Since fear is a subjective experience, its measurement cannot be exact and the numerous surveys of the past decade assessing the fear of crime are fraught with methodological flaws. Important considerations regarding both the causes and the forms of fear have been neglected in interpreting the studies. Fear should be differentiated according to cause: abstract fear, arising from a generalized, imprecise sense of endangerment and concrete fear from an actual, identifiable danger. Also crucial is the difference in the form of fear between spontaneous, irrational fear and fear upon reflection. While the former is an emotional response to vaguely perceived, imagined danger, the latter is a reasoned reaction to a representation of danger, involving rational risk assessment and planned protection measures. Having poorly defined these distinctions in the nature of human fear, fear-of-crime surveys have added to the confusion between the real crime threat and an exaggerated one, stimulated by the media, public debate, and indeed, the surveys themselves. These considerations help explain the divergent results between fear-of-crime surveys and likelihood of victimization assessments. Nine references are given.