Твитнуть
Всё там есть. Страницы 188, 189:
A more powerful test was made possible by data from a study unconnected
with astrology (Dean, forthcoming) involving 2,101 persons born in London
during 39 March 1958. They were born on average 4.8 minutes apart, so they
were precisely those for which Addey had predicted really exceptional similarities
of life and temperament. Measurements at ages 11, 16 and 23 had provided
for each person 110 relevant variables including test scores for IQ, reading and
arithmetic; teacher and parent ratings of behaviour such as anxiety, aggressiveness
and sociability; physical data such as height, weight, vision and hearing;
self-ratings of ability such as art, music and sports; and various others such as
occupation, accident proneness and marital status; all of which are supposed to
be shown in the birth chart. Included as a control were sixteen variables for the
mother such as age, blood pressure and length of labour; seven leading astrologers
agreed unanimously that these sixteen variables would definitely not be
shown in the birth chart of the child.
About 92% of birth times had been recorded to the nearest 5 minutes, the rest
to the nearest minute. Before analysis the subjects were arranged in chronological
order of birth. This gave 2,100 successive pairs of time twins; 73% were born
5 minutes apart or less, and only 4% were born more than 15 minutes apart. The
similarity between time twins for each variable was then measured as the serial
correlation between successive pairs AB, BC, CD and so on. Here serial correlation
is a direct measure of effect size and is extremely sensitive due to the large
sample size. Unlike the Roberts and Greengrass approach (count each time twin
against every other time twin within one day, which would have produced nearly
600,000 pairs), serial correlation counts each time twin once only, thus minimizing
the risk of artifacts. So the test conditions could hardly have been more conducive
to success. But the results are uniformly negative, see Table 2. The effect
size due to astrology is 0.00 ± 0.03.
A more powerful test was made possible by data from a study unconnected
with astrology (Dean, forthcoming) involving 2,101 persons born in London
during 39 March 1958. They were born on average 4.8 minutes apart, so they
were precisely those for which Addey had predicted really exceptional similarities
of life and temperament. Measurements at ages 11, 16 and 23 had provided
for each person 110 relevant variables including test scores for IQ, reading and
arithmetic; teacher and parent ratings of behaviour such as anxiety, aggressiveness
and sociability; physical data such as height, weight, vision and hearing;
self-ratings of ability such as art, music and sports; and various others such as
occupation, accident proneness and marital status; all of which are supposed to
be shown in the birth chart. Included as a control were sixteen variables for the
mother such as age, blood pressure and length of labour; seven leading astrologers
agreed unanimously that these sixteen variables would definitely not be
shown in the birth chart of the child.
About 92% of birth times had been recorded to the nearest 5 minutes, the rest
to the nearest minute. Before analysis the subjects were arranged in chronological
order of birth. This gave 2,100 successive pairs of time twins; 73% were born
5 minutes apart or less, and only 4% were born more than 15 minutes apart. The
similarity between time twins for each variable was then measured as the serial
correlation between successive pairs AB, BC, CD and so on. Here serial correlation
is a direct measure of effect size and is extremely sensitive due to the large
sample size. Unlike the Roberts and Greengrass approach (count each time twin
against every other time twin within one day, which would have produced nearly
600,000 pairs), serial correlation counts each time twin once only, thus minimizing
the risk of artifacts. So the test conditions could hardly have been more conducive
to success. But the results are uniformly negative, see Table 2. The effect
size due to astrology is 0.00 ± 0.03.
Комментарий