Much valuable information can be gained by analyzing the students' wrong responses. When a student answers a free response item she/he gives the response which she/he considers to be the correct one. Therefore, diagnosing the algorithm that led the student to his/her answer provides an important source of information for assessing his/her achievement. In light of recent developments in information processing made by cognitive psychologists, as well as by experts in artificial intelligence, it seems that psychometricians ought to incorporate error analysis as an integral part of their efforts toward improved measurement. This technical report provides some empirical results of error analysis in simple, signed number problems. Types of errors and their consistency are defined and discussed. Since a wrong algorithm may sometimes lead to a correct answer the conventional scoring system is compared with one based on error analysis in terms of reliability, latent trait estimates, and the underlying dimensionality. (Author/GK)