In my first article (1975), 20 sub areas of the emotion of anger were defined. In my second article (1976), the structure of anger was analyzed, and 7 factors were extracted. In the present article, I attempted to use such terms as more easily extract from subjects more concrete and real reaction, and then to compare the estimation based on such method with the estimation based on the Thurstone method. The results are as follows:

1) In regard to 106 emotional terms gathered from the free-description method, degree of anger are estimated according to the Thurstone method, and the median of each term is calculated. From the above 106 terms, 17 terms are selected. (From now on, I refer to these 17 t erms as “concrete terms”) It turns out that the medians of the concrete terms ranges from 0.3 to 8.5, that of semi-interquartile, from 0.5 to 2.2, and that the mean of semi-interquartile was 1.3.

2) It also becomes clear that, when I compare the estimation based on the concrete term method with that based on the Thurstone method, medians used Thurstone method are larger than means in the former estimation. This tendency is apparent as the medians become larger. That is to say, the answers given under the instruction “Indicate the degree of anger in each situation by checking a number from 0 to 9,” are different from the answers given under the instruction “From the terms given below, choose the emotional term you use habitually in each situation.” The tendency mentioned above is also more apparent in situations which are not desirable from a general point of view though the subjects are not directly involved in them, or situations which promise for the subject some sort of profit, or in situations which imply criticism to the subject. This fact reflects a disparity between words and actual reaction.

3) The concrete terms frequently used are as follows:

i) uninteresting

ii) unbearable

iii) can’t be helped

iv) gets on my nerves

v) unpleasant

vi) repulsive in an identifiable way

vii) can’t stand

4) Four patterns are presupposed from combinations of mean (M) of each subject in the concrete term method and standard deviation (SD). The H-H pattern means that both personal M and SD are higher than M and SD of the whole, and a person of this pattern is considered sanguine with a rather conventional conception of justice. The L-L pattern means that both personal M and SD are lower than M and SD of the whole, and a person of this pattern is apathetic to anything. The H-L pattern means that personal M is higher than M of the whole but personal SD is lower than SD of the whole, and a person of this pattern is considerably emotional. The L-H pattern means that personal M is lower than M of the whole but personal SD is higher than SD of the whole, and a person of this pattern is self-possessed with a remarkable degree of emotional stability.

Socials investigation, psychological tests and interviews were administered by request of the district court.

Psychopathology seemed to be denied and diagnosed as psychopathic tendency because of conspicuous character deviation. But the presence of mild visual, auditory hallucination and delusion of persecution necessitates the psychiatric examination.