A maximum likelihood estimation procedure is developed for the simple and the weighted additive models. The data are assumed to be taken by either one of the following methods: (1) categorical ratings--the subject is asked to rate a set of stimuli with respect to an attribute of the stimuli on rating scales with a relatively few observation categories; (2) pair comparisons--the subject is asked to judge which one of two stimuli presented at a time dominates the other in some respect; or (3) directional rankings--the subject is asked to rank order stimuli in a specific direction (i.e., from the smallest to the largest or the other way around). Although the procedure mainly focuses on these three experimental methods for data collection, it is by no means restricted to them. Practical uses of the procedure are reported with an emphasis on various advantages of the procedure as a statistical method. (Author/RL)