Quote:
Originally Posted by Eckelston
Just a few comments:
Your second trial completely failed to reproduce the finding of the first one about Sun/Uranus aspects. You looked at 39 aspects and one of them was unusual in that it happens once in 66 samples. In a sample of 39 this will happen 45% of the time. 2 out of 39 was less likely than 2 percent. This happens 18% of the time with random data. Basically you took a 100 sided dice, rolled it 39 times and got a 1 and a 2. Hardly unusual. You've got a good sample size and yet no signal. If there was the kind of effect you wrote in your first posts it would have had to show up in this kind of sample pretty clearly. Indeed if the original trend held the result would have been significant to 1-3.3*10-11.
I'm not sure I'm getting this right but you plotted the frequency in one subgroup (then on a different plot for the other subgroup) against the frequency (or rank in frequency) in the complete sample? If you did this you can't expect the trend lines to move to opposite directions. The aspect with the lowest frequency will have one of the lowest frequency in both subsamples. That's why it's lowest, it got unlucky twice. Same with the highest. You expect some aspects to get lucky twice and in turn you expect them to turn up in the upper right in both plots. You don't expect flat lines either because the expected distribution is not uniform.
|
Thank you for this analysis, it concurs with my comment that the result was positive but weak. However, I disagree with your assertion that the sample size is necessarily good, as your comment discounts the likelihood that a larger sample may detect real but weak signals. I plan to follow up by broadening the study. Granted, the Mars-Saturn linkage - evident in these scientist’s birth times sixty times more often than in the general population - is well within statistical expectations. The question remains whether a larger sample will deliver results that are significant by finding trends in this group that differ consistently from others. Comparison of this group with others may find real variance in frequency of specific planetary relations among different professions.
I checked on Richard Tarnas’ comments on Uranus/Sun. His argument is that this aspect often occurs at birth of outstanding innovators. It would not be expected to happen more often than chance among the eminent astronomers and scientists analyzed here, unless it were established that innovation was a feature of this group.
Thank you for the correction on my trend analysis, which as you have shown is not significant. It shows my statistics is a bit rusty. The underlying intention was that a bigger sample and longer time series would enable comparison of frequencies of specific aspects by decade, with potential to identify any trends.