Alternative methods of analysing gap acceptance data are considered with a view to removing the bias implicit in the situation in which minor road vehicle drivers exhibit consistent behavior and all accepted and rejected gaps are taken into account. Simulation techniques are used to show that some of these suggested methods are invalid, but verification is obtained for a method derived by the author in an earlier paper that simply requires the displacement of the gap acceptance curve time scale by an amount equal to the product of the variance of the observed gap acceptance distribution (sec2) and the major road volume (veh/sec). This, it is contended, is easier to apply than an alternative method of estimating the critical gap distribution parameters given by McNeil and Morgan. A theoretical relation is determined between the mean of the critical gap distribution and the Raff critical lag that shows the two to be equal only when the gap acceptance curve takes the form of a step function.