On the use of V/V(max) for gamma-ray bursts

As applied to the BATSE observations, the V/V(max) test has shown definitively that gamma-ray bursts are not distributed homogeneously. It is shown that variations in the detection threshold smooth the distribution of V/V(max) values, and consequently this distribution should not be used to determine the nature of the source inhomogeneity. Models fitted to the V/V(max) distribution assuming a constant detection threshold are quantitatively incorrect; the true source density has a more extreme radial decrease than previously found. Models should be fitted to the observed distribution of peak burst photon fluxes corrected for detection incompleteness.