Dear Eltjo,
your last public messages (July-August 2007), including the open letter we are replying to, anticipated our exchange in the Journal of Scientific Exploration. We think it is not correct at all to comment something still unpublished, because readers have no access to the text being commented and thus cannot cannot make their own informed opinion. Thus we waited. Our exchange was published in the October issue of the JSE, so this looks like a good time for our reply.
You didn't need to explain your concerns again; we got them right the first time and we already replied in the pages of the JSE. When you claimed that we handled the data as single samples (N = 1) instead of set averages (N > 1), we perfectly knew that you were not referring to the regressions, but only to the p-values in our tables; our reply was – and is – that you cannot split the two things, because we computed the p-values of those regressions. You should at least always handle your data either as single values, or as set averages, throughout the whole analysis; you cannot "switch" the multiplicity of the samples on and off at your convenience. We do agree with you that this is basic statistics theory.
Therefore, it escapes us completely why you wrote that we "don’t mention a word about" our p-values, nor about our "claims about the lack of statistical significance" (which is actually the same thing, since p-values are a measure of significance). We did write about the issue:
You wrote that your time is precious and scarce and that we should not expect any more reaction from your side. The only impact of your decision on us will be that we'll not need to reply any more, so saving our time, which is no less precious and scarce than yours, believe us. Causing reactions has never been our purpose; we wanted to comment your piece of "comments and opinions" published on Physiologia Plantarum, and we did it.
About your loss of faith in our integrity, it seems to us that you never had this faith since your first reply , so there is nothing new in your position.
Francesco Grassi
Ingegnere (Dr. Eng.), CICAP Experimentation Group
Claudio Cocheo
Centro di Ricerche Ambientali, Fondazione Salvatore Maugeri, CICAP Veneto
Paolo Russo
Programmer, CICAP Friuli Venezia Giulia
They all are members of the CICAP Study Team on Crop Circles
your last public messages (July-August 2007), including the open letter we are replying to, anticipated our exchange in the Journal of Scientific Exploration. We think it is not correct at all to comment something still unpublished, because readers have no access to the text being commented and thus cannot cannot make their own informed opinion. Thus we waited. Our exchange was published in the October issue of the JSE, so this looks like a good time for our reply.
You didn't need to explain your concerns again; we got them right the first time and we already replied in the pages of the JSE. When you claimed that we handled the data as single samples (N = 1) instead of set averages (N > 1), we perfectly knew that you were not referring to the regressions, but only to the p-values in our tables; our reply was – and is – that you cannot split the two things, because we computed the p-values of those regressions. You should at least always handle your data either as single values, or as set averages, throughout the whole analysis; you cannot "switch" the multiplicity of the samples on and off at your convenience. We do agree with you that this is basic statistics theory.
Therefore, it escapes us completely why you wrote that we "don’t mention a word about" our p-values, nor about our "claims about the lack of statistical significance" (which is actually the same thing, since p-values are a measure of significance). We did write about the issue:
It is curious that Haselhoff points out that we "incorrectly" handled the data as if they were single samples because that is exactly how he and BLT handled them in their regressions; should they have taken the number and variance of samples into account, the resulting correlation coefficients would have been much lower. Maybe the result would have been significant, but a significant mismatch. ... our goal was just to evaluate the significance of their regressions – just as they were.
You wrote that your time is precious and scarce and that we should not expect any more reaction from your side. The only impact of your decision on us will be that we'll not need to reply any more, so saving our time, which is no less precious and scarce than yours, believe us. Causing reactions has never been our purpose; we wanted to comment your piece of "comments and opinions" published on Physiologia Plantarum, and we did it.
About your loss of faith in our integrity, it seems to us that you never had this faith since your first reply , so there is nothing new in your position.
Francesco Grassi
Ingegnere (Dr. Eng.), CICAP Experimentation Group
Claudio Cocheo
Centro di Ricerche Ambientali, Fondazione Salvatore Maugeri, CICAP Veneto
Paolo Russo
Programmer, CICAP Friuli Venezia Giulia
They all are members of the CICAP Study Team on Crop Circles