The Disorder Model posits that individuals in environments characterized by physical or social disorder tend to exhibit higher levels of fear of crime. According to some researchers, fear of crime is a product of environmental cues, such as physical or social disorder in daily life or community settings that may suggest the possibility of criminal activity ( Skogan, 1990 ). Researchers in countries like the United States have conducted numerous studies on fear of crime, focusing heavily on communities. This focus is partly due to the early emergence of the modern community concept in these countries and partly because communities, as environments that overlap significantly with individuals’ daily lives, contain various factors that can greatly influence individuals. Box, Hale, and Andrews (1988) noted that individuals are more likely to exhibit a high level of fear of crime when their environment is filled with cues suggesting criminal activity or factors that easily lead to victimization judgments ( Box, Hale, & Andrews, 1988: pp. 340-356 ).

In previous studies, researchers generally viewed “disorder” as a physical or social obstacle, primarily referring to chaotic environmental factors, such as abandoned cars or transportation vehicles, piled-up garbage, damaged public property, graffiti-covered walls, and unoccupied or dilapidated houses. In addition to these physical aspects, there are also social ones, such as delinquent youths gathering on street corners, drunken individuals collapsed on the roadside, an excessive number of homeless people, sex workers, and gangs ( LaGrange, Ferraro, & Supancic, 1992: pp. 311-334 ). This seems remarkably similar to the Broken Windows Theory proposed in 1982.

The Broken Windows Theory posits that uncivil or deviant behaviors in the living environment, especially within communities, can influence the occurrence of criminal behavior through a socio-psychological mechanism ( Vilalta, Lopez, Fondevila, & Siordia, 2019 ). As a crime explanation theory rooted in the environment and community, the Broken Windows Theory has garnered significant attention from researchers since its inception ( Kramer, 2010: pp. 297-311 ), and has even greatly influenced law enforcement strategies and judicial models in some American cities ( Kelling & Coles, 1997 ). There is a plethora of research on the Broken Windows Theory, with several noteworthy aspects. Firstly, the theory emphasizes social and physical disorder. Previous studies have associated physical disorder with tangible items in the environment, such as abandoned cars (initially used in Broken Windows Theory experiments), graffiti, long-untouched garbage dumps, and overgrown wastelands ( Gau & Pratt, 2008: pp. 163-194 ). On the other hand, social disorder is more often defined as non-tangible behavioral activities, including public drunkenness, prostitution, gang violence, and drug trafficking ( Sampson & Raudenbush, 1999: pp. 603-651 ). Secondly, the causal mechanism of the Broken Windows Theory is of interest. Researchers believe that its causal mechanism arises from social psychology, but there is disagreement on its empirical validity. Some researchers argue that this causal mechanism cannot be directly observed ( Wilson & Kelling, 1982: pp. 29-38 ), while others hold the opposite view and have empirically demonstrated that this socio-psychological causal mechanism, although indirect, can be observed ( Kubrin, 2008: pp. 203-213 ). Based on this causal mechanism, Broken Windows Theory researchers contend that when physical and social disorder coexist, the risk of victimization for individuals in the environment increases significantly, and criminal opportunities in the environment also rise ( Wilcox et al., 2004: pp. 185-207 ).

This article will not discuss and verify the effect and influence of the Broken Windows Theory, but will focus on the relationship between the core of the Broken Windows Theory and the “Disorder Model” of fear of crime. Based on the understanding of the causal mechanism of the aforementioned Broken Windows Theory, this article believes that the “Disorder Model” explained by the fear of crime is an extension of criminal psychology based on the Broken Windows Theory. It focuses on the subsequent impact of physical and social disorder on individual cognition and emotional level, as well as on individual behavior, under the causal mechanism of social psychology.

The Disorder Model posits that social and physical disorder in the environment creates observable cues for the public, which in turn elevate individuals’ fear of crime ( Covington & Taylor, 1991: pp. 231-249 ). Is there a social psychological mechanism at play in the Disorder Model? The answer is affirmative, and researchers have put forward their views on this. Physical and social disorder explicitly makes individuals in the environment aware of the weakening social control, which leads to more disorder, further weakening, and ultimately collapse ( Hunter, 1978 ). The existence of this cycle actually signifies the impact of disorder on the individual’s surroundings. Most critically, it can distort the values of individuals in the environment ( Lewis & Salem, 1986 ), and decrease cohesion among individuals ( Lewis & Maxfield, 1980: pp. 160-189 ). According to the aforementioned literature, these factors clearly contribute to an increase in individuals’ fear of crime. Similar to the social psychological causality of Broken Windows Theory, the observability of the causal relationship between disorder and individual fear of crime has also been questioned. However, scholars of fear of crime research have supported the influence of disorder on individual fear of crime through empirical studies. A representative study is Brunton-Smith’s (2011) research on disorder and individual fear of crime among adolescents in England and Wales. This study lasted for four years and recorded data from nearly 5000 adolescents aged 10 - 25. Through analysis, it discovered the relationship between disorder and the level of fear of crime among British adolescents ( Brunton-Smith, 2011: pp. 885-899 ).

It’s worth noting that the current research on the Disorder Model of fear of crime is mostly concentrated in developed countries, primarily in the United States. Furthermore, there has been a shift in research direction, with researchers increasingly focusing on the social relevance of the Disorder Model in elevating individuals’ fear of crime. From this perspective, fear of crime appears to serve as a mediator between the Disorder Model and social issues ( Adams & Serpe, 2000: pp. 605-629 ). Researchers believe that this new perspective on the Disorder Model serves as a powerful tool for social policy development, leading to a series of measures aimed at combating low-level deviant behavior to control individuals’ fear of crime.

Given the current state of research progress, the number of supporting studies in developing countries remains surprisingly scarce. Presently, China lacks any supporting research on the Disorder Model, making it crucial to explore the model’s applicability. A review and discussion of the Disorder Model of fear of crime can provide substantial support for empirical research.

The victimization model suggests that individuals who have experienced victimization tend to exhibit higher levels of fear of crime compared to those without such experiences. The victimization model, as an explanation for the causes of fear of crime, is one of the earliest explanatory models in the field of fear of crime research. Garofalo (1977) focused on this direction and attempted to empirically establish individual victimization experiences as a significant predictor of fear of crime levels ( Garofalo, 1977 ). This was not long after the concept of fear of crime first emerged in academic research. From its early proposal to its subsequent widespread application in fear of crime studies, the victimization model has consistently served as a framework for many researchers to investigate and predict fear of crime.

After reviewing the aforementioned literature, it is evident that individuals who have been victimized by criminal behavior tend to exhibit higher levels of fear of crime compared to those who have not. Additionally, actual victimization significantly increases an individual’s sense of vulnerability ( Skogan & Maxfield, 1981 ). In early studies on fear of crime, when researchers utilized the victimization model to investigate fear of crime, most focused on direct victimization experiences of individuals. This is because some researchers believed that fear of crime should be a direct consequence of individual victimization, making individuals feel at a higher risk of being victimized again compared to others ( Garofalo, 1979: pp. 80-97 ). However, as the victimization explanation model gained more attention in research, some researchers discovered inconsistent results with previous studies. Several early studies, including later validations of the victimization model, found a significant correlation between victimization and fear of crime ( Warr, 1993: pp. 25-28 ). Nevertheless, in their study on fear of crime in communities, McGarrell, Giacomazzi, and Thurman (1997) tested the victimization model and found that the impact of victimization on fear of crime was not particularly significant, or even minimal ( McGarrell, Giacomazzi, & Thurman, 1997: pp. 233-250 ). This outcome prompted researchers to delve deeper into the relationship between victimization and individuals’ fear of crime, leading to the proposal of the paradox of the victimization explanation model ( Hale, 1996: pp. 79-150 ).

The paradox of the victimization model focuses on discussing what other factors within the model can influence fear of crime. As evident from the aforementioned review, the relationship between direct victimization, vicarious victimization, and the level of fear of crime has been extensively examined in research on fear of crime. These findings have prompted researchers to shift away from the traditional thinking that only direct victimization affects an individual’s fear of crime. Instead, they have begun to emphasize the impact of indirect communication on individuals’ fear of crime levels. Consequently, an increasing number of researchers have started to investigate the relationship between media reporting and fear of crime. Some researchers have pointed out that the media plays a significant role in disseminating information about crime and influencing individuals’ fear of crime levels. From its inception, mass media has had the characteristic of inducing fear of crime without actual victimization ( Heath & Gilbert, 1996: pp. 25-34 ). Therefore, researchers have been documenting and studying the relationship between media and fear of crime since the 1960s. With the development of the media, the popularity of crime-related programs, and reports on criminal behavior following the disclosure of crime data, individuals’ fear of crime is increasingly influenced. Schlesinger and Tumber (1994) suggest that media reports on certain criminal behaviors can lead to higher levels of fear of crime among individuals who closely follow such news reports. These individuals may not have had direct victimization experiences, but the more detailed the reports are, the greater the fear of crime they may induce ( Schlesinger & Tumber, 1994: pp. 221-229 ). Similar views are held by Heath’s (1984) research, which posits that media reports on major cases can increase fear of crime within a certain range. Reports on major cases tend to appear intensively over a short period, evolving from initial framework reporting to detailed reporting. The resulting fear of crime increases as the reports become more in-depth and specific, further influencing people’s daily behaviors ( Heath, 1984: pp. 263-276 ).

These studies positively respond to the hypothesis proposed by Gerbner and Gross (1976) that individual fear of crime is cultivated by the media ( Gerbner & Gross, 1976: pp. 173-199 ). In their study on the impact of crime-related news reports and realities on fear of crime, O’Keefe and Reid Nash (1987) pointed out that consistent empirical research results show a high correlation between individuals’ fear of crime and their attention to crime news and crime-related information ( O’Keefe & Reid-Nash, 1987: pp. 147-163 ). Individuals’ attention to crime-related information reports can alter their perception of victimization risk, leading to individuals exhibiting higher levels of fear of crime that do not align with reality ( Rountree & Land, 1996: pp. 1353-1376 ). Furthermore, Weaver and Wakshlag (1986) mentioned in their study that viewers without criminal experiences are more likely to have their fear of crime levels influenced when exposed to crime reports ( Weaver & Wakshlag, 1986: pp. 141-158 ). For those who believe they are more susceptible to crime, media reports on crime can also significantly increase their fear of crime ( Skogan & Maxfield, 1981 ).

This article argues that the victimization model, as the earliest explanatory model for the causes of fear of crime, should be further studied with the development of society and the advancement of technology. While focusing on the impact of personal and vicarious victimization experiences on fear of crime, researchers should also incorporate the transformation of media forms, such as the influence of streaming media on fear of crime, into their studies. Additionally, attention and consideration should be given to the changes in social communication modes and their impact on the dissemination of information among individuals, as well as their effect on fear of crime. In the field of fear of crime research in China, empirical studies should be prioritized. The applicability test of the victimization model in the Chinese research environment is the first concern and task for Chinese researchers studying fear of crime.

The Vulnerability Model is a classical explanatory model developed by early researchers of fear of crime when utilizing the Victimization Model to explain individual fear of crime. The core logical framework of this model aims to address the paradox present in the Victimization Model, which states that “individuals who are less likely to be actual victims can also exhibit higher levels of fear of crime” ( Garofalo, 1979: pp. 80-97 ). Based on this, Clemente and Kleiman (1977) shifted their focus to examining the relationship between gender and fear of crime ( Clemente & Kleiman, 1977: pp. 107-131 ). Subsequently, Lee (1983) further expanded the scope of research to include the relationship between age and fear of crime ( Lee, 1983: pp. 745-750 ). Some scholars suggest that individuals who recognize their physical, mental, and social vulnerabilities, which prevent them from keeping themselves away from criminal infringements, tend to exhibit higher levels of fear of crime ( Hale, 1996: pp. 79-150 ). Other researchers, after studying the Vulnerability Model, pointed out that women and older adults generally exhibit higher levels of fear of crime due to their weaker ability to recover from victimization ( Ferraro, 1996: pp. 667-690 ). These shifts in early research directions signify the beginning of the construction of the Vulnerability Model to explain the causes of fear of crime, with a focus on exploring the relationship between individuals’ physical, mental, or social vulnerabilities and the level of fear of crime they exhibit.

The Vulnerability Model suggests that individuals who are vulnerable from a physiological, psychological, or social perspective are more likely to exhibit higher levels of fear of crime. In simpler terms, when individuals face threats, those who recognize their physical, mental, or social disadvantages tend to show a greater fear of crime ( Scarborough, Like-Haislip, Novak, Lucas, & Alarid, 2010: pp. 819-826 ). Upon reviewing the research literature on fear of crime, it becomes apparent that the Vulnerability Model has become an indispensable explanatory model in such studies. It is closely related to demographic factors, and researchers often use this explanatory model to explore the relationship between certain demographic characteristics and fear of crime. The most common and typical characteristics examined include gender, age, race, income level, and physical illnesses. Psychological vulnerability, referring to psychological distress or mental illness, is considered a promising research direction within the Vulnerability Model.

Meanwhile, some researchers have explored more specific factors within the Vulnerability Model. Among them, this paper considers “altruistic fear” related to relational status as particularly representative. This viewpoint suggests that some individuals may exhibit a higher level of fear of crime, but this fear revolves around the concern that others may be victimized by crime. This perspective is generally used by researchers to explain the higher levels of fear of crime exhibited by married individuals or those with children. Researchers have also interacted with other vulnerability model factors based on this viewpoint. Scarborough, Like-Haislip, Novak, Lucas and Alarid (2010) proposed a hypothesis based on this, suggesting that married female immigrants with children, especially those in their later years, may experience the highest levels of fear of crime ( Scarborough, Like-Haislip, Novak, Lucas, & Alarid, 2010: pp. 819-826 ). However, this hypothesis has not yet been tested. Additionally, another highly representative and innovative study focuses on gender status, specifically sexual minorities. In their research on the relationship between sexual violence and fear of crime, Ceccato and Loukaitou-Sideris (2020) argue that gender status is a crucial factor in defining the varying degrees of fear exhibited by individuals under different levels of victimization ( Ceccato & Loukaitou-Sideris, 2020 ). Generally, sexual minorities are believed to endure more stress and face social disadvantages in many aspects compared to the general population. However, the relationship between this factor and fear of crime has not been studied or tested by researchers.

With the advancement of fear of crime research, the Vulnerability Model of causal explanation has been applied in most studies, reflecting its developing trend. Nowadays, there are few investigations specifically studying the relationship between vulnerability factors and fear of crime. Some researchers have noticed this shift, and Lewis (2013) put forward his viewpoint on this issue in his research. The individual’s gender, race, or social vulnerability alone is clearly insufficient to fully explain the level of fear of crime today. Despite the Vulnerability Model’s apparent potential in studying fear of crime, its factors need to be further crossed with other elements to explore a more comprehensive analytical potential of the fear of crime explanation model ( Lewis, 2013: pp. 869-892 ). Zhao (2013) also pointed out that the explanatory model of fear of crime needs to be more open. In addition to exploring and testing traditional factors, new elements should be continuously incorporated during its development to unleash new strengths ( Zhao, 2013: pp. 201-217 ).

The core viewpoint of the social integration model suggests that there is a significant negative relationship between social integration and fear of crime. The higher the level of social integration, the lower the fear of crime. Social integration serves as an effective means to combat fear of crime, and collective efficacy is an important factor in the relationship between social integration and fear of crime ( Hunter & Baumer, 1982: pp. 123-131 ). The “social integration” explanatory model for fear of crime emerged early in the study of fear of crime. This explanatory model is based on research examining the relationship between community factors and fear of crime. Lewis and Salem (1986) pointed out in their study on the relationship between social issues and fear of crime that individuals in communities with rich social relationships and high levels of social integration among citizens exhibit significantly lower levels of fear of crime ( Lewis & Salem, 1986 ). Similarly, Rountree and Land (1996) found in their study on the relationship between risk and fear of crime that social integration helps reduce residents’ perception of danger based on various factors in the community ( Rountree & Land, 1996: pp. 1353-1376 ). McGarrell, Giacomazzi and Thurman (1997) supported in their study on community factors and fear of crime that social integration not only combats individual fear of crime in the community environment but also plays a similar role in the entire social environment ( McGarrell, Giacomazzi, & Thurman, 1997: pp. 233-250 ). Ferguson and Mindel (2007) studied the relationship between community and social capital and fear of crime in Dallas, USA, pointing out that individual satisfaction with the community or society is an important predictor of fear of crime. The researchers mentioned that “overall, citizens who are more satisfied with their surroundings tend to exhibit lower levels of fear of crime” ( Ferguson & Mindel, 2007: pp. 322-349 ).

However, researchers’ attitudes towards the social integration model are not as unified as those towards other models. As research progresses, some researchers have found that the relationship between social integration and fear of crime is very weak. Baba and Austin (1989) reported in their investigation of social integration that measuring the impact of social integration on individual fear of crime through the number of friends and the length of time an individual has lived in the same location did not yield significant results ( Baba & Austin, 1989: pp. 763-780 ). Similarly contradictory results were also found in Riger, LeBailly and Gordon’s (1981) investigation of urban residents’ fear of crime, which showed no relationship between social integration and the level of fear of crime exhibited by women ( Riger, LeBailly, & Gordon, 1981: pp. 653-655 ). Austin, Woolever and Baba (1994) further investigated the relationship between the social integration explanatory model and fear of crime and once again obtained non-significant results, leading them to believe that social integration is not a significant predictor of individual fear of crime ( Austin, Woolever, & Baba, 1994: pp. 79-97 ).

However, these findings do not align with the majority of previous research outcomes. Confronted with such contradictory results, researchers began to ponder whether there might be some mediating variables influencing the relationship between social integration and fear of crime. Based on this hypothesis and contemplation, scholars shifted their focus to the collective efficacy brought about by social integration. Gibson, Zhao and Lovrich (2002) examined the relationship between collective efficacy and fear of crime across three cities. Their study revealed that collective efficacy serves as a significant predictor of individual fear of crime, and the impact of social integration on fear of crime is mediated by collective efficacy ( Gibson, Zhao, & Lovrich, 2002: pp. 793-807 ). This paper posits that the factors encompassed within the explanatory models of social integration and collective efficacy constitute key areas of development in fear of crime research. Amidst the backdrop of globalization and the pandemic, issues of integration and efficacy have become increasingly prominent. Challenges arise from differences in socio-cultural backgrounds and public perceptions across cities, which can significantly affect overall collective efficacy. Therefore, the continual expansion and investigation of various factors within this explanatory model deserve heightened attention from fear of crime researchers. A deeper exploration of this model will enable researchers to better integrate fear of crime issues from a holistic perspective, thereby promoting the enhancement of collective efficacy through a more comprehensive viewpoint and research approach.

This paper has reviewed and summarized the research history, differences, and various models explaining the causes of fear of crime in the context of developing the explanatory model for the fear of crime. Through this review, potential issues facing current research on fear of crime and its explanations in China have been identified. Although fear of crime is an indicator influenced by numerous variables and has not yet been quantified in any study, it holds significant importance for improving various social work practices. The review and discussion clearly highlight the dilemmas and paradoxes encountered in fear of crime research, which will guide future research directions in this field.

Research on fear of crime in China is still in its infancy, and many challenges remain to be addressed. One such challenge is the need for a unified definition of the concept, which may require further refinement based on China’s unique development context. Additionally, the applicability of research methods and explanatory models used in developed countries like the United States to the Chinese context is an urgent question that needs to be examined. A close examination of the research locations and volumes reveals that most studies have been conducted in developed countries like the United States, with developing countries accounting for a very small proportion of the research. However, developing countries face even more pressing issues and have a more urgent need for fear of crime research. As discussed earlier, fear of crime has a significant impact on individual behavior and societal progress. Therefore, fear of crime research and the control of fear of crime levels, as effective means of influencing individual and group states, deserve greater attention and focus from researchers.