Fox and her colleagues tested 44 people who claimed to suffer various illnesses as a result of exposure to mobile telephone signals, and 114 people who did not. They used a laboratory set-up containing a replica transmitter broadcasting both conventional and '3G' telephone signals.
When told that the transmitter was switched on for 50 minutes, 'sensitive' individuals reported higher anxiety, discomfort and tension. But when asked to tell whether the transmitter was on or off, only two of the 44 'sensitive' volunteers were correct six times out of six. Five of the controls were equally successful.
This is an extremely important point that many in the general public do not realize. The effects need to be large enough such that one can eliminate RANDOM CHANCE of a positive result. The fact that 2 of the 44 got it correct 6 out of 6 times does not mean anything beyond pure chance, because the 5 control subjects who are not sensitive to it also got it right! For something to be valid in science, it must be large enough to be above "background", in this case, random chance of getting it right 6 out of 6 times. This is the only way to get a convincing, scientific result.