Determining Chance

[Marc L]

I was reading the Swift commentary about the dowsing trials. In it, a preliminary test is reported on, included what was required for passing, and a comment on the dowsers performing below chance.

From the report:

Participants had to locate the water 4 or more times out of the 6 trials in order to pass the test. In line with previous controlled tests of dowsing ability, the results for each individual dowser were below chance expectations.

I have two questions:

1) How is the success rate determined (ie, 4 out of 6 as opposed to say, 5 out of 6 or 6 out of 6)?

2) How were the "chance expectations" determined.

Keep in mind that I'm not challenging anything, or trying to say the report was skewed. I'm just trying to figure out how the trial works.

Marc

 

[insomneac]

Short answer

In a statistical hypothesis test with a probability of success of 1/6 and six trials, you need at least four successful trials to say with either 95% or 99% certainty that the result was not due to chance.