The study involved 93 attempted homicide cases investigated by the Malmo Police Force (Sweden) between 1986 and 1995. In each case the offender had been convicted of a single attempted homicide against a single victim. A database of 108 variables (64 crime-scene variables, 27 offender-background variables, and 17 victim-background variables) was developed from a review of the police files. To explore the effects of victim resistance during the attempted homicide, a database of 49 variables was used. Attack variables were selected to assess the physical and verbal violence used by offenders during the homicide attempt. One analysis focused on the interaction between victim resistance and offender behavior, and the second analysis addressed styles of interaction. When victims resist an offender's attack, the offender is forced to make a decision about what to do, i.e., whether to change his/her behavior, abandon the attack, or maintain current behavior. The study findings indicate that this decision can be influenced by the perceived role of the victim as either a significant person, a vehicle to achieve a goal, or a depersonalized object. These roles correlated with the levels of violence displayed by the offender, with the most serious consequences being for victims who were viewed as objects; victim resistance led to an escalation of the violence. For victims perceived by the offender as significant people, offenders were more likely to change their behavior in an attempt to placate the victim. When victims were vehicles for the offender's aggression, there was no significant change in the offender's behavior in the face of resistance. Apparently the most favorable outcome occurred when offenders viewed victims as significant persons, either because there was a pre-existing relationship or because victims succeeded in personalizing themselves to the offender. 3 tables, 2 figures, appended description of crime-scene variables, and 26 references