Protective factors may be conceptualised as those variables which decrease the likelihood of an undesirable outcome, or increase the likelihood of a desired one. However, within the area of correctional psychology there is no consensus over a more exact conceptualisation, the mechanisms by which such factors work, or whether their inclusion in risk assessment is justified. Currently risk assessment measures focus almost exclusively on risk factors, which are conceptualised as those variables that make criminal behaviour more likely. The present research employed a narrative review of the current literature to critically examine the current usages of the term “protective factors”, and whether there is sufficient justification for the creation of a new concept. From 31 studies, four main definitions were identified. In opposing poles and main effects conceptualisations, protective factors are essentially the reverse of risk factors, and the concept is redundant. In an interaction definition, protective factors are separate, distinct variables, which operate according to the level of risk present. In a trichotomization model, protective factors are identified by comparing two ranges of a variable through a standardized statistical method. The latter two conceptualisations offer more promise for supporting the addition of this new concept. However, the field’s current lack of consensus regarding the mechanisms through which protective factors operate, the statistical framework for identifying them, and other related problems, will all need to be addressed before any progress will be made.