Note. Standard deviations are in parentheses. Highest scores that match with blood type characteristics, and p < 0.05 are highlighted in bold. * p < 0.05 after the Bonferroni’s correction.
In Analysis 1, we analyzed the results for the individual question items. However, since the sample sizes of the ”no-knowledge” group was relatively small, we decided to analyze the total of the result for the same blood type to reduce statistical errors like the TIPI (results for individual items are shown in Appendix C). In Survey 2’s ANOVA result (Table 8), all 4 items showed the same results as those shown for blood type characteristics in the preceding psychology papers (Tables 2-4), and 2 items were statistically significant at p < 0.05, and 1 item after the Bonferroni’s correction.

Analysis 3: Blood Type Prediction Using AI

We did not conduct the prediction using the data of Survey 1, because the sample size was too small. In survey 2, the accuracy rates were 45.0% (F1 = 0.450) in the group that had good knowledge of blood type characteristics (542 participants with scores equal to 3 or higher in both item Q3 (relation) and item Q4 (knowledge)), and 40.1% (F1 = 0.401) in the entire 1,859 participants (Table 9). When gender, age, and marital status were excluded from both the learning and training data, the accuracy rates fell to 42.3% (F1 = 0.423), and 39.6% (F1 = 0.396) respectively. The most common blood type among Japanese is type A, which accounts for 37.2% (692 persons) in this survey. Hence, the accuracy rate became 37.2% if all the participants were assumed to be the type. Amazon Machine Learning predicted the blood type at a higher accuracy than this value in all cases (presently, there is no standard statistical test for personality analysis using AI).