Once a suite of potential models is selected, corrections are made and the residuals (corrections) are regressed vs. mb and distance. p-values from the regression are examined to see if trends have been removed as a function of frequency. The best fitting models to not necessarily remove the trends. The figure on the left shows the results of a series of regressions for a model and associated F and p values. A large p value (e.g., > 0.05) indicates that the trends have been removed. A figure on the right shows a way that we visualize the number of p values that are greater than a given level of significance for different values of h
, Qo, and g.