Methods

Participants

The data used in this study (Appendix A and Supplementary Material File) was collected by the means of crowdsourcing. With the intention of avoiding ethical issues, question items on blood type characteristics of this study were extracted from reviewed academic articles (Sato et al. 1992; Watanabe 1994). These items were checked again at the previously stated crowdsourcing company which has been passed the Japanese privacy mark (JIS Q 15001). The company confirmed that there was no problem with the question items. It provides customers with anonymized data and obtains informed consent from participants (respondents) prior to its surveys.
Survey 1 was conducted in 2018 as a preliminary study with a sample size of 1,000. Japanese single individuals with the age of 20-39 were asked to rate a total of 8 items representing 4 blood type (A, B, O, AB) characteristics, each with scores of 1-5 for their personality traits (the larger the number, the more fitting the trait), and scores of 1 to 4 of relationship and their knowledge on the level between blood type and personality (the larger the number, the more related or informed). Survey 2 was conducted in 2019, and covered 2,000 Japanese people ages 20-59, further expanded blood type characteristics to 12 items, and asked the respondents to answer which blood type they thought these 12 items would categorize. In Survey 2, we added an item of marital status, since the age of participants spanned to 59, which meant that married persons were included.
Each participant’s blood type of was determined by self-report, because most Japanese people know their blood type. In Survey 1, the dataset included only those who knew their blood type. Hence, the valid sample size was 1,000. In Survey 2, 141 out of all the 2,000 respondents, or 7.1% of the total, did not know their blood type. Thereby the data of the remaining 1,859 participants with known blood types were used.
The result of preliminary Survey 1 suggested that people who were less interested in blood type had traits of introvert, unsociable, and less interested in personality. Subsequently we added three new question items in Survey 2 to confirm this. These items were determined by the same process that was used to extract question items on the characteristics of the blood type. All samples were equally allocated by age of 10-year increments, and gender.

Procedure and Analytical Strategy

We deliberately selected the most suitable characteristics that would certainly produce the differences, and then focused our analysis on whether personality self-fulfillment was occurring or not: with data from participants who “have no knowledge of blood type characteristics” (hereinafter abbreviated as “no-knowledge group”). Survey 1 had only 120 “no-knowledge” participants therefore we would mainly use the 325 “no-knowledge” participants of Survey 2, who were free of “self-fulfilling”, because they had no knowledge of blood type personality. Differences between groups with large and small characteristic differences according to blood type were also examined.
Our analytical methods were as follows:
Analysis 1: Cross-tabulation of blood type and personality ‒ Survey 1 and 2
Analysis 2: Cross-tabulation of knowledge and personality ‒Survey 1 and 2
Analysis 3: Cross-tabulation of knowledge, extroversion and sociability ‒ Survey 2 only
We set the alpha level to 0.05. The ANOVA, and the binominal test were used to analyze blood type characteristics. The effect sizes (Cohen, 1977) were also calculated. The dependent variables were analyzed by both the blood type factor and the covariates of gender, age (10-year increments), and marital status (since Survey 1 participants were all single, Survey 2 only). The distributions of blood types were almost equal to the Japanese average (Okubo, 1997), so the results of χ2 test for the bias in both surveys were not significant.