Japanese Psychologists' Reactions (2009)


02.gif (288 バイト)The psychology paper that definitely concludes "there are statistical differences"

When I read an article of Scientific American's guest blog last year (Feb. 15, 2011) --  You are what you bleed: In Japan and other east Asian countries some believe blood type dictates personality -- I found Mr. Sakamoto's paper (he is a very friendly person).

To my surprise, he clearly said that there are definite relationships between blood types and "self-reported" personality.

Sakamoto, A., & Yamazaki, K. et al. (2004), Blood-typical personality stereotypes and self-fulfilling prophecy: A natural experiment with time-series data of 1978-1988, Progress in Asian Social Psychology, Vol. 4. Seoul, Korea, Pp. 239-262. (pdf)
This indicates that blood-typical personality stereotypes actually influenced the personalities -- self-reported personalities, at least -- of individuals, and that they also operated as a self-fulfilling prophecy, although the greatness of that influence could be discussed.

Would it does not become a hot topic in Japan because this paper is written in English?
If I read this article before the article, I did not have a hard time.
Anyway, the debate -- whether or not there is a statistical difference -- was already over!!

In 2011, Mr. Muto and Mr. Nagashima (then National Nagasaki University) had performed a supplementary examination of Mr. Sakamoto's paper, which is  mentioned above.

The number of samples was 30,000 people in 11 years in Mr. Sakamoto's paper.  To my surprise, as many as 200,000 people in 30 years are investigated by Muto and Nagashima's Grant-in-Aid for Scientific Research (Japan) in 2011.  However, the number of samples and years and were not specified in this report for some reason, so that  I have to suppose these numbers.  The followings are excerpts:

We also demonstrate that significant difference on personal characteristics between blood-types by using the database on recent huge questionnaires survey.

From the Database of Grants-in-Aid for Scientific Research
FY2011 Final Research Report: A demonstrative and critical study on pseudo-science for scientific literacy construction at teacher education course
代表者 武藤 浩二 Principal Investigator MUTO, Cosy (National Nagasaki Univ.)
研究分担者 長島 雅裕 Co-Investigator MASAHIRO, Nagashima (National Nagasaki Univ.)

(October 11, 2015)

02.gif (288 バイト)Japanese Psychologists' Reactions (2009)

After the Japanese blood-type boom of 2004, the established blood-typical theory of psychology has become "there are statistical differences."
The conventional "established theory" --  "no statistical difference" was completely overturned!!
As a result, a psychological established theory became to accord for my results: "there are statistical differences."
By self estimation, at least ...
I feel very happy!!  Wow!!
I wonder why the situation changed so suddenly and easily?

For examle:

1. Kudo Eriko (Tokyo Woman's Christian University), Influence over self personality evaluation by blood-typical stereotypes, 73rd convention of the Japanese Psychology Association (2008).

--> Generally, it is recognized that self-evaluation of the subjects was equal to their blood-typical stereotypes.

自分の性格の評価に血液型ステレオタイプが与える影響 工藤 恵理子(東京女子大学) 日本心理学会第73回大会発表論文集 2008年

2. Yamaoka Shigeyuki (Shotoku University), Influence over self recognition of the blood-typical personality items by watching TV programs, 47th Convention of Japanese Society of Social Psychology (2006).

--> The main effect of blood types was recognized in 11 items in the high reception group, and the main effect of the blood types was recognized in 2 items in the low reception group.

血液型性格項目の自己認知に及ぼすTV番組視聴の影響 山岡 重行(聖徳大学) 日本社会心理学会大会第47回大会発表論文集 2006年

3. Kubota Kenichi (Nagoya City University), Implicit blood-typical beliefs and self information processing, 48th Convention of Japanese Society of Social Psychology (2007).

--> The main effect (F(1,32)=9.80, p.01) of personality items and the interaction (F(3,32)=3.22, p.05) of blood-typical-personality items by participants was significant ... After performing two similar factor analyses of variance about the result of definition exercises, the interaction of blood-typical-personality items by participants was significant.

潜在的な血液型ステレオタイプ信念と自己情報処理 久保田健市(名古屋市立大学) 日本社会心理学会大会第48回発表論文集 2007年
-->特性語の種類の主効果(F(1,32)=9.80, p<.01)と特性語の種類×参加者の血液型の交互作用(F(3,32)=3.22, p<.05)が有意だった…定義づけ課題の結果についても,同様の2要因分散分析を行ったところ,特性語×参加者の血液型の交互作用が有意だった.

Here is a Korean one:

4. Blood-type distribution; Beom Jun Kim, Dong Myeong Lee, Sung Hun Lee and Wan-Suk Gim; Physica A: Statistical and Theoretical Physics Volume 373, 1, January 2007, Pages 533-540

--> A psychological implication for the case of B-type males is also suggested as an effect of a distorted implicit personality theory affected by recent popularity of characterizing a human personality by blood types.

(August 6, 2012)

02.gif (288 バイト)Japanese Psychologists' Reaction (2013)

The 80th convention of the Japan Association of Applied Psychology (2013)

The 14P-41 Is "blood-type characterology" reliable ? (30th report) I
-- Are some characteristics of blood types seen in the members of the House of Representatives? --

Masao Ohmura (Nihon University)
Shuich Ukiya (Tokyo Fuji University)
Shuich Fujita (Nippon Sport Science University)

Type Os prevail among Japanese prime ministers, as shown in Table 4.

(March 3, 2015)

02.gif (288 バイト)Japanese Psychologists' Reaction (2014)

On the contrary, Mr. Nawata said that no statistical relationship was found.

No relationship between blood type and personality:
Evidence from large-scale surveys in Japan and the US
Kengo Nawata (Kyoto Bunkyo University)

Abstract Despite the widespread popular belief in Japan about a relationship between personality and ABO blood type, this association has not been empirically substantiated. This study provides more robust evidence that there is no relationship between blood type and personality, through a secondary analysis of large-scale survey data. Recent data (after 2000) were collected using large-scale random sampling from over 10,000 people in total from both Japan and the US. Effect sizes were calculated. Japanese datasets from 2004 (N= 2,878-2,938), and 2,005 (N = 3,618-3,692) as well as one dataset from the US in 2004 (N = 3,037-3,092) were used. In all the datasets, 65 of 68 items yielded non-significant differences between blood groups. Effect sizes (η2) were less than .003. This means that blood type explained less than 0.3% of the total variance in personality. These results show the non-relevance of blood type for personality.

Key words: blood-typing, blood type stereotypes, blood group, personality, pseudoscience.

The Japanese Journal of Psychology, 85(2), 148-156, 2014.

However, his conclusion may be wrong ...  Some differences are underestimated and/or ignored.
To the first,
he said "effect sizes (η2) were less than .003" but this is an underestimation. The adjusted value is 1 - 2 % because the number of type AB is very small in the US.
Secondly, the following question for Americans is
not calculated:
    Q22: If you make a critical decision, take risks or avoid
    -> Its effect size (
η2) is about 3%!
Interesting enough, tendency is totally opposite between Japanese and Americans. As for Japanese, type ABs are the most risk aversion people, while American ABs take risks most.

(March 3, 2015)

** APPENDIX **  (August 10, 2014)

Relationship between blood types and personality is real, not imaginary!

Mr. Masayuki Kanazawa [金澤正由樹] (August 10, 2014)

He wrote a book "An Introduction to Blood Type Humanics by Using Statistics.(血液型でわかる血液型人間学入門)" in 2014.

If relationship of blood types and personality are imaginary (i.e. caused by blood-typical briefs), scores of statistical differences must be proportional to those of trait (blood-typical personality) recognition.
But the reality is, there is little relationship. The coefficient of correlation is minus 0.10!
This result suggests that relationship of blood types and personality is real, not imaginary.

Mr. Masayuki Kanazawa analyzed blood-typical 28 question items of two surveys conducted by Japanese psychologists (Ms. Yoriko Watanabe and Mr. Shigeyuki Yamaoka *1).  The original surveys (i.e. combined results) said that 15 answers showed statistical significance with much blood-typical knowledge subjects (total number of questions is 28, as mentioned).  On the contrary, no answer showed statistical significance with little blood-typical knowledge subjects.  Many psychologists thought these differences were caused by "blood-typical beliefs" (conformation bias and/or self-fulfilling phenomenon) so that relationship of blood types and personalities are not real.

Strangely enough, seven questions out of those significant fifteen ones are "wrong answers".  For example, scores of type O people are sometimes higher than A people, even if questions are originally for type "A" -- not O.  So, he eager to check the phenomenon.

If relationship of blood types and personality is imaginary (i.e. caused by blood-typical briefs), scores of statistical differences must be proportional to those of trait  (blood-typical personality) recognition.

But the reality is, as shown thin dots and the thin solid line blow, there is little relationship. The coefficient of correlation is minus 0.10!
If there are any positive relations, the coefficient of correlation must be plus! (shown in the thick solid line below)

This result suggests that relationship of blood types and personality is real, not imaginary which is caused by blood-typical briefs.

*1 山岡重行 (2001) ダメな大人にならないための心理学

English Home Page

Last update: March 3, 2015. (Minor corrections on August 12, 2017)