designing a convincing prayer experiment

How will prayers be directed at the prayer group plants? Will the people praying be in close physical proximaty to the plants or remote like in some other prayer experiments? Do they have to know the "labels" of the plants or not?I'm interested in this because I think it is possible to have no one know which plants are in which group. The results can be spit out by a program without anyone breaking the grouping until after it is all finished and analysed.If people do not have to be in proximatity to the plants (like in the fertilization experiments) then the plants can be left mixed up. Numbered individually, grouped randomly by the computer into prayer and nonprayer groups. After the plants are evaluated the 'scores' are entered by number into the computer and the results are calculated. By protecting the computer's assignment the results don't have anyone to tamper with them. Tricky but possible.If they do have to be physically prayed over. Then the plants can be grouped, the numbers of a group entered into the computer, then numbers of the plants hidden and the prayer non-prayer assignment made by the computer. One group is labeled A and one B. The one chosen by the computer is prayed over.The plants are then mixed, evaluated the scores for each assigned by someone who doesn't know the number of the plant or which group it was in. Then, by someone else who doesn't know what the scores mean or the number mean, the scores and numbers are entered into the computer. The results are calculated.

---

Common sense isn't

How cynical! How right! The real danger is that with small "n" there will be the occasional such test that does produce a significant result if you run enough such experiments. The "usual" reaction (not just on the part of the faithful, I might add) is to grab that one as "proof" of something.This happens in the medical literature (personal communication with my brother) apparently.This is an example of the recognized publication bias problem where negative results do not get published as readily as positive ones.At some point there has to be a good statistician involved to determine what statistical protocols will be involved as well.

---

Common sense isn't