Fear of Crime in the U.S.
Examining What Makes Americans Afraid, and the Factors That Help Explain It
August 2026
By Ernesto Lopez, MPA, Council on Criminal Justice, and
Amanda Graham, PhD, University of Texas, San Marcos
Fear of crime is not new, nor is it rare. Across nearly 60 years of national data, at least one-third of Americans have said they are afraid to walk alone at night, one of the most common measures of fear of crime. In 2025, that meant about 90 million adults in the United States were living with some level of fear in their daily lives.
But fear of crime is multifaceted. People fear different offenses, assess their personal risk in different ways, and form different perceptions of how safe they believe their communities to be. Understanding those differences is essential to making sense of public perceptions of crime, especially at a time when many crime trends are moving in directions that do not always match how people feel.
A previous analysis of crime perceptions (i.e., whether crime levels differ from the prior year) and fear of crime from the Council on Criminal Justice (CCJ) revealed that several factors beyond crime may influence people’s feelings about it. The analysis found that individual attributes, such as race and political ideology, and neighborhood characteristics, such as poverty and the percentage of the non-white population, shape fear of crime. It also concluded that neighborhood characteristics are largely unrelated to people’s perceptions of crime, suggesting that fear of crime and perceptions of crime are distinct concepts.1 This research, as well as other scholarship in the area, leave many questions unanswered. Among them, one stands out: What factors are associated with specific crime fears?
This report, based on an online survey commissioned by CCJ, investigates that question. It examines which specific crimes people fear most and the perceived risk of victimization for those crimes. It also explores the degree of concern survey respondents feel about the victimization of certain groups, including family members and immigrants. Finally, it examines the direct relationship between individual characteristics and fear of victimization for select crime types.
Key Takeaways
- Asked to rank their fear of 27 offenses, survey respondents said identity theft, residential burglary (when the occupant is not at home), and online theft worried them most. One in 10 said they were very or extremely worried about being murdered—half of the share who expressed concern about identity theft.
- Past research shows a person may be worried about a crime while also recognizing that their chance of being a victim of that crime is unlikely. This analysis found that the largest gaps between fear and perceived risk were for murder, being punched, and being slapped.
- Respondents were more worried about a family member being victimized than themselves, friends, neighbors, immigrants, or local businesses.
- Women reported higher levels of worry than men for every offense except online theft. The largest difference was for sexual assault. Women’s average fear level for sexual assault was more than twice as high as men’s.
- General worries, such as concerns about personal health or losing a job, along with neighborhood disorder, such as litter and visible drug use, had the strongest associations with specific crime fears.
- Anti-immigrant sentiment, racial resentment, and neighbor visibility (the degree to which neighbors recognize each other and publicly interact) generally showed no association or very weak associations with offense-specific fears.
About the Survey
This report is based on a national survey commissioned by the Council on Criminal Justice and conducted by YouGov between March 31 and April 8, 2026. YouGov interviewed 2,121 respondents, and the sample was then reduced to 2,000 to produce the final dataset.
Respondents were matched to a sampling frame using the 2023 American Community Survey. Matching variables included gender, age, race, and education; this helped ensure that the final sample reflected the demographic composition of the U.S. adult population on those characteristics. All results presented in this report use weighted values unless otherwise noted. The survey measures respondents’ perceptions, concerns, and perceived victimization risks related to crime; it does not measure aggregate victimization rates or official crime trends. Additional details are provided in the supplemental methodology report.
Measuring the Fear of Crime
This report explores fear of crime by asking people how worried they are about someone committing a crime against them. Questions about fear were not asked directly because of measurement challenges; specifically, respondents might feel vulnerable if they said they were fearful, and thus respond that they were not, resulting in inaccurate levels of fear being reported. The use of worry in place of fear is consistent with past research in this area.2 The report uses fear, worry, and concern interchangeably.
A Close Look at Fear of Crime and Perceived Risk of Victimization
People Are Most Worried About Being a Victim of a Financially Related Crime
To learn which crimes people fear most, the survey asked respondents how worried they were about becoming a victim of 27 different offenses. The list included serious violent crimes, such as murder and sexual assault; property crimes; financial and online offenses; and lower-level interpersonal harms, such as verbal harassment.
Figure 1 shows the share of respondents who expressed at least some worry about each offense. Overall, the crimes that generated the highest levels of concern were financially related offenses. About three-quarters of respondents (77%) reported some level of concern about identity theft, with home burglary and online theft (both 71%) ranking as the next most worrisome crimes. A far smaller share of respondents expressed some level of concern about being murdered (43%), sexually assaulted (37%), or raped (35%).
Across offenses, most respondents who expressed any level of concern said they were “a little worried,” rather than “very” or “extremely” worried. For example, looking at the crime that caused the most significant concern for respondents, 20% said they were either very or extremely worried about identity theft, while 32% were “a little worried.” In other findings, 12% said they were very or extremely worried about verbal harassment and 10% expressed that level of concern about being murdered. Put differently, about 1 in 10 U.S. adults reported being very or extremely worried about being murdered, a level of intense concern only slightly lower than the share who expressed the same concern about verbal harassment.
Figure 1. How Worried People Are About Becoming Victims of Specific Crimes
The survey also showed that fears of different offenses tended to overlap, meaning that people who were highly worried about one type of crime were generally more likely to be highly worried about others as well. Although this pattern appeared across all 27 offenses, identity theft had the weakest relationship to other offenses. Worry about being harmed or threatened with a knife had the strongest relationships to worry about other offenses. The results for this analysis are presented in the supplemental methodology report.
Fear of Crime and Perception of Risk are Aligned for Some Offenses
Past research suggests that fear of crime and perceived risk of victimization do not always align.3 In other words, a person may be very worried about being murdered, while also recognizing that their murder is unlikely.
Figure 2 shows respondents’ perceived likelihood of being victimized by the same offenses shown in Figure 1. Generally, perceived risk was much lower for the most serious violent offenses. About 23% of respondents said they believed they were likely or very likely to be verbally harassed. But as seen with the fear-of-crime results, many of the offenses people considered most likely to affect them involved property, financial, or online victimization. For example, about one in five respondents said they were likely or very likely to be victims of identity theft or online theft. Six percent of respondents said they were likely or very likely to be drugged, murdered, sexually assaulted, or raped. Kidnapping had the lowest perceived risk, with 4% of respondents saying it was likely or very likely to happen to them.
Figure 2. How Likely People Think They Are to Become Victims of Specific Crimes
Figure 3 shows the relationship between fear and perceived risk by ranking which crimes generated the highest levels of fear and perceived risk. Because the points follow a generally diagonal path across the chart, they suggest a very strong relationship between fear and perceived risk.4 Points along the blue line indicate that the fear-risk ranking is the same. The further a point is from the blue line, the more misalignment there is between fear and perceived risk, on average. Points in the lower left corner are ranked higher than points in the upper right corner. More details on this approach are available in the supplemental methodology report.
Offenses that show some of the largest differences (a difference of 10 or more in rank) between fear and perceived risk are murder, being punched, and being slapped. The difference between fear and perceived risk should also be considered alongside the overall rankings. An offense may show a sizable gap between fear and perceived risk even when both fear and perceived risk are relatively low. For example, fear of being slapped is ranked 26th, but the risk ranking is 16. In contrast, verbal harassment ranks 8th in fear, but first in perceived risk. In general, more serious offenses had the largest gaps between fear and perceived risk.
Figure 3. Relationship Between Fear of Crime and Perceived Risk of Victimization
People Worry Most About Family Members Being Victimized
The figures presented above focus on respondents’ own fear of crime and perceived risk of victimization. Figure 4 broadens that lens by showing how often respondents worried that other people, or businesses, would become victims of crime. Overall, there was little variation in the degree of victimization concern expressed across the groups. One exception was that respondents cited family members as those they worried about more than other potential crime targets.
Most respondents were at least sometimes concerned about themselves or another group being victimized. Eighteen percent said they often or very often worried about becoming a crime victim, while 31% said they often or very often worried about family members being victimized.
Respondents also expressed concern about immigrants. Twenty-nine percent said they often or very often worried that immigrants would be victims of a crime. But the sentiments expressed about immigrants were polarized: Immigrants also had the highest share of respondents—22%—who said they never worried about that group being victimized.
Concern was lower for local businesses. About half of respondents (51%) said they never or rarely worried about businesses being victimized by crime.
Figure 4. Worry About Crime Victimization Among Self, Family, and Other Groups
Five Specific Crime Fears by Demographic and Political Characteristics
To simplify the analysis of factors associated with fear of crime, the following sections focus on five offenses: the three that generated the highest levels of worry overall—online theft, home break-ins, and identity theft—along with sexual assault and murder, two crimes often viewed as among the most serious. The analyses also include an average fear score for these five offenses.
Black Adults and Women Generally Had Higher Crime Fears
Figure 5 displays average fear levels and differences for these five offenses by race. Values range from 0, meaning “not worried,” to 4, meaning “extremely worried.” Across the five-offense average, White respondents had an average fear score of 1.03, while Black respondents had an average score of 1.34. Both scores fall between “a little worried” and “worried,” with Black respondents reporting a statistically significant higher average fear level than White respondents.
Differences in fear levels by race were also statistically significant for two specific crimes, sexual assault and murder. Black respondents and those identifying with other racial groups reported significantly higher levels of worry about those crimes than White respondents. White respondents’ fear levels for sexual assault were .43 and .28 lower, respectively, than those expressed by Black and “other race” respondents. For murder, the difference was .54 and .33 lower for White respondents than for Black and other racial group respondents, respectively.
Across all racial groups, identity theft generated the highest average level of concern. Black respondents were the most worried about identity theft, with an average score of 1.71.
Figure 5. Fear of Crime by Race for Selected Offenses
Figure 6 shows fear of crime levels by sex. Women reported higher levels of worry than men for every offense except online theft. These differences were statistically significant for the five-offense average measure and for sexual assault.
The largest difference was for sexual assault. Women’s average fear level for sexual assault was more than twice as high as men’s.
Figure 6. Fear of Crime by Sex for Selected Offenses
A Note on Correlations
Figures 7, 8, and 10 present results as correlations. Correlations show whether two measures are related and how strong that relationship is.
Correlation values range from -1 to 1. Values closer to -1 or 1 indicate stronger relationships between two variables, while values closer to 0 indicate weaker relationships. Correlations above zero indicate a positive relationship, meaning that both variables move in the same direction: As one increases or decreases, the other does the same. Correlations below zero indicate a negative relationship, meaning that the variables move in opposite directions: As one increases, the other decreases.
Correlations are useful for showing the strength of a relationship between two variables, but they have important limits. They do not show if one variable causes another to change, and they do not account for other factors that may help explain the relationship. Given this, some of the analyses presented here are not directly comparable to past research that controls for several factors at once.
Older Adults Generally Had Lower Crime Fear Levels
Figure 7 shows the relationship between age and fear of crime for selected offenses. Overall, this relationship was weak and negative, meaning that older respondents generally reported lower levels of worry.
This pattern appears to be driven by lower levels of worry about sexual assault and murder among older respondents. Fear of online theft had essentially no relationship with age, while fear of identity theft had a weak positive relationship. In other words, older respondents reported slightly higher levels of worry about identity theft.
Figure 7. Relationship Between Fear of Crime and Age for Selected Offenses
Political Characteristics Were Generally Unrelated to Specific Crime Fears
Figure 8 shows the relationship between political ideology and fear of crime for the five selected offenses. Political ideology was measured on a five-point scale, with 1 indicating “very liberal” and 5 indicating “very conservative.”
Overall, political ideology had little to no relationship with fear of crime across the offenses examined. The main exception was sexual assault, where the relationship was weak and negative. Because higher values on the political ideology scale indicate more conservative views, this means more conservative respondents reported slightly lower levels of worry about sexual assault. However, the relationship was weak.
Figure 8. Relationship Between Fear of Crime and Political Ideology
Figure 9 shows fear of crime levels by political party affiliation for the selected offenses. Across all party groups, identity theft generated the highest average level of worry, while sexual assault generated the lowest.
Democrats did not differ significantly from members of any other party-identification group on the offenses examined. Some differences emerged among other groups: Republicans had significantly lower levels of fear about identity theft than Independents, and significantly lower levels of fear about murder than respondents identifying with those from other political parties who are not Democrats or independents.
Figure 9. Fear of Crime by Offense and Political Party Affiliation
General Worries are Consistently Associated with Higher Levels of Fear of Crime
In addition to exploring how demographic characteristics influence fear of crime, the survey asked respondents about several broader experiences and attitudes that may be related. These include general worries, perceptions of neighborhood disorder, views about social groups, and neighborhood conditions. The full survey wording and list of items included in these measures are available in the supplemental methodology report.
Figure 10 shows the relationship between these measures and fear of the selected offenses. The results were generally consistent across offenses. General worries, such as concerns about personal health or losing a job, along with neighborhood disorder, such as litter and visible drug use, had the strongest associations with specific crime fears.
Neighborhood disorder was more strongly associated with fear of in-person victimization, including home break-ins, sexual assault, and murder, than with fear of online theft or identity theft. This suggests that visible signs of disorder in a person’s neighborhood may be more closely tied to worries about crimes that could occur in or near their physical surroundings.
Other measures showed little relationship with fear of crime. Anti-immigrant sentiment, measured through views such as whether immigrants are responsible for many social problems, generally had no association or very weak associations with fear of the selected crimes. The same was true for racial resentment, measured through views such as whether Black people should “try harder” rather than attributing racial inequality to broader social barriers. Neighbor visibility, which captures whether people recognize and interact with their neighbors, in ways such as saying hello, also showed no association or only very weak associations with any offense-specific fears.
Collective efficacy, which reflects shared community values and willingness to help neighbors, was negatively associated with fear of the selected offenses. In other words, respondents who reported higher levels of collective efficacy tended to report lower levels of fear. However, these relationships were generally weak or very weak.
Figure 10. Factors Associated with Fear of Selected Offenses
People’s Level of Concern About Crime is Similar to Their Concern About Other Negative Events
Placing crime-related fears alongside other everyday concerns reveals where crime ranks among the risks people worry about most. Figure 11 compares the worry levels of the five crime fears and the specific items that comprise the general worry measure used in Figure 10. Having a personal health crisis and being in a serious car accident had higher levels of overall worry, at 85% and 81%, respectively, than identity theft and residential burglary (when not at home), at 77% and 71%, respectively. At the lower end of the worry scale, respondents had about the same levels of overall concern about being murdered (43%) as they did about being a victim of police brutality (44%). Approximately 37% of respondents expressed some level of worry regarding sexual assault, about the same share (34%) of concern expressed about being struck by lightning.
Figure 11. Crime Victimization Worries and Other Worries
Conclusion
Understanding what influences fear of crime is important for improving people’s quality of life and informing public safety policy. Fear can shape how individuals move through their communities, how safe they feel, and what actions they believe governments and other actors should take in response to crime.
This is the second in a series of crime perceptions reports produced by the Council on Criminal Justice. Examining Gallup data from 2009 to 2021, the prior report determined, among other findings, that county-level property crime rates were associated with a greater likelihood that people were afraid to walk alone at night. This report, based on an online survey commissioned by the Council, builds on that work. It analyzes public fears across a detailed list of specific offenses and explores how individual characteristics, neighborhood conditions, broader worries, and other perceptions are associated with those fears.
One clear finding is that property-related crimes top the list of people’s concerns. Identity theft, home break-in, and online theft generated the highest levels of worry, demonstrating that fear of crime is not limited to anxiety about violence or physical harm. The report also found that crime fears vary across groups and contexts, underscoring the value of examining specific offenses rather than treating fear of crime as a single concern.
These findings may help identify which populations feel vulnerable in which circumstances, and they serve as a critical first step toward understanding what may or may not alleviate crime-specific fears. Future Council reports will expand this body of work by examining which factors people say make them feel safe or unsafe, how street-level contextual clues influence crime fears, and how everyday behaviors, such as shopping in a store versus online, relate to fear of crime.
About the Authors
Ernesto Lopez is a senior research specialist with the Council on Criminal Justice. His work currently focuses on analyzing and interpreting crime and criminal justice data to understand patterns and shifts across the broader crime landscape.
Amanda Graham is an associate professor in the Department of Criminal Justice and Criminology at Texas State University. Her research interests focus on policing, police-community relationships, public opinion, and measurement.
Acknowledgements
The authors thank Nicole Rader of Mississippi State University, Luzi Shi of the University of Rhode Island, and Jason Silver of Rutgers University for their guidance and input on the survey methodology used in this report, as well as DeRay Mckesson and his team at Campaign Zero.
Stephanie Kennedy, along with other members of the Council on Criminal Justice team, provided support on this report.
This paper was produced with support from the Harry Frank Guggenheim Foundation, Stand Together Trust, and CCJ’s general operating contributors.
Suggested Citation
Lopez, E., & Graham, A. (2026). Fear of crime in the U.S.: Examining what makes Americans afraid, and the factors that help explain it. Council on Criminal Justice. https://counciloncj.org/fear-of-crime-in-the-u-s/
Endnotes
1 The term perception refers to a person’s belief about whether the actual level of crime has increased, stayed the same, or decreased. Fear, on the other hand, is a psychological response typically centered on personal vulnerability.
2 Williams, F. P., McShane, M. D., & Akers, R. L. (2000). Worry about victimization: An alternative and reliable measure for fear of crime. Western Criminology Review, 2(2), 1-26. https://www.westerncriminology.org/documents/WCR/v02n2/williams/williams.html
3 See, for example, Rader, N. E., May, D. C., & Goodrum, S. (2007). An empirical assessment of the “threat of victimization:” Considering fear of crime, perceived risk, avoidance, and defensive behaviors. Sociological Spectrum, 27(5), 475-505. https://doi.org/10.1080/02732170701434591
4 The rank-order correlation (Spearman’s Rho) is equal to .74.
