowledge of self. It is of interest that, taking the salient questions on personality as a whole, there was found to be a distinctly higher percentage of reactionaries who failed to give a rating of themselves than of either of the other groups. The scores made upon the study of attitudes show that the reactionary group tend to be scientifically minded, snobbish, and somewhat cynical. The radical group, on the other hand, are idealists rather than mechanists. They are inclined, as we have seen, to be religious. They stand low in the score on scientific attitude, and high in moralistic, meliorative, and optimistic outlook upon life. The feminine qualities, conspicuous in sex differences, probably contribute to this characterization. Radical and reactionary have in common the following features. A higher percentage of both radical and reactionary groups failed to answer the question as to what they thought of the physical relation of the sexes than did the conservative. In many cases the question was marked as "too personal." One question was concerned with whether

(748) the subject believed he had the normal ability for making love, or whether he doubted his capacity. Among both reactionary and radical groups there was a higher ratio of the former to the latter than among the conservatives. The reactionaries especially, in the ratio of two to one, asserted their qualifications for the gentle art. The results suggest, but do not prove, the presence of an over-correction factor. Among the radicals 20 per cent failed to answer this question, 12 per cent considering it too personal. How far sex differences play a part in these results has not been ascertained. The cases studied by personal interview suggest that mental conflicts may provide the unrecognized forces behind atypical opinion at both reactionary and radical extremes. Finally, radical and reactionary agree in having a greater tendency than the conservative to differ from what they conceive to be the political views of their parents.


To return now to the problem stated at the opening of this paper, we may inquire, what are the psychological characteristics of individuals which lead them to adopt extreme social and political views. Our procedure has been, first, to measure and describe the distribution of opinion within a group; and secondly, to submit to psychological analysis individuals representing various regions of the logical scale and various degrees of typicality in the frequency of their views. The words reactionary, conservative, and radical have been used only to denote these positions upon the scale, and have not presupposed the existence of types bearing these names. This procedure has been justified by the results, for it seems that both the reactionary and radical, far from being considered as diametric opponents, are better understood by placing them in one group, which we may call the "atypical." That is, while there are characteristic differences between the two, there are fully as striking resemblances.

It must, of course, be remembered that our group was not sufficiently heterogeneous. The radicals may have been radical-sympathizers merely, and the reactionaries were few and overbalanced on the male side. The following conclusions, therefore,

(749) even though stated in general terms, must be taken to refer only to this group, and to serve as suggestions for further investigation.

The kindred nature of the reactionary and radical elements of our opinion curves is evidenced by the following facts. They lie upon the same side, rather than a-straddle, of the conservative group in self-rating on emotionality, rapidity, and self-reliance, in over-estimation of mental ability, possible over-estimation of capacity for making love, failure to react when asked whether they approved of or were averse to the sexual relation, lack of agreement with the conventional moral code, tendency to deviate from the views of their parents, and certainty and intensity of conviction upon a political issue. Additional agreement was found in the attitude study, not reported here in detail, in which radicals checked a number of reactionary items, and reactionaries checked a portion of the radical statements. The profiles show that they share one another's attitudes on diverse questions more fully than the conservative" shares the attitudes of either. The atypical individual, in other words, may be reactionary or ultraconservative in some things and radical in others. In the interviews certain recognized motivations of personality were seen to lead in some cases to the reactionary view, and in some to the radical, according to chance influences and the conditioning of these tendencies in the social environment.

Such apparent contradictions are familiar to most of us. Saul, the violent reactionary, was changed by conversion to Paul, the apostle of a radical gospel. The Ku Klux Klan, violently reactionary in principle, is violently radical in method. Wealthy and aristocratic persons, ultra-conservative by tradition, often become the champions of an agitation against the existing order. An ex-college president of note would radically alter and democratize our present educational system; yet in so doing he would go back for his models to the Athens of Pericles and the cloistered refinement of Epictetus.

It would be unfair, however, to leave unnoticed the differences between individuals likely to take the reactionary view and those prone to radical opinion. The study of mental conflict has taught us to expect irreconcilable differences within the same

(750) person. The reactionary believers of our group are mainly `tough-minded,' and mechanistic. They exceed the radicals in ratings of self-reliance and self-sufficiency, in certainty as shown in the opinion curves, and in lack of insight into their abilities and traits. Their opinions seem to be more decided, and their attitudes more pronounced. They remind us perhaps of the dogmatism of the Fundamentalist, of the 100 per cent American, and of some boards of censorship.

The radicals of our group seem more retiring in nature. They are tender-minded, religious, more aware of their inner motives and conflicts, less self-assertive, more `touchy' in personal matters, more moralistic and meliorative, and more sensitive to the opinions of others.

These differences remind us forcibly of a human contrast widely recognized by psychopathologists. Psychoanalysts divide their patients into extroverts and introverts. Although we have avoided type classifications, we may perhaps recognize in our reactionaries certain traits which coincide with those described for the extrovert; and in our radical we may discover the characteristics of the analysts' introvert. If extroversion and introversion are simply different ways in which people resolve their mental conflicts, does it follow that the common basis we have Found for reactionary and radical is really the existence of conflicts underlying the thinking of each? It is the purpose of the )resent paper to raise this question---but not to answer it.

The opinion scales employed in this study are reproduced below.


Place a cross (x) on the dotted line before the one item which most nearly expresses your own opinion.



Place a cross (x) on the dotted line before the one item which most nearly expresses your opinion. Mark only one item.


Place a cross (X) on the dotted line before the one item which most nearly expresses your own opinion. Mark only one item.



Place a cross (x) on the dotted line before the one item which most nearly expresses your opinion. Mark only one item.



Place a cross (x) on the dotted line before the one item which most nearly expresses your opinion. Mark only one item.


Place a cross (X) on the dotted line before the one item which most nearly expresses your opinion. Mark only one item.



Place a cross (x) on the dotted line before the one item which most nearly expresses your own opinion. Mark only one item.

The following scale of certainty and intensity of conviction was given accompanying each opinion scale:

Place a check above the item which applies to your answer:  
Extremely uncertain;
little more than a


Place a check above the item which applies to your answer:
I have practically no
personal interest or
feeling on this issue:
my opinion is given
My opinion is given
with only moderate
personal interest
in the issue.
I feel very strongly
upon this subject. I
am intensely interested in seeing the policy I have marked be put into effect.

Figure 1


Figure 2

Figure 3

Figure 4 Figure 5


 Figures 6 thru 8

Figures 9 thru 12

figures 13 thru 16 


  1. Read in abridged form before the joint session of the American Political Science Association and the American Psychological Association, at Washington, December, 1924.
  2. The complete set of scales will be found at the end of this article.
  3. It should be remembered that in irregular curves groups holding minority views may be found also in an intermediate position.
  4. The sex differences shown in the results will be treated in greater detail in a subsequent study.

Valid HTML 4.01 Strict Valid CSS2