We tried to follow those methods of psychological personality testing,
and deliberately selected the most suitable characteristics that would
explicitly display the differences: a) images of characteristics were
consistent to the preceding academic studies; b) showed large
differences in the academic studies and means were close to 50%; c) did
not show extreme values; d) images of characteristics were consistent to
the preceding surveys of other studies.
Analytical
Strategy
Firstly, we confirm whether a single or two-item question made a
difference in the preceding academic studies. We also focused our
analysis on whether personality self-fulfillment was occurring: with
data from participants who “have no knowledge of blood type
personality” or “do not believe in the relationship” (hereinafter
abbreviated as “no-knowledge group”).
Secondly, we confirm whether AI can predict a persons’ blood type more
than chance. We made to reference a facial recognition article written
by AI engineers (Nakamura & Iwasada, 2017) aimed at prediction because
there was virtually no preceding research available to analyze
personality with AI; it was for this reason we utilized AI as an
experimental method.
Procedure
Our analytical methods on personality were as follows:
Analysis 1: ANOVA of blood type and personality – Surveys 1 and 2
Analysis 2: ANOVA of blood type and personality for “no-knowledge
group” – Survey 2 only
We set the alpha level to 0.05. An analysis of variance (ANOVA) was
performed with personality characteristics scores as the dependent
variables and self-reported ABO phenotypes (A, B, O and AB). Before
ANOVA analyses, the normality of distributions was checked for each
personality characteristics score; this showed a normal distribution.
Effect sizes (Cohen, 1977) were also calculated. The data were analyzed
using jamovi software version 1.2.27 (The jamovi project).
Our analytical method on blood type predictions using AI was as follows:
Analysis 3: Blood type prediction using AI – Survey 2 only
All the data of survey 2 were stored in Amazon S3, and with Amazon
Machine Learning, all 12 characteristics, including gender, age, and
marital status, were used as training data for prediction targeting for
the blood type (since Masahiko Nomi claimed that these elements affected
personalities). Multinominal logistic regression algorithm was chosen
for the prediction. We divided the whole data into five groups of same
sample size. Each group was estimated as the prediction data, the rest
four groups as the training data, and then the average of the five
predictions was calculated. In these cases, since the sample sizes of
the AI training data were small (this means that the prediction errors
might become larger if we used the raw data of 1-year increment of age),
a dummy variable of 10-year increments was used [20s = 2, 30s = 3, 40s
= 4, 50s = 5].