Methods for examining the viability of assumptions underlying generalized linear models are considered. By appealing to the likelihood, a natural generalization of the raw residual plot for normal theory models is derived and is applied to investigating potential misspecification of the linear predictor. A smooth version of the plot is also presented, as are asymptotic standard errors appropriate for assessing the significance of apparent deviations from the presumed model. In attempting to devise procedures appropriate for exploratory analysis, it is important to balance the necessities of flexibility and computational economy against the desirability of a clear inferential framework. Methods based on residuals, particularly in combination with simple smoothing techniques are appealing and can be placed within a more formal parametric framework of the generalized linear model. This formalization provides indications towards procedures which are more efficient and informative. The paper states that completely satisfactory development remains a more or less distant, and perhaps, unrealistic goal. (Author/PN)