Risk assessment is commonly seen as the domain of physical and biological sciences, with social scientists focusing instead on risk management and communication. This division is unnecessary, and it may lead to errors in risk assessments. Social science input is needed for more accurate calculations of risk consequences and probabilities and for identifying potential biases created by certain risk assessment procedures, as well as in analyzing and explaining public responses to risk. Findings, moreover, suggest that the dichotomy between "real" and "perceived" risk is less "real" than is often assumed, particularly in cases involving controversial technologies.