The models include: (1) the victimization model, (2) the disorder model, (3) the community concern model, and (4) the subcultural-diversity model (Covington & Taylor, 1991; Katz, Webb & Armstrong, 2003). The victimization model explains fear of crime through personal victimization, experiences with others who have been victimized, and perceived vulnerability to victimization. Perceived vulnerability here refers to the information about crime one receives, such as media reporting of crime and reports of crime from friends, neighbors, and family members. Thus, the victimization model posits that one's direct and indirect experiences with crime and victimization affect fear of crime. For example, Kruger, Hutchison, and Monroe (2007) found that being a victim of assault increased peoples' level of fear. The victimization model recognizes that crime and victimization cues are ever-present in the environment, and some groups of people are exposed to larger number of these cues and people react more definitively to the cues.
The disorder model postulates that fear is produced by the amount of perceived disorder that people encounter in their neighborhood, or areas where they travel, such as work, shopping, and recreation (Hinkle & Weisburd, 2008). This model follows Wilson and Kelling's (1982) "Broken Windows" thesis. As people encounter signs of disorder—abandoned buildings, groups of people loitering on the streets, vandalism and graffiti, and the consumption of alcohol or drugs in public—they become more concerned about their environment and their own well-being. These indicators imply some level of perceived lawlessness and a feeling of deterioration of community values, which result in a greater level of fear of crime. Ross and Jang (2000) found that as disorder increases, people stay at home and reduce their ties with the neighborhood. People begin moving out of the neighborhood, further increasing the sense of disorder (Markowitz et al., 2001). Thus, neighborhood cohesion is directly affected by the perception of disorder, and as cohesion is reduced, fear increases. McGarrell, Giacomazzi, and Thurman (2000), Kanan and Pruitt (2002), and Franklin and Franklin (2009) found disorder to be one of the strongest causes of fear of crime. A factor to counter disorder is the length of time a person resides in a neighborhood. Roman and Chalfin (2008) discovered that people living in a neighborhood for a longer period of time have reduced levels of fear of crime. It seems that experience results in a better understanding of a neighborhood and actual victimization probabilities.
One of the primary indicators of disorder in major cities is the presence of gangs. In some neighborhoods, fear of gangs outweighs the fear of crime (Katz, Webb & Armstrong, 2003). Katz and his colleagues and Lane and Meeker (2000) found that minority neighborhoods are more fearful of gangs, than are White neighborhoods. This, to some degree, is attributable to gangs being more prevalent in minority neighborhoods. Lane and Meeker (2003) attribute this enhanced fear to the diversity of the neighborhood, the amount of disorder in neighborhoods that have gangs, and concern about the community. Along these lines, Katz (2001) found that African American residents, businesspersons, and police officers attributed high levels of crime in their neighborhoods to gang activity. Thus, the presence of gangs confounds the fear of crime problem by adding an additional dimension of fear.
The community concern model posits that fear of crime is related to community dynamics. The community concern differs from the disorder model in that it considers one's feelings or attitudes about factors such as social change, dissatisfaction with neighborhood life, and unease about life circumstances. Whereas the disorder model focuses on observed disorder and crime, the community concern focuses on social interaction and community processes. For example, Bursik (2000) and Markowitz et al. (2001) found that neighborhood cohesion contributed to fear of crime. Taylor (1999) found that awareness of neighborhood deterioration contributed to fear of crime. Thus, the dynamics of the broader community contribute to one's level of fear. The community concern model is the opposite of broken windows. Some researchers have approached the community concern model through social capital theory. For example, Kruger, Hutchison, and Monroe (2007) and Ferguson and Mindel (2007) found that neighborhood social capital mediates or reduces fear of crime. Ferguson and Mindel (2007) define social capital as consisting of a, "set of components found in social associations and interactions among people that, when activated, empower individuals and facilitate cooperation toward a mutual benefit. In essence, social capital refers to the social support networks, local institutions, shared norms of trust and reciprocity, and collective activities among community members that can be used to produce a common good" (p. 323).
It appears that disorder and social capital are not on a continuum, but represent two different community attributes. Perhaps the best way to understand them is to place them in the context of weed and seed. Attacking disorder and crime problems, weeding, will help to restore some resemblance of order in a neighborhood. However, to have a long-term positive effect, CPOs must seed the neighborhood through partnerships and community building; they must attempt to build the social capital in the neighborhood. This is accomplished by working with neighborhood groups, ministers, business people, and concerned community members. The objective is to get people in a neighborhood connected.
The subcultural-diversity model suggests that some measure of fear of crime is produced from persons living near others who have cultural backgrounds that are different from their own (Lane & Meeker, 2003). That is, as neighborhoods become more heterogeneous or diverse, there is an increase in the residents' level of fear. People from different cultures often do not understand the behavior of those from other cultures. This produces uneasiness and tension, which often leads to fear of crime. For example, Bennett and Flavin (1994) found that those who were racially different from their neighbors often were more fearful of crime. Chiricos, Hogan, and Gertz (1997) found that Whites who perceived themselves to be a minority in their neighborhood were more fearful of crime. This, however, was not the case for African Americans. One might perceive that multi-racial or diverse neighborhoods would result in less racism and fear of crime. However, Mears and Stewart (2010) found that inter-racial contact resulted in more fear of crime on the part of White people and higher income African Americans, but fear decreased for lower-income African Americans. Thus, diversity and contact with people of different races may not mediate fear of crime problems. It appears that community heterogeneity affects fear of crime in different ways based on income.
These models provide ample evidence that fear of crime is the result of numerous factors interacting with the individual. They also indicate that victimization does not necessarily predict fear of crime, and indeed, direct victimization may only play a minor role in most people's perception of crime and fear of crime. The research also shows that community concern, people's impressions of their neighborhood in terms of disorder and disintegration, are key to one's level of fear of crime. These models also infer that CPOs must address fear of crime comprehensively. That is people and neighborhoods must be examined using the SARA model to identify specific attributes of problems and responses implemented that attack the problems.