It is true that, in my analyses of results, answers are not necessarily on an equal footing, but I would not describe this as "cherry-picking".Note to everyone: Michel H has a long and consistent history of blatant and explicit cherry-picking in these tests; he will assign a greater credibility to results that agree with his hypothesis and a lower one to those that disagree, then declare the test a success. There is nothing to be gained by anyone from engaging with the process in any way whatsoever.
Dave
This reflects more a legitimate desire to make full use of the information provided by participants.
For exemple, an answer like:
would be considered as less credible than:This is a ridiculous test. I answer 33, but I warn you, this has nothing to do with extra-sensory perception.
.I believe you wrote 75 on your paper.
Credibility is in principle not related to the number given (if the target is a number) and people who read the analysis have an opportunity to agree or disagree.
This makes my tests probably much more reliable than those published in parapsychological journal, where authors could probably in many cases easily fake their data, that readers cannot verify.
It should be noted that there is also a possibility of pseudo-skeptics writing absurd and dishonest hostile comments, of which they themselves do not believe the first word, in order to try to sabotage the experiment.