Mystery


ABO FAN


Pencil_and_Paper32.gif (245 バイト)Detective ABO FAN -- Solve The Mystery of "blood type and personality"

** FIRST SEASON **  (September 20, 1998; slightly corrected on March 1, 2015)

As for relation between blood type and personality, results differs by questionnaires. Therefore, these results continued to bother researchers. Now, I might have find the consistent relation, if my reasoning is correct. But I'm not Holmes nor Poirot.
Please judge whether I am correct or not. Take it easy and enjoy my reasoning! It is you that is Sherlock Holmes!

Red_Ball12.gif (916 バイト)Is there relationship?

Many data are said to have proven the relationships between blood types and personality.  But three problems shown below are laying ahead.

1. There isn't reliability of data, because random sampling method is not carried out.
2. There are little reappearance of data.  Trends differ by questionnaires.
3. If trends appears, they are different from blood type personality.

If one can solve these three problems, we can say "the relationships between blood type and personality is proved completely," as Alexander the great cut the Gordian Knot.

Red_Ball12.gif (916 バイト)Five conditions

I think prerequisites like the following are necessary to get stable results.

1. Homogeneous subjects (social position, age, region etc.)
2. The number of subjects is more than several hundred (more than one thousand and ratio of each blood types are the same, if possible)
3. Choose personality descriptions of Mr. Nomi (do not use personality test)
4. Results do not correspond to Mr. Nomi's descriptions ("words" do not represent real "personality") -- also affected by culture and nationality etc.
5. Consider statistical errors carefully -- they are much larger than expected.

Red_Ball12.gif (916 バイト)Prerequisite 1: Homogeneous subjects

There appears only 10 - 20% differences -- at most -- of the results of questionnaires by blood types.  Needless to say, differences depend on contents of questions.  Moreover, there are fairly many factors that affect trends: age, area, sex, social position etc.  Even the same blood type person answered differently by these factors.  So, when "random sampling" is carried out, these factors sometimes overwhelm blood types.  In many cases, differences do not come out.  Never conduct "random sampling" that is the popular method of psychology!
So, clear differences have appears many times, when we look at data of university students.   Because there is few differences -- such as age, areas, social position -- and be very homogeneous data.

Red_Ball12.gif (916 バイト)Prerequisite 2: The number of subjects is more than several hundred

As I have mentioned, there appears only 10 - 20% differences, at most. Significant results of c2-test will not appear if the number of subjects is less than several hundred.  More than 1,000 subjects are recommended, if possible.

Red_Ball12.gif (916 バイト)Prerequisite 3: Choose personality descriptions of Mr. Nomi.

This is natural.  Because differences are difficult to come out in general personality test.  Question items of personality tests themselves are hard to measure the differences by blood types.  I don't know the reason yet ...

Red_Ball12.gif (916 バイト)Prerequisite 4: Results do not correspond to Mr. Nomi's description

As I have mentioned, there are fairly many factors that affect trends: age, areas, sex, social position etc.  Even the same blood type person answered differently by these factors.  "Words" do not represent real "personality" -- also affected by culture and country etc.

Red_Ball12.gif (916 バイト)Prerequisite 5: ConsEstimate statistical errors

They are much larger than you expect.  For example, a statistical error (yes-no question item) of 3,000 subjects is more than 7%!  Don't you think this is large?


Let's analyze concrete data!

Red_Ball12.gif (916 バイト)University students

The following data are presented by Mr. Ohmura. ("Chi no shonin no ejikini naruna -- detarame buri wa shomei sareta" [Do not become a prey of merchants of blood], The Asahi Journal, March 8, 1985, pp. 89-92

Subjects are 480 university students of Nihon University -- he is the professor of the university.

Are there characteristics of Type O? (%)    highest in red / lowest in blue

Items

Type O
115
Type A
216
Type B
104
Type AB
45

1. Realistic thinking

54.8 57.8 63.5 55.6

2. Romanticist

65.2 68.5 66.3 75.6

3. Strong group unity

60.6 50.0 51.9 53.3

4. Strongly independent

49.0 51.9 60.6 55.6

5. Open one's mind to friends

82.6 75.5 79.8 68.9

6. Watch out it in the first meeting

58.7 58.3 58.7 62.2

7. Logical

39.4 34.7 36.4 33.3

8. Judge things by instinct

61.3 64.4 71.2 60.0

There appear only 10 - 20% differences, at most.  As you see, these differences are not larger than usual people think.  In this case, there is no statistically significant differences.  Also prerequisite 4 seems to be applicable, because only three items out of eight showed the highest score.

Prerequisite 4: Results do not correspond to Mr. Nomi's descriptions

As I have mentioned, there are fairly many factors that affect trends: age, areas, sex, social position etc.  Even the same blood type person answered differently by these factors.  "Words" do not represent real "personality" -- also affected by culture and country etc.

Red_Ball12.gif (916 バイト)Regional Differences

The next data are regional differences.  NHK hoso bunka kenkusho [Japan Broadcasting Corporation broadcasting culture research institute] conducted nation wide research.

Representative items that showed differences Nation Tokyo Results of c2-test
Characteristics of Tokyo differ from other region 44.4% 51.9% p<0.01
Wish to join local events and festivals 46.7% 37.9% p<0.01
Associate around many relatives 54.3% 49.7%
Does not get deeply involved to each other 31.0% 36.9% p<0.01
Cut down usual life and want to leave money and property 42.1% 33.3% p<0.01
Number of subjects 29,620 580

from NHK hoso bunka kenkusho "Gendai no kenmin kishitsu [Modern regional traits]" NHK Publishing, 1997, pp. 127.

Interestingly enough, there are many statistically significant differences.  But differences themselves are the same as blood type.   Why?
Only because NHK collected many subjects!  Do you understand?

The Gordian Knot

Professor Matsui's paper is my Gordian Knot.

Matsui, Y. (1991). Ketsueki gata ni yoru seikaku no chigai ni kansuru toukeiteki kentou [Statistical consideration on personality difference according to blood types], Bulletin of Tokyo metropolitan Tachikawa junior college, 124, 51-54.

JNN Data Bank (Japan News Network Data Bank, a department of TBS, which is one of the major TV stations in Tokyo) conducts a large survey annually. The number of subjects are about 3,000 people each, from age 13 to 59 using random sampling method. Surveys of 1980, 1982, 1986, 1988 have items of blood type and 24 yes-no questions of personality.  More than 10,000 people were analyzed in total.

Table 3  24 yes-no question items of personality

1    I am ready to keep company with everyone.
2    I make an effort toward my goal.
3    I like to be a leader.
4    I am not particular about things.
5    I do not know how to refresh myself.
6    I am serious when I should be so.
7    I often tell jokes to make others laugh.
8    I do not change my idea after I present it.
9    I am not worried by what I was told.
10    I have a lot of friends.
11    I am often worried about things.
12    I sometimes indulge in fancies.
13    I am not good at getting along with others.
14    I like to have parties in my home.
15    I am careful when doing something.
16    I am often moved to tears.
17    I often change my mind.
18    I am a good loser.
19    I am persevering.
20    I cannot be quiet and I make merry when I am delighted.
21    I am shy.
22    I sometimes burst into a rage.
23    I like to think deeply alone more than to talk with others.
24    I do not like to visit someone without presents.

Table 8  Percentages of people that answered yes to Item 4: " I am not particular about things."  highest in red / lowest in blue

O A B AB

1980

31.8 30.6 37.8 34.3

1982

39.1 33.0 35.6 36.1

1986

39.5 32.4 38.8 39.9

1988

42.9 35.9 45.1 37.1

Only item 4 showed the same difference (Table 8). But the highest blood type(s) differed by year. The result lacked the consistency that exceeded one year. Therefore, the conclusion is that there is no blood-typical stereotypes in these data.

Only one out of 24 items showed significant difference ...  more than 10,000 subjects in total.  I was at a loss ...
There are many factors that affect the differences: age, area, sex distinction, social position etc. -- even the same blood type. Therefore, differences might not appear when "random sampling" is done ...

Red_Ball12.gif (916 バイト)University students again!

Let's try Mr. Matsui's another paper ...

Takuma, T., & Matsui, Y. (1985). Ketsueki gata sureroetaipu ni tsuite [About blood type stereotype], Jinbungakuho (Tokyo metropolitan University), 44,15-30.

The distribution of the blood types of the subjects is written in Table 1. 613 students without no answer are chosen out of 640 students who belong to the public/private university of Tokyo Metropolis / Kanagawa Prefecture. 345 men and 215 women. I think there may be no problem because it almost agrees with the average of Japanese.

Table 1  Distribution of ABO blood types of subjects (%)

Items

N O A B AB

Subjects

613 29.7 36.7 22.5 11.1

Average of Japanese

1,150 thousand 30.7 38.1 21.8 9.4

Percentages of people who answered "yes" are shown in Table 2. Prof. Matsui said "Items of personalities are carefully chosen from Nomi's book  (1984)".

From Table 2 (modified)  Percentages of people who answered "yes"  highest in red / lowest in blue

Items

O(182) A(225) B(138) AB(68) Result of c2-test

4. I esteem rules, customs and orders

57.7 54.7 50.7 41.2 p<0.001
10. I am optimistic to the future 46.2 52.9 52.2 67.6 p<0.05

Two out of twenty items showed significant differences ...   less than 1,000 subjects in total.
In this case, subjects are university students -- very homogeneous.  Therefore, virtually no difference appear when "random sampling" is done ... I might have cut the Gordian Knot.

All Items

1. I am particular with everything

2. I care about it to the person of the surroundings

3. I suppress my feeling and desire

4. I esteem rules, customs and orders

5. I am seeking my reason for living

6. I am not influenced by the surroundings

7. I do not like be controlled, constrained by person

8. I understand new things and flexible thought

9. I open heart to people

10. I am optimistic to the future
11. I think straightly
12. I have clear two-facedness of emotionally stable and unstable
13. I am distant to people
14. I am not able to be absorbed to anything
15. I am a good analyst and critic
16. I treat people with smile and carefully
17. I am both romantic and realistic
18. I care human relations, especially esteem trust of people
19. I have vitality
20. I go straight with persistence and achieve when the purpose is decided

Check the Reappearance of Data

Let's proof the order of group-dependence, which is O=A>B>AB (Valid only in Japan?).

Preliminary trial: questionnaire

The data (table 18, pp.71) from "Shin ketsueki-gata ningen-gaku" (New blood type humanics) published in 1978. Results of a questionnaire to first-class Japanese athletes (exact dates were not specified). The ratios of athletes who choose "Do the best for Japan" in a international game.

Blood type Number of persons Ratio
O 58 people 13.8%
A 73 people 13.7%
B 51 people 9.8%
AB 22 people 0.0%

It is the order of O=A>B>AB as expected. To my surprise, the ratio of Type AB is 0! Type-AB people doesn't identify themselves with their country -- Japan.

The Secondary trial: questionnaire

The next data from the same book (table 24, pp.211). The question is "Which do you feel best when eating with people?" The ratios of people who choose "Eat alone". Exact dates were not specified.

Blood type Number of persons Ratio
O 636 people 20.1%
A 739 people 19.6%
B 600 people 23.2%
AB 428 people 27.1%

It is the order of O=A<B<AB, too, as expected. It is clear that Type-AB people like to be alone.

The Third trial: Raymond B. Cattell's Study

Famous psychologist, Raymond B. Cattell's Study shows the clear relationships of blood types to personality traits. Click here!

Results of The ABO System

Australians are the same trend for the most part, although a level of significance is a little bit high, because there are few Type-AB people.

The last trial: Japanese psychologist

The same trend showed by Japanese psychologists. Surprisingly, probability is below 0.1%. The ratio of persons that chose "Esteem rule, custom and order" (Taketoshi TAKUMA and Yutaka MATSUI, 1985, About blood type stereotype, Jinbungakuho,172,15-30.).

Blood type Number of persons Ratio
O 29.7% 57.7%
A 36.7% 54.7%
B 22.5% 50.7%
AB 11.1% 41.2%
Total 613 people in all -

Even, this is the order of O=A>B>AB as expected.

Red_Ball12.gif (916 バイト)Do you think that there is relation between blood type and personality?

Almost three decades, more than 70% of Japanese people think that there likely are "relationships between blood types and personality" and the ratio is not changing.  However, less than 20% of people think that there is strong relation, while a little less than 50% of people think that there is some relation. This data agrees completely with my reasoning!
Namely, there is not strong relationships between "blood types and personality", but some relation.

1. Do you think that there likely is relationships between blood types and personality?
  -- 1,102 Tokyo metropolitan residents aged 15 - 69 by random sampling

Answers

Percentage

Yes 75.0%
No 18.4%
D/K 5.9%
N/A 0.6%

from NHK public opinions research (1986)

2. Do you think that there are relationships between blood types and personality?
  -- 2,320 men and women all around the nation residents aged more then 20

Answers

Percentage

Yes 18%
More and less 46%
No 21%
D/K 14%

N/A

1%

from the Mainichi Shimbun "Kokoro no Jidai [Times of Heart]" (1987) Nation-wide research

3. Are there relationships between blood types and personality/compatibility? (500 people)

Answers

Percentage

Yes 72%
No 18%
D/N 10%

from Kansai TV's "Aruaru Daijiten" (June 15, 1997)

4. How do you feel about relationships between blood types and personality (318 female university students)    more than 50% in bold

Items

Affirmative (%)

Personality differs by blood type

*

I trust it

37.5

Ii match my personality

53.8

I enjoy it

83.6

I like it

61.5

Useful for communications with other people

40.4

Useful in the first time meetings

26.0
Think people's blood type before action 5.8

Change action by blood type

4.9

Useful for understanding other people's action

14.4

First of all, I think compatibility by blood type

26.7

I often read articles about blood type

57.7
Useful for knowing oneself 26.0

Able to Understand unknown self

12.5

It does new discovery about myself

16.4

Able to see myself objectively

26.0
I don't like Type A people 5.8
I don't like Type B people 10.6
I don't like Type O people 1.0
I don't like Type AB people 10.6

* "Very much" (0%), "Much" (11.6%), "To some degree" (61.1%), "Not much" (16.8%), "Non-existent" (10.5%)
from Matsui, Y., & Kamise, Y. (1994). Functions and structures of blood-group stereotype. Bulletin of University of the Sacred Heart, 82, 91-111.

5. Positive and Negative Reasons for relationships between blood types and personality  (318 female university students)    more than 50% in bold

Items

Affirmative (%)*

1. It is scientific

11.6

2. Difference of components in blood influence personality

26.4

3. It is correct because it appears in many magazines

15.3

4. It fits well to the people around me

54.4

5. I can guess the blood type of other people somewhat

55.3

6. Personality of the same blood type as mine, resembles my personality

45.9

7. Personality is made by environment rather than blood type

93.1
8. Personality is not divided to four types by blood type 93.1

9. Personality is not such simple able to be judged by blood type

92.5

10. Personality of the same blood type differs

98.1

11. It does not fit well to some people around me

78.3

12. It does not fit well to me

21.3

* Percentages are the sum of "I think so" and "I think so to some degree"
from Matsui, Y., & Kamise, Y. (1994). Functions and structure of blood-group stereotype. Bulletin of University of the Sacred Heart, 82, 91-111.

Red_Ball12.gif (916 バイト)Cut the Gordian Knot -- Prof. Matsui's Paper

There is consistent differences although it was trifling!  Let's go!

2    I make an effort toward my goal.   highest in red / lowest in blue

Year

O A B AB Max. - Min.

1980

26.7 27.1 23.8 30.4 6.6

1982

27.7 30.7 26.8 28.5 3.9

1986

28.3 26.8 23.7 29.0 5.3

1988

27.8 28.2 27.8 28.6 0.8

Mean

27.6 28.2 25.5 29.1 3.6

4    I am not particular about things.   highest in red / lowest in blue

Year

O A B AB Max. - Min.

1980

31.8 30.6 37.8 34.3 7.2

1982

39.1 33.0 35.6 36.1 6.1

1986

39.5 32.4 38.8 39.9 7.9

1988

42.9 35.9 45.1 37.1 9.2

Mean

38.3 33.0 39.3 36.7 6.3

6    I am serious when I should be so.   highest in red / lowest in blue

Year

O A B AB Max. - Min.

1980

38.2 39.2 36.6 42.7 6.1

1982

41.6 41.2 37.0 44.9 7.9

1986

36.5 38.9 35.6 37.4 3.3

1988

39.3 39.5 35.0 39.0 4.5

Mean

38.9 39.7 36.1 41.0 4.9

9    I am not worried by what I was told.   highest in red / lowest in blue

Year

O A B AB Max. - Min.

1980

23.5 22.2 26.7 23.3 4.5

1982

28.2 24.3 24.4 25.0 3.9

1986

25.6 23.7 26.1 26.9 3.2

1988

27.6 24.2 27.2 28.3 4.1

平均

26.2 23.6 26.1 25.9 3.6

15    I am careful when doing something.   highest in red / lowest in blue

Year

O A B AB Max. - Min.

1980

32.1 29.8 25.9 29.1 6.2

1982

32.7 32.3 29.8 31.6 2.9

1986

29.3 33.6 28.7 33.6 4.9

1988

28.3 32.3 26.4 30.1 5.9

Mean

30.6 32.0 27.7 31.1 4.3

22    I sometimes burst into a rage.  highest in red / lowest in blue

Year

O A B AB Max. - Min.

1980

33.4 35.2 30.8 34.0 4.4

1982

34.2 34.1 28.4 33.5 5.8

1986

36.1 35.9 30.1 35.0 6.0

1988

36.0 35.0 34.7 34.9 1.3

Mean

34.9 35.1 31.1 34.4 4.0

Needless to say anything (laugh). Probability that only the certain blood type becomes the least rate is cube of 0.25 (0.016). So these results are not accidental, because there are six items that showed consistent trends.  Also c2-tests are significant.

Red_Ball12.gif (916 バイト)Cut the Gordian Knot -- Prof. Sakamoto's Paper

Prof. Sakamoto's paper analyzed the largest number of samples: more than 30,000 people using the random sampling method. Of course, it showed the clear difference between blood types.  But Mr. Sakamoto said "blood-typical stereotypes" caused the difference, not blood types themselves. So he thinks there is no relation between blood types and personality.

Yamazaki, K., & Sakamoto, A. (1991).  Ketsueki gata sutereotaipu ni yoru jiko joju gensho [The self-fulfillment phenomenon generated by blood-typical personality stereotypes: time-series analysis of nation-wide survey]. Paper presented at the 32nd annual convention of the Japanese society of social psychology. Tokyo (pp. 289-291).

JNN Data Bank (Japan News Network Data Bank, a department of TBS, which is one of the major TV stations in Tokyo) conducts a large survey annually. The number of subjects are about 3,000 people each, from age 13 to 59 using random sampling method. Surveys of 1978 - 88 have items of blood type and 24 yes-no questions of personality.

24 yes-no question items of personality

1    I am ready to keep company with everyone.
2    I make an effort toward my goal.
3    I like to be a leader.
4    I am not particular about things.
5    I do not know how to refresh myself.
6    I am serious when I should be so.
7    I often tell jokes to make others laugh.
8    I do not change my idea after I present it.
9    I am not worried by what I was told.
10    I have a lot of friends.
11    I am often worried about things.
12    I sometimes indulge in fancies.
13    I am not good at getting along with others.
14    I like to have parties in my home.
15    I am careful when doing something.
16    I am often moved to tears.
17    I often change my mind.
18    I am a good loser.
19    I am persevering.
20    I cannot be quiet and I make merry when I am delighted.
21    I am shy.
22    I sometimes burst into a rage.
23    I like to think deeply alone more than to talk with others.
24    I do not like to visit someone without presents.

Excerpts:

1. Age / Fig. 5

"A-B" score becomes higher as age becomes higher. In other words, senior people are more like "Type A".

2. Blood type / Fig. 6

"A-B" score of Type-A people (0.082) is significantly higher than that of Type-B people (0.027).

3. Year of the survey / Fig. 7

"A-B" score has gradually decreased in 11 years from 1978 through 1988. In other words, Japanese people are becoming more like "Type B".

4. Alternate action between blood types and year of each survey / Fig. 8

The alternate action between blood types and year of survey were detected. Type-A people become more like "Type A" and Type-B people become more like "Type B". This is the result that means a self-fulfillment phenomenon generated by blood-typical stereotypes.

stereo1.gif (48556 バイト)

Fantastic!

But, when you carefully watch Fig. 7 of the second paper, you should notice something strange.
Until now, many opponents of "blood type and personality" (especially, Japanese psychologists) have asserted that the difference of personality by blood types does not exist, because reliable data and/or results of questionnaire are not available.
However, the difference is very significant -- out of question -- in Fig. 7 of the second paper. Eleven years and more than 30,000 people by complete random sampling. Therefore, the former assertion, "reliable data and/or results of questionnaire are not available", is completely wrong!
Also read Ms. Yoriko Watanabe's paper.

Watanabe, Y. (1994). The roles of prototype and exemplar in the formation of the "blood type stereotype". Shakai shinrigaku kenkyu [Social psychology research], 10-2, 77-86.

Subjects generalized their personality traits as the traits typical to people with the blood type of their own, whether these personality traits are commonly and publicly assumed blood type-relevant personality traits or not.

This means answers were not affected by the knowledge of blood type personality.  Therefore, reliable data and/or results of questionnaire are available now, then there are clear evidence between "blood type and personality."  Wow!

** SECOND SEASON **  (March 20, 2007)

In the second season, I try to solve the remaining mysteries: Why differences don't appear by personality tests?  This time, you are Ellery Queen!

Red_Ball12.gif (916 バイト)Why not Mr. Nomi's items?

In the first season, we examined many questionnaire items, which are only made by psychologists.  Why not Mr. Nomi's items?  You are right!  Let's try.

Red_Ball12.gif (916 バイト)Ms. Yoriko Watanabe's Paper

I will introduce Ms. Yoriko Watanabe once again.

Watanabe, Y. (1994). The roles of prototype and exemplar in the formation of the "blood type stereotype". Shakai shinrigaku kenkyu [Japanese Journal of Social Psychology], 10-2, 77-86.

In this article, there seem little differences appear... so far.  But I noticed incidentally there might be great differences according to blood types.  So, I decided to recalculate and the result is....

Result by recalculation (Excerpts from Table 1: Relation between Correct Answer Rates of Each Questions and Blood Types)

Blood
 Type

Personality Traits

Excluding
One's Own Blood Type
One's Own Blood Type χ2 p
Type A

Introvert, solve problem by oneself

.250 .658 16.50 p<0.001

Considerate, careful about everything

.563 .842 8.39 p<0.005

Responsible

.344 .605 6.61 p<0.05

Esteem tatemae (principle) than honne (practice)

.422

.658 5.31 p<0.05
Type B

Shaken easily

.130 .440 11.10 p<0.001

Warm hearted

.182 .360 3.43 p<0.10

Less influenced by the surroundings with my pace style

.325 .640 7.81 p<0.01
Optimistic .312 .520 3.54 p<0.10

Lack of prudence

.390

.720 8.27 p<0.005
Type AB
Fairy-tale like .133 .333 3.20 p<0.10
Rational thinking .200 .583 8.41 p<0.005
Cool and dry .478 .750 3.14 p<0.10
Capricious .089 .250 2.85 p<0.10
Type O (Not Siginficant)
 p < 0.05 in red

It is natural, because these personality traits are referred to Mr. Nomi's (and other researchers) books, not pulled out from psychological tests!  Wonderful differences appear unless one does not use psychological personality tests. (laugh)

There may be another reason.  Personality diagnosis by blood type is very popular in Japan.  Therefore as many as 70 percent of Japanese people think that there are (some) relationships between blood types and personality.  Accordingly, in this case, as many as 70 percent of Japanese people would answer his/her personality that the person oneself is aware of -- in short, blood type's personality of the person oneself -- must be detected by the test, whether there are "real" relationships between blood types and personality or not.

These are logical conclusion.  Then, two mysteries were left.

1.  Are detected personalities are real or imaginary?
2.  Why psychological personality test cannot detect differences by blood types?

Mystery No.1 has already solved.  Ms. Yoriko Watanabe will explain to you, again.  From the above-mentioned article:

Difference between subjects that have exact knowledge and without it

Type of subjects

Percentages of
correct answers
Correlation coefficients of
exact knowledge
and subjects' knowledge
Conviction rate
of ones' knowledge
Subjects with exact knowledge

>=.75 (n=38)

.82 .90 (SD=.96)

>=.72 (n=46)

.73 1.05 (SD=.93)
Subjects without exact knowledge

< .75 (n=56)

-.06 1.32 (SD=.91)

< .72 (n=48)

-.10 1.34 (SD=.94)

< .65 (n=35)

-.25 1.48 (SD=.84)

Subjects that have exact knowledge of blood types and personality, the exact knowledge and their knowledge are agreeing for the most part (in blue). Therefore, there are strong positive correlations between them. Conversely, subjects without exact knowledge, must have "nonexistent knowledge" or "opposite knowledge". If there are relationships between blood types and personality, there would be negative correlations between "opposite knowledge" and exact one (in red).  Namely, it seems that there is some relationships between blood types and personality. But, it is not possible conclusion, just being suggesting possibility persistently, because there are not significant differences statistically. Furthermore:

If detected personalities are imaginary, people with exact knowledge must think that personality is decided with only blood types.  But this is not true.  Conversely, the opposite tendency seems to appear.  We must admit that detected personalities are real!  Wow!

Also Dr. Peter J. D'Adamo had got the same results.

Red_Ball12.gif (916 バイト)Dr. Peter J. D'Adamo

Dr. Peter J. D'Adamo performed a blood type-personality test on his website.  The result is as follows.  No comments needed.  Excerpts from his book "Live Right for Your Type".

The testing group demographics of www.dadamo.com

Total 20,635
Total Female 15,255
Total Male 5,380

The demographics by blood type

Type A Type AB Type B Type O
7,187 1,473 2,809 9,166
34.83% 7.14% 13.61% 44.42%

A Fairly reliable synthesis of personality characteristics by blood type

Type O Type A Type B Type AB
Extroverted
Strong
Leader
Confident
Pragmatic
Strategic
Patient
Logical
Introverted
Intense
Demanding
Perfectionist
Sensitive
Cooperative
Creative
Independent
Free Thinking
Resilient
Creative
Original
Subjective
Inveterate organizer
Intuitive
Emotional
Passionate
Friendly
Trusting
Empathetic

Dr. Peter J. D'Adamo also shown comparison with Mr. Nomi.

Blood Type-Personality Characterizations

  Type O Type A Type B Type AB
Masahiko Nomi Extroverted
Strong
Expressive
Introverted
Perfectionist
Restrained
Free-thinking
Independent
Lacking
ambition
Sensitive
Distant
Passive
Peter Constantine Extroverted
Outspoken
Introverted
Reserved
Calm
Pragmatic
Organized
Balance of extroversion and introversion
Raymond Cattell Stable Prone to anxiety Self-sufficient Alienated
Hans Eysenck Extroverted Calm Highly emotional Introverted
Blood Type-Personality Test as www.dadamo.com Extroverted
Practical
Decisive
Lives in the present
Introverted
Sensitive to the needs of others
Feeling
Flexible
Spontaneous
Subjective
Feeling
Intuitive

Red_Ball12.gif (916 バイト)Dr. Wu's study in Taiwan

Mystery No.2 has already solved, too.  Dr. Wu in Taiwan will explain the truth.

To my surprise, I find a blood type study in Taiwan!
Dr. Wu, the author, is a professor of medical college in Takao prefecture, Taiwan.

Personality and Individual Differences 38 (2005) 797–808.

Blood type and the five factors of personality in Asia
Kunher Wu, Kristian D. Lindsted, Jerry W. Lee

Summary

Personality diagnosis by blood type is popular in Taiwan, as well as in Japan.  Therefore about two thirds of high school students think that there are (some) relationships between blood types and personality, as well as Japan.

However, the most famous psychological test in English area is NEO-PI-R.  Its Chinese edition (of course, there is a Japanese edition, too)  which is standardized globally, detected little significant difference except type AB females, even using more than 3,000 high school students' subjects.

I think it is strange because there is not fictiveness scale (in other words, the person oneself tells a lie) in this personality test, so  the personality that the person oneself is aware of -- in short, blood type's personality of the person oneself -- must be detected by the test.

# The situation is the same as in Japan.  Few psychological tests detect differences of blood type, somehow.

It is said that NEO-PI-R psychological personality test has universality beyond languages, I may explain the past results (few significant differences appeared) without contradiction, including English versions!

The following is the explanation:

It is expected that significant differences must have appeared considerably, at least by individual answers (or the 30 facets).  However, significant differences do not come out if individual answers are transformed to five personality scales.  Therefore we might have to admit that NEO-PI-R test itself cannot detect the differences of blood type well, to explain the conventional data without contradiction.

Mr. Masahiko Nomi once said that psychological personality tests cannot measure the difference of blood type properly, and this might be right!

In addition, the NEO-PI-R's definition of personalities (ex: transformed by factor analysis etc.) are different from ordinary ones which we use everyday, It does not contradict to the conventional data.

Red_Ball12.gif (916 バイト)Blood Type in Asia (Excerpts)

 In this article, blood type personalities in Asia are expressed.  Item "Neuroticism" is mysterious here.  Type B seems to be most nervous somehow. In Japan, type As are said to be most nervous.  Why?

  Our study is the first published article to examine the relationship between blood type and personality based on the five-factor model in Asia, where the belief that blood type determines personality is widely accepted. According to D'Adamo and Whitney (2001), more than 70% of Japanese believe that blood type is directly related to personality. In Taiwan, books discussing the influence of blood type on temperament and interpersonal relationships are readily available (Chang, 1999; Cheng, 1999; Lin, 1999).
  Based on the literature and the popular beliefs in Asia, the following hypotheses were proposed:

  1. Blood Type B individuals will score higher on Neuroticism than will other blood types.
  2. Blood Type O individuals will score higher on Extraversion than will other blood types.
  3. Blood Type A individuals will score higher on Agreeableness than will other blood types.
  4. Blood Type AB individuals will score lower on the Extraversion, Conscientiousness, and Agreeableness than will other blood types.

 This is not fun for type AB persons, like me, anyway. :-p.  But, seeing from a medical view point, type Bs are most likely to be obsessive-compulsive neurosis (in Japan), perhaps this may affect.

Characteristics of Taiwan

  Then about subjects:

 After obtaining permission from the city of Kaohsiung in March 2000 to conduct surveys in all of their 29 high schools, we used stratified random sampling to select participating classes from each school...

Table 1
Distribution of gender, blood type, books, belief, and language separated by academic achievement level [abbreviated]

Males 1497
Females 1899
Type A 651
Type B 610
Type O 1282
Type AB 138
Books b
  Yes 2489
  No 851
Belief c
  Yes 2175
  No 1137

a Based on school enrollment, which depended on students' national test scores.
b Reponses to "Have you read any books about blood type and personality?"
c Reponses to "Do you believe there is a relationship between blood type and personality?"

 The ratio of persons who have read books about blood type is as high as 75%, and that of having belief is 66%.  We may think that the situation is almost similar to Japan.

Results

 Then I will write about an important result.

Table 2
Unstandardized regression coefficients with 95% CI for the blood types by gender on each of the five NEO-PI-R domains using the final multiple regression model [confidence intervals are omitted]

Blood types

  A B AB
Males      
Neuroticism  0.9   -1.3 4.1
Extraversion b
   Belief c (Yes) 3.2  1.3 -2.9
   Belief c (No) 3.8 -0.8  1.8
Openness 0.1 -0.6  0.2
Agreeableness b
  Belief c (Yes) -0.6  2.6 -4.4
  Belief c (No)  2.4  1.0  5.9
Conscientiousness  0.2 -1.8 -0.1
Females      
Neuroticism -1.1 -0.5  1.8
Extraversion -0.2 -0.0  0.3
Openness -0.9  0.2  0.6
Agreeableness  0.1  0.5 -0.9
Conscientiousness  0.2 -0.6 -5.2*

* P < 0.05.
a With blood type O as reference.
b The model included the interaction term between blood type and belief.
c Reponses to "Do you believe there is a relationship between blood type and personality?"

 It is just what it looked.  There is only one difference: type-AB girls' scores of conscientiousness are significantly lower than type-Os (level of significance 5%).

Red_Ball12.gif (916 バイト)Other studies, same results (March 1, 2015)

Conclusions of other studies that utilized Big-Five Factor model are the same; no correlation is detected.

In an article as shown below, Professor Sung (Yonsei University in Korea) said definitely that "studies that utilized Big-Five Factor model have shown no relationships at all (So Hyun Cho, 2005; Cramer, 2002; Rogers, 2003; Kunher, 2005). "

Sung Il Ryu, Young Woo Sohn (2007). A Review of Sociocultural, Behavioral, Biochemical Analyses on ABO Blood-Groups Typology, Korean Journal of Social and Personality Psychology, 21(3), 27-55.

These four articles are:

Professor Sung concludes like this:

The situation is much the same in Japan;  No (or little) correlation is detected in the following articles.

Abstract
 Correlation between blood types and personalities was investigated in university students (n=298) through two investigations by them administering (1) The Big Five personality test and (2) a questionnaire. The questionnaire consisted of items that described tendencies of each blood type that were quoted from a book about correlation between blood types and personalities. Another researcher conducted (1) above as a different study, in order to avoid response bias. Regarding (2), participants were required to estimate whether each item might agree or disagree with their own personalities. Then, their blood types were recorded. The results of an analysis of covariance indicated the following: (1) there was no significant difference in character tendencies based on the blood type. Regarding (2), items that participants regarded as agreeable with their characters and those they regarded as disagreeable did not have a significant correlation with blood types. The above results suggest that there is no correlation between blood types and personalities.

** APPENDIX **  (March 25, 2007)

By the way are there any scientific grounds?  Several scientist explained there hypotheses.

Dr. T. Abo (March 25, 2007)

   In addition, there is a study to understand activities of the sympathetic / parasympathetic nerve by examining the ratio of polymorph and lymphocytes of white blood corpuscles.

   According to Dr. Toru Abo, an immunologist who is playing an active role worldwide.
   In his book "medical care makes illness", he says about relationships between blood types and personality (page 17-19).

   I have mentioned that a person of sympathetic nerve superiority becomes an active "polymorph person", while parasympathetic nerve superiority becomes a generous "lymphocyte person".  However, in conjunction with this law, it was connected with blood type between human nature case.
   The connection with blood type and personalities has been pointed out for a long time, but few studies clarified the mystery scientifically till now.

   The scientific grounds are the following:

   When I summarize a level of a lymphocyte of peripheral blood of the adult whom I measured by clinical survey in every blood type.  Type O, 39%; Type B, 37%; Type A, 36%; Type AB, 34%; (object: 5,000 people).  Dr. Fukuda Minoru, my respected friend clarified this scathing observation...

Brain Waves (March 25, 2007)

   Several experiments were televised.  For example, "Spa Spa Ningengaku (Humanics)" on October 7, 2004...

   On the TV program, instantaneous exercises (dumbbell exercises) and stamina exercises (bicycle machines) were investigated, for the purpose of elucidating the reason that people cannot continue exercises, according to blood type. The next results were provided by measuring brain waves during exercises.

  Instantaneous Exercise Stamina Exercise
Type A Much stress Little stress
Type B Little stress Much stress
Type AB Different by person Different by person
Type O Little stress Little stress

   Very different by blood type.

   Another example, "Aru Aru Daijiten" on October 3, 2004...

   I was surprised by experiments of brain waves, this time, too (professor Munetaka Haida, faculty of medicine, Tokai University).  It is interesting to measure compatibility using brain waves.  On the program, he used machinery made by Hitachi Medico and measured photons which appeared from brain.  We can easily find what part of brain is activated by detecting photons.   Is it a high grade lie detector?  I have a feeling that Hitachi, Ltd. did that kind of announcement. (Mr. Haida's experiment was introduced on Korean TV -- SBS in 2006, too.)

   Another experiment by professor Noriyuki Yoshida, faculty of human and social environment, Hiroshima International University, was introduced on the same TV program.  This is also very interesting experiment to measure differences of brain waves of 40 subjects (both males and females) according to blood types.  Subjects push a button if a light is on and release it if the light is off with one's eyes closed.  Brain waves among these two states are analyzed.

     Point 1: How do brain waves react to light stimulation?
     Point 2: After light stimulation, how long does it take for brain waves to return to the original state?

   The results were fantastic!

  [Added on March 25, 2007]

   When we assume degree of brain waves excitement as [ quantity β band / (quantity of α band + quantity of β band) ], the degree was type A > type O > type AB > type B in order.  In other words, to react to stimulation of the outside world in this order.   Difference of brain wave excitement by blood type means difference of optic nerve and/or brain reactions by blood types.  Finding differences of reaction by blood type would be demonstrated at cerebrum physiology level (aside from mechanism) for the first time!  Great!!

   Type As and type Os strongly reacted to lights and remains strong reactions of brain waves, even after lights have faded.  Mr. Yoshida commented that type As and type Os are "easy to be influenced when something happens" and "worried about it".  Immediately the narrator says "always set up one's antenna to grasp the situation"; "attentive";  "considerate".

   As for type Bs, reaction is small and being restored immediately; Mr. Yoshida, says "changed one's feeling immediately"; "able go to a meal immediately even if in quarrel".  The narrator says "carry through one's pace without being controlled by surrounding situations";  "seem to be self-centric"; "keep calm".

   As for type ABs, reaction is bigger than type Bs (smaller than type As and Os) and to be restored immediately.  Mr. Yoshida says "mind and take stimulation and can do postprocessing".  "Show reactions in one's surrounding condition, but return to one's pace immediately"; "said to have two-facedness"; "deal with things calmly" says the narrator.

    If the result are shown properly as above, these experiments may be appreciated properly.  Then positive feedback might happen, and studies of relationships between blood types and personality may advance steadily in the near future.  I really hope so.

Brain Waves in Korea (March 1, 2015)

From the Journal of the Korea Academia-Industrial cooperation Society / v.12 no.6, 2011, pp.2554-2560

Title: A study on the effects of one's blood type on emotional character and antistress of adults
[http://kiss.kstudy.com/journal/thesis_name.asp?tname=kisskw&key=50117192]

Authors: Choong-Shik Kim 1 and Seon-Gyu Yi, Dept. of Information Management Seoul University of Venture & Information

Abstract This study is focused on the effects of one's blood type on emotional character and anti-stress by comparing each individual's brain wave. Observed records are from those kids who volunteered for KRIJUS(Korea Research Institute of Jungshin Science)'s brain wave measurement from September 2006 to December 2009. The whole 4636 adult(20-59 age) group consists of 3221 female and 1415 male. Blood types are as follows; A>B>O>AB dominately ordered. The result shown no relations between the blood types of adult and their emotional character. But meaningful difference had been revealed between the blood type and stress resistancy; type O rated higher scores in awareness and stress resistancy then other types, which shows that type O has more ability to reduce its stress. This study revealed meaningful relation between blood types and emotional character among several categories, while suggesting follow-up researches figuring out more
detailed explanations for its effect and genetic role.

Appendix (October 12, 2015)

Recently two medical theses had appeared.

Shoko Tsuchimine, Junji Saruwatari, Ayako Kaneda, Norio Yasui-Furukori (2015), ABO Blood Type and Personality Traits in Healthy Japanese Subjects
[http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0126983]

Abstract
There is no scientific consensus that a relationship exists between the ABO blood group and personality traits. However, a recent study hypothesized that the dopamine beta-hydroxylase (DBH) gene is in linkage with the ABO gene. The sample population consisted of 1,427 healthy Japanese subjects who completed the Temperament and Character Inventory (TCI). Each subject’s ABO blood type was determined by genotyping the rs8176719 and rs8176746 ABO gene single-nucleotide polymorphisms (SNPs) using a TaqMan genotyping assay. The relationships between the six ABO genotypes or four ABO phenotypes and personality traits were examined using a multivariate analysis of covariance (MANCOVA), controlling for age and sex. The MANCOVA data showed a significant difference in TCI scores among the ABO genotype groups (F [7, 1393] = 3.354, p = 0.001). A subsequent univariate analysis showed a significant difference in the mean scores for Persistence among the genotype groups (F = 2.680, partial η2 = 0.010, p = 0.020). Similarly, dividing the ABO blood type into four phenotypes revealed a significant difference among the phenotype groups (F [7, 1397] = 2.529, p = 0.014). A subsequent univariate analysis showed a significant difference among the phenotype groups in the mean scores for Persistence (F = 2.952, partial η2= 0.006, p = 0.032). We observed a significant association between ABO blood group genotypes and personality traits in a large number of healthy Japanese subjects. However, these results should be regarded as preliminary and should be interpreted with caution because it is possible that the association between ABO blood group genotype and the Persistence trait is relatively weak.

Donna K. Hobgood (2011), Personality traits of aggression-submissiveness and perfectionism associate with ABO blood groups through catecholamine activities
Medical Hypotheses, Volume 77, Issue 2, Pages 294-300.
[http://www.sciencedirect.com/science/article/pii/S0306987711002106]

Abstract
 There is no scientific consensus that a relationship exists between the ABO blood group and personality traits. However, a recent study hypothesized that the dopamine beta-hydroxylase (DBH) gene is in linkage with the ABO gene. The sample population consisted of 1,427 healthy Japanese subjects who completed the Temperament and Character Inventory (TCI). Each subject’s ABO blood type was determined by genotyping the rs8176719 and rs8176746 ABO gene single-nucleotide polymorphisms (SNPs) using a TaqMan genotyping assay. The relationships between the six ABO genotypes or four ABO phenotypes and personality traits were examined using a multivariate analysis of covariance (MANCOVA), controlling for age and sex. The MANCOVA data showed a significant difference in TCI scores among the ABO genotype groups (F [7, 1393] = 3.354, p = 0.001). A subsequent univariate analysis showed a significant difference in the mean scores for Persistence among the genotype groups (F = 2.680, partial η2 = 0.010, p = 0.020). Similarly, dividing the ABO blood type into four phenotypes revealed a significant difference among the phenotype groups (F [7, 1397] = 2.529, p = 0.014). A subsequent univariate analysis showed a significant difference among the phenotype groups in the mean scores for Persistence (F = 2.952, partial η2= 0.006, p = 0.032). We observed a significant association between ABO blood group genotypes and personality traits in a large number of healthy Japanese subjects. However, these results should be regarded as preliminary and should be interpreted with caution because it is possible that the association between ABO blood group genotype and the Persistence trait is relatively weak.

also read Japanese Psychologists' Reactions (2009)


English Home Page


Last update: October 12, 2015

E-mail: mailto:abofan@js2.so-net.ne.jp