• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

Randi on Sheldrake - small sample size?

davidsmith73 said:
What if one person had done 1000 trials and the results were based on this? Would there be cause for concern?
Click to expand...
Depends on the question, of course. The 1000 trials could not be considered independent, because they are one person's data. The N for this hypothetical study would still be 1. For Sheldrake's question, I think T'ai Chi is quite right (haven't read this particular new study in question, but I have read Sheldrake's earlier stuff; besides, TC's statement here--"Trials aren't the same as a sample size"--is simply a general statement, and a corrrect one).
 
Mercutio said:
Depends on the question, of course. The 1000 trials could not be considered independent, because they are one person's data. The N for this hypothetical study would still be 1. For Sheldrake's question, I think T'ai Chi is quite right (haven't read this particular new study in question, but I have read Sheldrake's earlier stuff; besides, TC's statement here--"Trials aren't the same as a sample size"--is simply a general statement, and a corrrect one).
Click to expand...


But does the n number matter for Sheldrake's experiment?

Isn't he just testing whether the number of guesses is more than 25%? I don't see why the number of participants needs to be taken into account.
 
More Bad Scepticism from Randi?

Nobody here seems to know (or care?) whether Randi's comments on Sheldrakes experiment were fair.

Randi's comments:

"There’s a problem here. The sample size was small on both trials, just 63 people for the controlled telephone experiment and 50 for the email...

Sheldrake obtained a 43% hit rate on the email experiment which was calculated using the exact binomial test. Its my understanding that this test does not need to take into account the number of participants when applied to Sheldrakes experiment (as far as the 43% hit rate is concerned). All Sheldrake would need to take into account is the overall number of trials performed, which were over 500. Is this correct?
 
As near as I can tell, Sheldrake did the computations correctly, considering each trial independent regardless of who the participant was. I agree, the number of participants isn't really an issue in this study.
 
Beth said:
As near as I can tell, Sheldrake did the computations correctly, considering each trial independent regardless of who the participant was. I agree, the number of participants isn't really an issue in this study.
Click to expand...

Which leads me to conclude that Randi either did not read the details of Sheldrakes experiment or he did not understand it. Either way, its bad journalism but also what I've come to expect from the JREF :(
 
We used three different procedures, involving progressive simplifications. In all cases, when a trial was taking place, when the participant picked up the telephone he or she immediately indicated the person guessed by saying that person’s name. The caller then revealed his or her identity, so the participants received immediate feedback.
Click to expand...

This is all a joke right?
 
davidsmith73 said:
Which leads me to conclude that Randi either did not read the details of Sheldrakes experiment or he did not understand it. Either way, its bad journalism but also what I've come to expect from the JREF :(
Click to expand...

That's silly.

So your point is what exactly? Are you pointing out that there were worse flaws in the experiment that Randi should have mentioned instead of focusing on the small number of people involved?

If so, then you're right. Randi failed to slam Sheldrake as hard as he should have, but he succeeded in bringing it to the attention of other people who have the time to point out more flaws.
 
I personally think its all a load of hogwash but am prepared to be amazed.

Why don't we just run our own test on this thread? Should be enough of us to record all of our calls for the next couple of days and make some sort of yes/no/maybe telepathy statement. Honesty would be expected of course???
 
I also think Randi could have done a better job on that article. Here's a much better article on the same subject:

http://www.timesonline.co.uk/article/0,,2-2344196,00.html

The article doen't point out the small sample size, but that the study was presented without challenge from other scientists. It also mentions the inadequate controls used in Sheldrake's experiments. Both of these points would have made Randi's article much stronger and more interesting to skeptics.
 
Pennywise said:
That's silly.

Are you pointing out that there were worse flaws in the experiment that Randi should have mentioned instead of focusing on the small number of people involved?
Click to expand...


Not worse flaws but different flaws. Remeber that the small sample size issue is not a flaw at all. I also think there are inadequate controls in his experiment. But my point was to highlight that Randi made quite a serious error in SWIFT which was based on either not understanding pretty basic stats, or not reading the paper properly. This is supposed to be a sceptical organisation so I was just showing that we need to be sceptical of sceptics too, and not take what anyone says for granted. Check the facts!
 
You must log in or register to reply here.

Back
Top Bottom