carlson test and debunking randi

A shy extrovert would be somebody who seeks out the company of others and defines himself by his relationships with others, yet feels embarrassed and uncomfortable doing so.

It would be like being devoted to a hobby you don't enjoy, but at least it would be possible. It also shows you why you can't give someone a horoscope and ask them how accurate it is. It's too subjective. I can't think of any way of testing astrology that doesn't involve comparing one group of horoscopes to some sort of know data. The two cases we suggested are personality tests or questionnaires and simply asking people which horoscopes they like.

well tell us what test you want.
note that astrology doesnt tell the future, it shows trends, like, one day youre likely to be angry, so, youll be more prone to accidents:D
 
well tell us what test you want.
note that astrology doesnt tell the future, it shows trends, like, one day youre likely to be angry, so, youll be more prone to accidents:D

Another avoidence.... It's your claim, you suggest a test and sucess/failure criteria.

Also, we had agreed on what astrology does...."trends" was never mentiond
 
Last edited:
A shy extrovert would be somebody who seeks out the company of others and defines himself by his relationships with others, yet feels embarrassed and uncomfortable doing so.

It would be like being devoted to a hobby you don't enjoy, but at least it would be possible. It also shows you why you can't give someone a horoscope and ask them how accurate it is. It's too subjective. I can't think of any way of testing astrology that doesn't involve comparing one group of horoscopes to some sort of know data. The two cases we suggested are personality tests or questionnaires and simply asking people which horoscopes they like.

No, just giving someone a horoscope and asking them to rate its accuracy is not a good test. The Forer effectWP comes into play here, meaning that people will tend to rate the horoscope as good description of them, even if it is in fact the horoscope of someone else (a mass murderer for instance, as in the test done by Michel Gauquelin that is described about half-way down this page).

But good tests are possible and have already been done. What is needed is a situation where each person has to choose the horoscope that fits them best out of a certain number of different ones. This is how one of the Carlson tests was organised: each subject received three horoscopes, one of which corresponded to their birth data. Each subject had a 1 in 3 chance of choosing their own horoscope. Of 83 people, 28 chose the horoscope that corresponded to their own birth data: exactly what would happen if the people had chosen at random.
 
MichaelC - It will be interesting to see what idunno suggests test wise; its his claim, he needs to come up with a testing process.

That said, something like the Carlson test you mention would be easy enough to do on line I'd think.
 
well tell us what test you want.
note that astrology doesnt tell the future, it shows trends, like, one day youre likely to be angry, so, youll be more prone to accidents:D

Well, a couple of tests have been suggested here so far. The astrologer should select the tests though, because he has the best knowledge and can maximize success.

For example, would a test that describes what kind of day I had yesterday be more reliable than a test that describes what kind of person I am? The test should look at whichever is the most reliable.
 
In order to do a test like that on line, we need a person who is accepted as arbitrator, who centralises data and only passes on anonymous birth data to the astrologer and anonymous horoscopes to the test subjects (similar to Tricky's proposition higher up in this thread). Both the astrologer and the test subjects would have to trust this person not to help one side or the other. How do we organise that?
 
Here's an idea:

3 people provide their birthdates/times/whatever data is needed for a chart. This should be provided in the thread.

idunno can then cast these dates and make the three horoscopes. DO not post them here, however. Instead, we should have a third party identified as the "escrow holder". The three horoscopes are sent to him, labelled with the names of who they belong to. He, in turn, will remove the names (relabelling them A,B,C) and send the three out to each of the three birthday providers. Each provider will review the three, and send the escrow holder the letter for the one they feel matches them most closely. The escrow holder can then post the results.

Of course, three people is not really a significant sample size, but it's a start on a baseline for the protocol, in an online form.

ETA: Actually, MichaelC's idea of keeping even the birthdates anonymous is a good one. Otherwise, there's an oppurtunity for research into the poster to make a horoscope more accurate. HAve to rule out all possibilities for error or deciet.
 
No, just giving someone a horoscope and asking them to rate its accuracy is not a good test. The Forer effectWP comes into play here, meaning that people will tend to rate the horoscope as good description of them, even if it is in fact the horoscope of someone else (a mass murderer for instance, as in the test done by Michel Gauquelin that is described about half-way down this page).

But good tests are possible and have already been done. What is needed is a situation where each person has to choose the horoscope that fits them best out of a certain number of different ones. This is how one of the Carlson tests was organised: each subject received three horoscopes, one of which corresponded to their birth data. Each subject had a 1 in 3 chance of choosing their own horoscope. Of 83 people, 28 chose the horoscope that corresponded to their own birth data: exactly what would happen if the people had chosen at random.

oh im aware of forer effect, thats why i dont trust the judgement of believers. i agree with huntsman and Christine.Lets get on with it:boxedin:
 
I have to run, but if nobody else volunteers I will be either arbitrator or one of the test subjects. I'll also calculate the odds for getting 3 of 3 (or whatever we end up with) by chance.

I'll be back in around five hours.
 
I have to run, but if nobody else volunteers I will be either arbitrator or one of the test subjects. I'll also calculate the odds for getting 3 of 3 (or whatever we end up with) by chance.

I'll be back in around five hours.

and me in 2 hours:)
 
Here's an idea:

3 people provide their birthdates/times/whatever data is needed for a chart. This should be provided in the thread.

idunno can then cast these dates and make the three horoscopes. DO not post them here, however. Instead, we should have a third party identified as the "escrow holder". The three horoscopes are sent to him, labelled with the names of who they belong to. He, in turn, will remove the names (relabelling them A,B,C) and send the three out to each of the three birthday providers. Each provider will review the three, and send the escrow holder the letter for the one they feel matches them most closely. The escrow holder can then post the results.

Of course, three people is not really a significant sample size, but it's a start on a baseline for the protocol, in an online form.

ETA: Actually, MichaelC's idea of keeping even the birthdates anonymous is a good one. Otherwise, there's an oppurtunity for research into the poster to make a horoscope more accurate. HAve to rule out all possibilities for error or deciet.

The birthdates must be anonymous. It won't do if idunno knows which posters are participating, nor if the subjects themselves know who the other subjects are. A possibility would be to use spouses, family members or friends as test subjects. Three posters could agree to give the birth data of someone they know well, who accepts to participate in the test. We would know who the three posters are, but we wouldn't know who they had chosen. Each of the three posters would send the birth data of the person they had chosen to the escrow holder, who would communicate this data, labelled only A,B,C to idunno.
 
Each of the three posters would send the birth data of the person they had chosen to the escrow holder, who would communicate this data, labelled only A,B,C to idunno.

I will happily volunteer to be the escrow holder. I would also suggest that all "escrowed" communication happen via forum PM, since that way (if necessary) one of the admins like
Darat can confirm the validity of the contents.
 
Well, a couple of tests have been suggested here so far. The astrologer should select the tests though, because he has the best knowledge and can maximize success.

Exactly. I think we are putting the cart before the horse here....idunno needs to take the lead and define a test, not folks here.

Comments like "seems ok to me" and "lets get on with it" are simply to vague.

Idunno - if you want to do a test, speak up and tell us what you'd like to do, vs just give a vague "seems ok to me". If you are going to continue to be vague, or avoid questions because of time constraints you might have that is fine....just speak up and we can find someone else.
 
Exactly. I think we are putting the cart before the horse here....idunno needs to take the lead and define a test, not folks here.

Comments like "seems ok to me" and "lets get on with it" are simply to vague.

Idunno - if you want to do a test, speak up and tell us what you'd like to do, vs just give a vague "seems ok to me". If you are going to continue to be vague, or avoid questions because of time constraints you might have that is fine....just speak up and we can find someone else.

I suggest Huntsman test with the corrections mentioned by Michael:cool:
 
Here's an idea:

3 people provide their birthdates/times/whatever data is needed for a chart. This should be provided in the thread.

idunno can then cast these dates and make the three horoscopes. DO not post them here, however. Instead, we should have a third party identified as the "escrow holder". The three horoscopes are sent to him, labelled with the names of who they belong to. He, in turn, will remove the names (relabelling them A,B,C) and send the three out to each of the three birthday providers. Each provider will review the three, and send the escrow holder the letter for the one they feel matches them most closely. The escrow holder can then post the results.

Of course, three people is not really a significant sample size, but it's a start on a baseline for the protocol, in an online form.

ETA: Actually, MichaelC's idea of keeping even the birthdates anonymous is a good one. Otherwise, there's an oppurtunity for research into the poster to make a horoscope more accurate. HAve to rule out all possibilities for error or deciet.
The holder also needs to scan the readings to remove any references to planetary positions or anything else that might give information as to which date the reading is for. The final horoscopes should contain only character traits. Even predicitions and/or postdictions could give hints as to the age of the person it is for.
 
What sort of data do you need? Birth date, time, and location?

Most predictions and postdictions can be handled by rewording comments to say things like "you have or will have lots of children." The only question is whether some comments like that would be too contrived to be applied to anyone.
 
I am clearly an inappropriate candidate for either a test subject or escrow holder, but would be willing to assist in any way. As a skeptic who knows a bit about astrology and can generate charts in a variety of methods, there may be ways I can help avoid any unintentional cheating. For example, I can proof-read the horoscopes to see if any "hints" as to birthdate are left in. In this case, I would notify the escrow-holder who could then ask idunno to review and revise if necessary.

Idunno, please understand I am not accusing you of cheating, merely wanting to make sure that no accusations can be leveled at you after the fact. If you have any suggestions, or if you would prefer that I am completely uninvolved, I can respect that.
 
please hurry up cause ill be busy learning guitar tunes this weekend before going on holiday monday:)
 

Back
Top Bottom