A review of the literature reveals only one standardized measure of children’s attitudes toward firearms and violence: the Attitudes Toward Guns and Violence Questionnaire (AGVQ), developed by Shapiro and his colleagues (1997) at the Applewood Centers in Cleveland, Ohio. The AGVQ demonstrates satisfactory internal consistency (Cronbach’s alpha = .94) and concurrent validity, with 23 items relating to violence, guns, or conflict behavior answered on a 3-point Likert-type scale (disagree, not sure, agree). A factor analysis of the AGVQ revealed four factors associated with participants owning or wanting to own a gun: (1) aggressive response to shame: the belief that shame resulting from being insulted can be undone only through aggression; (2) comfort with aggression: general beliefs, values, and feelings about aggression and violence; (3) excitement: feelings of being excited and stimulated by guns; and (4) power/safety: feeling the need to carry a gun to be powerful and safe on the streets. Shapiro and his colleagues (1998), administering the AGVQ to 1,619 children and adolescents, found that the measure was useful for predicting gun ownership. Validity coefficients were lower for girls in elementary school.

Measuring behavior in the presence of firearms is more difficult and rarely done as part of the evaluation of firearm violence programs. When behavior is measured, one of two sources of information is typically obtained:

• Community-wide or school-wide measures of the consequences of gun-carrying or gun violence—for example, school suspensions, mortality and morbidity rates, arrest rates for firearm-related offenses, suicide attempts using firearms. The behaviors that firearm violence programs are typically designed to modify or prevent are often rare events (e.g., accidental firearm deaths), so from a program evaluation point of view it is difficult to assess the effectiveness of a program designed to keep something of low frequency from actually happening. This is because data must be collected from a large number of individuals and often over a long period of time to obtain adequate numbers for analysis.

• Program participants’ description of their experiences around firearms through focus groups, class discussions, or questionnaires. Younger children may be asked if they have ever seen or touched a gun, and adolescents may be asked if they carry a gun or if they would use a gun in certain situations. While this information may be of interest, self-reports are subject to biases that may lead to underreporting, particularly when children and adolescents are asked about socially sensitive behaviors (Moskowitz, 1989).

The most direct outcome measure of behavior is an unobtrusive observation of children and adolescents when they encounter a gun. None of