I can provide more details from wherever it was I originally found them to copy and paste under the book review debate mentioned.
"She summarized her results in the Noetic Sciences Bulletin, Autumn 1991, pg
1. She also presented her findings in full as her thesis at the University
of Nottingham but never published them in a journal."
And he following quotes from Sheldrake: "The crossword puzzles she used were from the London Evening Standard, not the New York Times,and in the experiments she tested groups of subjects before and after the crossword puzzles were published in the Evening
Standard on Feb 15th 1990. Each group of subjects also did a control
crossword which had been published ten days earlier in the Evening Standard.
This was to estimate their ability to solve crosswords of this kind. The
results were scored blind. She found that the results for the Evening
Standard 'easy crossword' showed that the subjects performed better after
the crossword had been published in London, relative to scores before
publication. This difference was significant at the 5% level, using the
one-tailed t test. This effect was not detectable with the Evening Standard
'quick crossword' which the students found much harder to do and in which
they completed fewer clues.
"Obviously this result is only a preliminary one, but it does suggest that
experiments with crosswords provide a feasible way of testing for morphic
resonance. There is however one problem with this general method which only
became apparent to me when I enquired more about how crossword puzzles are
compiled. Crosswords are made up by professional compilers and it seems
that quite often they recycle parts of old crosswords and so this means it
is not always the case that a crossword is original and has not been done by
anybody before even if it has not yet been published in a particular
newspaper.
"The reason Monica England thought of doing this experiment in the first
place is that there is a folk-lore among people who do crosswords,
especially difficult ones like the Times of the Daily Telegraph, that these
crosswords are easier to solve if they're done the next day or in the
evening rather than on the morning of the day they are published, suggesting
a possible influence from others who have done them."
As an interesting aside, I thought I'd paste in something else Sheldrake
mentioned in his post:
"I wonder if, on your discussion list you have yet brought up the subject of
rises in IQ. In the appendix to my book 'Dogs That Know When Their Owners
Are Coming Home' I discuss the Flynn effect and show data for these rises in
IQ. I predicted this effect in the early 1980s but could find no data about
it. Then Flynn discovered that there had in fact been rises in IQ. He and
a colleague, William Dickens, have recently tried to explain this by what
seems to me a highly complicated and artificial argument, published in the
Psychological Review (Vol. 108, 2001). (Unfortunately I don't have the
page numbers because I only have a proof copy of the article).
"The fact that Dickens and Flynn have been driven to this tortuous attempt
to explain the phenomenon is because all previous attempts to explain it
have failed, and yet the data are very solid and have been replicated in at
least 14 different countries. The paradox is that IQ test improvement is
not paralleled by any other indication that intelligence really is
increasing. I think this is happening because people are simply getting
better at doing IQ tests because so many people have done them before. In
other words, it's a morphic resonance effect. They have already tested
possibilities about it being due to more TV, increasing test sophistication,
etc, etc, and none of these have been shown to explain it. This might be an
interesting subject to take up on your discussion list, if you have not done
so already."
