Antecedent Variable of Fear of Crime with University Sample in Shenzhen City ()
1. Multi-Level Definition of Fear of Crime
Fear of crime (FoC) is commonly understood as an emotionally charged process. Gabriel & Greve (2003) conceptualized it as an individual’s perceived crisis due to the real threat of criminal acts (Gabriel & Greve, 2003: pp. 600-614) . Sundeen & Mathieu (1976) viewed it as a fear of victimization, encompassing an individual’s anxieties about becoming a victim of crime (Sundeen & Mathieu, 1976: pp. 211-219) . Other scholars, such as Merry (1981) , describe FoC as an inward psychological process, presenting a comprehensive definition that encompasses cognitive, emotional, and behavioral dimensions (Merry, 1981: p. 10) . The cognitive aspect involves an individual’s understanding of the risk or susceptibility to crimes, including murder or serious harm (Warr, 1990: pp. 891-907) . The emotional dimension encompasses the fear generated by potential criminal threats (De Donder, Verté, & Messelis, 2005: p. 363) . Behaviorally, it refers to actions such as purchasing firearms, avoiding nocturnal walks, or installing security measures, aimed at diminishing the FoC or enhancing personal security (Covington & Taylor, 1991: pp. 231-249) .
The concept of FoC is nuanced, encompassing multiple levels and dimensions. Schafer, Huebner, & Bynum (2006) have identified several key aspects: First, there’s the fear of criminal behavior itself, manifesting as anxiety over the likelihood of becoming a victim. This can be specific to a particular crime or a general apprehension about criminal acts (Schafer, Huebner, & Bynum, 2006: pp. 285-301) . Second, the reaction to this fear often drives individuals, especially in areas where owning firearms is legal, to adopt precautionary measures such as avoiding solitary nighttime outings, steering clear of poorly lit areas, enhancing home security, and being vigilant of one’s psychological well-being (Drakulich, 2012: pp. 322-355) . Third, fear stemming from past victimization is crucial and noteworthy (Skogan & Maxfield, 1981: pp. 15-23) , including not only personal experiences but also those of close associates like friends and family, which significantly impacts one’s fear levels (Greve, 1998: pp. 277-309) . Fourth, the self-evaluation of public safety and environmental factors plays a significant role. This includes assessing the adequacy of surveillance coverage, lighting, patrolling efficiency, as well as the complexity of community demographics and crime incidence in surrounding areas (Williams, et al., 2000: pp. 1-28) . Previous experiences of fear in certain situations also predispose individuals to similar reactions in the future (Rader, 2004: pp. 689-704) . Additionally, according to Stovic, Fischhoff, & Lichtentien (1987) , FoC encompasses various emotional states and attitudes, ranging from distrust and anxiety about perceived risks to concerns over community dynamics and the public environment (Stovic, Fischhoff, & Lichtentien, 1987: pp. 158-175) . Hence, FoC is a multifaceted concept involving an emotional response to the potential of victimization and the behavioral efforts undertaken to mitigate such fear.
2. Antecedent Variable: Media and FoC
Heath & Gilbert (1996) highlighted the media’s significant influence on societal FoC, attributing this impact to a variety of factors. Key among these are the nature of crime exposure in the media, including coverage of severe violent crimes and property crimes, as well as the media’s demographic targeting, such as focus on ethnic minorities and white-collar workers (Heath & Gilbert, 1996: pp. 25-34) . Schlesinger & Tumber (1994) further analyzed the presentation and interpretation of criminal news through interviews with journalists and editors and case studies. They posited that media reports on certain criminal acts could heighten specific individuals’ FoC, particularly those who follow the news closely, regardless of their personal experiences. They argued that the level of detail in these reports directly correlates with the extent of fear they may provoke (Schlesinger & Tumber, 1994: pp. 221-229) .
Heath (1984) posits that media coverage of serious incidents can escalate FoC within a specific scope. Such reports, initially concentrated over a brief period and evolving from basic outlines to detailed accounts, intensify and become more precise over time. This escalation in the FoC, as the reporting deepens, subsequently influences individuals’ daily behaviors (Heath, 1984: pp. 263-276) .
3. Antecedent Variable: Daily Activities and FoC
The nexus between individuals’daily behavior, their living environments, and the FoC constitutes a pivotal focus within the realm of FoC studies. Grounded in routine activity theory, the premise is that crime arises from opportunities for offending that are embedded within the societal matrix of everyday life (Cohen, 2008: pp. 325-353) . Warr (1990) posited that the locus of research on FoC should concentrate on identifying which demographic groups harbor the greatest fear of victimization. Such research should empirically investigate these groups to delineate the environmental features they encounter, unveil potential criminal opportunities within these settings, and elucidate the correlation between such opportunities and the FoC. Moreover, advanced a nuanced approach to analyzing the environment, suggesting that the multifaceted nature of an individual’s surroundings be considered in assessing their FoC. He proposed that discerning specific environmental aspects inducing fear could enhance understanding of the issue (Warr, 1990: pp. 891-907) . In categorizing environments based on their relationship with the FoC, he identified two primary types: “novel” and “unfamiliar”. A novel environment, he argued, could provoke societal FoC (Jackson & Stafford, 2009: pp. 832-847) . Furthermore, the elements of environmental analysis were delineated into objective factors—such as the presence of strangers or the level of darkness—and subjective factors, including the degree of unfamiliarity felt by individuals (Melde, Berg, & Esbensen, 2016: pp. 481-509) .
As theories of daily activities and environmental analysis have evolved, the study of FoC has increasingly incorporated the characteristics of community environments and the individual behavior patterns within these settings. Robinson (1998) suggests that the presence or proximity to a significant number of potential criminals in an area can markedly influence the FoC experienced by community members (Robinson, 1998: pp. 19-32) . For instance, within university settings, students are frequently the primary targets of potential criminal acts by their peers. Consequently, this places all students at heightened risk of victimization, as they are part of the regular activities of these potential offenders. The presence of such individuals within the student body significantly amplifies the overall sense of vulnerability to crime among students.
4. Antecedent Variable: Risk Perception, Victimization Perception and FoC
Warr & Stafford (1998) argued that the FoC inherently implies an individual’s awareness of being at risk of victimization. This is exemplified by situations such as the presence of a serial killer, which elevates the perceived risk of becoming a victim, as individuals might believe they are more susceptible to victimization compared to random crimes. Consequently, distinguishing between the assessment of risk perception and the FoC itself is critical (Warr & Stafford, 1998: pp. 1033) . Ferraro (1996) ’s study on female victims and crimes specifically targeting women expanded upon this discussion by demonstrating a strong correlation between FoC and personal perceptions of risk and victimization (Ferraro, 1996: pp. 667-690) .
In addition to examining risk perception factors, certain studies have concentrated on investigating the perceived factors influencing victimization severity. Nagao & Davis (1980) undertook an empirical investigation into the perceived factors determining victimization severity using simulated judgments. Their findings revealed that prior experience and personal knowledge predominantly influenced individuals’ perceptions of victimization severity. Specifically, individuals with previous experiences of rape crimes perceived the severity of victimization significantly higher than those who had no such experiences (Nagao & Davis, 1980: pp. 190-199) .
5. Antecedent Variable: Demographic Factors and FoC
Lagrange & Ferraro (1989) utilized a national crime survey to assess the level of FoCand eleven alternative indicators of specific FoCs, examining the relationship between age, gender, and FoC. Their research indicated that both women and the elderly experience a heightened FoC, with women’s perceived risk and fear consistently surpassing that of men, regardless of the fear measurement approach. The FoC among the elderly presents a paradox in terms of age (Lagrange & Ferraro, 1989: pp. 697-720) . Conversely, Sutton & Farrall (2005) challenged the longstanding theory that women inherently fear crime more than men by suggesting that social pressures could skew men’s reported fear levels, implying that, in the absence of gendered societal pressure, men’s fear might be comparable to women’s (Sutton & Farrall, 2005: pp. 212-224) .
Age has consistently been recognized as a crucial factor in the FoC (Fisher & May, 2009: pp. 300-321) . Clememte & Kleiman (1976) conducted a comparative study on FoC between elderly and non-elderly individuals, considering gender, race, socioeconomic status, and community size as primary variables. Their findings indicated that urban elderly respondents exhibited a heightened FoC. However, it is important to note that the elderly population is not uniform; certain segments display extreme fear, while others show relatively little concern (Clememte & Kleiman, 1976: pp. 207-210) . This early observation led to the theory of the paradox of FoC, suggesting variability in fear levels across age groups. Despite previous studies acknowledging this paradox, research into the relationship between age and FoC continues to be a frequent pursuit (Farrall, Gray, & Jones, 2021: p. 35) . Tulloth (2000) emphasized the significance of age as a determinant of FoC, employing both quantitative modeling and qualitative analysis to explore how different age groups perceive this fear. Through his research, Tulloth aims to mitigate FoCamong various age demographics, highlighting the ongoing importance of understanding this dynamic (Tulloth, 2000: pp. 451-467) .
6. Research Paradigm of Social Investigation on FoC in Universities
The seminal study on the FoC within a university setting, titled “University FoC: A Study of a Southern University” by McConnell (1997) , categorizes the determinants of FoCinto two main groups (McConnell, 1997: pp. 22-46) . These independent variables encompass student demographics such as race, age, gender, chosen curriculum, involvement in campus activities, academic standing, and the size of their hometown, among others. The multifaceted notion of FoC is quantified through a self-administered scale comprising 17 items. McConnell’s findings illuminate the intricacies surrounding the concept of FoC.
Firstly, the characteristics of a location notably influence students’ FoC; a majority of students report feeling fearful of criminal activity when frequenting particular areas on campus (Caruso, 2011: pp. 455-463) . Secondly, student demographics have a considerable impact on crime apprehension; gender emerges as a significant determinant, with women exhibiting a markedly higher FoC. Beyond gender, individual experiences and the prevalence of specific crimes in one’s hometown also shape students’ perceptions of safety and their FoC (Agostino, Sironi, & Sobbrio, 2013: pp. 724-727) .
Carmen, Polk, Segal, & Bing (2000) conducted a social survey examining students’ FoC, incorporating event variables before and after a crime’s occurrence to ascertain differences in the factors influencing this fear (Carmen, Polk, Segal, & Bing, 2000: pp. 21-36) . The study included 186 students prior to a violent crime event and 374 students afterwards, revealing that students’ field of study significantly impacts their FoC. Specifically, students who majored in criminal justice exhibited a lower FoC compared to their peers in non-criminal justice fields. Additionally, the occurrence of campus crime markedly influenced students’ fear levels: prior to a campus violent crime, 31.7% of students expressed fear of violent crime, which increased to 41.2% after such an incident (Price, Evans, & Bates, 2003: pp. 91-109) .
7. Research Method
This study focuses on a university in Shenzhen City, selecting 179 participants from the faculties of finance, law, and adult education for an anonymous questionnaire survey and before the questionnaire survey, read out the ethical requirements and informed consent to the respondents. The sample includes 59 individuals from the finance faculty, 60 from the law school, and 60 from the adult education faculty. The questionnaire comprises two sections, consisting of a total of 34 closed-ended and anonymous questions. The primary questions utilize a scoring model based on degrees, where participants rate their fear level on a scale from 1 to 5.
The cognitive-emotional measurement method assesses an individual’s FoC by analyzing their understanding of and emotional reactions to crime. This approach is widely adopted in both domestic and international research, characterized by unique problem settings (Otis, 2007: pp. 198-217) . Key survey questions include “whether individuals feel comfortable walking alone at night” and “their sense of security when alone at home,” which are pivotal in gauging FoC (Ferraro & Grange, 1987: pp. 70-97) . Notably, this method was employed by the Ministry of Public Security in China for public security surveys conducted in 1991, 1994, and 2004. The research on cognitive-emotional measurement methods has spurred advancements and refinements in methodological approaches. Beyond the primary issues, this method also evaluates FoC through perceived fear levels towards various crime types, such as fraud, robbery, murder, sexual assault, stalking, and theft (Ferraro & Grange, 1992: pp. 233-244) .
In this study, the data will undergo cross-checking and be analyzed using an independent sample t-test in accordance with the research hypothesis to determine whether there is a significant relationship.
8. Research Questions and Assumptions
This research have four hypothesis as follows:
Research Hypothesis 1 mentioned that this study investigates whether gender differences significantly impact students’ FoC across various types of criminal acts.
Research Hypothesis 2 explores whether experiences related to victimization significantly affect students’ FoC across various criminal activities. Previous studies indicate that experiencing victimization does not consistently result in significant differences in an individual’s FoC.
Research Hypothesis 3 posits that the impact of crime data and media news attention on students’ FoC might vary significantly across different types of criminal acts. Given that news media attention and its potential over-reporting have consistently been significant variables in numerous studies, the shift in students’ media consumption patterns toward more dispersed and less engaged interaction raises questions about the continued influence of news media on perceptions of crime.
Research Hypothesis 4 posits a significant relationship between students’ behavior, specifically their tendency to avoid poorly lit areas, and their FoC.
9. Data Analysis
9.1. Gender and FoC
A cross-comparison and independent sample T-test were conducted to assess if there were significant gender differences in students’ fear of various types of criminal acts. The results, detailed in Table 1, reveal that in the T-test examining the relationship between gender and the fear associated with different crimes, the mean group statistics significantly differ for all categories except sexual crimes. Specifically, there is no notable gender difference in the fear of sexual crimes, while clear disparities exist in other areas. For violent crimes, the mean fear levels were 3.3088 for men and 4.1622 for women; for property crimes, 2.6618 for men versus 3.3153 for women; and for threat-related crimes, 2.7941 for men compared to 3.6847 for women. Despite the modest differences in these mean values, it is evident that women consistently reported higher levels of fear across the assessed crime categories.
Upon conducting additional independent sample t-tests for data on gender and various crime categories, it becomes evident that the outcomes of the variance tests differ, as illustrated in Table 2. This table unequivocally demonstrates the influence of gender on individuals’ FoC across different crime categories, summarized as follows:
• The variance homogeneity test for the fear of sexual crimes indicates a significance level (Sig) of 0.208, surpassing the 0.05 threshold, suggesting equal variance. Conversely, the t-test yields a Sig of 0.000, falling below 0.05, thereby revealing a significant gender-based difference in the fear of sexual crimes. Specifically, women exhibit a notably higher fear level compared to men regarding these crimes.
• The homogeneity of variance test for the FoC associated with violent offenses indicates that the significance value (Sig) is 0.080, which exceeds the threshold of 0.05, confirming equal variance. Conversely, in the t-test, a Sig value of 0.000, falling below the 0.05 benchmark, demonstrates a statistically significant disparity in the fear of violent crimes between genders. Specifically, women exhibit a markedly higher level of fear of violent crimes compared to men.
![]()
Table 1. FoC between different genders and different crime categories.
![]()
Table 2. Gender and fear of different crime categories.
• The homogeneity of variance test for fear associated with property crimes indicates a significance level (Sig) of 0.330, exceeding the 0.05 threshold, and thus suggesting equal variance. Conversely, the t-test result, with Sig = 0.001 (Sig < 0.05), reveals a significant gender-based disparity in the fear of property crimes, with women exhibiting a substantially higher level of fear than men.
• The homogeneity of variance test for the FoC resulting from threats indicates a significance (Sig) value of 0.030, which is less than the 0.05 threshold, demonstrating that the variance is not equal. Furthermore, the corresponding T-test yields a Sig value of 0.000, well below the 0.05 significance level, revealing a significant gender difference in the level of FoC due to threats. Women experience a significantly higher fear of such crime than men.
9.2. Victimization Experience and FoC
Research indicates that experiencing victimization does not invariably affect an individual’s FoC, presenting a somewhat paradoxical relationship with this antecedent variable (Ziegler & Mitchell, 2003: pp. 173-187) . For instance, victims of violent crimes may not necessarily exhibit a greater fear of such incidents compared to individuals who have never experienced violent crimes. Accordingly, this study focuses on examining this paradox and assessing the relevance of this antecedent variable in the Chinese context. It involves a comparative analysis and an independent sample T-test of survey data, with specific attention to the impact of sexual crime victimization experiences, the results of which are detailed in Table 3.
The comparative analysis of the sample comprising victims of sexual crimes or individuals with related experiences, cross-comparison their levels of fear toward sexual crimes, highlights an insignificant disparity in average values between victims or those with related experiences (3.4286) and those without such experiences (3.3358).
Furthermore, an independent sample T-test was conducted to explore the relationship between experiences of sexual crimes or related incidents and the fear of such crimes. The results, depicted in Table 3, indicate that the homogeneity of variances for the fear of sexual crimes yielded a significance value (Sig) of 0.058, with Sig > 0.05, suggesting equal variance. In the T-test analysis, the significance value was 0.728, exceeding the 0.05 threshold, indicating no significant difference between experiences of sexual victimization or related experiences and the fear of sexual crimes. This implies that having experiences related to sexual crimes does not significantly influence the level of fear of these crimes among the sample population (see Table 4).
This finding underscores the paradox observed in the examination of sexual crime victims or related experiences and the concern regarding sample representation in sexual crime research. The study proposes several reasons for this
![]()
Table 3. FoC between victims of sexual crimes or related experiences and FoC of sexual crimes.
![]()
Table 4. Sexual crime victims or related experiences and fear of sexual crime.
anomaly. Firstly, there is a lack of comprehensive understanding of sexual crimes among students. Secondly, the term “experience” used in the survey typically denotes direct victimization, where a noticeable reluctance to self-report such incidents exists. Lastly, the sample size is limited. If future studies address these issues, the paradox could be locally re-examined and potentially clarified.
A cross-comparison and an independent sample T-test were conducted to assess the relationship between experiences of violent crimes or victimhood and the fear of such crimes. The findings, presented in Table 5, reveal no significant difference in the level of fear between individuals who have experienced violent crimes (mean = 3.7778) and those who have not (mean = 3.8707). This indicates that experiencing violent crimes does not necessarily result in a heightened fear of violent crimes.
An independent sample T-test was conducted to analyze the fear of violent crime, with results presented in Table 6. The variance homogeneity test yielded a Sig value of 0.657 (Sig > 0.05), indicating equal variances, and the T-test resulted in a Sig of 0.608 (Sig > 0.05). This suggests that there is no significant difference in the fear of violent crime between individuals with personal or vicarious experiences of violent crime and those without such experiences.
Cross-comparison and independent sample T-tests were conducted to assess the impact of property crime victimization or related experiences on the sample’s fear of property crime. Table 7 indicates that the mean scores for fear of property crime do not significantly differ between those with property crime experiences (mean = 2.9500) and those without such experiences (mean = 3.1007).
![]()
Table 5. FoC between victims or related experiences of violent crimes and FoC of violent crimes.
![]()
Table 6. Victims or related experiences of violent crimes and fear of violent crimes.
![]()
Table 7. FoC between victims or related experiences of property crimes and FoC of property crimes.
Subsequent analyses included a test for homogeneity of variances and an independent samples t-test on the fear of property crime. The results, presented in Table 8, indicate that the homogeneity of variance was not violated (Sig = 0.627, p > 0.05), suggesting equal variances between groups. The independent samples t-test yielded a Sig value of 0.513 (p > 0.05), indicating no significant difference in the level of fear of property crime between individuals with and without experience of being victimized by property crime.
A cross-comparison and independent sample T-test were conducted to compare the levels of fear of threatening behavior within the sample. According to the results presented in Table 9, the analysis reveals no significant difference in fear of threatening behavior between individuals who have experienced threatened victimization or related incidents, with an average of 3.4444, and those who have not, with an average of 3.3217.
Subsequent analyses, including the Independent Samples T Test and the test for homogeneity of variance, were conducted to assess fear of criminal threats. Table 10 displays these results. The homogeneity of variance test yielded a significance value (Sig) of 0.259, with Sig > 0.05, indicating equal variances. In the T Test, the significance value was 0.568, again exceeding 0.05, suggesting that there is no statistically significant difference in the level of fear towards criminal threats between individuals who have and have not experienced such threats.
9.3. Crime Data and News Concern and FoC
Drawing upon earlier research and sociological theories, this study examines if there are notable differences in individuals’ concern for crime data and media coverage and their fear of various types of crimes. Prior research has consistently identified the extent of media attention, particularly its potential for over-reporting, as a key pre-determinant of public fear. However, with the current generation’s increasingly fragmented and subdued engagement with media, it is imperative to reassess the media’s influence as a significant factor. Therefore, this investigation categorizes concerns into two distinct types: those pertaining to crime data and those related to news media coverage.
The comparative analysis and Independent Sample T-test examining the relationship between participants’ attention to crime statistics and their fear of various types of criminal behavior reveal the findings in Table 11. Notably, there is no significant difference in the average group statistics for different types of criminal fears. The fear of sexual crimes yielded average scores of 3.2766 for
![]()
Table 8. Victims or related experiences of property crime and FoCof property crime.
![]()
Table 9. FoC between threat victimization or related experience and criminal FoC of threat.
![]()
Table 10. Threatened victimization or related experience and fear of threatened crime.
![]()
Table 11. Concern about crime data and FoC.
those concerned and 3.4286 for those unconcerned. In the category of violent crimes, the averages are 3.8085 for concerned participants and 3.8571 for those unconcerned. For property FoCs, concerned individuals had an average of 2.9574, whereas the unconcerned group’s average was 3.1667. Finally, fear of threatening crimes showed averages of 3.3936 for concerned individuals and 3.2738 for those not concerned.
Subsequent analyses, including homogeneity of variance and independent sample testing of survey results, reveal variations across different crime categories, as illustrated in Table 12. Despite these variations, the core findings related to concern over crime data and FoC remain largely consistent, indicating an absence of significant correlation.
The homogeneity of variance test for FoC related to sexual offenses indicates that with Sig = 0.250, which is greater than 0.05, the variances are equal. Additionally, in the t-test, Sig = 0.503 (also greater than 0.05), suggests there is no significant difference in the FoC related to sexual offenses based on whether individuals pay attention to crime data. Consequently, focusing on crime data does not significantly influence an individual’s fear of sexual crimes.
![]()
Table 12. Attention to crime data and FoC.
The homogeneity of variance test for FoC associated with violent offenses indicates a significance value (Sig) of 0.269, with Sig > 0.05, suggesting equal variance. Similarly, the t-test yields a Sig of 0.779, Sig > 0.05, demonstrating no significant difference in the degree of FoCrelated to violent crimes based on whether individuals pay attention to crime data. Hence, paying attention to crime data does not significantly influence an individual’s fear of violent crime.
The homogeneity of variance test conducted on FoC in relation to property crimes indicates that with Sig = 0.660 (Sig > 0.05), the variances are homogenous. Furthermore, the independent samples t-test yields Sig = 0.275 (Sig > 0.05), suggesting no statistically significant difference in the levels of fear of property crime when comparing individuals who pay attention to crime data with those who do not. This outcome implies that attention to crime data does not markedly influence an individual’s fear associated with property crimes.
The homogeneity of variance test for FoC in response to threatening behavior indicated that the significance level (Sig) is 0.241, which is greater than 0.05, suggesting equal variance. Additionally, in the t-test, the significance level was found to be 0.487, exceeding the 0.05 threshold, indicating no significant difference between the consideration of criminal data and the level of FoC due to threatening behavior. This demonstrates that attention to criminal data does not significantly influence an individual’s FoC associated with threatening behavior.
In China, there appears to be an absence of significant correlation between reported crime data and individual apprehensions regarding crime. This phenomenon could be attributed to the semi-public status of crime data disclosure in the country. Notably, a considerable segment of the population, particularly students, infrequently access such data. Additionally, individuals are generally not inclined to actively seek out crime-related information unless compelled by specific circumstances.
Secondly, the T-test examining the association between attention to newspapers and news reports and fear of various criminal acts, as presented in Table 13, reveals negligible differences in the mean statistics across all categories of criminal fear. Specifically, for fear of sexual crimes, the mean scores are 3.2807 for concerned individuals and 3.4923 for those unconcerned. For fear of violent crimes, the mean scores for the concerned group is 3.7807, compared to 3.9385 for the unconcerned. Regarding property FoCs, mean scores of 3.0263 for the concerned are marginally lower than 3.1385 for the unconcerned. Lastly, for fear of threatening crimes, the mean scores are 3.2368 for the concerned group versus 3.5385 for those without concern.
Further analysis using independent sample tests on the survey results reveals a significant difference in the variance, as illustrated in Table 14.
The test for homogeneity of variance in FoC related to sexual offenses yielded a significance value (Sig) of 0.206, indicating that the variance is equal since Sig > 0.05. Furthermore, in the T-test, a Sig value of 0.367, also greater than 0.05, suggests no significant difference in the level of FoC related to sexual offenses based on whether individuals pay attention to newspapers and news reports.
![]()
Table 13. Concern about newspapers and news reports and FoC.
![]()
Table 14. Attention to newspapers and news reports and FoC.
Therefore, the attention given to newspapers and news reports does not significantly influence the level of FoC concerning sexual offenses.
The homogeneity of variance test for FoC related to violent offenses reveals a significance level (Sig) of 0.025, indicating non-homogeneous variance as Sig < 0.05. This contrasts with the results from the unequal T-test, where Sig = 0.358, demonstrating Sig > 0.05, and thus, no significant difference exists between the attention paid to newspapers and news reports and the level of fear of violent crimes. Consequently, the extent to which individuals pay attention to newspapers and news reports does not have a significant impact on their fear of violent crimes.
The test for homogeneity of variances concerning the fear of property crime yielded a significance value (Sig) of 0.660, which exceeds the 0.05 threshold, indicating equal variances. Additionally, a t-test produced a significance level of 0.574, also surpassing the 0.05 mark, suggesting no significant difference in the level of fear of property crime between individuals who pay attention to newspapers and news reports and those who do not. This implies that attention to such media sources does not substantially influence an individual’s fear of property crime.
The variance homogeneity test for FoC associated with threatening behavior yielded a significance (Sig) value of 0.825 (Sig > 0.05), indicating equal variances. Furthermore, the t-test produced a Sig value of 0.091 (Sig > 0.05), suggesting no significant difference between the level of FoC related to threatening behavior and the degree to which individuals pay attention to newspapers and news reports. This implies that consuming newspaper and news report content does not significantly influence the level of fear individuals experience regarding criminal threatening behavior.
9.4. Insufficient Lighting Avoidance and FoC
The relationship between student behavior patterns, particularly their avoidance of poorly lit areas, and their FoC has been a subject of discussion in the fields of environmental crime analysis and the study of FoC. This preemptive factor has a direct impact on individual behavior and feelings in specific situations.
A comparative analysis investigating whether the sample actively avoids poorly lit areas in response to different types of criminal behavior reveals no marked differences in the mean statistical groupings of criminal fears, as evidenced in Table 15. The mean fear scores for sexual crimes are 3.4516 for the active avoidance group and 3.1455 for the non-avoidance group. For violent crimes, the averages are 3.8145 for the active group and 3.8909 for the inactive group. In the context of property FoCs, scores are 3.0565 for the actively avoiding individuals and 3.0909 for those who do not actively avoid. Lastly, the mean scores for fears of threatening crimes are 3.4032 for the active group and 3.2182 for the inactive group.
Subsequent independent sample testing of the survey data reveals distinct variance test outcomes, as presented in Table 16.
The homogeneity of variance test for the FoC related to sexual offenses indicates a significance level of Sig = 0.022, which is less than the 0.05 threshold, signifying unequal variances. This contrasts with the results from the equal variance T-test, where Sig = 0.240, exceeding the 0.05 cutoff, thus demonstrating
![]()
Table 15. Actively avoiding areas with insufficient lighting and FoC.
![]()
Table 16. Active avoiding under lighted areas and FoC.
no significant difference between the propensity to actively avoid poorly lit areas and the level of fear associated with sexual crimes. Consequently, the decision to avoid areas with insufficient lighting does not significantly influence an individual’s fear of sexual crimes.
The homogeneity of variance test for FoC associated with violent incidents reveals that the significance value (Sig) is 0.953, which, being greater than 0.05, indicates equal variance. Additionally, the t-test yields a significance value (Sig) of 0.684, again surpassing the 0.05 threshold, signifying that there is no significant difference in the fear of violent crimes between individuals who actively avoid areas with inadequate lighting and those who do not. Consequently, the decision to avoid dimly lit areas does not have a substantial impact on an individual’s fear related to violent crimes.
The homogeneity of variance test for the FoC associated with property crimes yields a significance (Sig) value of 0.164, which is greater than 0.05, indicating equal variances. Furthermore, a t-test produces a Sig value of 0.868, also exceeding the 0.05 threshold, demonstrating no significant difference between the inclination to actively avoid areas with insufficient lighting and the level of fear regarding property crimes. Consequently, whether an individual chooses to avoid poorly lit areas does not significantly influence their fear of property crimes.
The test for homogeneity of variance regarding FoC due to threatening behavior indicates a significance level (Sig) of 0.201, where Sig > 0.05, suggesting equal variance. Furthermore, in the t-test, Sig is noted as 0.321, also exceeding the 0.05 threshold. These results demonstrate no substantial difference in the degree of FoC related to threatening behavior between individuals who actively avoid areas with poor lighting and those who do not. Hence, actively avoiding poorly lit areas does not significantly impact an individual’s FoC associated with threatening behavior.
10. Conclusion
First and foremost, gender emerges as a predominant factor in discussions surrounding the influence on FoC. Across various types of crime under investigation, female students consistently exhibit higher levels of fear compared to their male counterparts. This disparity is particularly pronounced in the context of sexual crimes, as evidenced by the average values. Such findings underscore the critical and consistent role of gender as a pre-exist determinant in shaping FoC of student. Secondly, the impact of victimization-related experiences on students’ fear of various criminal acts revealed no significant effect. This finding aligns with Young (1992) , who pointed out that some victimization experiences do not alter the individual’s fear of encountering similar crimes in the future, illustrating a paradox in FoC (Young, 1992: pp. 419-441) . Furthermore, traditional media no longer plays a significant role in shaping students’ FoC. The shift towards self-media, including casually consumed news and mobile applications, marks a changing landscape.
From the data obtained from the analysis of this study, we can see that among the antecedents of college students’ FoC, gender and victimization-related experiences should be paid more attention to. When facing college students or engaging in work related to college students’ safety, we should first pay more attention to female college students, and formulate corresponding help measures to create a brighter and safer environment, so that women’s FoC under the influence of gender will not be too prominent, which will affect the efficiency of female college students. Secondly, among college students, the experience of direct victimization accounts for a small proportion after all, and its data expression of college students’ FoC is not particularly obvious. However, there is a paradox in this antecedent variable, so we should include indirect victimization, mainly pay attention to the FoC brought to college students by the abnormal spread of victimization experience and its impact on efficiency, establish relevant mechanisms, promptly guide the direct victims, and block the indirect victimization experience that may cause discomfort in time.