Abstract
When a statement on the unsafety of a location is needed, accident frequency is often to lows to make reliable estimates and additional data is then needed. Conflicts observed during some short period of time are often used as if they were past accidents to estimate the accident potential. The justification for so using a TCT depends on its reliability and validity. These concepts are highly related to proper definition of conflicts and proper severity scaling. Validation studies are difficult to accomplish and very expensive.Calibration of conflict techniques is the first step in the process of validation and will be useful to design better studies and interpret future findings.