Mar-Sat aspects for 3646 scientists
16 views
Skip to first unread message
Ray Murphy
Aug 21, 2009, 9:07:29 AM8/21/09
to Tropical Astrology Research
Hi All,
If we look at the 3646 Gauquelin data for the scientists who were born
during the 124 year period of the dataset from 1794 to 1918 it is easy
to see that people *consistently* gravitated towards becoming
scientists at 25% above the average rate if they had MAR sqr SAT <
11.25 degs and consistently avoided becoming scientists if they had
MAR con or opp SAT < 11.25 degs at 23.6% below the average rate.
When I say "consistently" I'm referring to the progressive results at
any given time in the 3646 data.
Note: The 11.25 degree orb is not meant to be anything special. It is
simply a good fit when the circle is divided into 16 parts to show the
distribution of 16H aspects.
The key to all of this is the ~consistency~ and not just different
scores.
Ray
If we look at the 3646 Gauquelin data for the scientists who were born
during the 124 year period of the dataset from 1794 to 1918 it is easy
to see that people *consistently* gravitated towards becoming
scientists at 25% above the average rate if they had MAR sqr SAT <
11.25 degs and consistently avoided becoming scientists if they had
MAR con or opp SAT < 11.25 degs at 23.6% below the average rate.
When I say "consistently" I'm referring to the progressive results at
any given time in the 3646 data.
Note: The 11.25 degree orb is not meant to be anything special. It is
simply a good fit when the circle is divided into 16 parts to show the
distribution of 16H aspects.
The key to all of this is the ~consistency~ and not just different
scores.
Ray
Axel
Aug 21, 2009, 10:22:48 AM8/21/09
to Tropical Astrology Research
Ray, I like the consistency approach because it puts things in
historical perpective. Even an INconsistent result, provided it is
consistent within several contiguous time periods, can be useful; e.g.
one can look for other factors which caused the change.
historical perpective. Even an INconsistent result, provided it is
consistent within several contiguous time periods, can be useful; e.g.
one can look for other factors which caused the change.
Graham Douglas
Aug 21, 2009, 6:20:09 PM8/21/09
to tropical-astr...@googlegroups.com
| Hi Ray, Yes it's interesting, but remember that MA especially tends to appear more often in some signs than others. I looked at it also in heliocentric coords and of course the angle changes from 90/270 to 60/240 but the significance also drops greatly with a Chi-sq of 17.5 with Df = 11. But its still interesting. Graham. --- On Fri, 21/8/09, Ray Murphy <raym...@tpg.com.au> wrote: |
Ray Murphy
Aug 22, 2009, 2:52:20 AM8/22/09
to Tropical Astrology Research
On Aug 22, 7:20 am, Graham Douglas <ondastropic...@yahoo.co.uk> wrote:
> Hi Ray,
> Yes it's interesting, but remember that MA especially tends to appear more
> often in some signs than others. I looked at it also in heliocentric
> coords and of course the angle changes from 90/270 to 60/240 but the
> significance also drops greatly with a Chi-sq of 17.5 with Df = 11. But
> its still interesting.
> Graham.
Hi Graham,
> Hi Ray,
> Yes it's interesting, but remember that MA especially tends to appear more
> often in some signs than others. I looked at it also in heliocentric
> coords and of course the angle changes from 90/270 to 60/240 but the
> significance also drops greatly with a Chi-sq of 17.5 with Df = 11. But
> its still interesting.
> Graham.
It was a big error. I've looked further and found the same trend in
random data and it is even more prominent in data for every 2 days for
the 124 years.
The reason for me overlooking such an obvious thing is because I've
been doing a huge amount of work for months with up to 140 categories
of occupation etc simultaneously - where some categories have a factor
quite strongly and others don't. They are the types of surveys that
often don't need random data for comparisons, and I overlooked using
it after seeing that a category like Gauquelin Sportsmen [roughly in
the same time frame - with a 20 year offset] didn't have it.
Incidentally, it would have been 8 Df for 9 bars in a graph -
00
22.5
45
67.5
90
112.5
135
157.5
180
Also, I've found that we are almost guaranteed to never get anywhere
in astrology with Chi square the way it's been handled for decades.
I've run a series of tests and found that we could conceivably run
10,000 experiments and still not get a decent p-score - which is why
we haven't been seeing them.
It seems that the only way we can get any decent p-scores in astrology
is to compare factors with "non-factors" - such as Aries and non-
Aries. When that is done we get computer simulated results that
conform to he theory that certain p-scores should be seen at certain
intervals. With the current system we rarely (if ever) see it
happening.
Ray
Graham Douglas
Aug 22, 2009, 5:49:20 AM8/22/09
to tropical-astr...@googlegroups.com
| Hi Ray, Yes, I just looked at it with 12 when I did the Chi2 because I've been in the habit of using 12. Can you say a bit more about why Chi2 doesn't work I don't follow. Graham. |
--- On Sat, 22/8/09, Ray Murphy <raym...@tpg.com.au> wrote: |
|
Ray Murphy
Aug 22, 2009, 6:27:34 AM8/22/09
to Tropical Astrology Research
[....]
here:
Take this graph for the ADB category of "Long Life" where Moon in 7th
house Placidus is extremely prominent and also the best graph I've
ever come across in about 1,000,000 I've looked at in astrological
research.
http://tiny.cc/czVn3
Jigsaw gives a p-score that says we can find a graph like that 1 in 3
times we look (on average) purely by chance, but it's simply not true.
If anyone doesn't believe this, try generating random random numbers
from 1 to 12 (to represent the 12 houses) and then score them and see
how many hundreds or thousands of random data sets it *actually* takes
to find a graph that good.
I'm sure you would agree that if some astrologer rocked up here with
an observation that had yielded a p-score of 0.05 you would be saying
"Look it's interesting but that can happen easily by chance 1 in 20
times you look (on average) - in fact it is SUPPOSED to happen, and
something would be awfully wrong if it didn't", but if someone like me
rocks up with my Moon in the 7th observation and a p-score of 0.3, you
could be excused for saying "Get outa here -- 0.3 is nothing" and yet
when we look at the graph we can see it is extraordinarily rare to see
such a thing in astrology.
Ray
> Can you say a bit more about why Chi2 doesn't work I don't follow.
RM: There are several ways to approach this problem, but let's start
here:
Take this graph for the ADB category of "Long Life" where Moon in 7th
house Placidus is extremely prominent and also the best graph I've
ever come across in about 1,000,000 I've looked at in astrological
research.
http://tiny.cc/czVn3
Jigsaw gives a p-score that says we can find a graph like that 1 in 3
times we look (on average) purely by chance, but it's simply not true.
If anyone doesn't believe this, try generating random random numbers
from 1 to 12 (to represent the 12 houses) and then score them and see
how many hundreds or thousands of random data sets it *actually* takes
to find a graph that good.
I'm sure you would agree that if some astrologer rocked up here with
an observation that had yielded a p-score of 0.05 you would be saying
"Look it's interesting but that can happen easily by chance 1 in 20
times you look (on average) - in fact it is SUPPOSED to happen, and
something would be awfully wrong if it didn't", but if someone like me
rocks up with my Moon in the 7th observation and a p-score of 0.3, you
could be excused for saying "Get outa here -- 0.3 is nothing" and yet
when we look at the graph we can see it is extraordinarily rare to see
such a thing in astrology.
Ray
Graham Douglas
Aug 22, 2009, 7:16:29 AM8/22/09
to tropical-astr...@googlegroups.com
| Hi Ray, I guess what you are saying is that since nearly all the Chi2 score comes from just one point that point is very significant ? I'm not a statistician, but I imagine what they would say is: either 1) you can predict PI as the most frequent sign for MO in which case you have indeed got something impressive , doing a Chi2 with just Df = 1 and only the one-tailed version which divides the p value by 2 again, OR 2), this is the first time you found it and if you run loads more and find nothing then this just becomes one among many and the p increases each time. In this case finding similar or varied versions of the same pattern in other data sets would be more important than just comparing with the control. Cheers for now, |
Graham. --- On Sat, 22/8/09, Ray Murphy <raym...@tpg.com.au> wrote: |
|
Graham Douglas
Aug 22, 2009, 7:23:33 AM8/22/09
to tropical-astr...@googlegroups.com
| Hi Ray, BTW I am interested in the aspect data or perhaps its phases, especially in relation to the earth. Have a look at ER-MA in heliocentric coords for Sports after breaking the data into subsets according to the MA placing in GQ sectors (geocentric of course). I think pre-natal conjunctions or other aspects might be involved in outbursts of solar activity which could influence the foetus during the last 3 months of pregnancy. Regards, |
Graham. --- On Sat, 22/8/09, Ray Murphy <raym...@tpg.com.au> wrote: |
|
Ray Murphy
Aug 22, 2009, 3:44:58 PM8/22/09
to Tropical Astrology Research
> I guess what you are saying is that since nearly all the Chi2 score comes from just one point that point is very significant ? I'm not a statistician, but I imagine what they would say is:
> either 1) you can predict PI as the most frequent sign for MO in which case you have indeed got something impressive , doing a Chi2 with just Df = 1 and only the one-tailed version which divides the p value by 2 again,
> OR 2), this is the first time you found it and if you run loads more and find nothing then this just becomes one among many and the p increases each time. In this case finding similar or varied versions of the same pattern in other data sets would be more important than just comparing with the control.
> Cheers for now,
> Graham.
>
RM: It was MO in 7th (PI was the other discussion for scientists).
> either 1) you can predict PI as the most frequent sign for MO in which case you have indeed got something impressive , doing a Chi2 with just Df = 1 and only the one-tailed version which divides the p value by 2 again,
> OR 2), this is the first time you found it and if you run loads more and find nothing then this just becomes one among many and the p increases each time. In this case finding similar or varied versions of the same pattern in other data sets would be more important than just comparing with the control.
> Cheers for now,
> Graham.
>
The unusual thing about this is that stats don't matter much because
the strong trend would have been visible in the early data about 80
years ago if anyone had looked, and it has remained well above the
average ever since (in the available data).
I hasten to add that consistent trends can disappear. For example I
noticed in the Gauquelin Sportsmens' data that Mars in Sector 4 (as a
peak) was non-existent after the first 74 years of data (in birth
sequence) at the 326th chart. If anyone wishes to check this - make a
copy of the Sportsmens' file and then sort it into birth order then
keep only the first 326 cases.
One of the big questions for astrologers is - How long must a trend
hold up consistently before it seems obvious that it's working? The
example above for Gauquelin Sportsmen indicates that (where 12
divisions are involved) the figure is at least 326 cases.
Ray
Ray Murphy
Aug 22, 2009, 4:45:08 PM8/22/09
to Tropical Astrology Research
On Aug 23, 4:44 am, Ray Murphy <raymu...@tpg.com.au> wrote:
[...]
> I hasten to add that consistent trends can disappear. For example I
> noticed in the Gauquelin Sportsmens' data that Mars in Sector 4 (as a
> peak) was non-existent after the first 74 years of data (in birth
> sequence) at the 326th chart. If anyone wishes to check this - make a
> copy of the Sportsmens' file and then sort it into birth order then
> keep only the first 326 cases.
>
> One of the big questions for astrologers is - How long must a trend
> hold up consistently before it seems obvious that it's working? The
> example above for Gauquelin Sportsmen indicates that (where 12
> divisions are involved) the figure is at least 326 cases.
[Extra]
After writing the above message I decided to have a look at another
random Gauquelin sector to see if a trend was not consistent and chose
to look first at Saturn in Sectors 1 or 4. I haven't looked at Sector
1 but Sector 4 (as a peak) didn't exist for the first 45 years and the
first 878 samples (in birth sequence). It was actually a trough until
that point.
This indicates that we probably shouldn't take too much notice of any
consistent looking trends up to the 878 mark.
Ray
[...]
> I hasten to add that consistent trends can disappear. For example I
> noticed in the Gauquelin Sportsmens' data that Mars in Sector 4 (as a
> peak) was non-existent after the first 74 years of data (in birth
> sequence) at the 326th chart. If anyone wishes to check this - make a
> copy of the Sportsmens' file and then sort it into birth order then
> keep only the first 326 cases.
>
> One of the big questions for astrologers is - How long must a trend
> hold up consistently before it seems obvious that it's working? The
> example above for Gauquelin Sportsmen indicates that (where 12
> divisions are involved) the figure is at least 326 cases.
After writing the above message I decided to have a look at another
random Gauquelin sector to see if a trend was not consistent and chose
to look first at Saturn in Sectors 1 or 4. I haven't looked at Sector
1 but Sector 4 (as a peak) didn't exist for the first 45 years and the
first 878 samples (in birth sequence). It was actually a trough until
that point.
This indicates that we probably shouldn't take too much notice of any
consistent looking trends up to the 878 mark.
Ray
Graham Douglas
Aug 23, 2009, 6:13:59 AM8/23/09
to tropical-astr...@googlegroups.com
| Hi Ray, Thanks for these figures...... or it could mean that the effect comes and goes which would be very interesting. I am already thinking that the two big key sectors may have separate trains of causation. |
| Graham. --- On Sat, 22/8/09, Ray Murphy <raym...@tpg.com.au> wrote: |
|
Ray Murphy
Aug 23, 2009, 7:59:30 AM8/23/09
to Tropical Astrology Research
On Aug 23, 7:13 pm, Graham Douglas <ondastropic...@yahoo.co.uk> wrote:
>Hi Ray,
>Thanks for these figures...... or it could mean that the effect comes and
>goes which would be very interesting. I am already thinking that the two
>big key sectors may have separate trains of causation.
I've noticed that these trends (where something appears to be working)
>Hi Ray,
>Thanks for these figures...... or it could mean that the effect comes and
>goes which would be very interesting. I am already thinking that the two
>big key sectors may have separate trains of causation.
come and go quite frequently with random data - and after much
experimentation with computer simulation, found that the frequency is
very close to what we would expect it to be.
For example, I found an apparent Mars-Saturn connection for
astrologers, but an examination of the trend over a long period showed
quite clearly that it was a transitory thing.
The key to it all is the consistency of the above-average-percentages
over a long period, while bearing in mind that small amounts of data
will often cause wild fluctuations in percentages in the early stages
of the progressive graphs.
Have another look at this graph on this group's website, where it
shows that people have gravitated towards becoming astrologers more
often if they had Sun in tropical Aquarius. Now that graph shows that
the trend "disappeared" after the first 100 or so, but it was far more
likely to have been the volatility of percentages rather than a real
'bombing-out' of the trend.
http://tiny.cc/RU7UC
A big question for us is "How much data should be used before a trend
is considered to have disappeared?"
As indicated above, it's not hard to find trends in date-sequenced
data but the hard part is finding trends that are present all along
(apart from that wild fluctuation at the beginning of graphs).
Ray
Ray Murphy
Aug 23, 2009, 3:23:10 PM8/23/09
to Tropical Astrology Research
On Aug 23, 7:13 pm, Graham Douglas <ondastropic...@yahoo.co.uk> wrote:
> Hi Ray,
> Thanks for these figures...... or it could mean that the effect comes and goes which would be very interesting. I am already thinking that the two big key sectors may have separate trains of causation.
> Graham.
I've been thinking the same thing because the Comedians and TV
> Thanks for these figures...... or it could mean that the effect comes and goes which would be very interesting. I am already thinking that the two big key sectors may have separate trains of causation.
> Graham.
personalities in Astrodatabank scored much higher than the Gauquelin
sportsmen for Sector 1 but had nothing happening in Sector 4.
I've just made a polar graph in Jigsaw after combining the Comedians
and TV personalities and removing the duplicates - giving 547 cases.
The GIF has been placed on this group's Files section - http://tiny.cc/tdFFX
I've also had a quick look in Jigsaw to see what percentages of the
Sportsmen and the Comedians/TV personalities had Mars in Sector 1
above the average. They were 130% and 175% respectively.
Ray
Graham Douglas
Aug 23, 2009, 5:30:47 PM8/23/09
to tropical-astr...@googlegroups.com
| Hi Ray, This is quite impressive though I'd like to know how accurate the data was. By the way, how do you manage to add those comments to the Jigsaw graphs ? Graham. |
--- On Sun, 23/8/09, Ray Murphy <raym...@tpg.com.au> wrote: |
|
Ray Murphy
Aug 23, 2009, 5:56:23 PM8/23/09
to Tropical Astrology Research
On Aug 24, 6:30 am, Graham Douglas <ondastropic...@yahoo.co.uk> wrote:
> I've just made a polar graph in Jigsaw after combining the Comedians
> and TV personalities and removing the duplicates - giving 547 cases.
> The GIF has been placed on this group's Files section -http://tiny.cc/tdFFX
> I've just made a polar graph in Jigsaw after combining the Comedians
> and TV personalities and removing the duplicates - giving 547 cases.
> The GIF has been placed on this group's Files section -http://tiny.cc/tdFFX
> Hi Ray,
> This is quite impressive though I'd like to know how accurate the data was.
RM: Yes, the data accuracy needs to be questioned, but the trend is
> This is quite impressive though I'd like to know how accurate the data was.
strong enough to make it worth following up. I'm inclined to believe
that most of it is fairly accurate, particularly because those two
categories of Comedian and TV Host came out so close together as very
high percentages amongst 100+ categories. The category of
"Disciplined" came in between them.
t shouldn't be too hard to start gathering similar data from non-
Astrodatabank sources to see if the same thing is happening in places
like Spain, France, Germany, Canada, and the U.K. etc, and if the
trend is real, it wouldn't take a lot of data to begin seeing it
repeated :-)
> By the way, how do you manage to add those comments to the Jigsaw graphs ?
(1) Screen capture is done with Gadwin Printscreen to give a bitmap of
2305k
(2) Image taken into Photoshop and alterations made
(3) Image saved as a Compuserv .gif (proper perfect quality gifs) at
21k
It's best to save the screenshot as a bitmap to capture the perfect
image, rather than capturing with el-cheapo gifs which are all a bit
fuzzy.
Yell out if you need any important images modified for publication
etc.
Ray
Ray Murphy
Aug 23, 2009, 8:07:14 PM8/23/09
to Tropical Astrology Research
[....]
> I've just made a polar graph in Jigsaw after combining the Comedians
> and TV personalities and removing the duplicates - giving 547 cases.
> The GIF has been placed on this group's Files section -http://tiny.cc/tdFFX
Hi Graham,
Your question about the accuracy of the charts for Comedians and TV
hosts led me to check on some other categories. The result was very
interesting because I saw that the categories of Composers and
Instrumentalists both scored low for Mars in Gauquelin Sector 1 at
about the same level, so I joined the two sets of data and then
removed the duplicates and had 732 to work with. They scored only 80%
of the average rate, compared with the 175% of the average rate of the
Comedians and TV Hosts -- Less than HALF!
Ray
> I've just made a polar graph in Jigsaw after combining the Comedians
> and TV personalities and removing the duplicates - giving 547 cases.
>
> I've also had a quick look in Jigsaw to see what percentages of the
> Sportsmen and the Comedians/TV personalities had Mars in Sector 1
> above the average. They were 130% and 175% respectively.
[Extra]
> I've also had a quick look in Jigsaw to see what percentages of the
> Sportsmen and the Comedians/TV personalities had Mars in Sector 1
> above the average. They were 130% and 175% respectively.
Hi Graham,
Your question about the accuracy of the charts for Comedians and TV
hosts led me to check on some other categories. The result was very
interesting because I saw that the categories of Composers and
Instrumentalists both scored low for Mars in Gauquelin Sector 1 at
about the same level, so I joined the two sets of data and then
removed the duplicates and had 732 to work with. They scored only 80%
of the average rate, compared with the 175% of the average rate of the
Comedians and TV Hosts -- Less than HALF!
Ray
Graham Douglas
Aug 24, 2009, 5:24:00 PM8/24/09
to tropical-astr...@googlegroups.com
| Hi Ray, Good sounds encouraging. Graham. |
--- On Mon, 24/8/09, Ray Murphy <raym...@tpg.com.au> wrote: |
|
Graham Douglas
Aug 24, 2009, 5:26:38 PM8/24/09
to tropical-astr...@googlegroups.com
| Hi Ray, Aha thanks, for the info about capture. Regarding TV hosts etc, of course it would be good to know who was born without any medical intervention. Cheers, Graham. |
--- On Sun, 23/8/09, Ray Murphy <raym...@tpg.com.au> wrote: |
|
Ray Murphy
Aug 25, 2009, 1:27:37 PM8/25/09
to Tropical Astrology Research
On Aug 25, 6:26 am, Graham Douglas <ondastropic...@yahoo.co.uk> wrote:
[...]
here's something that tends to support my contention that birth
inducement makes no difference to the way peoples' charts work:
It's something I've just spotted in the 2174 astrologers data where
they have consistently had Moon in Sector 10 at 20% above the average
rate - right through the data as it was plotted progressively.
That has led me to start systematically checking bodies in all the
sectors for 100 categories of occupation etc and it's beginning to
shape up like an astrology house system - where meaning seems to be
associated with each sector. It was fascinating to see the category of
'Psychologist' sitting right next to astrologers for Moon in Sector
10.
Ray
[...]
> Regarding TV hosts etc, of course it would be good to know who was born without any medical intervention.
RM: That could be hard to ascertain for minor surveys like these, but
here's something that tends to support my contention that birth
inducement makes no difference to the way peoples' charts work:
It's something I've just spotted in the 2174 astrologers data where
they have consistently had Moon in Sector 10 at 20% above the average
rate - right through the data as it was plotted progressively.
That has led me to start systematically checking bodies in all the
sectors for 100 categories of occupation etc and it's beginning to
shape up like an astrology house system - where meaning seems to be
associated with each sector. It was fascinating to see the category of
'Psychologist' sitting right next to astrologers for Moon in Sector
10.
Ray
Graham Douglas
Aug 28, 2009, 7:09:42 AM8/28/09
to tropical-astr...@googlegroups.com
|
Ray Murphy
Aug 28, 2009, 12:22:46 PM8/28/09
to Tropical Astrology Research
On Aug 28, 8:09 pm, Graham Douglas <ondastropic...@yahoo.co.uk> wrote:
> Hi Ray,
> I know you use JS so I wonder if you could see if yours does this ?
> Just open any file and go to Tabulate and Graph, then select Natal details at the bottom followed by Time(Local).
> When I display the result as a table the Total entry is always zero [......]
> Hi Ray,
> I know you use JS so I wonder if you could see if yours does this ?
> Just open any file and go to Tabulate and Graph, then select Natal details at the bottom followed by Time(Local).
> I can tell you about another bug that can really mess up your data handling if you are interested, maybe we should have a special group and complain together.
and I agree that we should get together to complain and offer
suggestions. This group would be a good place to discuss and collate
all known problems in Jigsaw or in any other software that is related
to astrological research. At one time I started a Yahoo group for
Jigsaw but no one had anything to say.
I'm sure we could all move ahead a lot quicker with our various
research interests if we included discussion about software that is
all connected as a "research net" - you know, Jigsaw, Astrodatabank,
Excel, MS Access, Solar Fire and various home made programs and add-
ons to commercial programs.
Ray
Reply all
Reply to author
Forward
0 new messages