Prevalence is the proportion of individuals within a population who have experienced a phenomenon. It is typically expressed as a percentage. For example, in the first study of its kind, Russell found that 38% of women reported that they had experienced contact sexual abuse prior to the age of 18. Prevalence must be distinguished from incidence, which is the rate of individuals per thousand who have been exposed to some phenomenon over a given time period. In the violence literature, incidence is often expressed as the number of individuals per 1,000 exposed to a particular type of violence. For example, approximately two female children per 1,000 experience sexual abuse each year.
Prevalence is the better figure for defining the scope of a particular problem (e.g., underage drinking), whereas incidence is better for understanding the impact of a problem during a given time period. The best estimates of prevalence often come from methodologically rigorous national prevalence studies of the given problem. These studies often construct a prevalence estimate from the individuals’ lifetime exposure to the problem.
Issues With Prevalence
Although prevalence can provide an approximation of the scope of a problem, it has limitations based upon the rigor of the studies gathering the information. An issue of great concern is the number and types of screen questions used to assess for the type of violence. Especially with sexual violence against females, behaviorally specific questions are better at soliciting disclosures. More numerous questions also tend to be better at soliciting disclosures. There are mixed findings regarding the type of survey that elicits better disclosures, although it is generally assumed that multiple responding formats are better (e.g., face-to-face and self-administered questionnaire).
Another issue of concern in prevalence studies is the accuracy of retrospective recall when adults are asked to report whether they experienced specific violence in childhood. Although findings are mixed, the issue of underreporting appears to be of much greater concern than the issue of overreporting. For example, Williams found that adults with known histories of childhood sexual abuse did not always recount the known history of abuse to researchers, even though they sometimes recalled other incidents of abuse. Although much research remains to be done in this area, researchers tentatively conclude that retrospective prevalence studies provide relatively valid estimates of a problem, although they may underestimate that problem somewhat if the violence is recalled retrospectively.
