Victim survey data and official survey data can be used in the same analysis. Victim survey and police samples are made comparable so that they contain noncommercial robbery incidents, not victimizations, occurring to Chicago residents within the city in 1974-75. Both citizens and the police make decisions as to whether a robbery incident should continue to become an official robbery statistic or whether it should be eliminated from the system. Thus, robbery data at a later step of the transformation process are systematically different from robbery data at an earlier step. A comparison of these data sets provides information about the decisions that must have produced them. The estimates of the transition possibilities of the hypothetical case suggest that (1) overall, about one-fourth of Chicago noncommercial robbery incidents become official robbery statistics; (2) neither citizen nor police decisions to eliminate a robbery from the system are random; and (3) the completion of the robbery as compared to the attempt only is the main factor in the victim's decision to notify the police. Conclusions about robbery incidents, therefore, should be based on incident data, while conclusions about founded robberies should be based on founded data. Footnotes, tables, and a bibliography of approximately 25 references are provided.