Coincidences in Gravitational Wave Experiments

The experimental information about gravitational waves is practically nonexistent. The only available experimental information comes from the binary pulsar PSR 1913+16 by J.H. Taylor and J.M. Weisberg(1), which shows indirect evidence for gravitational waves emitted by a continuous source and their quadrupole nature. We believe it is likely that the direct observation of gravitational waves will show phenomena at present unpredictable, as most of the times have happened in science for new discoveries.In designing the experiment we must certainly take into consideration the present theories of gravity , but we must be careful not rely on them too much. We must try to keep our mind open. Our search for gravitational waves will be based on experimental data obtained by applying the technique of coincidences among two or more detectors and by the statistical treatment of the data. We believe that this is one of the most delicate problems.In the following we shall discuss an experimental method to estimate the probabilities, already used by Joe Weber and by the gravitational waves hunters who had operating detectors. The outcome of this method we call experimental probability. We shall indicate the care needed when applying to the raw data filtering procedures for detecting small signals embedded into noise and finally we shall discuss the importance that the probability be estimated only on 'a priori' bases.