Classics in the History of Psychology

An internet resource developed by
Christopher D. Green
York University, Toronto, Ontario
ISSN 1492-3173

(Return to Classics index)

PERSONALITY TRAITS: THEIR CLASSIFICATION AND MEASUREMENT [1]

By Floyd H. Allport & Gordon W. Allport (1921)

First Published in Journal of Abnormal and Social Psychology, 16, 6-40.

THOSE who have been active of late in measuring intelligence have made great progress in the development of measuring scales but very slight progress in the actual definition of intelligence. In dealing with the elusive term "Personality" we may well expect still less satisfactory clarity of definition, no matter what success we have in its measurement. In the measurement of intelligence we have at least the advantage of scales of performance in various mental functions standardized into age or point scale groups. We have, in other words, a means of comparing an individual with his fellows in certain abilities, even though we may not be so bold as to term those abilities intelligence. We may seek, moreover, for a person's mental level in his relative success of adjustment, either to the problems of the school curriculum or to the general problems of life. In this manner a quantitative statement of at least an hypothetical intelligence may be obtained.

The measurement of personality, however, embraces none of these advantages. Individual differences are so great and personal traits so vaguely related to the solution of problems that the notion of an age scale in personality has no significance. Moreover, personalities of divers sorts succeed equally well in the general adaptation to situations of practical life. It may be added that differences of personality are of a qualitative rather than a quantitative sort. These difficulties stand in the way of the development of a personality measurement based. on the correlation between tests and familiar objective criteria such as those of intelligence. We must strive toward a descriptive treatment rather than quantitative. Our aim is personality study and description rather than personality testing.

Since, however, description itself demands a definite point of [p. 7] view, and a definite appraisal of elements, it is necessary to seek some criterion of these elements -- some means of stating as objectively as possible the personality as an entity apart from the tool of analysis which we employ. The true criterion of personality is without doubt to be found in the field of social interaction. We are incapable of giving a complete popular description of personality without indicating the manner in which the personality in question stimulates or influences other human beings and the manner in which the behavior of other human beings produces adjustments or responses in the personality in question. In describing this personality we inevitably take the view-point of those "other human beings." Robinson Crusoe, alone on a desert island, undoubtedly displayed a very measureable degree of intelligence in his adaptation to his environment. It was only with the advent of Friday, however, that his personality could be said to stand forth in its full significance. Not only is the language of personality a social one, but, the problems arising from the interaction of various personalities are in the truest sense social problems. They include every form of social maladjustment -- from the whims of the eccentric to the worst deeds of the criminal. In general it may be said that the aim of personality measurements is the establishing of adjustments between an individual and his fellows which are a benefit to both.

It follows from what has been said of the necessity of a social sphere for the exhibition and description of personality that our objective criterion of what a personality actually is must be the description in quantified terms of that personality by a group of persons with whom he is wont to come into contact. It is with the average of the ratings of such a group of persons that our instruments of analysis must be correlated. Human beings to be sure are subject to prejudice, to quick deduction from scant data, and to great differences in powers of discrimination and judgment. These sources of error must be overcome by improvement in the technique of personality rating; they do not constitute a ground for discarding what must be admitted the only available objective criterion of personality.

Without going into detail as to the removal of these difficulties it may be in place here to note two suggestions of method:

(1) The raters of personality in any given experiment must be fairly numerous, in order to give certainty to the average or median, and to permit the discarding of widely variant ratings.

(2) The raters must have had actual opportunity to observe [p. 8]

Having defined the objective standards of personality to which our methods of measurement must conform, it is necessary now to determine tentatively at least the fundamental traits with which we have to deal. Many of the studies up to the present time have made the error of a superficial and hasty selection of traits. They have been content to rate subjects on such attributes as truthfulness, neatness, conscientiousness, loyalty, perseverance, tactfulness, and the like. It is of course possible to rate individuals with respect to these traits, and for practical purposes such a rating would be valuable. It will generally be found, however, that attributes of this nature are subordinated on the one hand to particular sets of conditions under which perseverence [sic], truthfulness, loyalty, etc., are manifest, and on the other hand to some of the more pervasive, more deeply lying, and far less evident, tendencies of the personality. Neatness, for example, may be due to such diverse causes as (1) the persistence of the parental ideal, and passive attitude toward parental authority, (2) a phobia toward dirt, arising as a defense reaction against infantile habits, (3) the compensatory striving of a plain-looking girl to make herself attractive in all ways possible, (4) an extreme sensitivity to the social behavior and attitudes of one's fellows. Thus we see that the deeper and more pervasive tendencies are of far greater importance than the superficial attributes which themselves are merely the product of more fundamental tendencies in their play upon the particular environment. The currents of the river are more significant than the eddies and bubbles which arise through the irregularities of the river bed and shore.

Character is the interplay of the fundamental personality tendencies in the social and economic environment, as seen from the point of view of ethics and legal right. Though fundamental from the practical standpoint, it is superficial in that there are deeper currents beneath it. The tragedies of Hamlet, Macbeth, and Othello indicate what divers trends of personality may underlie a single act which from the standpoint of character alone would brand each of the three protagonists as a murderer. [p. 9]

Let us now consider a tentative outline of those fundamental and pervasive tendencies which constitute the main currents of human personality. The following scheme has been used in the Harvard Psychological Laboratory as a working basis.

PERSONALITY

I. Intelligence

II. Temperament

III. Self Expression (Strength)

IV. Sociality

While it should be emphasized that the traits of the above list lay no claim to final comprehensiveness, they have however the advantages of being fairly exclusive of one another and of standing for fundamental and dynamic forces underlying behavior. We shall proceed to the task of their definition in connection with the possibility of measuring. them.

At the start the investigator is only vaguely aware of the things he intends to measure, and he can only guess at test problems and procedures which will indicate the traits which he selects. The persons who do the rating with which the test results are to be correlated in their turn have but a feeble notion of the characteristics to be sought, and are inexperienced in interpreting human behavior in the light of [p. 10] these characteristics. It is not surprising therefore that at the start correlations should be infrequent and low. A correlation of .30 or .40, although far below the requirement of a more advanced measurement technique, may be taken with a considerable degree of optimism as justifying a further development of the test.

The ratings upon which most of the correlations of the following account were based were obtained in the following manner: Each student of a class of fifty-five men, representing all academic grades, but particularly Sophomores and Juniors, was given three copies of the Personality Rating Scale shown in Appendix A. These scales were given by the student to three associates with the request to rate him as indicated, and in as frank and careful a manner as possible: Envelopes bearing a distinguishing mark were also given out, and the rater in each case was toy sign his report, seal it, and return it by the student to the experimenter. These instructions were carefully carried out by nearly all of the students, and their hearty co-operation throughout the tests was secured. It may be said, however, that three ratings for each student is an inadequate number to establish the reliability of tests with much certainty.

For each of the subjects the average of the three estimations of, each trait was computed. The names of all the subjects were listed with the averages, one list on a separate sheet for each trait. This procedure facilitated computing the rank and the correlation with the ranking obtained in the test for that trait. It also rendered easier the many trials for ascertaining correlation by inspection with various other data. Owing to the small number of raters, in all cases where there was a discrepancy of twenty-five between any two of the three ratings, the results of the individual concerned were discarded for that trait.

Turning now to the measurement procedure, as might be expected, many more methods of personality testing were devised and tried out than were found in the end successful. The writers lay no claim therefore to any series of personality scales which will adequately cover the classification given above. If they shall be able to demonstrate a few tolerably successful tests with some helpful suggestions for the elaboration of others, they will have accomplished their purpose.

In addition to the procedure described each student was required to write the answers to the questionnaire, selected questions from which are exhibited in Appendix B. It may be remarked that the [p. 11] questionnaire method, tough somewhat under suspicion at present, can be made of service in supplementing the evidence from the tests and ratings. It also provides a glimpse of the personality in action, a human touch which adds color and meaning to the quantitative, analysis of the various traits. Valuable results, however, can be obtained only if the questionnaire is stated in "behaviortistic" rather than "introspective" terms. To ask the subject whether he is honest, moral, thoughtful, literary in tastes, etc., or to analyze himself by inward searching; is only to encounter the obstacles of carelessness; rationalization, and defense reactions. The questions asked, should be in terms of what the subject actually does in his daily life; let the subject judge himself as another person might -- by his habitual behavior.

Let us proceed now to a consideration of the traits and tests in detail. The order will follow that of the classification given in Part II.

INTELLIGENCE

It was considered important to provide a test of intelligence for purpose of correlation with the various other categories. Professor W. F. Dearborn's Group Test of Intelligence, General Exam., No. 5, ;was used. This intelligence test, though excellent for immature subjects for whom it was designed; proved to be inadequate as a test of intelligence of our group. The differentiation in the results of individuals had to be obtained by limiting the time, so that the results indicated little difference but that of ability in speed of performance. There was no correlation between the scores thus obtained, and the ranking in any of the traits. Intelligence, however, should be considered as one of the most significant factors in personality, since it determines the quality and success of so many of the general adjustments of the individual.

TEMPERAMENT

1 & 2 Emotional Breadth and Strength. The traits numbered one and two in the classification are intended to cover a rather modern treatment of temperament. They occupy the second and third places respectively in the personality rating scale (Appendix A). The rather vague notion of quick and slow in emotion is, replaced by the breadth or spread of emotionality. It is thought that the number and variety of objects which to the individual are emotionally toned presents a [p. 12] valuable dimension in the analysis of his temperament. It will be seen later than there are relations between the emotional, tendency and the extroverted and ascendant types.

As to the tests employed there was little that was found to have value. The Pressey Affective Spread Test (see Bibliography) was used, but no significant correlation obtained.

SELF EXPRESSION (STRENGTH)

Under our third large division of 'Personality are included a number of rather loosely related traits indicating the subject's most general type of "ego-expression" and adjustment. The essential contrast made in this division is that between the assertive, self-expressive, and generally expansive and dominant type on the one hand, and the withdrawing, secretive, and yielding personality on the other. The former type is one which a new acquaintance readily grasps, and which he finds convincing and forceful. In popular language he is a man "with a personality." With the latter type one is rarely en rapport, and the personality seems baffling or else purely negative and weak. It must be understood that these traits are by no means always associated in this manner in individuals we meet, but the distinction presented has a certain rough practical value.

Extroversion-Introversion. The extroverted person is one whose mental images, thoughts, and problems find ready expression in overt behavior. Mental conflicts trouble him but little, and he appears to have nothing to repress or to avoid. The introvert, on the other hand, dwells largely in a realm of imagination, creating inwardly a more desirable ideal world rather than adjusting himself outwardly to the real one. He is not always a misfit, however, for given sufficient ability, his internal or covert reactions may be the vision of the poet or artist. On the whole he takes many things too personally, is anxious and self-searching; if not actually afraid of the repressions and conflicts which have not found a salutory neural outlet or resolution.

This valuable distinction, derived from Freudian psychology and first made clear by the writings of McDougall and Jung, has within recent years opened up a remarkable vista for the understanding of humanity. Owing to its importance many tests were sought by us, and many possible correlations scrutinized, but with little result. A general difficulty lies in the impossibility of knowing whether a certain negative reaction in a test is due to a repression or to an actual absence [p. 13] of that element in the individual concerned. This opposition between Freudian and non-Freudian reactions pervades, a great, deal of the work in personality study, and renders many apparently ingenious tests almost impossible to interpret. Another reason for the lack of correlation lies in the ratings by associates. This trait is not ordinarily recognized by any but trained raters who have an intimate knowledge of the subject. It is not so much the absence of overt behavior that marks the introvert as the presence of rich and persistent internal responses; and the latter of course is the most difficult to detect, being accessible only to one who is subtle enough to read the subject from small and unconscious clues.

4. Ascendance-Submission. When a person comes into a face to face relation with another person whom we may suppose is his equal in every way there is generally a real, though sometimes scarcely conscious, conflict between the two egos. This is true both because each is endowed with the fundamental drives of humanity which are essentially self-seeking, and also because two persons are rarely sufficiently alike to establish a perfect harmony of responses. Social behavior is not a process like the movements of cog-wheels in machinery, but a conflict and adjustment of variant attitudes of individuals. In conflicts, generally speaking, there emerge a victor and a vanquished; and this is true of the face to face dual reactions described. One of the two opponents becomes the master; his interest dominates, and he carries his point. The other yields and accordingly is dominated, though by no means always against his will. The former personality we describe as ascendant. In terms of social behavior he is active. The latter is said to be submissive; his attitude in the face to face relation is passive. (This trait is No. 1 in the Personality Rating Scale.)

It must be understood that a person may be ascendant in some situations and submissive in others. The most dominant man among his peers may be thrown into the passive attitude even by the recollection of his parents or early teachers. We may safely say, however, that in dealing with equals, and in the aggregate of their responses, most men may be said to fall in one or the other of these two classes. The principle is well illustrated among children by dynamometric strength contests. In contests of two boys facing each other, each with an instrument in his hand, it is found that, almost at the beginning of the trial, one boy, feels overcome by the masterful strength of the other, and adopts the attitude of making a respectable score rather [p. 14] than of standing at the top. In every walk of life the "leading" type and the "following" type may be readily noted.

Test of Ascendance-Submission. For the purpose of measuring this trait an Active-Passive Reaction Study was devised. This study consisted of the description of a number of typical situations in each of which the ascendant-submissive relation was involved. The subject was asked to react in a spontaneous, emotional manner to these situations, and to write down immediately the way in which he would conduct himself if faced with the conditions described. This type of test, to be sure, presupposes the co-operation of the subjects, and an interest of the part of each in actually analyzing and truthfully presenting his own type of behavior rather than in merely making a good impression.

Following are two of the typical situations presented in the Active-Passive Study. They will serve to give the reader a general idea of the test method as well as of the nature of the ascendant submissive relation which we are now considering.

(2) Suppose you are to years of age now, but with the traits you actually had at that age. You are playing war with some boys the same age. (a) Are they likely to make you fight on the side of Germany? (b) If they do, what will you do about it?

(3) Upon leaving college you become a salesman and are trying to sell a life insurance policy to a middle-aged financier of great note. He says, "Young man, I don't know how long you have been in the game, but you will never succeed unless you acquire more experience and confidence in yourself." What will you say or do?

In the first situation (fighting on the side of Germany) an interesting series of scores was derived. It was found that the individuals who appeared on the basis of their answers in the Reaction Study as a whole to fall in the ascendant group reacted usually by recording an active struggle to keep from being made a German, or at least a refusal to play the game under such humiliating circumstances. The submissive individuals were, in three cases out of every four, those who simply acquiesced and played the rôle of the German. Another response indicating the trait of submission, though to a less degree, was the attitude of acceptance with the idea of making Germany win.

The second situation proved likewise to be suggestive. The answer indicating an undisturbed persistence in trying to sell the insurance scored +12 (i.e. appeared twelve times as often in the replies of the ascendant individuals as in the replies of the submissive); taking the defensive scored +6; acknowledging the criticism with thanks scored -4 (i.e. appeared four times as often in the [p. 15] replies of the submissive individuals as in the replies of the ascendant) ; apologetic attitude, -6; emotional reaction, angry leave-taking, etc. scored -10.

It is significant that when all the situations had been scored there was usually a sufficient incidence of the ascendant and submissive answers in the case of any individual subject to allow for a fairly consistent and certain decision as to the type to which the individual belonged. When the subjects were all ranked from the most ascendant to the most submissive on the basis of these scores the correlation with the ranking by Personality Rating was found to be .40. Considering the tentative nature, both of the situations used and the method of scoring, as well as the inadequacy of the technique of rating, this correlation is high enough to justify further development of tests of this nature.

5. Expansion-Reclusion. This trait is easily recognized, but requires careful definition. As stated in the Personality Rating Scale (trait 8) the expansive person is one whose "ego", or whose "personal touch" enters into all that he says or does. The successful minister or politician, as well as the executive and artist in the general sense, belong to this type. When, however, the person is not gifted, and is by chance aggressive as well as expansive, he is socially objectionable. The reclusive person either consciously or otherwise keeps himself in the background. He fulfills his office in a perfunctory manner without extending himself into his work. He is by no means, however, necessarily secretive or introverted, or even submissive. The expansive person usually writes a questionnaire or other personal documents in a manner charged with personal feeling, thought, and even reference. There is a certain richness in this sort of reaction; we feel a fullness of contact with the person. He may be said to have a definite or outstanding personality. The writings of the reclusive individual, however adequate objectively, are poor in self-feeling and expression. His manner as well as his style leave us unsatisfied or in doubt; we have not made a satisfactory contact with the person. Probably this trait, expansion-reclusion, more than any other is the key-note in the judgment of personality by one's fellows.

Test of Expansion-Reclusion: We endeavored to get a measurement of this quality by asking the subjects each to write a letter answering an advertisement of a position, a letter in which one would be free to tell as much or as little about one's self as desired. The advertisement which was answered was worded as follows:

Wanted: Young men for detective work; good government positions ahead. No experience required. Address, Ganor, St. Louis, Box 777.

The instruction was to answer this advertisement as seriously and with as natural a reaction as if actually seeking the position. The subject could say as much or as little as he chose.

The letters produced were read and scored by a group of twelve graduate students in psychology on the basis of the trait of expansion reclusion. The score of 1 was given to a letter indicating the most expansive personality which they could imagine to occur in a group of 50; the score of 50 indicating the most reclusive. The following expressions of this trait were borne in mind in rating the letters: a considerable number of references to self, statement in detail of qualifications, particularly those of a rather personal sort, and the development of subjective ideas, feelings, and interests in relation to detective work. The absence of these characteristics combined with general meagerness or brevity, and the general impression of conveying little about the personality, were grounds for scoring the letter on the reclusive side of the scale.

Examples of the expansive and reclusive types of letter are given below

Letter of D (Expansive)

Mr. Ganor, Box 777, St. Louis, Dear Sir:

In replying to your ad, I should like to say that I am desirous of undertaking this work because I feel introspectively capable of doing the work. I think that this line of occupation is one that cannot fail to keep up my interest -- is one extremely variegated and opening up new channels of adventure at each succeeding step -- of which, I must say, I am extremely fond. I do not hesitate to say that I am positively certain of being able to do the work knowing as I do that I am gifted for it and shall undertake it, if successful in obtaining the position, in the full confidence that I will make good. I must acknowledge, however, that I have had no practical experience whatever in this line, but I think my lack in this respect will be compensated for by my enthusiasm for undertaking this work. Hoping to hear favorably from you.

Very sincerely yours,

Letter of R (Reclusive) (See Personality Graph IV of same subject).

Mr. Ganor, Box 777, St. Louis, Mo. Dear Sir:

Having seen your advertisement for men for detective work, I am writing you that you may consider me as one of the men desirous of entering that work. At present I am a senior at Harvard University expecting to finish my course about the last of June. Until that time I would not be able to consider entering your service, but if you do not expect to enroll men immediately, I would be glad if you would keep me in mind, knowing that you could count on me surely by the end of June, 1921.

Sincerely,

The correlation between the average ratings of the letters made by the twelve judges and the Personality Rating was .34., which again must be considered suggestive in the light of the crude and tentative methods.

In connection with this, important trait a few suggestions may be added. In certain cases we have known the handwriting of expansive persons to be unusually large and, to use Dr. June E. Downey's expression, uninhibited or impulsive. In regard to motor expression in general, certain expansive persons when tested have shown pronounced tendencies to over-estimation in kinaesthetic reproduction of distances; certain cautious persons are prone to underestimate. The exact relation between these kinaesthetic judgments and the personality trait we are now considering needs to be worked out in greater detail before we can generalize. Another indicator of expansive egotism is found in the number of "I's" used, or other direct references to self, in reports given by the various students in regular seminary courses. The number of ego-references were obtained only when the reports of the individuals were on equally objective topic and were considered per unit of time. A count of this kind, the victims of course being unaware, was made. Certain individuals make with surprising constancy as high as eighty qr [sic] ninety ego-references per, one half hour of speaking; others make as low as four or five; the median approximates the lower end of the scale, being about ten.

6. Compensation: The trait of compensation is not mentioned in the Personality Rating Scale, nor have we any satisfactory tests to offer. Its presence is indicated in thoroughly answered questionnaires, but owing to the fact that it can be understood only through a knowledge of the complete history of the individual, it is extremely difficult to bring to a focus in any measurement procedure. The method of discovering whether compensations are present must take into account two factors: first, the limitation or defect, physical, mental, social, or financial, which was or still is, present in ;the case before us, and secondly, the nature and perseverence of the process of overcoming the difficulty, or satisfying the fundamental demands of ,the ego through vicarious channels. A complete discussion of this trait would take us too far afield. It is not clear whether there is such a [p. 18] thing as a generally compensating personality or whether the compensation is dependent only upon the peculiar circumstances, defects and abilities, of a given case: The self-made man is probably one of the best examples, and suggests a general compensatory attitude. Few of the deep-lying currents of the individual's life are of such importance as this. It is the chief basis of what is popularly known as "character building."

The following case illustrates this process. It was only after a careful study that the compensatory mechanism was discovered. For such practical issues as the one here involved, and for understanding people generally, a test of this trait would be most desirable.

G. L. came to the office greatly agitated over the fact that his mark in psychology course was reported as a D+. He said that he must have a C- in order to be retained on the list of the Federal Board, the government's agency for furnishing an education to young men who had fought in the war and had incurred some wound or other considerable sacrifice. On one previous occasion G. L. had drawn attention to himself. The class was given some group intelligence tests, and he had tried the first problem only, for finding that he was very slow, he had refused to finish the tests, writing at the top a lengthy but rather stupid excuse, saying that he was in no condition, the test would not be fair, etc.

In general appearance G. L. was slow-moving, stodgy, and rather oppressed looking. His complexion was pale, and had the unwholesome inelasticity of putty. His reactions, both mental and physical, were very slow. He was considerably above the age of the average college student. The general impression he created was certainly not favorable; every aspect of his personality was against him. An investigation of his case revealed the following facts:

Born of humble parentage, of low caste socially, with an inferior physique, and exceedingly slow in mental reactions-though not distinctly defective, this young man had struggled for years to maintain employment. His efforts were persistent and conscientious, but without avail. Though faithful and patient G. L. was repeatedly laid off when he failed to keep up with a rush in business. His chief drawbacks in the business world (as stated objectively by himself) were (1) slowness and (2) "lack of personality"

During the war, after serving near the front for two or three months, a shell burst a little distance away from him, and he developed a nervous break down, a typical case of shell-shock with some paralysis. After recovery he was discharged, but entered government employ again as a crude laborer in a munitions factory. One day, his story goes, he was carrying a large shell when it slipped and dropped on both his feet injuring him painfully. After being again discharged, G. L. worked for months to be placed upon the Federal Board and so incur the advantages of the disabled soldier. These advantages G. L. told off earnestly and significantly on his fingers: the government would agree to give the wounded ex-service man a college and technical education, aggregating if necessary 5 years, paying all expenses, tuition, books and living; and if the man made good the government would keep him in some form of employment with good remuneration the rest of his life. Furthermore, and most significant, in all civil service examinations ex-service men of his class would have a 10 per cent. lower passing grade than other men. (just the handicap which G. L. figured that he needed to secure a position in competition with brighter men) [p. 19] The occupation which he himself suggested -- and which in this round about way he was laboring to attain -- was an assistantship of some sort keeping records of some government museum. This certainly would be an ideal sinecure for an educated inferior mind of great slowness.

The mechanism here is fairly obvious. There was in G. L.'s case recognized inferiority, of personality, social caste, physique, and mental reactions, for which he compensated by a -- remarkable program of events, most if not all of which rested more fully in his own hands than he himself would admit. We may perhaps hazard the assumption that the Freudian mechanism of wish-fulfillment underlay his shell-shock experience as well as the accident wherein the disability was incurred which placed him permanently under the protection of the Federal Board.

This shrewdness in so shaping events as to compensate for inherent defects may perhaps not win our full approval. And yet is it not remarkable how effectively the compensatory tendency has operated so as to enable the man by cunning and perseverance to provide a position for himself where he might become a respectable citizen in the career of government employ, rather than the casual laborer, mendicant, or even criminal, to which his defects would otherwise condemn him? In place of the cowardly inertia which appeared at first sight, we find all the dynamic energy of the evolutionary struggle for existence condensed into one human personality. To the Socratic dictum "Know Thyself" he has added "and Compensate Therefor."

7. Insight and Self-Evaluation. The term insight indicates the degree to which a person is able correctly to appraise his traits of personality. (See Trait No. 5 in Personality Rating Scale.) It is derived from the use in psychopathology in which it signifies the extent to which a person realizes that his delusion or hallucination is or has been a fiction of the imagination. A person with good insight is not likely to be deceived by his own rationalization and by the self-extenuation of his acts by refusing to recognize their motives. The possession of this trait or the modification of the personality so as to acquire it has deep practical significance. The process of reformation of a criminal or of character improvement in the socialized individual, is possible only when one's personality is revealed to one's own eyes. The measurement of this trait was provided for as follows: Each of the subjects made out for himself a personality rating scale. In his ratings he differs more or less from 'the average of the estimations' of the three raters in the various traits. The average of his deviations for the various traits gives a practical index of his insight, the lower the average the higher the insight.

Closely related to Insight is the trait of Self-Evaluation, indicated in the Rating Scale as No. 9. The individual here is asked to rate his abilities, for example, his general intelligence, rather than his traits of personality. This estimation, of course, must be made prior to the giving of the intelligence test. The self-evaluation index of each individual is expressed by the difference between his actual rank in the [p. 20] group in intelligence test score and his self-ranking, prefixing a + or -- according to whether he over or under-rated himself.

There are some pertinent conclusion from done in the Harvard Laboratory in pervious similar investigations. An interesting relation exists between the degree of intelligence and the self-evaluation index.  There is a striking tendency for those who are high in intelligence to have a negative index, that is, to underestimate themselves, and for those who are low intelligence, to have a high index. This finding is in harmony with the result of Hollingworth in his self-evaluation in various desirable traits.  The possession of the train in question seemed to insure a more accurate recognition of it, both in self and others; while those low in the desired qualities tended to overrate themselves. The actual inverse correlation between intelligence and self evaluation index which we obtained from an earlier experiment upon thirty subjects was -.67.

The following char, Fig. 1, shows this inverse relation of intelligence to self-evaluation.  The dotted line indicates the relative intelligence of a group of sixteen subjects (according to the Otis Group Tests), and the blocks their corresponding self-evaluation indices.  The horizontal line represents both the median of the Otis scores and the point above which the self-evaluation, and below which this index shows negative or under self-evaluation.

Sociality

The final group of traits in intended to call attention to the individual in his rôle as a unit of society. The self-expressive group dealt with the general types and levels of adjustments of the individual to his world.  Sociality, however, is intended to bring ou the specific reactions which show how much and individual is dominated in his behavior by considerations purely of self and by aggressive self-interest which overrides the interests of other and seems to be incapable of modification by social stimulation and control. The other side of the picture of course is an actual interest in the welfare of others and an inclination to be stimulated by their presence and behavior. In this group lie the foundations of the ethical aspect of character, high ideals, principles, and the like.

8. Social Participation. In this practical field actions must speak louder than words. A most important indication of sociality is the amount of time and energy (not money or good wishes) which the

[p. 21]

[p. 22] individual gives to social activities. By "social activities" is meant not merely attendance at dances, clubs, and the like, where the behavior of others is a matter of secondary interest only; but more especially actual occupation with the guidance, understanding, control, and amelioration of other human beings. It is an interest in the social behavior and welfare as such. Participation is the trait sought, not mere passive interest. It is indicated as trait No. 6 on the Personality Rating Scale.

An analysis of the questionnaire reports indicated that 56 per cent. more individuals in the group expressed an interest in social activities than actually took active part in them. Naturally enough there appears to be no reliable test of this trait other than observing the habits of the person or accepting his statement therefor. A correlation was found between the Personality Rating for extroversion and that for social participation amounting to .4o. In comparing the results of the questionnaire (questions 15 and 17) with the graphs o f personality, to be described later, the following relations were found:

Of the nine reporting participation in social work,

6 were ascendant; 3 were submissive, (2 of these expansive).

7 were of strong type; 2 were of weak type.

By "strong" and "weak" type are meant individuals who were respectively high and low in the group of traits under self-expression taken as a whole.

9. Self-seeking and Aggressive Self-seeking. Selfishness without being distinguished from aggressive selfishness is the tenth trait on the Personality Rating Scale. A test of this trait is much "to be desired. Upon it depends very largely the success or failure of a person as a citizen of the community. When the self-seeking person is incapable of forming habits modifying his natural egoistic tendencies to accord with social influences, and when as a result he pursues these tendencies in opposition to the socially recognized rights of others, he becomes a criminal. It seems quite possible that a tactful adaptation of the principle used in the Active-Passive Reaction Study might afford a useful measurement.

10. Susceptibility to Social Stimuli. A great deal might be written about the degree to which a personality is habitually sensitive to the behavior, suggestions, gestures, emotional expression, and physiognomy of others. Obviously one can neither understand men nor. control his own actions in accord with the social regime unless he can [p. 23] be stimulated to respond to these objects. Inasmuch as such stimulability is a prerequisite for the successful adjustment of an individual to his social environment it may be termed a kind of "social intelligence." Our efforts toward the measurement of this ability have consisted of a test of judging facial expressions. The material and possibilities of such a test, as well as the nature of the process of facial interpretation; are discussed in two articles by Professor. H. S. Langfeld (see Bibliography). We have followed his suggestion and devised a test, not however in the nature of naming the facial expressions, as this involves linguistic ability, but by underlining the name of cite expression most fitting the picture out of a list of eight possible expressions: Fourteen pictures, photographed from sketches of a German actor representing various moods and emotions were chosen as material.

The only correlation which we could detect in this test was' that between ability in judging facial expressions and artistic ability, particularly of the literary type. Of the nine men who made the highest scores in the facial expressions test, seven had literary ability to the extent of actually having produced and gained some recognition by publication or otherwise. There was only one case of a low score in' facial expressions where the individual had also a record of literary production.

Aside from the main purpose of developing tests or measurements of personality one of the most valuable possibilities in a study of this sort is representing in graphical terms the degrees of the various traits which go to make up that unique thing -- a human personality. Personalities, like faces, have no duplicates; each one is a unique mixture of varying degrees of divers [sic] traits. It is at least conceivable, however, that there may be a general agreement of individuals in the rough pattern produced by graphing their qualities. We may perhaps expect certain traits to follow the same general level as others. Where this is true we may be said to have discovered types. Our chief aims, therefore, in graphical representation shall be: first, a picture of the individual personality and its checking up by the questionnaire or other information; second, the discovery of striking or unusual personalities and strongly contrasting personalities; and [p. 24] to find out what general types if any the various combinations of traits may reveal.

Inasmuch as the tests themselves are in too primitive a state of development to justify their graphical representation, the ratings in the various traits plotted on the following charts were based on the average of the three personality ratings of a given individual. The first point in the graph, corresponding to Intelligence, however, was taken from the actual intelligence tests. As suggested above, owing to the fact that the test was not adapted to his particular group, this point on the graph must not be taken too seriously. Both the point corresponding to the degree of insight and the point denoting self evaluation are Personality Ratings on these points and not the indices described above.

The trait of compensation was not rated and hence is not shown in the graphs. Aggressive self-seeking was not differentiated in the ratings or graphs from simple self-seeking. There was a general tendency to rate the individuals too high in this trait, that is, as too unselfish. The true median therefore lies somewhat above that shown on the charts. Susceptibility to social stimuli is indicated by the actual rank in the class in the score of the facial expressions test.

Explanation of the Personality Graphs. The four main groups of traits, namely Intelligence, Temperament, Self-expression, and Sociality are indicated as sections in the order stated from left to right, separated by heavy vertical lines. It will be convenient often to regard these sections as wholes. The facial expression test column is separated from the other members of the sociality group by a heavy line, because it seemed better in this case to leave it out of account in considering social and non-social types.

The manner in which the graphs were plotted is as follows: In the vertical columns are given the various traits, at the top, one extreme (that coinciding with a rating of 1 in the Personality Rating Scale), and at the bottom the opposite extreme (50 in the Rating Scale). The horizontal line through the middle indicates the median rating (25) in the trait concerned. To take an example in the first graph, subject F was among the lowest five in the group in the intelligence test score, was rated two or three above the median in the strength of emotions, about 10th from the top in extroversion, and in self evaluation about 30th from the top (i.e. from the most over self-evaluated). [p. 25]

Following is given a key to the abbreviations used in the vertical columns of the charts:

The seven graphs reproduced below were selected from those of fifty-five subjects and serve to illustrate some of the individual characteristics. When the graphs were sorted two main types were apparent. It was found that the traits under Self expression went fairly uniformly together. There were of course exceptions. On the whole, [p. 26] however, we can distinguish the strong type (e.g. F and S.) and the weak type (e.g. H. and R.), with the self-expression traits as a whole, respectively above and below the median. In the following explanations these two types are designated as extroverted and introverted, this trait being regarded as most salient. A further classification can be made in which are to be found distinct, types, i. e. the social and a-social. The former consists of those having social participation and non-self-seeking above the median, the latter of those having these traits below the median. Four fairly distinct groups (with a few transitional cases) may thus be defined as follows:

Other special relations of traits are indicated below.

A few remarks may be made upon certain relations which were found to exist among the various large divisions of the personality traits. . In order to discover whether a given type of temperament went with a self-expression (or strength) type, the graphs were sorted x into piles on the basis of both the breadth and strength of emotions. The number of cases showing the different sides (extroverted-introverted, etc.) of the self-expressive traits selected, was counted in each pile. In this way any distinct predominances, for example, of the narrow over the broad emotionality in introversion, was shown. Table 1 indicates these numerical relations; and the general description of the emotional traits predominantly accompanying the various traits of self-expression is given at the right. The figures indicate the number of individuals having their emotions characterized by the letter at the top of the vertical columns who belong also to the strength type indicated at the left of the line in which the given figure occurs. [p. 27]

GRAPH I.

Subject F belongs to the extroverted social type. The narrow breadth of the emotions accompanied by strength is, as will be later shown, characteristic of the strongly extroverted type. The extroverted social type is more numerous than the extroverted a-social.

GRAPH II.

Subject Fo illustrates the extroverted a-social type. His personality does not seem to be as strong as that of F owing to the slight reclusion.

Note again the rise of the curve in emotionality, from N. to S. Emotions.

Questionnaire says he is "easily embarrassed." This fact as well as the low sociability seem to accord with reclusive (but not introverted) habits. [p. 28]

GRAPH III.

Subject H. belongs to the rather exceptional introverted social type. He is somewhat abnormal, almost a psychopathic personality. Neurotic heredity.

This subject, possessing some insight, is very depressed as well as introverted. His low S. E. index would indicate this. He is totally pessimistic, saying in his questionnaire that he would commit suicide if it were not for his parents. Life holds no interest or attraction for him.

The social interests are to be explained probably as persistences of strong social welfare leanings, political interests, and boys' club activities around the age of ten and in high school. Since entering college he has lost interest in the world about him, become depressed and apathetic. This marked introversion probably was connected with adolescence.

H's questionnaire reveals some very enlightening facts. First of all, he asked his physician if he ought to write the answers, because having just recovered from a nervous collapse he thought it would be unwise to dwell on his personal troubles too much. He is bashful in the presence of his elders and with girls. It is difficult for him to recover his composure. He feels a nervous tingling and a burning of the cheeks for two or three hours when angry, and does not trust himself to say or do anything for a time. Emotions easily upset. Has day dreams of physical prowess (is actually a frail, weak youth) in which he is fighting a crowd, kicking a goal with a football from any angle within the fifty yard line, shooting goals in basket ball with his back to the basket. States himself these fantasies may be a compensation for poor physique. Has many "fussy" mannerisms, such as biting or tearing nails, biting lips, tearing papers, etc. The sex relation "shocks his sensitivity -- unpleasant mental images, etc." He is entirely apathetic toward girls. Has only one girl friend who is five years older than himself. His roommate has had a very detailed book on sex on his desk for several weeks, but H. says he 'never troubled to open the cover. In his sex reactions H. is decidedly a-typical. [p. 29]

GRAPH IV.

Subject R, is an example of the introverted a-social type. This is a very remarkable personality, the only one in the group below the median in all traits. We may suppose it is very difficult to establish any "contact" with this man. The extreme lowness in both reclusion and under self-evaluation are worthy of note. His answers to the questionnaire are very short and laconic. Socially considered, this is a very poor personality.

The low emotionality may of course be due to repression of outward evidence of emotion. .

The repression and "touchiness" of the introvert are shown in this man by the fact that he was the only subject in the group of fifty-five who considered any of the questions "too personal" to answer. He placed cross marks (x), the designated sign, in place of answers to questions touching his reactions when angry and events in his sex life. [p. 30]

GRAPH V.

Subject S. stands out in marked contrast to subject R. Here is evidently a strong, dominant personality. His questionnaire was particularly expansive, and charged with self feeling and reference, though of an extroverted sort.

In S. we find a combination of literary achievement and an ability to read facial expressions. His perfect self evaluation is worthy of, remark.

In his interests he is active, loving prize fights, games of strength, etc. He writes his disagreement with the author in margins of books.

Probably the clue to this titanic individual is found in compensation. Evidently thrown early upon his own resources he has had all kinds of jobs, some very menial, in which he asserts that he has learned human nature. He has probably done more than this in building up a strong personality. [p. 31]

GRAPH VI.

Subject P. appears from his graph to be a somewhat stormy and troubled soul. And this is the case. His outstanding emotionality and expansiveness, coupled with his moody introversion and self-depreciation give his personality a uniqueness whether for weakness or for strength.

His type is particularly of interest in showing the combination of deep introversion with wide expansion. We all know a few people of this sort who expand and spread their introverted knot of personality out over the social world. Students of this type often show expansion by asking questions in class which owing to their introverted bearing seem to others unintelligible or even absurd. On the other hand a touch of genius might produce from such a combination a Byron or a Tchaikowsky. P.'s questionnaire, ten pages in length, expands in an introverted outcry against the tedium of college life, and the worthlessness of academic studies leading only to "a piece of sheepskin" with his "name on it in Latin." He wants to get away, get married, to go west, and make money. His introversion is best shown in what he says he would do m imagination in a case of balked anger. His imaginary maltreating of his enemy (which he says he would repress in actual life) is indeed fearful to read. When falsely accused of a misdemeanor he tempered his sullen resentment with a mild glory in his martyrdom.

The discrepancy between the intelligence test and facial expression test scores is interesting. [p. 32]

GRAPH VII.

Subject L. shows the combination, also not infrequent, of submission and extroversion. Of the opposite type, ascendant introverted, we have not found a single clear example.

This individual probably presents a ready social contact, having a combination of traits which should render him easy-going and well liked. He is not troubled by exaggerated emotions; he yields readily; following rather than wishing to lead; he is devoid of the baffling and disturbing repressions of the introvert; he expands his ego forming ready points of contact; and he is socially inclined and unselfish. [p. 33]

The tendencies expressed in Table 1 are somewhat more clearly shown in Figure 2, the two extremes of the two emotional traits respectively, being indicated by the letters in blocks. The crossing of the emotional traits in the extrovert and introvert is perhaps the most significant point. Submissive and reclusive differ from introverted in having a narrower range of emotional response. [p. 34]

The answers to certain items of the questionnaire were compared with the personality ratings and with other questionnaire items so as to discover any possible further indications of the inter-relation of personality traits.

Compensations were indicated, mostly arising from childhood tendencies, in about eight or nine cases. They were generally of the athletic sort, and were closely associated with persistent childhood influences (Q. 1) , and with creative ability (Q. 5).

Those who reported the foster child fantasy (Q. 25) strangely enough seemed to be of the extroverted type. They were also somewhat ascendant though often reclusive.

There were about eight cases of a-typical sex response and interest (i. e. eight cases recognized as a-typical from their reports). They consisted of those to whom there was something repulsive about the sex relation and those showing unusual indifference or repression in sex events. Of these eight a-typical sex reaction types:

The following list showing the frequency of occurrence of various types of answers to the questionnaire may prove of interest. Doubtless many of the figures should be higher because of the failure of the subject to remember or truly to analyze himself. The figures indicate that per cent. of the total number of subjects whose answers clearly indicated the presence of the trait or conditioned named.

VII

This paper has endeavored to present a working classification for the study of personality including the main divisions of Temperament, Self-expression or Strength, and Sociality. The traits selected stand, we believe, for fundamental forces in the human life, and form the needed basis of sociability, habit formation, and character. It may be said that personality means the definitely fixed and controlling [p. 36] tendencies of adjustment of the individual to his environment. They generally have a long and important history in the life of the individual. If we include in our definition the adjustment to non-societal objects, intelligence must also be comprised in our classification. Personality, in the stricter sense, however, may be said to be the adjustment tendencies of the individual to his social environment. It is essentially social in its bearings. Of the various traits, intelligence and temperament are probably for the most part inborn. The qualities of self-expression and sociality are probably in the main acquired by the reaction of the hereditary structure of the individual upon his social surroundings. In this field the persistence of childhood and adolescent influences are of vast importance, and the need of an understanding of personality formation by parents and teachers correspondingly imperative.

A well controlled process of rating individuals by associates is probably an adequate means of obtaining an objective notion of a group of personalities with which the results of tests devised for this sort of measurement may be correlated. The most promising suggestions for such tests and studies resulting from our investigation are as follows: a reaction study of miniature situations for ascendancy and other traits; a letter writing, ego-reference, and motor expression test for expansion; a test of judging facial expressions; the insight and self-evaluation index; an objectively and specifically stated questionnaire.

When the traits of personality are either rated or (better if possible) tested, they can be conveniently plotted in graphical form giving a picture of the individual. A refined impression is thereby gained by which we can know and deal with the person in an intelligent manner.

Certain types seemed to fall out from our fifty-five graphs; viz., the strong type of personality (extroverted) either social or asocial, and the weak type of personality (introverted) either social or a-social. We find interesting relations between these major types of personality and temperament itself (emotionality), e. g. the tendency of the extroverted individual to have narrow and strong emotions and of the introvert to have broad but superficial emotions. In general the emotionality of the strong type is of greater breadth and strength than the emotionality of the weak type.

The present classification and research is to be regarded as but the beginning of the investigation of personality. The development [p. 37] of a complete and sensitive instrument of individual measurement for personality as well as for intelligence is a distant but perhaps not an unattainable goal. Progress must be made along two lines': first, the theory and genetic study of personality and the isolation of recognizable traits which are truly fundamental, and second, the practical technique of defining impressions in the rating of individuals and of devising crucial tests of the various traits.

SELECTED REFERENCES

Personality Tests and Scales:

Bureau of Salesmanship Research, Carnegie Institute of Technology. Schenley Park, Pittsburg.

Test Number 9, Will-Temperament Test, Salesmanship Series (Modification of Downey Test.)

Downey, June E., The Will-Profile, A Tentative Scale for Measurement of the Volitional Pattern.

University of Wyoming Bulletin. Vol. 16, No. 4b, November, 1919.

Fernald, G. G.: The Defective Delinquent Class: Differentiating Tests. Am. Journal of Insanity, Vol. 58, No. 4, Apr., 1912. 5.23-592.

Moore, H. T., A Method of Testing the Strength of Instincts.

Am. Journal Psychology, Vol. 27, Apr., 1916, 227-233.

Morgan, Eleanor, and others. An Attempt to Test Moods or Temperaments of Cheerfulness and Depression by Direct Recall of Emotionally toned Experiences. Am. Journal of Psychology, July, 1919. 302-304.

Pressey, S. L. and L. W., "Cross-out" Tests, with Suggestions as to a Group Scale of the Emotions, Journal Applied Psychology, Vol. III, 1919, pp. 138-150.

Pressey, S. L. and Chambers, O. R.,' First Revision of a Group Scale Designed for investigating the Emotions, with Tentatative Norms. Journal Applied Psychology, Vol., IV, i. Mar., 1920, 97-104

Questionnaires and Outlines of Systematic Observation of Personality:

Fernald , G. G.: Character as an Integral Mentality Function. Mental Hygiene, Vol. II, No. 3, July, 1918, 448-462.

Franz, S. I., Handbook of Mental Examination Methods, pp- 170-176. [p. 38]

Hoch and Amsden, A Guide to the Descriptive Study of Personality, Rev. of Neurology and Psychiatry, 11, 1913, 577-587

Partridge, G. E., Outline of Individual Study.

Watson, J. B. Psychology from the standpoint of a Behaviorist, Chs. VI, XI.

Wells, F. L., Mental Adjustments, Chap. VII.

Wells, F. L., The Systematic observation of the Personality -In its Relation to the Hygiene of Mind. Psychological

Review, Vol. 21, 1914. 295-332 (A good classification).

Experimental Researches:

Downey, June E., Some Volitional Patterns Revealed by the Will Profile: Journal of Exper. Psychology. Vol. III, NO. 4, Aug., 1920, 28I-30I.

Hollingworth, H. L. Vocational Psychology, Chs. II, VI, VII.

Langfeld, H. S., The Judgment of Emotions from Facial Expressions. Journal Abnormal Psychology, Vol. 13, 191819. 172184.

Langfeld, H. S., Judgments of Facial Expression and Suggestion. Psychological Review, Vol. 25, 1918. 488-494.

General Articles:

Fernald, G. G., Character vs. Intelligence in Personality Studies. Journal Abnormal Psychology. Vol. I5, No. 1, Apr., 1920. 111.

Rosanoff, A. J. A Theory of Personality Based Mainly on Psychiatric Experience. Psychological Bulletin, 17, 1920. 281-299.

APPENDIX A

PERSONALITY RATING OF ..............

THESE RATINGS WILL BE TREATED CONFIDENTIALITY

Read these directions carefully. Rate the person named above on a scale of from 1 to 50, according to the place you believe he would occupy in a group of go average college students, in each of the following ten classifications. The extremes (1 and 50) are described below. Under "rating" place the number (any number from 1 to 50 inclusive) which you think indicates the relative position of the individual in the group. There will be 10 such ratings in all. Be sure you understand what is meant by each of the descriptions. Read them carefully. [p. 39]

NO. TRAIT RATING

I.

II.

IIII. [sic]

IV.

V.

VI.

VII.

VIII.

IX.

(Kindly place in envelope and seal. This information is confidential, and will not be disclosed to the person rated.) [p. 40]

APPENDIX B QUESTIONNAIRE

1. Show as specifically as you can how certain persons and events in your life have determined your present interests, habits, and traits of character.

2. (a) What vocation or interest made the strongest appeal to you in childhood? (b) At the age of 16? (c) At the present?

3. What subjects do you like best in college?

[Classics Editor's note: no question number 4]

5. (a) Did you ever design or plan a new object or write an original production?

If so, what? (b) How did it succeed? 40

[Classics Editor's note: no questions number 6 and 7]

8. How many times have you been so angry at a person that you could not forgive him (or her)?

9. (a) How many times have you been angry in the past week? (b) For what causes?

10. Give instances of what you do in thought or imagination or otherwise when you cannot vent your anger by definite action toward its object.

11. Are your emotions stable or are you easily upset?

12. (a) Are you given to day dreaming? (b) What is the content of such day dreams and what is their effect on you?

13. (a) How many times have you been punished unjustly? (b) For what causes? (c) What did you do about it? (d) How long did it bother you?

14. What mannerisms do you have that you know of? (e. g. Putting objects in your mouth, shuffling your feet, automatic movements while thinking, talking to yourself, etc.).

[Classics Editor's note: no questions number 15 and 16]

17. (a) Are you interested in social conditions and social service work? (b) What have you done along that line?

18. (a) Have you any hobbies or special interests? (b) Name them.

19. (a) Do you ever feel that people are talking about you behind your back? (b) What do you imagine they say?

20. In a group do you consider yourself reserved or assertive?

21. (a) Are you or have you ever been bashful? (b) At what age?

22. (a) Are you easily embarrassed? (b) Under what circumstances? (c) How easily do you recover your composure?

[Classics Editor's note: no question number 23]

24. (a) Have you ever written comments in the margins of library books or other books not belonging to you? (b) About how many times?

25. (a) Have you ever fancied that you were a foster child -- that is, that your sup posed parents are not your real parents? (b) At what age and over what period of time? (c) Did you ever really believe this?

26. (a) How did you gain your knowledge of sex? (b) At what age?

27. (a) Is there anything repulsive to you about the sex relation? (b) If so, why?

28. Are there any events in your sex life which have influenced your life history? Specific statements are desired if you are willing to make them.

Footnotes

[1] Adapted from material presented before the National Association of Principals of Schools for Girls, Atlantic City, February 26, 1921.